Pre-cursors to physical employment standards for rural tanker-based bushfire suppression work.

By

Matthew Phillips

B.Sc (Hons)

Submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

Deakin University

October, 2010

i

ABSTRACT

Studies which are contained in this thesis serve as pre-cursors to the development of physical employment standards (PES) for multi-day rural tanker based suppression (RTBS) work. RTBS work is performed by 206,954 predominantly volunteer bushfire fighters (BFF) Australia-wide in response to Australian bushfire incidents which regularly devastate regions in excess of one million hectares resulting in property destruction, rural stock loss and human casualties. The physical demands of RTBS work had not been previously identified or PES developed for this occupational category. Given fire fighting work which involved similar tasks, such as urban structural fire fighting (SFF) and land management fire fighting (LMFF), has been characterised as physically demanding, it was likely RTBS work would also be characterised as physically demanding. Critical SFF and LMFF work tasks performed by aging BFF personnel could potentially generate relatively high oxygen consumption ( O2max) values. Given an estimated 55% of Australian BFF personnel are aged over 40 years old, it is important PES are developed to ensure Australian BFF are physically capable of safely performing RTBS work.

The first study (Chapter 3) conducted a job task analysis of 55 commonly performed RTBS work tasks using the ratings of 31experienced BFF subject matter experts (SME). The aim of this study was to characterise and identify physical demanding tasks or criterion job tasks using SMEs perception of RTBS work. SME rated RTBS work tasks using a number of variables including duration, frequency, load, operational importance, operational difficulty, the actions and types energy systems involved in each task. SME identified seven critical RTBS tasks from 55 commonly performed RTBS tasks. SME rated the relative importance and frequency of RTBS work tasks, charged hose advance and full charged repositioning more highly than other tasks identified as physically demanding. Work tasks characterised as physically demanding were also able to differentiated in two broad categories, work involving a hose or work involving a rakehoe (rake). 90% of physically demanding work tasks by frequency was considered to involve hose work and the remaining 10% rake work. SME experts identified the majority of RTBS work tasks was strength endurance work and involved carry or drag actions. Physically demanding and critical job tasks from a job task analysis such as this would traditionally have been selected for further study as criterion job tasks. The accuracy of SME ratings was compared to actual movement and physiological analysis conducted during task simulations and large-scale, single-day ‘controlled’ bushfires later in the thesis (Chapter 7).

The second study (Chapter 4) examined the cardiovascular and aerobic demands of simulated RTBS tasks. The aim of this study was to characterise and identify the most physically demanding RTBS task using

RTBS task simulations. Oxygen consumption, heart rate and global positioning system (GPS) movement was recorded in 26 BFF performing simulated trials 15 common RTBS tasks. O2max during simulated RTBS tasks ranged from 10.2 ± 2.9 mL.kg-1 ·min-1 to 33.1 ± 5.1 mL.kg-1 ·min-1. Solo hand tool, team hand tool, spot fire

iii rake, charged hose advance, uphill hose advance and knapsack RTBS tasks recorded the highest O2max usage of between 26.7± 4.8 and 33.1 ± 5.1 ml-1.kg·min-1. These tasks were identified as the most physically demanding RTBS task simulations. Corresponding relative intensities were recorded of 94.8 ± 7.8 %HRmax

-1 -1 observed up to and 89% O2max. The author estimated PES for RTBS work of 28.6 ml .kg·min for rake work tasks and 25.3 mL.kg-1·min-1 for hose work, are likely to reflect the minimal peak metabolic demands of RTBS work. The accuracy of detailed movement and physiological analysis conducted during RTBS task simulations was compared against similar analysis during large-scale, single-day ‘controlled’ bushfires and SME ratings later in the thesis (Chapter 7).

The third study (Chapter 5) conducted an analysis of the activity, movement and cardiovascular demands in 28 BFF during a single shift of RTBS work at a single day ‘controlled’ fire. The aim of this study was to characterise RTBS work performed in a single day simulation. Heart rate, activity and GPS movement was recorded continuously and later synchronised with video footage of tasks. Continuous, sequential tasks with a collective focus and performed without break were analysed as a work bout. Hose and rake work bouts comprise 87 and 13% of the relative duration of an average fire ground work bout. Analysis of 393 hose work bouts observed mean hose work bout was composed of 4.8 ± 7.2 tasks, had a duration of 192.0 ±

- 310.6 seconds, a heart rate equivalent to 67.0 ± 0.1% HRmax and was conducted at a speed of 1.2 ± 1.3 km.hr 1. Five individual RTBS tasks accounted for 78% of all RTBS work including Support operator, blacking out work, lateral repositioning and operating RTBS tasks. Mean and peak heart rate corresponded to 58.8±

11.4%HRmax and 78.2 ± 15.3 %HRmax respectively. The large time spent in sedentary activity and lower intensity heart rate zones across numerous fire sites indicate a substantial portion of RTBS work is inactive in nature but characterised by tasks which potentially generate high heart rates and given the results of study two, relatively high oxygen uptake levels. The accuracy of detailed movement and physiological analysis conducted during RTBS single day ‘controlled’ bushfires was compared against similar analysis during task simulations and SME ratings later in the thesis (Chapter 7).

The fourth study (Chapter 6) characterised the activity, movement and cardiovascular demands of multi-day campaign RTBS work. The aim of this study was to characterise the physical demands of campaign fire suppression RTBS work and compare these demands against single day controlled fire RTBS work (Chapter 5). Heart rate, activity and GPS movement was recorded continuously during operational deployment at the

2006-07 North East bushfires in 38 BFF. Mean and peak heart rate corresponded to 59.9± 10.7%HRmax and 91.0

± 19.1 %HRmax respectively. Campaign RTBS work contained less frequent, higher activity work bouts for similar observed mean average and cumulative activity counts as single-day RTBS work. Single day RTBS work generated similar mean heart rates as campaign RTBS work but over-predicted activity intensity at lower intensities. Observed mean average speed and portion of time spent over 8 km.hr-1, were higher during campaign RTBS work, most likely due to higher amounts of sedentary truck travel encountered during single day RTBS work. It is likely the composition and intensity of physically demanding work is relatively consistent iv between single day and campaign RTBS work. It is likely fire fighters, particularly older and less fit individuals, encountered significant cardiovascular demands during single day and campaign RTBS work given peak heart rate was recorded as 78.2 ± 15.3 %HRmax 91.0 ± 19.1 %HRmax respectively.

The collective findings presented in this dissertation demonstrated RTBS work generated comparable peak physical demand with other previously observed SFF, LMFF and military fire fighting research (Chapters 4,5 & 6). The lower absolute and relative heart rates observed during Australian RTBS work when compared to SFF or LMFF are likely due to reductions in loaded bearing work and increased duration of operational deployment. Subject matter expert ratings overestimated task duration and frequency of occurrence during the large-scale, single-day ‘controlled’ bushfires model and emergency multi-day bushfires (Chapter 7). Simulated work representations of RTBS tasks accurately generated the duration and intensity encountered during the large-scale, single-day ‘controlled’ bushfires model and likely emergency multi-day bushfires (Chapter 6).

The findings and accompanying discussion presented in this dissertation characterised and provided a foundation for continued development of PES for RTBS work. This thesis has taken steps toward the development of an Australian PES model where the duration of a physical selection test can be indexed back to the job. This approach has been taken with specific regard to the OHS requirements of Australian workplace law and has been designed to achieve compliance in this domain. Given the findings in this thesis, it is reasonable for any rural fire agency to expect fire fighters may be exposed to physical/ more arduous work for around four percent or thirty minutes of every 14 hour shift. The high relative exercise intensities associated with RTBS tasks may prompt Australia’s rural fire agencies to consider implementing PES prior to each fire season, particularly given the ageing nature of their workforce.

v

ACKNOWLEDGEMENTS

I would like to acknowledge the contribution of a number of people who have offered the time and support for the work in this thesis.

First and foremost, I would like to sincerely thank my supervisors Dr Brad Aisbett, Professor Warren Payne, and Dr Aaron Russell. They have provided me with support and encouragement throughout my candidature, for which I am very grateful. Their patience, guidance and knowledge have been invaluable. I would also like to make mention of Dr Glenn McConell. His support of the project and my candidature before it moved to Deakin University was fundamental to the continuing success of the work.

I am appreciative of the financial and industry support as a recipient of the Bushfire Cooperative Research Centre (Bushfire CRC). I would like to acknowledge the role David Nichols of the Country Fire Authority played in facilitating industry support and managing the Bushfire CRC project. Peter Langridge and Audrey Birmingham deserve special mention as one of the first industry supporters of work, providing unique access to fire grounds and comment on ideas or methodological problems.

A number of industry based individuals have worked to facilitate research work. Many of these individuals have conducted work with the project voluntarily or in addition to normal work duties. Shane Cramer (Greendale Rural Fire Brigade), James Dullard (Fiskville Training College, CFA) and Chris Beeck (FESA) have played a fundamental role in advising methodological considerations, championing the work throughout the industry and providing access to research participants

Finally but not least, I would like to thank my family and friends. Thank you to my friends Steve, Nick, Charles and my cousin Christopher. I am very fortunate to have a loving and supportive family and would like to thank my mother, father, sister, brother-in-law and partner Larissa for their endless love, encouragement and understanding. I promise you all, I will not be enrolling in any further university degrees for at least a few years.

vi

PUBLICATIONS RELATED TO THIS THESIS

Aisbett, B., Phillips, M., Sargeant, M., Gilbert, B. and Nichols, D. (2007). "Fighting with fire: How bushfire supression can impact on fire fighters' health". Australian Family Physician 38(12): 994-997.

Peterson, A., Payne, W., Phillips, M., Netto., K., Nichols. & Aisbett, B. (2010) Validity and relevance of the pack hike wildland firefighter work capacity test: a review. Ergonomics 53(10): 1276-1285.

Phillips, M., Petersen, A., Abbiss, C.R., Netto, K., Payne, W., Nichols, D. & Aisbett, B. (2011). Pack Hike Test finishing time for Australian firefighters: pass rates and correlates of performance. Applied Ergonomics 42(3): 411-418.

SCIENTIFIC & INDUSTRY COMMUNICATIONS RELATED TO THE THESIS

Phillips, M. The Bushfire Fighting Test: Quantifying a safe standard of fitness and health in CFA volunteers. Fire Managers Research Workshop, Wollongong, 2006.

Phillips, M. Characterising the physical demands of tanker based bushfire fighting. Australasian Fire and Emergency Service Authorities Council (AFAC) Conference, Hobart, September 2007.

Phillips, M. The aerobic energy demands of simulated common tanker based wildfire tasks. Australian Association for Exercise and Sports Sciences Conference, Melbourne, April 2008.

Phillips, M. The Aerobic Demands of Simulated Common Tanker Based Wildfire Fighting Tasks. The American College of Sports Medicine Annual Meeting, Indianapolis, May 2008.

Phillips, M. Predicting job performance in volunteer (TBFF), structural (UFF) and land management (LMFF) in Western Australian fire fighting populations: Future implications for nationwide fit for duty testing. AFAC Conference, Adelaide, September 2008.

Phillips, M. The pack hike as a job fitness assessment. ACT Fire managers meeting, Canberra, February 2009.

Phillips, M. Fire Fit: The burning question? Cooperative research centres association annual conference, Canberra, May 2009.

Phillips, M. The importance of Physical employment standards. Cooperative research centres association Parliamentarian’s breakfast, Canberra, May 2009.

vii

Table of Contents

ABSTRACT ...... iii ACKNOWLEDGEMENTS...... vi PUBLICATIONS RELATED TO THIS THESIS ...... vii SCIENTIFIC & INDUSTRY COMMUNICATIONS RELATED TO THE THESIS ...... vii Table of Contents ...... viii LIST OF FIGURES ...... x LIST OF TABLES ...... xiii ABBREVIATIONS ...... xv CHAPTER 1: INTRODUCTION ...... 1 Overview: Motivators for physical employment standards ...... 1 CHAPTER 2: LITERATURE REVIEW ...... 5 Legislative forces and considerations behind the development of employment selection tests in North America ...... 5 Australian anti-discrimination law in relation to PES: Differences between Australian and North American law ...... 7 Justification for the implementation of employment standards in Australian fire agencies...... 10 Australian fire fighting and RTBS ...... 13 Urban or SFFs ...... 14 Land management, conservation and state government networked fire fighters ...... 17 Rural bushfire agencies ...... 20 Physical factors differentiating Australian volunteer fire fighters from land management or SFFs...... 24 Developing a PES for Australian RTBS ...... 29 Job task analysis ...... 30 Limitations of physical demands analysis in the PES model for rural bushfire fighting ...... 34 Summary ...... 37 CHAPTER 3: Job Task Analysis of Crew Member Duties during Australian Tanker-Based Bushfire Suppression Work...... 39 INTRODUCTION ...... 39 METHODS ...... 40 RESULTS ...... 41 Characterisation of Tasks ...... 41 Identification of Critical Tasks...... 50 DISCUSSION ...... 55 CONCLUSION ...... 57 CHAPTER 4: Aerobic demands of simulated bushfire suppression tasks performed by tanker-based fire fighters ...... 58 INTRODUCTION ...... 58

viii

METHODS ...... 59 RESULTS ...... 64 DISCUSSION ...... 74 CONCLUSION ...... 76 CHAPTER 5: Detailed work patterns and physiological responses of Australian tanker based bushfire fighters to rural bushfire fighting...... 78 INTRODUCTION ...... 78 METHODS ...... 80 Activity counts were divided into three ranges specified by Actical® software; ...... 83 RESULTS ...... 84 Characterisation of work demands ...... 85 Characterisation of work bouts ...... 106 DISCUSSION ...... 114 Characterisation of work demands ...... 114 Characterisation of work tasks and work bouts ...... 116 CONCLUSION ...... 118 CHAPTER 6: Work demands and physiological responses of Australian tanker based bushfire fighters to campaign bushfire fighting...... 119 INTRODUCTION ...... 119 METHODS ...... 120 RESULTS ...... 123 DISCUSSION ...... 140 Characterisation of work demands ...... 140 Comparison of Single day and Campaign bushfire suppression ...... 141 CONCLUSION ...... 143 CHAPTER 7: DISCUSSION ...... 145 Characterising Australian tanker based bushfire fighting...... 145 Comparison of SME expert opinion, task based simulations, single day and campaign fire activity as a basis for a job task analysis ...... 147 OHS considerations with further development of PES for RTBS work ...... 154 Operational issues for fire agencies which can be informed by science ...... 156 LIMITATIONS ...... 158 FUTURE DIRECTIONS ...... 160 REFERENCES...... 161 APPENDICES ...... 182

ix

LIST OF FIGURES

Figure 2.1, Current pathways to the development of PES (adapted from Payne & Harvey, 2009; Rayson, 2000c) ...... 30 Figure 2.2, Detection accuracy of low Vs high experienced incumbents (from Richman & Quinones, 1996) ...... 33 Figure 2.3, Novel potential additions to the job task analysis process (adapted from Payne & Harvey, 2010; Rayson, 2000c; Jamnik et al., 2010a) ...... 36 Figure 3.1, Overall years of fire industry experience across six experimental conditions...... 43 Figure 3.2, Operational importance Vs. perceived relative frequency of all tasks...... 51 Figure 3.3, Operational importance Vs. perceived frequency of perceived physically demanding tasks...... 51 Figure 3.4, Operational importance Vs. perceived relative frequency of perceived physically demanding tasks...... 52 Figure 3.5, SME perceived action type and frequency of physically demanding RTBS tasks...... 53 Figure 3.6, SME perceived activity type and frequency of physically demanding RTBS tasks...... 54 Figure 4.1, Oxygen consumption and duration of hose work tasks...... 68 Figure 4.2, Oxygen consumption and duration of rakework tasks...... 69 Figure 4.3, Oxygen consumption and duration of miscellaneous work tasks...... 70 Figure 4.4, Velocity of bushfire suppression task simulations...... 71 Figure 4.5, Peak and average percentage of predicted Maximal heart rate encountered during tasks...... 72 Figure 4.6, Mean oxygen consumption and duration categorized by hose (N = 54), rake(N = 47) and miscellaneous work (N = 24) tasks found to elicit peak metabolic demand...... 73 Figure 5.1. Map of Tasmanian fire data collection sites; 1, Sorell region (N=7); 2, Houn valley region (N=7); 3, Ansons Bay (N=7); 4, Ansons Bay (N=7). Google maps, Accessed November 2009...... 81 Figure 5.2, Data collection techniques used in Tasmania fire sites . A) Heart rate GPS and Activity monitor positioning; B) Digital video camera synchronisation and analysis & C) Remote monitoring of Heart rate, GPS Speed and Distance...... 82 Figure 5.3. Age and experience differences between participants and Hobart and St Helens fire sites...... 87 Figure 5.4, Mean average activity data recorded during single day fire work of Hobart, St Helens and all fire sites...... 88 Figure 5.5, Mean two hour activity data recorded during single day fire work of Hobart, St Helens and all fire sites...... 88 Figure 5.6, Cumulative whole shift activity data recorded during single day fire work of Hobart, St Helens and all fire sites...... 89 Figure 5.7, Mean cumulative 2 hr activity data recorded during single day fire work of Hobart, St Helens and all fire sites...... 89 Figure 5.8, Mean intensity distribution of activity data recorded during single day fire work of Hobart, St Helens and all fire sites...... 91 Figure 5.9, Mean intensity distribution of activity data recorded during subsequent 2 hour periods of single day fire work of all fire sites...... 92 x

Figure 5.10, Mean average heart rate recorded during single day fire work of all fire sites...... 93 Figure 5.11, Mean heart rate data recorded during two hour segments during single day fire work of all fire sites...... 93 Figure 5.12, Mean Maximal heart rate related to age predicted Maximal recorded during single day fire work of all fire sites...... 94 Figure 5.13, Mean heart rate related to age predicted Maximal recorded during single day fire work over two hour segments at St Helens, Hobart and all fire sites...... 94 Figure 5.14, Average ACSM classification % Maximal heart rate data recorded during single day fire work of all fire sites...... 96 Figure 5.15, Mean average ACSM classification of heart rate data recorded during single day fire work of all fire sites...... 97 Figure 5.16, Mean speed recorded during single day fire work of all fire sites...... 98 Figure 5.17, Mean time spent in speed classifications recorded during single day fire work of all fire sites...... 99 Figure 5.18, Percentage of individual work bout comprising hose work RTBS tasks during single day fire work ...... 111 Figure 5.29, Percentage of individual work bout comprising rake work RTBS tasks during single day fire work ...... 112 Figure 5.20, Percentage of rake (A) and hose (B) tasks by duration composing rake or hose work bouts during single day fire work respectively...... 113 Figure 6.1, Map of campaign fire data collection sites; 1, Mugdee (N=6); 2, Jamieson (N=12); 3, Mt Buller Ski Resort (N=6); 4, Swifts Creek (N=6); 5, Bairnsdale (N=4); 6, Mansfield (N=8) (DSE fire incident map, accessed July 2009)...... 121 Figure 6.2. Mean Activity counts during campaign bushfire shifts...... 128 Figure 6.3. Mean Activity intensity distribution over two hour periods during campaign and single day bushfire shifts...... 129 Figure 6.4, Mean Activity intensity distribution during campaign bushfire shifts...... 130 Figure 6.5, Mean Activity intensity distribution over two hour periods during campaign bushfire shifts...... 131 Figure 6.6, Mean and peak (absolute and relative to age predicted Maximal) heart rate during campaign and single day bushfire shifts...... 133 Figure 6.7, Heart rate comparisons between single day (N= 26) and campaign fire sites (N= 32). 134 Figure 6.8, Mean ACSM classification distribution during campaign bushfire shifts...... 135 Figure 6.9, Comparison of mean ACSM classification distribution during the first six hour period of campaign and single day bushfire shifts...... 136 Figure 6.10, Mean speed recorded during single day fire work of all fire sites...... 137 Figure 6.11, Mean time spent on truck Vs. Time spent off truck recorded during campaign and single day fire work...... 138 Figure 6.12, Mean time spent in speed classifications recorded during single day fire work of all fire sites...... 139

Figure 7.1, Comparison of average %HRmax observed during simulated tasks and at single day fires for the same tasks...... 150

xi

Figure 7.2, SME rated task duration of RTBS tasks (Chapter 3) V. task duration observed during task based simulations (Chapter 4) & single day fire observations (Chapter 5)...... 153 Figure 7.3, Speculative acceptable development range for Australian PES showing the range between a minimum PES (a decreased risk of job related injury) compared with reasonably practicable PES (subjectively determined productivity or profitability)...... 155

xii

LIST OF TABLES Table 2.1, American Fire Fighter Age at time of Fatal Incident 1990 – 2000 (adapted from Burton, 2007) ...... 12 Table 2.2, Energy expenditure, physiological, and subjective responses to different jobs performed during fireline construction (Budd et al., 1997b; Budd et al., 1997d) ...... 19 Table, 2.3. Response status during call outs of volunteer fire fighter to fire incidents (CFA, 2008). 21 Table 2.4, Variations in finish time and calculated rates of travel during the Pack and No Pack trials over a course of 660.5m long with a vertical rise of 137m (Ruby et al., 2003)...... 26 Table 2.5: Physical performance scores for male and female incumbent fire fighters (Misner et al., 1989b) ...... 27

Table 2.6, Previous recommendations for Minimum acceptable O2max standards of fire fighting.28 Table 2.7, Interrater reliabilities and intrarater coefficient alpha reliabilities of overall job performance rating between peer (incumbent) and supervisors (adapted from Viswesvaran et al., 1996) ...... 32 Table 3.1, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as hose work...... 44 Table 3.2, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as rake work...... 45 Table 3.3, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as miscellaneous work...... 46 Table 3.4, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as integrated and involving a number of individual tasks...... 48 Table 3.5, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as physically demanding...... 49 Table 4.1, Specifications of RTBS tasks simulations...... 66 Table 4.2, Physical Subject Characteristics...... 67 Table 5.1, Video analysis tasks categories for Single day fire sites as determined by SME in Chapter 4...... 86 Table 5.2. Physical Subject Characteristics ...... 87 Table 5.3, Activity data recorded during simulated fire work of Hobart, St Helens and all fire sites...... 87 Table 5.4, Mean task values from all work tasks performed during RTBS shifts...... 101 Table 5.5, Mean task values from 25mm hose work tasks performed during RTBS shifts...... 102 Table 5.6, Mean average task values from 38mm hose work tasks performed during RTBS shifts...... 103 Table 5.7, Mean average task values from miscellaneous work tasks performed during RTBS shifts...... 104

xiii

Table 5.8, Mean average task values from rake work tasks performed during RTBS shifts...... 105 Table 5.9, Descriptive statistics between hose work bouts at Hobart and St Helens fires during single day fire work...... 108 Table 5.10, Descriptive statistics between rake work bouts at Hobart and St Helens fires during single day fire work...... 109 Table 5.11, Descriptive statistics of combined hose and rake work bouts during single day fire work...... 110 Table 6.1: Descriptive data for the sample of Australian campaign BFFs studied during 2006-07 fire season...... 127 Table 7.1, Subject matter expert (SME) perceived frequency per season of RTBS tasks V. Predicted frequency per season RTBS tasks extrapolated from single day (SD) fire observations...... 149 Table 7.2, Pilot testing for the aerobic power of Western Australian fire fighters. (Phillips et al., 2011)...... 156 Table 7.3, Number of participants sampled by gender and percentage of overall sample across Chapters 3 to 6...... 159 Table A.1, Task list of RTBS work...... 182 Table A.2, Potential options and definitions of variables discussed by participants for each task on the task list. Participants could select multiple options for variable seven, action category. For all other variables, participants could only select one variable...... 184

xiv

ABBREVIATIONS

= equals > greater than < less than greater than or equal to less than or equal to ABS Australian Bureau of Statistics ACSM American College of Sports Medicine ACTRFS Australian Capital Territory Rural fire service ANOVA analysis of variance AUD Australian Dollars BFF Bushfire fighter BFOR Bona Fide Occupational Requirement BMI Body Mass Index CFA Country Fire Authority CFS South Australia Country Fire Service CI confidence interval CPAT Candidate physical ability test DEC Department of Environment and Conservation DSE Department of Sustainability and Environment EEOC Equal Employment Opportunity Commission e.g. for example

%HRmax percentage of maximal heart rate i.e. that is GPS Global Positioning System kg kilograms kJ kilojoules km kilometres km.hr-1 kilometers per hour L.min-1 litres per minute LMFF land management fire fighter min minutes mL millimetres mL.kg-1.min-1 millilitres per kilogram per minute N sample size/ number of observations

xv

NSWRFS New South Wales Rural Fire Service OHS Occupational Health and Safety PES Physical employment standards RTBS Rural tanker based bushfire suppression SD standard deviation sec seconds SE standard error SFF structural fire fighter SME Subject Matter Expert TFS Tasmanian Fire Service

O2 oxygen uptake

O2max peak oxygen uptake WFF Wildland fire fighter

2

CHAPTER 1: INTRODUCTION

Overview: Motivators for physical employment standards

Wildland fires impact on many parts of North America, Europe and Australasia annually throughout the summer months (Hunter, 2004; Hyde et al., 2008; Schmuck et al., 2004; Australian Bureau of Statistics (ABS), 2004; Department of Sustainability and Environment (DSE), 2008). In Australia, rural wildland fires (known as bushfire) often burn areas in excess of 1 million hectares, resulting in large scale property loss, rural stock loss and rural community causalities between December and March each year (Hunter, 2003; ABS, 2004; Department of Environment and Conservation (DEC) Annual report, 2007; DSE Living with Fire Report, 2008). Combined urban structural fire fighters (SFF) and bushfire fatalities in Australia occur, on average, at a rate of 0.7 per 100,000 people, but this rate can climb to 1.5 per 100,000 a year with the occurrence of significant fire events (Australian Injury prevention bulletin, 1997). Since 2003, bushfire has resulted in 193 casualties throughout Australia, many of them occurring in recent catastrophic fires (Haynes et al., 2008).The frequency and severity of large scale fire events has worsened with the occurrence of severe long term drought in many areas throughout Australia (ABS, 2004; DSE Living with Fire Report, 2008). Multi day (campaign) suppression efforts, managed by land management and rural fire agencies, can last for 70 days and involve 28,000 work hours. In a single campaign bushfire, logistical requirements can involve over 15,000 personnel and direct suppression efforts cost in excess of $170 million Australian dollars (AUD) (Hunter, 2003; ABS, 2004; DSE Living with Fire Report, 2008). A large population of rural volunteer fire fighters across Australia provides the bulk of the work force for work hours during prolonged bushfire suppression (McLennan, 2004). This review will focus primary on identifying important issues for physical employment standards (PES) development in bushfire fighter (BFF) personnel.

Australian wildfire (hereafter known as bushfire) suppression is performed by rural volunteer BFFs using tanker based suppression techniques (Hunter, 2003; Country Fire Authority (CFA), 2006). Currently the demands of rural tanker based bushfire suppression (RTBS) and the physical capabilities of fire fighters completing this work are unknown. Common operational practices of RTBS work incorporate a number of physically demanding fire fighting techniques used by SFF and land management fire fighters (LMFF) (Sothmann et al., 2004; Gledhill & Jamnik, 1992a; Gledhill & Jamnik, 1992b; Sharkey, 1997; Ruby et al., 2003; Heil et al., 2002; Ruby et al., 2002; Bilzon et al., 2001a). RTBS work is likely characterised range of operational, equipment and personnel differences which suggest it is a unique style of fire suppression and may be similarly physically demanding (CFA, 2006; McLellan, 2004). In recent years, there has been acknowledgement by Australian fire agencies, of the need to protect the safety of ageing fire fighters and generate valid PES which ensure a matched work capacity of Australian fire fighters (DEC, 2006; Aisbett et al.,

1

2007; Gledhill, 2001; Thornton et al., 2008; Payne et al., 2005). The development of valid PES for RTBS requires a detailed task analysis of the work, physical demands analysis of this work and assessment of PES validity in relation to the work (Rayson, 2000c).

In order to comply with Occupational Health and Safety (OHS) acts (OHS Act, 2004), employers must “eliminate risks to health and safety so far as is reasonably practical” and if unable to eliminate risks, “reduce those risks so far as is reasonably practicable”. Bushfire suppression agencies, despite their status as voluntary bodies, are subject to the risk assessment and risk reduction strategies of all other Australian businesses or organisations. Australian fire agencies have a duty, so far as is reasonably practicable, to protect the health and safety of all fire fighters by: assessing workplace risk, applying strategy to reduce workplace risk, training fire fighters to understand and competently manage their workplace risk (OHS Act, 2004).

The principle OHS risks faced by BFFs are likely to be musculoskeletal injury, heat stress, and smoke inhalation (Aisbett et al., 2007). Cady et al., (1979) found injury rates for least fit fire fighters were 6.3% higher than for most fit fire fighters. Rodriguez and Eldridge (2003) demonstrated a direct inverse relationship between increases in aerobic power and/or muscular strength and decreased probability of injury in fire fighters. PES provide a practical and widely accessible means of controlling the potential risk of over-exertion injury by assessing the job specific capabilities of an employee against the required demands of work (Rayson, 2000c; Rayson, 2000b). Cost effective savings, work performance increases and substantial reductions in injury rate have been demonstrated with the introduction of wellness programs into most physically demanding occupations (Cady et al., 1985; Aasa et al., 2005; Aasa et al., 2008; Maniscalco et al., 1999; Musich et al., 2001).

For the purpose of this review, I will refer only to the RTBS duties of these agencies and the volunteer personnel who conduct this work. Rural fire agencies respond to between 4000 and 15500 fire incidents per year (TFS, 2008; CFS, 2008; CFA, 2008). The DSE, Living with Fire report (2008) stated more than 80% of fire incidents are contained to less than five hectares. The remaining 20% of larger incidents resulted in 90% of area burnt annually. Long term or campaign fire incident response often involves the remote deployment of personnel from neighbouring regions, urban areas, interstate or even overseas depending on the size of the fire incident (NSWRFS, 2008; CFA, 2008). Over 200,000 Australian volunteer fire fighters are not required to meet physical performance standards to perform fire work. No studies have assessed the physical demand of RTBS or the capabilities of rural fire fighters to safely complete this work. This review and thesis will focus on identifying the physical demands of RTBS work with particular emphasis on campaign bushfire suppression.

The level of aerobic power and overall fitness required to work safely and productively can be established through PES. In the absence of any detailed reports on the operational behaviours of Australian rural BFFs, 2 a preliminary interview with an experienced operations officer (Bosua, 2006) suggests a common individual campaign deployment lasts for between three to five days. BFFs crew a fire tanker with three to five individuals from the same fire station and usually have some degree of familiarity with their fellow crew members. Each tanker is commonly deployed with four other trucks, called a strike team (Bosua, 2006; CFA, 2006). The method of fire suppression and individual crew tasking within a strike team, is highly dependent on fire behaviour and the protection of assets located within the area (Bosua, 2006; CFA, 2006). Australian rural fire crews’ use a combination of ‘wet’ and ‘dry’ fire suppression techniques to contain fire spread (Bosua, 2006; CFA, 2006). The Wildfire Firefighter: Learning Manual (2006) describes a number of tasks performed on the fire ground including; ‘wet’ hose tasks to distribute extinguishing mediums, ‘dry’ manual rake tasks to create mineral earth lines or contain small fires, and miscellaneous tasks which focus on using a portable knapsack spray pump, filling the fire tanker with water and working with other vehicles, aircraft or heavy machinery. Similarly to SFF techniques, ‘wet’ fire fighting techniques involve the suppression of bushfire using a fire truck to deploy water and other extinguishing mediums directly or indirectly onto a bushfire. These techniques when applied directly, extracts the heat from a burning fire or indirectly, lower the flammability of areas around it. Common operational practices of RTBS incorporate a number of fire fighting techniques used by urban SFFs and LMFFs (Sothmann et al., 2004; Gledhill and Jamnik, 1992a; Sharkey, 1997) but with a range of operational, equipment and personnel differences suggest RTBS is a unique style of fire suppression (CFA, 2006; McLellan, 2004).

Many vital components of a job task analysis have not been quantified for RTBS including work (task) intensities, task frequency, task duration, subjective measurement of task importance and the nature by which tasks are performed in sequence to achieve a targeted goal, a work bout. Without characterisation of these variables, identification of the components of a PES for BFFs completing RTBS work is problematic and unachievable.

The aims of this thesis are to:

Identify the physically demanding tasks (during campaign bushfire suppression) Characterise the operational importance of the physically demanding tasks (as above) Quantify the intensity of the above tasks Quantify (subjectively and objectively) the frequency and duration of the above tasks

These aims will be achieved by:

Chapter 3: Focus group discussion with subject-matter experts Chapter 4: Measurement of the aerobic energy demands of simulated versions of the physically demanding tasks.

3

Chapter 5: Remote monitoring of task intensity, frequency, and duration of the physically demanding tasks during a large-scale controlled 'live-fire' work shift. Chapter 6: Remote monitoring of work intensity during emergency bushfire suppression shifts.

4

CHAPTER 2: LITERATURE REVIEW

The focus of this literature review and the thesis which follows is to identify and characterize the physically demanding tasks performed during Australian RTBS. The review begins by establishing the process for identifying and characterizing physically demanding tasks, a job task analyses and how a job task analysis underpins the development of scientifically valid and legislatively sound PES. The review will then evaluate the process required to identify physically demanding tasks, and establish their operational importance (or criticality), frequency, intensity, and duration.

Legislative forces and considerations behind the development of employment selection tests in North America

Canadian PES law is governed by the Canadian Human Rights Act (1977) which expressly forbids employment discrimination on the grounds of “race, national or ethnic origin, colour, religion, age, sex, sexual orientation, marital status, family status, disability and conviction for which a pardon has been granted”. Employers are entitled to select employees for physically demanding occupations with an exemption which suggests “it is not discriminatory practice if any refusal, exclusion, expulsion, suspension, limitation, specification or preference in relation to employment is established by an employer based on a Bona Fide Occupational Requirement (BFOR)”. A PES will not be considered discriminatory if the employer objectively demonstrates the existence of inherent requirements of the job, which are fundamental and constant to successful performance of the job and the job cannot be altered in a way which accommodates otherwise disadvantaged workers (Shephard & Bonneau, 2002; Canadian Human Rights Act, 1977; Kuruganti & Rickards, 2004). This principle is known as the “duty to accommodate”. Employers are excused from the duty to accommodate if the work environments possess intrinsic and unalterable characteristics, objectively demonstrated BFOR and the demands of accommodating a worker would place significant unreasonable “undue hardship” on the employer (OHRC, 2000; McCallum, 2004; Shephard & Bonneau, 2002; Berkowitz & Muller-Bonami, 2006; Jamnik et al., 2010c). Undue hardship is demonstrated by significant inverse relationships with financial costs of accommodation or health and safety concerns would outweigh accommodating a worker.

Canadian employment law in relation to PES has been significantly defined by judicial proceeding, in particular the Meiorin case, British Columbia (Public Service Employee Relations Commission) v. British Columbia (Government and Services Employees Union), (1999) 3 S.C.R. 3; Jamnik et al., 2010c). Decisions delivered in Meiorin (1999) cases held that an employer’s ability to implement PES is fundamentally linked with three considerations. Employers must prove:

5

1) The existence of a BFOR which proves a direct rational connection between the requirement and job performance, i.e., workers who can pass the job test will be safe and successful at the occupation. 2) The requirement was adopted by the employer in good faith and with the intention that the requirement relates directly to its intended work related purpose 3) The requirement is reasonable necessary for the accomplishment of the occupation, i.e., Unsuccessful applicants should not possess the same characteristics as successful applicants tested in PES. This three step model has been used as criteria for evaluating workplace discrimination in numerous employment law cases since Meiorin throughout Canada (Canadian Human Rights Commission, 2007; McCallum, 2004; Kuruganti& Rickards, 2004) and marks a simplifying shift away from the American method of evaluating direct and indirect discrimination (EEOC, 1978; Jamnik et al., 2010c). Employment law in America is governed by a multitude of legislation, notably: Title VII of the Civil Rights Act (1964), Americans with Disabilities Act (1990) and Age Discrimination in Employment Act (1967). These laws collectively act to prohibit discrimination, including employment discrimination, based on ethnicity, religion, sex, age or a physical disability (Rayson, 2000c; Hogan & Quigley, 1986).

The development of PES for physically demanding occupations throughout the USA is structured according to the Uniform Guidelines on Employee Selection Procedures (EEOC, 1978). The uniform guidelines provide a “best practice” approach to the selection of workers for physically demanding occupations and legislative resistant research practices (U.S. Department of Labor, 1999). A wide range of criterion-related, content and construct validity studies can be used to develop PES if supported by appropriate evidence (Hogan & Quigley, 1986). Employment law in America shares some similar concepts to those discussed in the previous section relating to Canadian employment law. American employers have a duty to accommodate any legally disabled worker with a known physical or mental disability (Rayson, 2000c). Undue hardship may be claimed by an American employer if accommodation results in significant difficulty or expense disproportionate to the expertise or financial resources of an employer (EEOC, 1979; Rayson, 2000c). Job discrimination in America is classified under a number of areas;

1) Disparate Treatment or direct discrimination: unfair or prohibited employment practices intentionally applied to an individual (Hogan & Quigley, 1986; Berkowitz & Muller-Bonami, 2006). 2) Disparate Impact (adverse Impact) or indirect discrimination: unfair employment practices or policies unintentionally applied to a group of individuals who share a similar ethnicity, religion, sex, age or a physical disability (Rayson, 2000c; Hogan & Quigley, 1986; Berkowitz & Muller-Bonami, 2006). Adverse impact has been an important feature of a number of judicial rulings which have affected the process of PES development in America (Rayson, 2000c; Hogan & Quigley, 1986; Jackson, 1994). 6

Adverse impact occurs when an ethnic, gender, religious or disability group is unable to pass an employment related level at four-fifths the rate of another group with the highest pass rate (EEOC, 1978; U.S. Department of Labor, 1999). Any employment practice established by a plaintiff as displaying adverse impact must be supported by adequate evidence based criterion, content or construct validity (Jackson, 1994; EEOC, 1978; Hogan & Quigley, 1986). An employer has the responsibility to prove the work standard is related to a BFOR and highly related to job performance. This defence is referred to as “business necessity” and implies the introduced selection measure does cause adverse impact but is a valid representation of successful job performance (Rayson, 2000c; Hogan & Quigley, 1986; Jackson, 1994). Even with valid testing measures, a test can still be ruled as discriminatory if the employee fails to accommodate workers by investigating “alternative selection” measures. Employers must show evidence investigating alternative testing protocols, job redesign or relocation and adoption of selection measures displaying the least possible adverse impact (EEOC, 1978; U.S. Department of Labor, 1999; Rayson, 2000c; Hogan & Quigley, 1986; Jackson, 1994).

Australian anti-discrimination law in relation to PES: Differences between Australian and North American law

Australian employers face similar issues to the rest of the world but have less defined legislative guidelines when compared to America or Canada (Uniform Guidelines on Employee Selection Procedures, 1978; Meiorin, 1999; Jamnik et al., 2010c). PES development in Australia is governed by state based Anti- Discrimination or Equal Opportunity Acts (Ronalds, 2008). Discrimination for employment selection, employment training, or employment dismissal on the grounds of sex, marital status, pregnancy or potential pregnancy, breastfeeding, family responsibilities, race, disability, age, sexuality, religious belief or activity, political belief or activity, trade union activity, criminal record or association is prohibited in most states or territories due to these laws. Direct discrimination cases involving sexual harassment or racial vilification is also illegal (Ronalds, 2008). In contrast to Canadian employment law, employer motive is not seen as a relevant factor in determination of indirect discrimination (Anti-Discrimination Act, 1991; Equal Opportunity Act, 1995; Anti-Discrimination Act, 1998). Similarly different from American employment law, adverse impact in Australia is not recognised as occurring at a set level and is instead given a broad meaning which suggest “indirect discrimination occurs if a person imposes, or proposes to impose, a requirement, condition or practice that is not reasonable” (Anti-Discrimination Act, 1991; Equal Opportunity Act, 1995). Reasonable is a term which relates to the “relevant circumstances” of the case including the consequences of failing to comply with the requirement, cost of alternative requirements and financial circumstances of the person or company imposing the requirement (Ronalds, 2008; Anti-Discrimination Act, 1991; Equal Opportunity Act, 1995). As many American PES are built to comply with American laws on adverse impact, the methodological differences involved with assessing indirect discrimination between Australian and America is a major validity concern for

7 directly applying American PES in Australia. To a lesser extent there are also validity concerns with Canadian tests applied to Australia due to the subtle differences in legislative forces behind the development of Canadian PES and major focus on the concept of a BFOR and intent of the employer.

Australian employment law share some exceptions to America and Canada which enable employers to create PES. A “genuine occupational qualification” features in many state based Anti-Discrimination or Equal Opportunity Acts as an exception from discrimination attributes. This exception is often legislated as a means to isolate occupations which require distinct commercial attributes or which are only acceptable for a single gender or race, i.e. specifying female cleaners at a female toilet (Anti-Discrimination Act, 1998; Equal Opportunity Act, 1995; Anti-Discrimination Act, 1977; Anti-Discrimination Act, 1991; Equal Opportunity Act, 1984; Equal Opportunity Act, 1984). Few states use the term to mean a vital inherent physical component to successful performance of an occupation (Equal Opportunity Act, 1995; Anti- Discrimination Act, 1998) and even in these circumstances the term is used to specify differences in gender. Consequently genuine occupational qualification in Australian employment law is different from BFOR in North American law. Closer to the BFOR, is a provision made solely in the Disability Discrimination Act (1992) termed “Inherent requirements for job” to exempt employers from accommodating disabled workers which are unable to perform inherent requirements of an occupation (Ronalds, 2008). For Australian employers to use this exemption they must demonstrate the requirements of an occupation should be objectively measurable and essential to the performance of the employee (Cosma v Qantas Airways Limited, 2002; Qantas Airways Ltd v Christie, 1998). Very few employers have successfully claimed the genuine occupational qualification as a defence for employment selection decisions (Ronalds, 2008).

Australian judicial findings have ruled processes identifying the inherent requirements of any work should include analysis of normal operating conditions and additional analysis of less frequent “foreseeable circumstances” within an occupation (X v Commonwealth of Australia, 1999). This approach advocates the “Universality of service” concept used by the Canadian Armed Forces (National Defence and the Canadian Forces, 2006a, National Defence and the Canadian Forces, 2006b) and states any employee “must at all times and under any circumstances perform any functions that they may be required to perform”. To effectively accommodate for this range of functions, the development of PES must logically reflect both high frequency tasks and potentially more physically demanding tasks encountered less frequently.

Essential components of the job are vital to job performance and by definition a BFOR or inherent requirement (Payne & Harvey, 2010). Fire fighters, for example, will not be able to control their work environment and therefore may be required to perform a variety of tasks (Gledhill & Jamnik, 1992a; DEC, 2006; CFA, 2006; Sharkey, 1997; Bos et al., 2004; Sothmann et al., 1992a) at any given incident. It is also logical universality of service may be important in work situations where a team of workers rely on each other’s work performance in order to complete individual or sequenced physically demanding tasks, such

8 as; a military patrol, a police SWAT penetration, a fire rake hoe line or a fire fighting truck (Bowers et al., 1994; Budd et al., 1997a; CFA, 2006). To achieve an overall performance, each individual would be required to complete their assigned tasks satisfactorily but also complete them at a speed and competence which facilitates a combined objective (Sharp et al., 1997; Lee, 2004). Each individual within any given work crew should be capable of at least completing all work tasks, as self selection away from physically demanding work increases the individual load on every other worker in the team. Universality of service protects individuals and integrated teams from individuals of low job capability. Judicial findings also make a distinction between operational employer requirements and the inherent requirements of an occupation. Business considerations such as efficiency and productivity generally do not provide reasonable grounds for use of any anti-discrimination exemption; particularly in cases where hardship cannot be determined. However economic viability and profit considerations may serve as factors in a decision which demonstrated unjustifiable hardship as an exemption to discrimination laws (Ronalds, 2008). The “grey” area in legislation allows employers to develop a productivity component to employee work which may not be practicable in unpaid, volunteer workers.

An Australian employer has the duty, were reasonably possible, to adapt or modify the work environment to accommodate disabled workers otherwise unable to work in an existing environment (Disability Discrimination Act, 1992). Australian employment law allows exemption on the grounds of “unjustifiable hardship” if the demands of accommodating a worker would be impractical, unreasonable or if the job contains “inherent requirements” (Anti-Discrimination Act, 1977; Anti-Discrimination Act, 1991; Ronalds, 2008). The unjustifiable hardship exemption is similar in focus to the “alternative selection” and “undue hardship” concepts from North America but is limited to disabled workers interaction with their environment. The majority of Australian employment law in relation to genuine occupational qualification, inherent requirements and unjustifiable hardship relates to interactions between employers and employees (Anti-Discrimination Act, 1977; Anti-Discrimination Act, 1991; Equal Opportunity Act, 1995; Anti-Discrimination Act, 1998; Equal Opportunity Act, 1984; Equal Opportunity Act, 1984). Organisations which utilise volunteer workers, like emergency service agencies and are established from a state or commonwealth law (CFA annual report, 2008; NSW RFS annual report, 2008; TFS annual report, 2008) are referred to as “voluntary bodies” (Disability Discrimination Act, 1992). Volunteer workers fall outside these interactions in all areas except the provision of services (labour), and compliance with OHS laws. A voluntary body must not discriminate against a volunteer providing a service in the provision of equipment, training or design of facilities (Disability Discrimination Act, 1992; Equal Opportunity Act, 1995; Anti-Discrimination Act, 1991) but no legislative or judicial discrimination considerations appear to apply to the introduction of PES in Australian voluntary bodies to volunteer workers.

The primary legislative compliance issues with developing PES for Australian volunteer workers are likely to focus on OHS compliance rather than discrimination law compliance as has been the case in North 9

America. For this reason, and due to subtle differences in the legislative forces behind the development of PES in North America, it is important to highlight PES developed in other countries may not be directly valid and reflective of the work environment in Australia. Valid PES need to be developed which reflect compliance with Australian law and represent the content or physical demand of the targeted occupation in the way it is performed in this country.

Justification for the implementation of employment standards in Australian fire agencies.

The higher risk of injury during fire fighting work (Nuwayhid et al., 1993; Walton et al., 2003; AFAC, 2002; Fabio et al., 2002; Guidotti & Clough, 1992; Matticks et al., 1992), similar to other physically demanding occupations (Knapik et al., 2001a; Knapik et al., 2001b; Bell et al., 2000), has been a significant factor motivating the implementation of wellness programs or PES (Rorke, 2002; Sothmann et al., 1990; Barknekow-Bergkvist et al., 2004; Stewart et al., 2008; Cady et al., 1985; Brownlie et al., 1984; Sluiter & Frings-Dresen, 2007; Shephard & Bonneau, 2003). Cost effective savings, work performance increases and substantial reductions in injury rate have been demonstrated with the introduction of wellness programs into most physically demanding occupations (Cady et al., 1985; Aasa et al., 2005; Aasa et al., 2008; Maniscalco et al., 1999; Musich et al., 2001). Introduction of these programs into fire fighting samples can produce variable overall health results but still serve to highlight the connection between increased levels of general fitness and lower work injuries (Cady et al., 1979; Green & Crouse, 1991). Cady et al., (1979) found injury rates for least fit fire fighters where 6.3% higher than for most fit fire fighters. In subsequent work Cady et al (1985) observed individuals with the low fitness managed to improve their physical work capacity by 16% after completing a prescribed exercise program. From 1977 to 1995, the leading cause of all American SFF fatalities was heart attack or stroke (Washburn, 1996) many of which may have resulted from increased physical work demands during work.

Fahy (2005) reported sudden cardiac events encountered during a fire shift were responsible for 440 American SFF deaths between 1995 and 2004. Fahy & LeBlanc (2006) reported 60% of the American fire fighter mortality rate at age 40 or over was cardiac related in 2005. Mortality rate for American fire fighters in their fifties and sixties was observed 66% and 350% higher than the average for all age groups respectively. Burton (2007) showed 59% of all fire fighters who died between 1990 and 200 were over 40. Fahy (2005) observed 70% of deaths were in volunteer fire fighters suggesting volunteer status increases the chance of fire fighter mortality when compared to career status. Heart attack was the cause of death in 42% of all US volunteer wild land fire fighter fatalities from 1990 – 1998 (Mangan, 1999). The volunteer mortality rate is most likely due to the older operational age of volunteer fire fighters and the increase in relative work demands encountered with age related functional declines (Fahy, 2005; Chan & Koh, 2000; Sluiter & Frings-Dresen, 2007; Sothmann et al., 1990; Lusa et al., 1994; Sothmann et al., 1992a). Fahy (2005) suggested career fire fighters were generally younger during their operational involvement with physically 10 demanding occupations and do not remain active after their retirement like volunteer fire fighters. This idea is supported in fitness studies show career fire fighters are significantly aerobically fitter in the earlier stages of their careers when compared to volunteer. Swank et al., (2000) showed American career SFF aged 30 years old had aerobic power levels of 46 mL.kg-1.min-1. An aged matched sample of volunteer SFF had a significantly lower aerobic power level of 36 mL.kg-1.min-1. Swank et al., (2000) also showed the career fire fighters do not maintain a high level of aerobic power as they are and are comparable to volunteers after the age of 40.

At the age of 40, both career and volunteer fire fighters appear to have comparable aerobic power levels to the general public of around 33 mL.kg-1.min-1. Similar findings for both age groups have been observed in Canadian SFFs (Horowitz and Montgomery, 1991), American SFFs (Byrd & Collins, 1980) and Australian SFFs (Reaburn, 2000). Some groups of fire fighters have shown matched or lower aerobic power levels compared to age matched groups in the general population. McFadyen (1996) found higher aerobic power rake hoe crews could rake, carry and deploy hose at rates of 27%, 9% and 14% respectively than lower aerobic power crews indicating aerobic power is important for fire fighting work. Rodriguez and Eldridge (2003) demonstrated a direct inverse relationship between increases in aerobic power and/or muscular strength and decreased probability of injury in fire fighters. Over exertion is likely a factor which contributed to fire fighter heart attack and death (Fahy & LeBlanc, 2000; Fahy & LeBlanc, 2006), especially in individuals predisposed to cardiac injury such as older fire fighters and volunteers. Volunteer fire fighters are exposed to the same absolute job demands as all other fire fighters. Due to lower aerobic power levels throughout their careers (Swank et al., 2000) and often increased age (McLellan, 2004; Fahy & LeBlanc, 2006; Fahy, 2005), Australian volunteer fire fighters may face a high risk of injury whilst performing physically demanding work on the fire ground.

11

Table 2.1, American Fire Fighter Age at time of Fatal Incident 1990 – 2000 (adapted from Burton, 2007) Age Percent of Fatalities

Under 21 years 4% 21 – 30 years 18%

31 – 40 years 19%

41 – 50 years 24%

51 – 60 years 20%

61+ years 15%

Australian fire agencies spend between one and two percent of their annual budget on compensation claims or compensation insurance in relation to workplace injuries. (CFA annual report, 2008; NSW RFS annual report, 2008). Around 1% of all volunteer fire fighters lodged compensation claims compared to the lodging rate of 6% in career fire fighters (CFA annual report, 2008; TFS annual report, 2008). Few Australian rural fire fighters are killed on duty during operational activities compared to American fire fighters (Burton, 2007; Washburn, 1996; Fahy & LeBlanc, 2006). Australian fire fighting mortality rate is estimated at only 1.5 fire fighters per year over the past 15 years or 0.7 per 100,000 Australian BFF (Aisbett, 2005). Injury data from NSW during the 2002-2003 fire season, showed seven fire fighters were killed and up to 400 fire fighters injured through heat stress, dehydration, smoke inhalation, cuts, strains and sprains (ABS, 2004). Brotherhood et al., (1997) has suggested injuries of this type which occur on the fire ground are signs of fatigue and may lead to more severe injuries. The lower mortality rates in Australia may be attributed to operational and work demand differences between Australian fire fighting and fire fighting overseas (Sothmann et al., 1992b; CFA, 2006; Sharkey, 1997; Gledhill & Jamnik, 1992a; Cuddy et al., 2007), but these interactions are yet to be investigated. To the authors’ knowledge, Australian fire agencies have released no published statistics on the overall rates or causes of fire fighter injury and have a poor understanding of injury or mortality in their populations.

It is reasonable to assume Australian RTBS will be composed of physically demanding elements like other observed fire fighting work. Although the fire fighter mortality rate in Australia is low, injuries possibly linked to over exertion, such as heat exhaustion occur frequently on the fire ground (ABS, 2004) and warrant the development of PES for RTBS (Mangan, 2002). Not only does a PES serve a risk mitigation purpose they also achieve compliance with legislative OHS purposes (Rorke, 2002; Jackson, 1994).

The bushfire suppression environment is inherently dangerous requiring long working hours in a volatile operational situation, extreme radiant heat, and often thick smoke (CFA, 2006; Mangan, 1999; DSE, 2008; Hunter, 2003). All fire fighting work is composed of a range of heavy lifting and manual handling tasks

12

(Sothmann et al., 1990; Sharkey, 1997; Brotherhood et al., 1997; Lusa et al., 1994; Deakin et al., 1997; CFA, 2002). Fire fighting tasks requiring repetitive or sustained application of force, awkward postures, unbalanced loads are classified in Australia as hazardous manual tasks (National Standard for Manual Tasks, Australian Safety and Compensation Council, 2007). In order to comply with OHS acts (OHS Act, 2004), employers must “eliminate risks to health and safety so far as is reasonably practical” and if unable to eliminate risks, “reduce those risks so far as is reasonably practicable”. Bushfire suppression agencies, despite their status as voluntary bodies, are subject to risk assessment and reduction standards of all other Australian businesses or organisations. Australian fire agencies have a duty, so far as is reasonably practicable, to protect the health and safety of all fire fighters by: assessing workplace risk, applying strategy to reduce workplace risk, training fire fighters to understand and competently manage their workplace risk (OHS Act, 2004).

PES provide a practical and widely accessible means of controlling the potential risk of over- exertion injury by assessing the job specific capabilities of an employee against the required demands of work (Cady et al., 1985; Rayson, 2000c; Rayson, 2000b). A PES engineered on safety grounds does not appear to contradict any anti-discrimination laws. Australian employment law allows for discrimination against an individual, employee or volunteer worker, on the basis of impairment or physical features if the discrimination is reasonable necessary to protect the health or safety of any individual, members of the public or property (NSW Anti-Discrimination Act, 1977; QLD Anti-Discrimination Act, 1991; Victorian Equal Opportunity Act, 1995; Tasmanian Anti-Discrimination Act, 1998; WA Equal Opportunity Act, 1984; SA Equal Opportunity Act, 1984). In Australia, the implementation of PES is therefore the practical compromise between OHS law and employment discrimination law. Organisations or employers may design PES determined ‘reasonably practicable’ and ‘reasonably necessary’ to successfully protect the health of a person on a work site, members of the public or property. The precise definition of reasonably practicable has been left unclear in both the OHS and anti-discrimination acts leaving organisations the power to determine the definition of successful or appropriate job performance. The resulting ‘grey area’ enables employers to set a level of productivity based on “reasonable” expectations of employee work behaviour, beyond those required to simply maintain safety and indicative of economic considerations. Although Australian fire fighters likely have relatively low rates of occupational fatality when compared with American fire fighters (Fahy, 2005; ABS, 2004), Australian fire agencies have a similar legislative duty to manage the physical risk faced by working personnel on the fire ground. This includes physical work which may place older and less fit fire fighters with pre existing cardiovascular conditions at the risk of over exertion related injury.

Australian fire fighting and RTBS

13

A large population of rural volunteer fire fighters across Australia contribute the bulk of work hours worked during a prolonged bushfire. Their voluntary contribution is the primary reason behind the relatively cheap cost of fire suppression efforts in Australia compared globally to other fire agencies (Hyde et al., 2007). This saving is highlighted in one of the world’s largest volunteer-based emergency services organisations, Victoria’s Country fire Authority (CFA). Fire related disruption to the Victorian power grid in 2006 resulted in a single day economic impact of $500 million AUD dollars (DSE Living with Fire Report, 2008). The cost of replacing volunteer manpower with alternative commercial manpower for the summer of 2000-01 was valued at $470 million AUD and the annual overall contribution to the Victorian economy was estimated at $840 million AUD in 2008 (Handmer & Ganewatta, 2008; Gledhill, 2001; ACSES, 2007). When balanced against the statutory contribution of $268.6 million AUD to CFA funding in 2007-08 (CFA annual report, 2008), a rural fire agency provides a cost effective return on investment for the Victorian community. A clear priority of Australian government and fire agencies is the retention, safety and maintenance of the infrastructure supporting volunteer fire fighters (Gledhill, 2001). It is important for the purposes of this review to draw clear distinctions between the three distinctly different fire fighting groups in Australia; urban or SFFs, LMFFs and rural BFFs. This review will focus primary on identifying important issues for PES development in BFFs completing RTBS work.

Urban or SFFs

The majority of urban fire fighters in Australia are full time employees working within major cities in Australia (Metropolitan fire service South Australia Annual report, 2008; Metropolitan fire and emergency services board Annual report, 2008; NSW fire brigades Annual report, 2006-07). In Australia’s two largest cities, Melbourne and Sydney, structure fire agencies are responsible for fire fighting, rescue, automotive incident or other emergency response in inner city urban areas. Individual state structural fire agencies typical respond to between 3,600 and 24,000 fire incidents a year, dependent on the size of the population they service (Metropolitan fire service South Australia Annual report, 2008; Metropolitan fire and emergency services board Annual report, 2008; NSW fire brigades Annual report, 2006-07). Structural fire agencies, such as Melbourne’s Metropolitan Fire and Emergency Services Board (MFB) employ 1,702 operational fire fighters who operate in high density urban areas containing two million residents in around 1,000 square kilometres (MFB Annual report, 2008). Primary fire suppression work on large scale fires can involve up to 240 fire fighters and is unlikely to last for more than five hours before fire behaviour is controlled (NSWFB annual report, 2008). The average operational age of a SFF reflected in research work is between 30 and 37 years of age (Williford et al., 1999; Gledhill & Jamnik, 1992a; Gledhill & Jamnik, 1992b; Lusa et al., 1994; Bos et al., 2004; Dawson et al., 2000; Lemon & Hermiston, 1977; Dotson et al., 1978; Byrd and Collins, 1980; Deakin et al., 1997; Rhea et al., 2004; Michaelides et al., 2008; Richmond et al., 2008; Davis et al., 1982).

14

Photo: SFFs engaged in hose spray (left) and direct attack at a structural entry point (right). SFF are wearing SCBA in both pictures, Burton, 2007.

Most structural fire agencies measure the job related physical competency of their personnel prior to employment using a Candidate Physical Ability Test (CPAT). Evolutions of the CPAT and live structure fire incidents are almost always conducted wearing self contained breathing apparatus (SCBA) and protective equipment which collectively weigh around 22- 25kg (Michaelides et al., 2008; Gledhill & Jamnik, 1992a; Williford et al., 1999; Rhea et al., 2004; Davis et al., 1982). Most variations of the CPAT used through the world are generally developed from task analysis assessments (Gledhill & Jamnik, 1992b) and consist of core elements including: Ladder extension and climb, Charged hose advance, Simulated victim rescue, Forcible entry with sledgehammer or against weighted sled, Multi-story stair climb, Equipment hoist, Hydrant assembly and Equipment carry (AFAC, 2002; Michaelidis et al., 2008; Davis et al., 1982; Rhea et al., 200; Williford et al., 1999; Deakin et al., 1997; Harvey et al., 2008). Williams-Bell et al., (2009) showed the aerobic demands of successful completing the CPAT were 38.5 and 36.6 mL.kg-1.min-1 in men and women respectively. A defence developed variant of this test, which uses similar tasks, the CF-DND, has shown aerobic demands of 36.6 mL.kg-1.min-1 (Dreger & Petersen, 2007). The average physiological stress values recorded in the CPAT are equivalent to between 64 and 73% O2max and up to 91% Maximal heart rate depending on the sequence of tests used (Williams-Bell., 2009; Harvey et al., 2008). In combination with the CPAT, Australian SFFs are also required to obtain an aerobic standard of 45 mL.kg-1.min-1 or 9.6 on the multi stage shuttle run to satisfy the physical standards of the selection process (AFAC, 2002). The multi stage shuttle run performance is indexed to the hardest SFF task, high rise pack and halligan tool carry up multiple flights of stairs, which has an aerobic demand of 44 ± 1.5 mL.kg-1.min-1 (Gledhill & Jamnik, 1992a).

Few studies have evaluated actual or “real” fire suppression work in SFF or other fire fighting populations. Bos et al., (2004) observed 85 real SFF shifts were composed of a low number of sporadic, high

15 intensity incidents per shift, an average of 1.5 incidents, with short durations of operational time, an average of 88 ± 76 min. Operational time at deployment was determined to comprised 47% preparation, 18% inside structure tasks, 16% SCBA tasks, 11% water-collection, 4% operation of equipment and 4% extinguishing tasks. Up to four times as many participants were involved in preparation, inside structure and water collection work as SFF completing SCBA, operation of equipment or extinguishing tasks. Moderate %HRR values were observed for SCBA (58.4 ± 19.1%HRR), inside structure (43.4 ± 18.6%HRR), water collection (37.7 ± 15.0%HRR), preparation (32.1 ± 10.3%HRR) and extinguishing tasks (31.9 ± 8.9%HRR). Sothmann and colleagues (1992b) reported high peak heart rates during actual fire suppression duties of up to 88 ± 6 %HRmax over durations of 15 ± 7 minutes. The average work heart rate during work were 157 ± 8 b.min-1, equivalent to a fire fighters’ predicted oxygen consumption rate of 25.6 ± 8.7 mL.kg- 1.min-1.

Due to the problematic nature of emergency service work, simulated work studies are often conducted in the place of real studies. Oldham and colleagues (2000) observed simular muscular force is applied during performance of simulated and real common SFF tasks such as lifting ladders and dragging hose. Heart rate responses (Mol et al., 2003; Selkirk & McLellan, 2000) and oxygen consumption rates encountered during simulated drills (Holmér & Gavhed, 2006; Sothmann et al., 1990) may however, over- represent the actual physical demand of real SFF work. During a multi stage SFF work bout, Sothmann et al., (1990) measured heart rates of 173 ± 9 b.min-1 and oxygen consumption rates of 30.5 ± 5.6 mL.kg-1.min-1 and

-1 2.5 ± 0.5 L.min , equivalent to 76% of fire fighters O2max for around 9 minutes of work. On a similar multi stage SFF drill, Holmér and Gavhed (2006) observed an average heart rate of 168 ± 11.7 b.min-1, rating of perceived exertion (R.P.E) of 16.4 ± 1.2 and oxygen consumption rates of 33.9 ± 4.2 mL.kg-1.min-1 or 2.8 ± 0.3 L.min-1. Bilzon et al., (2001a) demonstrated even higher aerobic demands in naval fire fighters conducting

-1 -1 simulated work tasks. Peak metabolic demand of 43 mL.kg .min or 82% O2max was encountered during a simulated drum carry task. Only one task in this study, a hose spraying task generated oxygen consumption

-1 -1 less than 77% fire fighter O2max (23 mL.kg .min or 47% O2max,).

Smith et al., (2001) studied the cumulative effects of three repeated 7 minute strenuous live fire drills on recruit SFFs. High heart rates of 174.8 ± 8.3, 186.0 ± 9.1 and 189.3 ± 10.5 b.min-1 were observed after each trial, but these heart rates did not differ significantly from each other, indicating no increase in already near maximal heart rates with cumulative strain. Smith et al. (1996) also reported high heart rates after 8 minute live fire training drills in an elevated temperature facility. Advanced a charged fire hose up one flight of stairs and then advanced the hose to various positions on the second, generated a heart rate of 170 ± 2 beats·min-1 and an RPE of 13.4 ± 2.6. The authors reasoned these values would equate to an exercise intensity of 65% O2max. Williford et al., (1999) and Manning & Griggs (1983) reported close to near maximal mean heart rates of 92% relative Maximal and 90 to 100% Maximal respectively, during simulated SFF drills between 150 and 430 seconds in duration. In longer 14.5 minute simulated fire drills, Louhevaara et al., 16

-1 (1994) observed a mean heart rate range of 122 to 158 beats·min or 66 to 86%HRmax. Estimations of oxygen consumption during this drill of 26 mL.kg-1.min-1 and 2.1 L.min-1 were similar those predicted in Sothmann et al., (1990). Gledhill & Jamnik (1992) performed physical demands analysis over two studies on a number of individual SFF tasks demonstrated aerobic demands between 16.8 and 44 mL.kg-1.min-1. Peak mean aerobic demands of 44.0 ± 1.5mL.kg-1.min-1 were observed during 128 second duration equipment carry up high rise stairs tasks and a peak mean heart rate of 181 ± 9 during a 64 second pitched roof ventilation exercise. The

- authors highlighted “90% of the fire fighting operations investigated required a mean O2 of 23.4 mL.kg 1.min-1 indicating the importance of determining the frequency of occurrence of more physically demanding jobs. Both actual and simulated SFF are likely to generate mean peak heart rates ranging from 80 -100

-1 -1 %HRmax and oxygen consumption rates commonly exceeding 40 mL.kg .min . It is clear the highly energetic heart rates and metabolic work rates observed in a number of tasks highlight the sporadically occurring, physically demanding nature of SFF work.

Land management, conservation and state government networked fire fighters

In response to major bushfires, Australian government agencies tend to work together under a Networked Emergency Organisation model and deploy similarly trained personnel to perform fire fighting functions (DSE, 2008). As these workers perform similar functions, I will refer to all of these fire fighters as LMFFs for the remainder of this review. Single state land management agencies may employ 2,500 full time staff and an additional workforce of up to 700 seasonal fire fighters for the duration of the summer fire season (DSE, 2008). LMFFs work outside urban areas of Australia and manage the ecology of public land, national and state parks year round. Employees of Australian land management agencies are required to perform regular activities such as tree felling, tree planting, prescribed/ fuel reduction burning and management of walking trails or infrastructure within parks (DEC, 2007; DSE, 2008) . An essential function of the LMFF occupation is to perform, when required, bushfire suppression and prevention activities throughout the fire season. Land management fire agencies respond to between 480 and 600 fire incidents each year (ABS, 2004; DEC, 2007). Over 3000 staff can be deployment to large scale or multiple consecutive fire incidents lasting between 50 and 150 days (ABS, 2004; Hunter, 2003).

17

Photo: LMFFs engaged in team line build (left) and burn out ignition (right).

The average operational age of a LMFF observed in research work is between 24.5 and 31.4 years of age (Heil, 2002; Ruby et al., 2002; Brotherhood et al., 1997; Brown et al., 2007). The composition of LMFF work is much closer to the North American wildland fire fighter (WFF) than any other type of Australian fire fighter. Both groups work with limited water storage vehicles, land clearing machinery, remote area deployment teams and conduct load carriage, emergency response work, manual rake tasks, such as rake hoe mineral earth line construction, using an axe or chain sawing tasks (DSE, 2008; Sharkey, 1997; Payne et al., 2005). Using a rake to generate fire line generates an aerobic cost of 22.5 mL.kg-1.min-1 (Sharkey, 1997). LMFFs are assessed for physical capacity using the 4.8 km hike loaded with 20.5 kg of weight in a time faster than 45 minutes, known as the pack hike test (DEC, 2006; Sharkey, 1997; Payne et al., 2005; Brown et al., 2007; Sharkey & Rothwell, 1996; De-Lorenzo-Green & Sharkey, 1995; AFAC, 2002). This test was designed to test aerobic power, a factor assessed as the primary factor limiting sustained work performance in LMFF work (Sharkey, 1999). Successful completion of the pack hike replicates a sustained aerobic demand of 45 mL.kg-1.min-1 for 45 minutes (DeLorenzo-Green & Sharkey, 1995; Sharkey, Rothwell & DeLorenzo-Green, 1994; Sharkey & Rothwell, 1996), a figure twice the aerobic demand of the hardest fire fighting task, fire line construction using a hand tool. The pack hike has alternatively shown to produce lower sustained aerobic demands of 29.6 and 42.3 mL.kg-1.min-1 in men and women respectively (Brown et al., 2007). The Uniform guidelines (EEOC, 1979) state “where two or more selection procedures are available which serve the user’s legitimate interest in efficient and trustworthy workmanship, and which are substantially equally valid for a given purpose, the user should use the procedure which has been demonstrated to have a lesser adverse impact”. The pack hike has shown to have lower rates of adverse impact than tests of simulated line construction, a high level of test reliability, and significant correlations with muscular strength tests or tests of simulated line construction (DeLorenzo-Green & Sharkey, 1995). The pack hike was selected for implementation in North America because it is comparable with a simulated line construction test but has a smallest adverse impact (DeLorenzo-Green & Sharkey, 1995).

18

Table 2.2, Energy expenditure, physiological, and subjective responses to different jobs performed during fireline construction (Budd et al., 1997b; Budd et al., 1997d) Chainsaw Measurement Raker Slasher Crew Leader Operator

% of total time in primary task 57.8 ± 15.1 75.6 ± 15.9 65.9 ± 15.2 -

Oxygen Uptake -

L·min-1 1.9 ± 0.3 1.4 ± 0.2 0.9 ± 0.2 -

mL·kg-1·min-1 27.2 ± 3.8 19.5 ± 2.8 12.7 ± 2.5 -

a Exercise Intensity (% O2max ) 57.6 ± 7.1 39.7 ± 6.1 27.0 ± 5.2 -

Heart rate (beats·min-1) 152 ± 10 160± 14 143 ± 13 126 ± 14

RPE (6 – 20)b 13.2 ± 2.1 14.0 ± 1.9 14.2 ± 1.7 10.6 ± 1.9

All data are means ± standard deviation. %, percentage; L·min-1, liters per minute; mL·kg-1·min-1, milliliters per kilogram body mass per minute; O2max, maximal oxygen uptake; min, minutes; a, O2max predicted by extrapolating O2 during normal and fast rake-hoeing to age-predicted Maximal heart rate; b, Noble, B. J. and R. J. Robertson (1996). Perceived Exertion. Human Kinetics. Campaign, IL.

Average total energy expenditure rate of 4160 ± 884 and 4768 ± 478 kcal day-1 and activity energy expenditure of 2101 ± 716 and 2585 ± 406 kcal day-1 have been observed in American WFFs using doubly labelled water and physical activity monitor techniques respectively (Ruby et al., 2002; Heil, 2002). Heil (2002) suggested the energy expenditure rate of LMFF “rarely exceeded 8 kcal.min-1 and tended to oscillate between 4 and 6 kcal min-1 most frequently”. Cuddy et al., (2007) observed 66% of fire ground time is spent in sedentary activity intensities and 33% in light activity ranges. LMFFs spent 74.2 ± 21.6 and 79.1 ± 21.1 min.2hr-1 of shift time in a sedentary activity range and 45.7 ± 21.6 and 40.8 ± 21.0 min.2hr-1 of shift time in a light activity range when supplemented with carbohydrate and placebo respectively (Cuddy et al., 2007). The same carbohydrate and placebo supplemented fire fighters generated 551 ± 185 and 451 ± 133 mean activity counts per minute respectively, both levels classified as “light” work intensity (Cuddy et al., 2007). Table 2.2 displayed similar results in Australian LMFFs. In simulated work roles of raker, slasher and chainsaw operator, the authors reported 57.8 ± 15.1, 75.6 ± 15.9, 65.9 ± 15.2% of total time in these role was spent involved in the primary task. The same roles were shown to be working at an exercise intensity equivalent to 57.6 ± 7.1, 39.7 ± 6.1 and 27.0 ± 5.2 % O2max. The role of raker, likely the most common for

19

LMFF, contained the lowest portion of time performing a primary task but required the highest sustained aerobic consumption of 27.2 ± 3.8 mL.kg-1.min-1 or 1.9 ± 0.3 L.min-1. Taken in combination, LMFF work is characterised by significant periods of sedentary time but the potential for tasks to generate high heart rates, aerobic demand and relative work rates at any time throughout a shift.

McFadyen et al., (1996) observed job performance differences between low fit and high fit WFF. Even though, both low and high fit WFF groups exceeded the minimal required level of aerobic power with

-1 -1 O2max values of 48.8 ± 3.0 and 55.7 ± 3.0 mL.kg .min , the authors reported high fit fire fighters were able to sustain a more productive pace throughout a shift. High fit WFF completed more handline raked per hour (27%), hose carry (9%), hose deployment (14%) and worked at a higher oxygen consumption rate for a lower Maximal heart rate. An initial rake hoe suppression pace of 3.2 ± 0.7 and 2.5 ± 0.9 L.min-1 was reported in high and low fit WFF. The same authors reported a sustained all day rake hoe pace of 2.3 ± 0.4 and 2.0 ± 0.6 L.min-1 for the same high fit and low fit WFF respectively, suggesting higher fit aerobic individuals have greater productivity in rapid and sustained manual rake work. Brotherhood et al., (1997a) observed multiple physiological variables during simulated self paced 5 minute ‘slow’, ‘normal’, ‘fast’ and 23 minute ‘crew’ rake hoe tasks. ‘Slow’, ‘normal’, ‘fast’ and ‘crew’ rake hoe tasks generated an aerobic demand of 22 ± 6, 32 ± 4, 41 ± 6 and 31 ± 4 mL.kg-1.min-1; heart rates of 137 ± 16, 161 ± 17, 180 ± 13 and 158 ± 17 b.min; and Maximal relative raking work rate of 47 ± 10, 70 ± 12, 88 ± 10 and 67 ± 16% respectively. Interestingly the slow rake task resulted in a high oxygen usage for less ground raked, a value

-1 -2 -1 -2 of 17 ± 6 ml.O2 kg m compared with comparable measures of 13 ± 3, 13 ± 3 and 13 ± 2 ml.O2 kg m for normal, fast and crew rake hoe tasks respectively. The authors suggested LMFFs self pace themselves at sustainable work rates which balance fire line productivity against relative physiological cost. Using this logic and findings from the aforementioned studies, LMFFs work are exposed to arduous short duration

-1 -1 -1 work rates of between 31 - 41 mL.kg .min or 67 -88% O2max, and heart rates of 158 to 180 beats·min roughly 60% of their shift time. It is likely LMFFs self select their pace in individual and crew rakes to maintain a level of productivity relevant to the fire behaviour or the task at hand.

Rural bushfire agencies

Rural bushfire agencies provide a community-based, rapid first attack and prolonged response to rural structure fire, bushfire and rescue emergencies across Australia with the primary purpose of protecting life, property and environmental assets (TFS, 2008; CFS, 2008). It is important to highlight around 99% of the 206, 954 personnel utilised by Australian rural bushfire agencies are composed of community-based volunteer workers (CFS, 2008; CFA, 2008; McLellan et al., 2004). For the purpose of this review, I will refer only to the RTBS duties of these agencies and the volunteer personnel who conduct this work. Rural fire agencies respond to between 4000 and 15500 fire incidents per year (TFS, 2008; CFS, 2008; CFA, 2008). The DSE, Living with Fire report (2008) stated more than 80% of fire incidents are contained to less than five 20 hectares, with the remaining 20% of larger incidents resulting in 90% of area burnt annually. CFA annual report (2008) indicated 54% of deployments over the last five years were primary deployments, meaning the personnel were deployed close to the origin of the fire and most likely able to contain the fire to a small size. The same report indicated 45% of deployments were support deployments, a long term campaign fire incident in response to unsuccessful primary deployments.

Photo: RTBS personnel engaged in hose spray mobile patrol tanker (left) and 38 blacking out work (right), CFA (2008).

Due to the logistical impracticalities of conducting research on local, first response deployments which have the potential to be spread out over a 20 million hectare area or more (CFA, 2008), this review and subsequent research work will focus on large scale campaign fire deployments. Long term or campaign fire incident response often involves the remote deployment of personnel from neighbouring regions, urban areas, interstate or even overseas depending on the size of the fire incident (NSWRFS, 2008; CFA, 2008). Currently only 370 of Australian RTBS personnel, from the smallest rural fire agency, are required to meet a PES, the pack hike (ACT RFS, 2007; McLellan, 2004; Payne, 2005). All other Australian BFF volunteers are required to meet only equipment operation standards which assess an individual’s ability to work within a fire crew, understand factors which affect the bushfire behaviour and maximise their survival chances in the event of a catastrophe (CFA, 2006). No studies have assessed the physical demand of RTBS or the capabilities of rural fire fighters to safely complete this work. This review and thesis will focus on identifying the physical demands of RTBS work with particular emphasis on campaign bushfire suppression.

Table, 2.3. Response status during call outs of volunteer fire fighter to fire incidents (CFA, 2008). 2004 2005 2006 2007 2008 5 year average Primary 58% 57% 53% 52% 54% 55% Support 42% 43% 47% 48% 46% 45%

21

In the absence of any detailed reports on the operational behaviours of Australian rural BFFs, a preliminary interview with an experienced operations officer (Bosua, 2006) suggests a common individual campaign deployment lasts for between three to five days. Fire fighters generally crew a fire tanker with three to five individuals from the same fire station and usually have some degree of familiarity with their fellow crew members. Each tanker is commonly deployed with four other trucks, called a strike team (Bosua, 2006; CFA, 2006). The method of fire suppression and individual crew tasking within a strike team, is highly dependent on fire behaviour and the protection of assets located within the area (Bosua, 2006; CFA, 2006). Australian rural fire crews’ use a combination of ‘wet’ and ‘dry’ fire suppression techniques to contain fire spread (Bosua, 2006; CFA, 2006). The Wildfire Firefighter: Learning Manual (2006) describes a number of tasks performed on the fire ground including; ‘wet’ hose tasks to distribute extinguishing mediums, ‘dry’ manual rake tasks to create mineral earth lines or contain small fires, and miscellaneous tasks which focus on using a portable knapsack spray pump, filling the fire tanker with water and working with other vehicles, aircraft or heavy machinery. Similarly to SFF techniques, ‘wet’ fire fighting techniques involve the suppression of bushfire using a fire truck to deploy water and other extinguishing mediums directly or indirectly onto a bushfire. These techniques when applied directly, removes the heat from a burning fire or indirectly, lower the flammability of areas around it.

During 4 minute simulated hose work tasks, Bilzon et al., (2001a) demonstrated aerobic demands of

-1 -1 38 mL.kg .min in naval fire fighters, equivalent to 77% O2max. Gledhill and Jamnik (1992a ) observed 15 metre and 61 metre uncharged hose advance/lay tasks in SFFs which lasted 21 and 65 seconds; peak heart rates of 146 ± 11 beats·min-1 and 123 ± 1 beats·min-1 and aerobic demands of 23.4 ± 1.5 mL·kg·min-1 and 25.7 ± 3.0 mL·kg·min-1 respectively. Additionally the physical demands of 68mm charged hose advance and 100mm lateral hose repositioning tasks. Charged advance and lateral repositioning tasks lasted 39 and 46 seconds; peak heart rates of 166 ± 7 beats·min-1 and 173 ± 5 beats·min-1 and aerobic demands of 30.9 ± 3.0 mL·kg·min-1 and 31.7 ± 2.8 mL·kg·min-1 respectively. The increased in oxygen consumption and duration of completion is most likely due to the 16.4kg difference in force required to advance two sections of charged 68mm hose compared to the same hose configuration in an uncharged state. McFadyen et al., (1996) observed the mean absolute oxygen costs of high fit American WFFs carrying hose, deploying fire hose and working with charged hose were reported as 2.1 ± 0.5, 1.9 ± 0.5 and 3.8 ± 0.5 L.min-1 respectively. The low fit crew completed the same tasks with similar oxygen costs of 2.2 ± 0.4, 1.9 ± 0.5 and 3.2 ± 0.5 L.min-1 respectively but may have conducted this work slower than the high fit fire fighters. If similar to previously investigated fire fighting tasks, it is likely ‘wet’ bushfire suppression techniques will elicit similar task durations, heart rates and oxygen consumption rates of between 21 to 65 seconds, 123 to 174 beats·min-1 and 23.4 to 31.7 or 1.9 to 3.8 L.min-1 respectively. To successfully complete ‘wet’ fire fighting tasks, average male and female aged matched members of the Australian population may need to work at between 61 –

84% O2max and 77 – 104% O2max respectively (Gore & Edwards, 1992).

22

Similarly to LMFF techniques, ‘dry’ fire fighting techniques utilise manual hand tools, earth moving machinery or burnt ground to create physical barriers devoid of fuel sources around a bushfire (Bosua, 2006; CFA, 2006). Both Sharkey (1997) and De-Lorenzo-Green & Sharkey (1995) stated prolonged use of manual rake to generate fire line generates an aerobic cost of 22 mL.kg-1.min-1. As mentioned earlier in this review, Brotherhood et al., (1997a) found ‘normal’, ‘fast’ and ‘crew’ rake hoe trials generated higher aerobic demands of 32 ± 4, 41 ± 6 and 31 ± 4 mL.kg-1.min-1; 2.3 ± 0.4, 2.9 ± 0.5 and 2.1 ± 0.3 L.min-1; heart rates of 161 ± 17, 180 ± 13 and 158 ± 17 beats per minute; and Maximal relative raking work rate of 70 ± 12, 88 ± 10 and 67 ± 16% respectively. Both ‘normal’ and ‘crew’ rake hoe trials are reflective of the sustained all day rake hoe pace of 2.3 ± 0.4 and 2.0 ± 0.6 L.min-1 reported by McFadyen et al., (1996) for high fit and low fit fire fighters. ‘Dry’ bushfire suppression techniques are likely to elicit heart rates and oxygen consumption rates of 158 to 180 beats·min-1 and 31 to 41 mL.kg-1.min-1 or 2.1 to 2.9 L.min-1 respectively. Successful completion of previously investigated ‘dry’ fire fighting tasks, by average male and female aged matched members of the Australian population, will require between 82 – 108% O2max and

102 – 134% O2max respectively (Gore & Edwards, 1992).

The portable knapsack sprayers used by the bushfire fire fighter consist of a 6.5 kg 16-litre polythene tank, fitted with two shoulder straps, a hand pump and nozzle (Ghugare et al., 1991; CFA, 2002).

Gledhill and Jamnik (1992a) identified SFF water backpack carry elicited a mean O2 of 29.7 ± 1.3 mL·kg·min-1 and a peak heart rate of 134 ± 16 beats·min-1 in 27 seconds of task duration. Sharkey (1997) reported similar energy costs of sustained packing of heavy loads in American WFF. Load carriage of pumps, hose packs and 5 gallon water bags required 22.5 mL·kg·min-1 on flat terrain and 29.5 mL·kg·min-1 on hill terrain. Ruby et al., (2003), Table 2.4, observed maximal simulated backpack hiking escape with a 16kg LMFF line pack over a 660.5m course with 137m of vertical gain. Male and female fire fighters completed the course in 10.7 ± 1.4 and 13.7 ± 1.3 minutes whilst generating oxygen consumption of 41.1 ± 6.0 and 32.5 ± 6.6 mL·kg·min-1 respectively. Subjects in either trial did not spray water from their backpack which remained at a constant weight for the trial. It is therefore likely carrying sustained knapsack loads in RTBS work would require 29 mL·kg·min-1 (Gledhill & Jamnik, 1992a; Sharkey, 1997) and reduce as the water weight is utilised for fire suppression. Gledhill and Jamnik (1992a) reported setting up a portable water pump lasted for 213 seconds, generating aerobic demands of 22.5 ± 3.6 mL·kg·min-1 and a peak heart rate of 172 ± 4 beats·min-1. Miscellaneous bushfire suppression tasks are likely to elicit heart rates and oxygen consumption rates of 134 to 172 beats·min-1 and 22 to 29 mL.kg-1.min-1 respectively. Average male and female aged matched members of the Australian population, will require between 58 – 77% O2max and 72

– 95% O2max respectively to successfully complete knapsack spraying and pump preparation tasks (Gore & Edwards, 1992).

Australian BFFs regularly utilize a number of ‘wet’ or ‘dry’ suppression techniques similarly to SFF and LMFF techniques (Bosua, 2006). There are however distinct differences in the operational terrain the tasks 23 are performed in, the equipment used, operational age of personnel or likely operational performance of the tasks when compared to either SFF or LMFFs (Bosua, 2006). For this reason, the physical demands of these fire fighting occupations, even though they may contain similar work tasks, may not be valid representations of the physical demands of Australian RTBS work.

Physical factors differentiating Australian volunteer fire fighters from land management or SFFs.

RTBS has task similarities with urban SFF and LMFF work. The operational performance of similar tasks in a rural BFF, and therefore the physical demand of these tasks, may however be considerably different due to a number of factors. LMFF is conducted in the same terrain and with similar fire behaviour to RTBS (DSE, 2008). The lower water carrying capacity on a land management primary vehicle necessitates the increased use of dry fire fighting techniques such as rake hoe, back burning and small diameter hose work when compared to BFF in the same situation (DSE, 2008; CFA, 2006). Urban SFF work is conducted in very different terrain but utilise similar wet suppression techniques (Gledhill & Jamnik, 1992a) as rural BFFs within close proximity of a fire tanker. Urban SFFs commonly use larger 64mm diameter hoses on a smooth asphalt surface compared with rural BFFs which use 25mm or 38mm hoses in heavily forested terrain (CFA, 2006). Gledhill & Jamnik (1992a) observed only small differences in the oxygen consumption of SFFs advancing a charged 64mm hose or laterally repositioning a 102mm hose. Richmond et al., (2008) observed no physiological difference in SFFs handling a charged 45mm and 75mm hose in a simulated search and rescue trial but did report fire fighters accepted the smaller diameter hose was easier to handle. The author suggested standard operational procedures allow additional fire fighters to assist in moving the larger diameter and therefore heavier hoses, which would result in little additional physiological load.

Rural bushfire crews are often remotely deployed, unable to access additional man power for performance of long length charged hose work. Working in heavily forested, loose ground terrain may also play a factor in changing the physical work rate of hose work. The additional energy cost of walking in sand has been measured at 2.1 to 2.7 times walking on hard ground (Lejeune et al., 1998). For the purpose of this review, this author will suggest although the common working surface for bushfire suppression is not a soft as sand, the uneven working surface of litter, shrub and log material (Brotherhood et al., 1997) will be significantly softer than urban hard ground and produce a higher drag coefficient during hose work.

Both urban SFFs and remote deployed LMFFs regularly conduct fire suppression work wearing a self contained breathing apparatus (SCBA) or line pack respectively (Sharkey, 1997; Davis et al., 1982; Payne, 2005). A major consideration when comparing both LMFFs and SFFs to BFFs is the frequency of occupational work involving load carriage. Load carriage during occupational work has been shown to increase heart rate, oxygen consumption, blood lactate, ratings of perceived physical exertion and core body temperature (Ricciardi et al., 2008; Beekley et al., 2007; Keren et al., 1981; Skoldstrom, 1987; Hooper

24 et al., 2001) and decrease work tolerance (White & Hodous, 1987). SFF self contained breathing apparatus (SCBA) and protective equipment weighs around 22- 25kg (Hooper et al., 2001; Michaelides et al., 2008; Gledhill & Jamnik, 1992a; Williford et al., 1999; Rhea et al., 2004; Davis et al., 1982). SCBA is worn routinely during operational work to protect urban fire fighters from hazardous gases or other respiratory particulates which could otherwise pose a hazard to the lungs (Musk et al., 1982). Hooper et al., (2001) demonstrated significant physiological advantages of wearing a modern lightweight 10kg SCBA when compared to traditional steel 22kg SCBA in urban SFFs. Fire fighters working with lighter SCBA showed lower heart rates and oxygen consumption at work rates ranging from 300 to 400 kg m.min-1. Richmond et al., (2008) observed significant increases in core temperature and %HRR during operational work in fire fighters wearing a 9kg heavier SCBA. O’Connell et al., (1986) demonstrated similar results in a 5 minute

-1 -1 simulated stair climbing activity by observing O2 increased from 21.6 ± 2.1 to 38.6 ± 2.8 mL.kg .min with the addition of SCBA and protective clothing. This equates to a relative % O2max increase of almost 35% in these fire fighters. SFFs tend to achieve near maximal work rates regardless of load carried during work tasks but wearing SCBA reduces O2max, work tolerance and increases the time to completion for any given work task (Manning & Griggs, 1983; Dreger et al., 2006; White & Hodous, 1987).

25

Table 2.4, Variations in finish time and calculated rates of travel during the Pack and No Pack trials over a course of 660.5m long with a vertical rise of 137m (Ruby et al., 2003). Pack No Pack

Male

Finish time (min) 10.7 ± 1.4 8.4 ± 0.7*

Average pace (m min-1) 62.6 ± 8.1 79.6 ± 6.8*

-1 -1 Average O2 (mL.kg .min ) 41.1 ± 6.0 46.0 ± 6.1*

Average energy expenditure (kJ min-1) 70.4 ± 11.3 77.5 ± 12.1*

Female

Finish time (min) 13.7 ± 1.3 10.1 ± 0.6*

Average pace (m min-1) 48.7 ± 4.4 65.8 ± 4.2*

-1 -1 Average O2 (mL.kg .min ) 32.5 ± 6.6 35.1 ± 4.7

Average energy expenditure (kJ min-1) 44.4 ± 9.6 48.2 ± 7.5

Values are means ± standard deviation; * P<0.05 for No Pack vs. Pack

LMFFs are often deployed to remote areas carrying packs or heavy equipment (Sharkey, 1997). Ruby et al., (2003), Table 2.4, demonstrated commonly used 16kg line packs slowed the hiking speed of LMFFs by 17 meters per minute or up to 26% during a simulated escape from an approaching fire front. The lack of increase in O2 and energy expenditure in females subjects of this study, Table 2.4, may indicate females were working at near maximal rates in both trials. Keren et al., (1981) supported these results by demonstrating load carriage of 20kg increased oxygen consumption from 2.0 ± 0.1 to 2.4 ± 0.1 during 6.4km/hr treadmill walking at 5% gradient. Lyons et al., (2005) observed a significant relationship between

O2max and lean body mass (kg) in the performance of load carriage during 20kg and 40kg treadmill walking trials at 0% and 9% gradients. These relationships were not present at 0% and 9% gradients when no load was carried. It is clear working with SCBA or a line pack significantly increases the physiological load on the body during fire ground work. BFF do not work with SCBA or a line pack. PES developed for SFF and LMFF reflect the increased physiological demands of work tasks wearing load carriage equipment, may overstate the physical demands of similar work without load equipment. Similarly, rural BFFs are not required to perform a range of operational tasks (CFA, 2006) which have shown to be some of the most demanding SFF task; stair climb carrying equipment, forcible entry or ladder climb which generate peak aerobic demands of 44.0 ± 1.5, 30.5 ± 1.7 and 30.6 ± 0.3 mL.kg-1.min-1 respectively (Gledhill & Jamnik, 1992a). Rural bushfire are also not required to perform victim rescue tasks, which have a moderate aerobic demand of 24.6 ± 1.4 mL.kg-1.min-1 but a high individual muscular load of 111lbs (Gledhill & Jamnik, 1992a). PES developed of the

26 peak aerobic demand during stair climb carrying equipment for SFF work may overstate the physical demands of RTBS work performed without these tasks.

Table 2.5: Physical performance scores for male and female incumbent fire fighters (Misner et al., 1989b)

Male Female Percentage Variable (n = 37) (n = 25) diff. between genders

Stair Climb 13.0 ± 1.9 17.6 ± 2.9 26.1 Hose Couple 27.8 ± 5.0 31.9 ± 6.7 13.0 Flexed Arm Hang 65.0 ± 16.9 47.0 ± 20.6 27.7 Lift and Carry 16.9 ± 1.4 21.1 ± 2.5 14.9 Modified Stair Climb 14.0 ± 2.0 23.6 ± 6.9 18.5 Ladder Lift 7.9 ± 2.9 11.9 ± 4.5 40.7 Forcible Entry 9.4 ± 3.4 21.3 ± 13.9 33.6 Dummy Drag 8.8 ± 1.5 11.1 ±1.6 55.9 Obstacle Run 84.0 ± 8.6 103.1 ± 15.9 20.7

Values are means ± standard deviation in seconds; n, number; diff, difference

Gender effects on job performance, or the broader gender-related issue of “adverse impact”, have been a motivator for many judicial rulings in North America (Hogan & Quigley, 1986; Shephard & Bonneau, 2002). Significant differences in body size, injury rate, muscular strength and muscular power have been observed between males and females involved in multiple physically demanding occupations (Knapik, 1997; Misner et al., 1989b; Misner et al., 1989a; Bell et al., 2000; Knapik et al., 2001). Misner et al., (1989b), observed female SFFs performance worse on a range of fire-related tasks commonly used in the CPAT test, particularly on tasks utilising high individual muscular force application such as dummy drag, forcible entry and ladder lift (Gledhill & Jamnik, 1992). Female fire fighters are likely to face higher relative muscular demands for the same tasks in this study as they were, on average, 16kg lighter with 21.4kg less fat-free mass than male subjects (Misner et al., 1987a). Bilzon et al., (2001a) observed 73 ± 10% O2max and 84 ±

10% O2max during the same fire hose work task in male and women naval fire fighters respectively. Only 27% of female fire fighters were able to successfully complete all assigned fire fighting tasks. Gore & Edwards (1992) highlighted the likely difference in aerobic power between Australian males and females, which is likely to be 8, 8.7, 7.4 and 7.2 mL.kg-1.min-1 in the age ranges of 18-29, 30-39, 40-49 and 50-59 years old, respectively. The majority of PES are based on the aerobic power of male workers due to the lack of female subjects in development studies (Bilzon et al., 2001a; Payne, 2005). McLennan (2004) suggested women make up 17% or 35, 704 of the 206, 954 Australian RTBS personnel. 22, 494 women in operational

27 roles. 63% of women rural fire fighters are likely to be engaged in operational roles on the Australian fire ground, a lower portion when compared to 85% of men. The development of valid PES for Australian RTBS work must reflect all personnel performing operational roles, including female workers, as they may encounter gender-related increased work demands during RTBS activities and potentially a higher risk of injury.

Table 2.6, Previous recommendations for Minimum acceptable O2max standards of fire fighting.

- a a c -1 mL.kg %O2max %O2max % O2max Fitness standard F.F. Type L.min 1 -1 .min AUS male AUS female BFF

Kilbom, 1980 SFF 2.8 – 3.0 - 88 – 94% 110-118% 88 – 94%

Louhevaara et al., 1985. SFF 3.5 - 109% 137% 109% Von Heimburg et al., 2006 SFF 4.0 - 125% 157% 125%

Sothmann et al., 1990 SFF - 33.5 88% 110% 88%

O’Connell et al., 1986 SFF - 39.0 103% 128% 103%

Lemon & Hermiston, 1977 SFF - 40.0 106% 131% 106%

Davis et al., 1982 SFF - 42.0 111% 138% 111%

Gledhill & Jamnik, 1992a SFF - 45.0 119% 148% 119%

Dawson et al., 2000 SFF - 45.0 119% 148% 119%

Dreger & Petersen, 2007 Military - 34.1b 90% 112% 90%

Bilzon et al., 2001a Military - 41.0 108% 134% 108%

Deakin et al., 1997 Military - 44.0b 116% 144% 116%

DeLorenzo-Green & LMFF - 45.0 119% 148% 119% Sharkey, 1995

MGMT, management; AUS, Australian; a, % O2max values are based on estimated O2max of the average 40-49 year old Australian male and female (Gore & Edwards, 1992); b, based off successful completion of an 8 minute fire fighting circuit time; c, Pilot testing of 21 FESA rural fire fighters (Phillips et al, 2010).

Large body size has shown to be an advantageous factor in the performance of occupational tasks involving load carriage (Brotherhood et al., 1997; Vanderburgh, 2008; Payne, 2005). Larger individuals may face a smaller relative work rate due to higher musculature and longer biomechanical levers to perform the same absolute work. Brotherhood et al., (1997) suggested self pacing played a significant role in the physiological response of LMFFs performing prolonged rake hoe work. In trials where fire fighters were able to self pace their effort, fire fighters with higher fat mass had higher productivity in contrast to the 28 predicted productivity from a stepping O2max assessment. Brotherhood et al., (1997b) suggested fixed paced, work against steep terrain, emergency or other circumstances requiring a sustained near maximal effort would promote adverse effects in higher fat mass fire fighters. From the figures of three large rural fire agencies, McLennan (2004) estimated as much as 55% of volunteer BFF personnel are older than 40 years of age. The body mass index (BMI) from aged matched members of the Australian population would suggest Australian rural fire fighters may have overweight average BMIs of 25.9 ± 3.5, 26.2 ± 3.1 and 26.2 ±

2.9 for ages 40-49, 50-59 and 60-69 respectively (Gore & Edwards, 1992). The O2max of aged matched members of the Australian population would also suggest Australian rural fire fighters are likely to have average aerobic capacities of 37.9 ± 8.5, 33.6 ± 5.8 and 31.1 ± 9.6 for the same ages 40-49, 50-59 and 60-69 respectively (Gore & Edwards, 1992).

Lusa et al., (1994) observed frequency of physically demanding tasks is not dependent of age, suggesting fire fighters of all ages are exposed equally to physically demanding work on a fire ground. It is therefore reasonable to assume Australian rural fire fighters will be exposed to operational demands regardless of age. Assuming the average Australian BFF volunteer is similar in aerobic power to the Australian population, Table 2.6, suggests SFF and LMFF PES would elicit aerobic work rates of between 88 and 119% O2max. It is likely a substantial number of fire fighters, especially older or female fire fighters, would not pass the physical requirements of this PES. PES applied to BFF will unfairly discriminate against higher fat mass and older fire fighters if the PES over-estimates the demands of their work. It is important to develop valid PES for RTBS retains and utilises as much of the capable existing workforce as possible.

Developing a PES for Australian RTBS

The first parts of this review have focused on the legislative considerations; practical justifications and the potential physical demands of RTBS. The remainder of this review will focus on the process and potential methodological advancements in development of valid PES for Australian fire fighters. The primary goal of a PES is to establish valid, reliable and predictable links between job performance of an occupation and the procedures used to select or assess workers within that occupation (Rayson, 2000c; Jackson, 1994; Hogan & Quigley, 1986; Kuruganti& Rickards, 2004; Jamnik et al., 2010a; Jamnik et al., 2010b; Jamnik et al., 2010c). As discussed earlier in the review, the acceptable strength of these links is often defined by legislative motivators within employment discrimination or OHS law (Rorke, 2002). As there is great variation between job descriptions and factors affecting optimal job performance in different occupational settings, valid PES are likely to depend on the characteristics of the occupation being studied. Scientific approaches to the development of PES for fire fighting populations have focused on a ‘best practice’ approach outlined in Figure 2.1 (compiled from Shephard, 1991; Jackson, 1994; Taylor & Groeller, 2003; Kuruganti& Rickards, 2004; Gledhill-Shaw, 2003).

29

Figure 2.1, Current pathways to the development of PES (adapted from Payne & Harvey, 2009; Rayson, 2000c)

The remainder of this review will focus on methodological issues involved with the first step of the process outlined in Figure 2.1, Job task analysis. A valid and comprehensive task analysis, detailing the sub text outlined in Figure 2.1, will serve as a strong foundation for ongoing development of a PES in rural fire fighters but discussion of these elements beyond step one in Figure 2.1 is outside the scope of this review.

Job task analysis

A job task analysis is considered a content valid method of determining the range of job specific tasks conducted within an occupation or role within an occupation and a logical starting point for the process of PES development (Kuruganti & Rickards, 2004; Taylor & Groeller, 2003; Rayson, 2000c; Rayson, 2000b; Jackson, 1994; Jamnik et al., 2010a; Jamnik et al., 2010b). A job task analysis attempts to systematically and comprehensively collate information defining a broad range of work behaviours involved in a job, types of job skills or characteristics required by an employee to successfully complete a job or an individual component of a job (Rayson, 2000c; EEOC, 1978). The overall focus of a job analysis is to identify 30 important or critical measures of job performance (EEOC, 1978; Richman & Quinones, 1996; Fleishman, 1988; Rayson, 2000c; Sanchez & Levine, 1989; Jamnik et al., 2010a; Jamnik et al., 2010b) and tasks which involve specialist knowledge, skill, ability or other worker characteristics (KSOA) (Maurer & Tross, 2000; Morgeson et al., 2004). Critical tasks are usually isolated using a balance between relative importance and frequency of occurrence (Sanchez & Fraser, 1992; Sanchez & Levine, 1989; Rayson, 2000c; Taylor & Groeller, 2003). Critical tasks are then used as criterion tasks which predict the physical capability required for successful performance of a physically demanding occupation (Sothmann et al., 2004; Arvey et al., 1992; Nottrodt & Celentano 1987; Rayson, 1998; Gledhill & Jamnik, 1992b; Jamnik et al., 2010b).

A job inventory and psychophysical job analysis is a valuable method of determining critical physical tasks for any occupation (Rayson, 2000c; Jackson, 1994). A job task analysis leading toward the development of PES starts by creating a job inventory to define the job domain (Kuruganti & Rickards, 2004; Taylor & Groeller, 2003; Jamnik et al., 2010a; Jamnik et al., 2010b). A job inventory contains a list of tasks or work which a particular role or trade within an occupation is likely to encounter. Judicial rulings in Australia have suggested this list should be comprehensive of all reasonable circumstances and not limited to average daily operating procedures. Sources of information which inform a job inventory for emergency services are varied and include training manuals, interviews with training officers or field operations managers, annual reports, incident summary reports and current assessment procedures (Rayson, 2000c).

Hughes et al. (1989) suggests the most accurate and valid job task analysis is conducted through direct, real time observation of physical demands and work practices. Real time observation is difficult in a rural fire fighting environment where incumbent workers are spread across large operational areas and may engage in BFF activities infrequently. Rayson (2000c) suggested these techniques are particularly valuable for occupations with short repetitive work cycles but may require extended observation in unstructured occupations. Direct and detailed measurements of incumbents’ work patterns and physical demands, are not always feasible due to significant workplace risk, logistical or operational considerations, often facilitating the need to develop alternative methodology (Wilson, 1997). In work settings, where direct observation is not practical, such as the military and emergency services, job tasks can be evaluated by experienced incumbents or subject matter experts (SME) (Viswesvaran et al., 1996; Richman & Quinones, 1996; Landy & Vasey, 1991). Qualitative data from SME can subjectively assess frequency, criticality, difficulty and relative importance of all tasks within an occupation. Rayson (1998) and Sharp et al., (1980) highlighted SME data represents experienced opinion rather than observed practice.

Methodology use to access the knowledge of a SME sample may affect the validity of job performance ratings. Many studies have utilised surveys of large numbers of individual incumbents or supervisor SME to obtain job inventories and job analysis ratings (Rayson, 1998; Lusa et al., 1994; Reilly et al., 2006; Mueller & Belcher, 2000; Davis et al., 1982, Kaczmarek & Packer, 1996). Others have focused on

31 isolating smaller sample SME committees or a number of committees to act as representative opinion of the occupation (Deakin et al., 1999; Payne, 2005). Maurer & Tross (2000) observed likely agreement, or convergence, in critical job task and KSOA assessment between the collective opinions of selected SME committees and larger surveys of individual SME groups for “Relative time spent”, “Importance” and “distinguishes performance levels”. The same authors cautioned against a SME committee approach if there was potential bias in the political or social perception of the job analysis process between groups. It is feasible too many individuals with shared experiences on a SME committee could potentially bias the collective experience if not appropriately counteracted with an appropriate sample size.

Table 2.7, Interrater reliabilities and intrarater coefficient alpha reliabilities of overall job performance rating between peer (incumbent) and supervisors (adapted from Viswesvaran et al., 1996)

N Mwt SD Munwt SD 80% CI 80% Cred

Interrater

Supervisor 14 650 0.52 0.10 0.68 0.15 0.50– 0.54 0.41 – 0.63

Incumbent 2 389 0.42 0.11 0.44 0.16 0.37 – 0.47 0.30 – 0.54

Intrarater

Supervisor 17 899 0.86 0.14 0.84 0.15 0.84 – 0.88 -

Incumbent 1 270 0.85 0.12 0.81 0.12 0.80 – 0.90 -

N = sample size of group; wt = sample size weighted; unwt = unweighted or frequency weighted; CI = confidence interval; Cred = credibility interval.

The validity of SME assessments which define a job domain or critical components of real occupations are fundamentally linked to; task engagement, variation in ratings from supervisor or incumbent sources; interrater reliability, the capacity to obtain the same job performance ratings obtained on the same employee by different SME; and intrarater reliability, the capacity to obtain the same job performance ratings on the same employee by different equally qualified SME (Pine, 1995; Viswesvaran et al., 1996; Richman & Quinones, 1996; Wilson, 1997; Green & Veres, 1990). Morgeson & Campion (1997) opined intrarater disagreement will represent either genuine individual variation of SME occupational experience or be indicative of an inaccurate job analysis. Mueller & Belcher (2000) observed similar relative task criticality ratings between fire captains and supervisors of 5.1 ± 1.7 and 5.8 ± 1.5 respectively. Viswesvaran et al., (1996) demonstrated reasonably similar overall job performance ratings between experienced incumbents and supervisors for coefficient alpha intrarater and interrater reliabilities, Table 2.7, indicating both can be used in a task analysis with similar reliability. Not surprisingly intrarater, rather than interrator ratings of job performance were more reliable reflecting 80% confidence intervals of 0.8 – 0.9 and 0.5 – 0.5 or 0.4-0.5 respectively. These results suggest it is reasonable to assume 10 – 20% and 45 – 63% of the observed variance in responses of SME used for a job task analysis may be attributed to 32 intrarator or interrator differences of opinion respectively, depending on the methodology. These differences may not be entirely due just to differences in rater opinion or reliability but also false reporting. Pine (1995) identified 45% of incumbent correctional officers falsely reported many tasks with are important for their job performance. False reporting to this magnitude would contribute significantly to the variance in interrator reliability.

Figure 2.2, Detection accuracy of low Vs high experienced incumbents (from Richman & Quinones, 1996)

Experience and operational context also plays roles in accuracy of a job task analysis. Richman & Quinones (1996), Figure 2.2, analysed the accuracy of task frequency ratings across two domains; task engagement and rater experience. Although the participants were not occupational workers and the findings may be of limited questionable application to a real workplace, the authors observed low experience performers and supervisors of basic building tasks were more accurate at determining frequency for these tasks than higher experience counterparts. This finding requires further research but is in opposition to the historical view of task analysis which encourages the use of experienced incumbents or supervisors with an in-depth understanding or familiarity with the physical work demands (Hahn & Dipboye, 1988; Gledhill & Jamnik, 1992a; Rayson, 2000b; Rayson, 2000c).

An inaccurate job analysis may be caused by the pressures of increased recruiting costs, work compensation costs, a poor understanding of training processes or lack of resources (Mueller & Belcher, 2000; Morgeson & Campion, 1997). Morgeson & Campion (1997) opined “task based surveys, coupled with relative-time-spent ratings, appear to maximise within-job differences” in contrast which construct rating measures (e.g. difficulty to successful completion, relative importance) which “typically find fewer differences”. Lindell et al., (1998) found high interrater agreement for ratings of relative importance in emergency management SME, but not ratings of time spent performing each task. The authors suggest a high level of genuine within-job variability, or a systemic context-related error in the task analysis. Job 33 context was significant correlated with ratings of time spent performing each task but not ratings of relative importance, indicating operational difference could be due to genuinely different SME job experience of tasks. Collectively these studies raise many potential questions regarding optimum selection of SME or the validity of SME opinion used in PES development to define critical job tasks. It is clear current PES development processes may not fully comprehend interactions within or between SME raters affect the validity of a job task analysis such as; task engagement, reliability, experience and job efficiency. To the authors’ knowledge, few published investigations have been made to investigate the accuracy of SME derived job task analysis against actual emergency operational deployment. This thesis will attempt to index actual emergency rural bushfire work against heavily monitored simulated work shifts in order to ascertain the accuracy of SME ratings for development of PES for Australian RTBS work. Through this work detailed measurements of operational task frequency and the determination of operational variation in tasks or activities should be attainable.

Limitations of physical demands analysis in the PES model for rural bushfire fighting

As discussed in earlier sections of this review, a job task analysis aims to define the range of tasks encountered during operational work and isolate which of these tasks are “critical” for successful job performance. A physical demands analysis completes a task analysis by characterizing the physical stress and strain encountered during performance of critical tasks or important work (Rayson, 2000c; Taylor & Groeller, 2003). This process is a vital step in matching the physical demands of the job to the required physical capabilities of a worker (Payne & Harvey, 2010; Jackson, 1994; Rayson, 2000b). Ideally all job task analysis, including physical demands analysis is conducted with observation during actual work activities (Rayson, 2000c; Hughes et al., 1989). As discussed earlier in this review, the increased workplace risk, logistical or operational considerations often necessitate physical measurement in a more controlled, safe setting. Sensitive scientific equipment used to characterize physical demand of work is expensive, can limit incumbent performance (mass, bulk, range of motion) and often is not designed for a wide range of uses within highly variable environments (Rayson, 2000c). Consequently, physical demands analysis often occurs in a simulated environment (Rayson, 1998; Reilly et al., 2006; Bilzon et al., 2001a; Deakin et al., 1997).

Shephard (1991) opined minimum PES for police recruits, arbitrarily set at 60% of population norms, would have little chance of defending their validity in court. The same author also suggested a common approach to the establishment of an age- accommodating PES is to “set standards so that the average 45-year old-employee is working at 80% of an acceptable loading”. This approach expected an employee of average fitness should be able to physically perform the job up to the retirement age of 65. It is difficult to justify this approach to PES development both legislatively and scientifically. From a scientific standpoint this approach makes a large effort to accommodate older workers but little effort to directly relate a PES to acceptable job performance. It is likely a PES developed using this methodology would result 34 in high rates of false negatives in younger workers, and therefore violate discrimination law. These workers may become true negatives with age-related fitness declines before the end of their career (Sluiter, 2006; Chan & Koh, 2000; Sluiter & Frings-Dresen, 2007; Sothmann et al., 1990; Lusa et al., 1994) but many may be capable of acceptably meeting the physical work demands at a younger age and it would be discriminatory practice to terminate their employment until they fail validated annual fitness tests.

The pack hike is a loaded march PES which replicates a work rate of 45 mL.kg-1.min-1 or twice the average 22.5 mL.kg-1.min-1 cost of LMFF duties (DeLorenzo-Green & Sharkey, 1995; Sharkey, 1997; Sharkey, Rothwell & DeLorenzo-Green, 1999). Sharkey (1999) reasoned a PES, which generates a prolonged “whole- shift” standard, must be established at this level as “workers cannot sustain day-long work rates above 50% of their Maximal capacity”. This reasoning does however make assumptions about the consistent nature of work across a full shift. Firstly, the assumption is made were workers will need to sustain prolonged aerobic work at 50% O2max consistently across a full work shift. The most demanding individual physical work bouts encountered by LMFFs are likely to elicit between 22.5 and 27.2 ± 3.8 mL.kg-1.min-1, or up to 57.6 ±

7.1 5 % O2max (Budd et al., 1997c; Brotherhood et al., 1997a; DeLorenzo-Green & Sharkey, 1995; Sharkey, 1997). Although this level of expenditure is possible throughout a shift, as discussed in early sections of this review, indications from Cuddy et al., (2007) are 66% of fire ground time is spent in sedentary activity intensities, likely well below 50 % O2max.

Shephard & Bonneau (2002) highlighted the commonly accepted minimum aerobic PES for public safety officers would allow for 8 hours of work at twice the standards operating level. The authors opined this methodology bears little relevance to an occupation composed of short duration “critical incidents” of 1-5 minutes. It is likely the majority of RTBS work is made up of distinct ‘critical’ work bouts rather than prolonged whole shift work at any given metabolic rate (Bos et al., 2004; Cuddy et al., 2007). A PES intended to allow workers to work below 50 % O2max throughout a shift may be of questionable content validity for Australian RTBS work, particularly if work demands are predominantly sub maximal and maximal levels occur infrequently. Traditionally, Douglas bags are used to monitor oxygen consumption over shorter task orientated work (Gledhill & Jamnik, 1992a) or to sample steady state oxygen consumption at distinct points during work (Budd et al., 1997c; Reilly et al., 2006). Portable metabolic analysers can be used to continuously monitor oxygen consumption during actual work (Nag et al., 1980), simulated work trials (Ruby et al., 2003; Bilzon et al., 2001a; Holmér & Gavhed, 2006; Dreger & Petersen, 2007; Louhevaara et al., 1994; Deakin et al., 1999) and through interface with SCBA (Sothmann et al., 1990). The simulated model used by Gledhill & Jamnik (1992b), to characterise individual task demands including; duration, heart rate, oxygen consumption and load, is likely to provide good accuracy in the characterisation of physical demand for intermittent fire fighting activities. The use of portable metabolic analysers or Douglas bags for similar research in real life situations is problematic due to technical or oxygen storage difficulties respectively (Rayson, 2000c; Holmér & Gavhed, 2006). Without clearly defined means of measuring job 35 performance, the content validity of a standard developed using a “whole-shift” 50% O2max approach indexed to incumbent characteristics could be problematic to objectively assess against job performance and is essentially arbitrary. A more accurate approach for Australian RTBS work. Figure 2.3, may be to define a work standard directly of the most physically demanding components of the occupation.

Figure 2.3, Novel potential additions to the job task analysis process (adapted from Payne & Harvey, 2010; Rayson, 2000c; Jamnik et al., 2010a)

The advance of computer technology has seen a variety of lightweight and non invasive data collection devices become available for use in the area of occupational physiology. In particular, portable metabolic systems (Ruby et al., 2003; Bilzon et al., 2001a; Holmér & Gavhed, 2006; Dreger & Petersen, 2007; Louhevaara et al., 1994; Deakin et al., 1999), global positioning system (GPS) technology (Larsson & Henriksson-Larsen, 2001; Edgecomb & Norton, 2006; Larsson, 2003; Demczuk, 1998), radioactive isotopes (Ruby et al., 2002; Ruby et al., 2003), wireless transmission/ temperature monitoring systems (Richmond et al., 2008) and computer analysis programs (Anton et al., 2003) have been used to collect detailed

36 physiological data were previously impossible. The remote logging capabilities of many of these devices enable data to be collected without the presence of a researcher, which is a safer for research teams who previously had to collect data in emergency situations. Average total energy expenditure rates of 4160 ± 884 and 4768 ± 478 kcal day-1 and activity energy expenditure of 2101 ± 716 and 2585 ± 406 kcal day-1 have been observed in American WFFs using doubly labelled water and physical activity monitor techniques respectively (Ruby et al., 2002; Heil, 2002). Ruby et al., (2002) suggested doubly labelled water technique requires a measurement period of between four and sixteen days in order to provide optimally accurate results.

It is unlikely Australian RTBS personnel are likely to be deployed for longer than three days (Bosua, 2006) and as such, the use of doubly labelled water is not useful practical. Heil (2002) proposed a cheaper, viable alternative in the use of physical activity monitors, which he showed produced comparable estimates of daily energy expenditure in American WFFs with the capability to sample work rate each minute for 21 days. Activity monitors may be of more use for measurement of Australian RTBS work due to the higher sensitivity to account for the diversity of work tasks across a smaller timeframe. It is fair to suggest the accuracy of physical activity monitors to predict energy expenditure has not been calibrated against portable metabolic devices in complex work environments (Trost et al., 2005; Plasqui & Westerterp, 2007). Prediction of energy expenditure for a single shift or short duration deployments is most likely to be problematic, inaccurate and outside the scope this review. Trost et al., (2005) suggested a low within-and between unit variations within activity monitors across longer durations, although it appears higher variation occur with very short durations (Welk et al., 2004). If the output is similar for the same work with this technique, activity monitors may be more useful as a means to compare the range of activity within and between individual work shifts rather than describe overall energy expenditure rates. This thesis will instead attempt to analyse live fire baseline measures of heart rate, activity count and GPS movement as a basis for comparison with single day “campaign simulated” trial tasks to determine the accuracy of rural bushfire simulations for more detailed analysis.

Summary

Australian fire agencies are required to identify and develop strategies to minimise work related risks to the health of their workers. The physical demands of Australian rural bushfire suppression are currently unknown but are likely to be physically demanding. No Australian rural bushfire agency has conducted a peer reviewed or published psychophysical job analysis to determine the critical tasks of RTBS. Pre-existing tests of fire fighting performance, currently used in Australian fire fighting populations, are unlikely to be valid for use with this work. A PES should be developed and validated for Australian RTBS work for this reason.

37

The scientific mechanisms currently used to determine duration on a PES may not be sufficiently accurate for developing valid PES in occupations characterised by intermittent physically demanding work. It is clear duration, either of tasks or of work bouts, has been undervalued as an important content valid component of job performance. Particularly in an operational setting, when it determines the performance of a worker or success of the work activity. There is likely an inverse relationship between success of fire suppression work and duration of fire fighting bouts during all fire fighting, i.e. any given fire will be more difficult to extinguish given a longer duration to burn and grow before suppression efforts commence. The test duration enforced on a PES is intangibly connected to the acceptable operational practice, in much the same way as specific metabolic or musculoskeletal loads reflect operator physical demands of content valid PES. This thesis will attempt to develop methodology, Figure 2.3, incorporating task duration, frequency and work bout composition in the development of a PES for Australian RTBS work.

38

CHAPTER 3: Job Task Analysis of Crew Member Duties during Australian Tanker-Based Bushfire Suppression Work.

INTRODUCTION

Within the last decade, large scale wildfires (called bushfires in Australasia) have impacted significantly on communities in Australia, Southern Europe and North America (Hunter, 2004; Hyde et al., 2008; Schmuck et al., 2004). Fire tankers are commonly used by fire agencies in these regions as a bushfire suppression tool. Operational techniques of rural tanker based bushfire suppression (RTBS) work (Wildfire firefighter learning manual, 2006) involve the combination of traditional structural fire fighting (SFF) tasks (Gledhill & Jamnik, 1992a; Sothmann et al., 2004) in an environment similar to North American wildland fire fighting (WFF) (Sharkey, 1997). SFF and North American WFF work has long been identified as physically demanding (Bilzon et al., 2001a; Gledhill & Jamnik, 1992a, Ruby et al., 2002; Brotherhood et al., 1997) but the physical demand of RTBS work, a hybrid of other fire fighting work styles, is yet to be investigated. The characterization of RTBS work is particularly important in regions such as Australia, where rural fire agencies, comprised of up to 207, 000 volunteer fire fighters (McLellan, 2004), perform this suppression work in parallel with rake crews at major bushfire incidents. The workforce using RTBS techniques may be up to two and three quarters the personnel who use more traditional North American wild land suppression techniques (Hunter, 2004). As RTBS work is potentially composed of many elements of other fire fighting styles, it would be reasonable to expect this type of work will be of similar physical demand.

A job inventory and psychophysical job analysis is a valuable method of determining critical physical tasks for any occupation (Rayson, 2000c; Wilson & Corlett, 1990; Jackson, 1994; Jamnik et al., 2010a; Jamnik et al., 2010b). Detailed job task analysis of the general operational duties within physically demanding occupations has facilitated the development of physical employment standards (PES) throughout the world. Qualitative data from experienced job incumbents or subject matter experts (SME) assessing task frequency, criticality, difficulty and important can be used to demonstrate critical tasks (Sanchez, 1989) as a precursor to the development of valid PES (Rayson, 2000b). Task analysis focuses on identifying cognitive and decision making factors which can be used for psychological screening prior to employment (Kaczmarek & Packer, 1996) or distinct content related criterion tasks which comprise the critical spectrum of physical activities required for successful occupational performance and corresponding content valid job PES (Sothmann et al., 2004; Arvey et al., 1992; Nottrodt & Celentano 1987; Rayson, 1998; Gledhill & Jamnik, 1992b).

Criterion PES are a fundamental focus of job selection for physically demanding occupations (Sothmann et al., 2004; Arvey et al., 1992; Jamnik et al., 2010a; Jamnik et al., 2010b). Australian employers

39 face similar issues to the rest of the world but have less defined legislative boundaries (Uniform Guidelines on Employee Selection Procedures, 1978). The Victorian Equal Opportunity Act (1995) and OHS Act (2004) suggest employers are able to “discriminate against another person, employee or volunteer worker, on the basis of impairment or physical features if the discrimination is reasonable necessary to protect the health or safety” of and to set genuine occupational requirements for employees. Genuine occupational requirements are generally considered to be an exception from state anti-discrimination legislation and employers are capable of setting PES determined ‘reasonably practicable’ to fulfil the physical requirements of the occupation or ‘reasonably necessary’ to protect the health of a person on a work site.

No fire agency has ever conducted a published job task analysis to determine the critical tasks for job performance in RTBS. As such, the vast majority of rural fire agencies enforce no annual or recruitment PES upon entry into RTBS crews. Validated PES for safe RTBS work cannot be developed until there is greater understanding of the physical demands of RTBS work. The aim of the present study was, therefore, to conduct a job task analysis on crew member duties experienced during RTBS work. From this analysis, we aimed to determine a critical list of job tasks and characterize the RTBS work for further physiological analysis.

METHODS

Subjects. Thirty one (31M) Australian volunteer BFF (50.1 ± 11.6 yrs of age, 20.9 ± 12.4 yrs of fire agency membership) participated in this study. All participants were operationally accredited for RTBS work, volunteer members of the Country Fire Authority (CFA), Victoria, Australia and had participated in at least two major multi-day fires within the last ten years. All participants had attended a variety of multi-day bushfires within the last ten years and were qualified in operational fire ground leadership roles above the level of crew member. Ethical and technical approval for the study was obtained from the Deakin University Human Research Ethics Committee and the Australasian Fire Authorities Council respectively.

Experimental Protocol: Six discussion sessions were run throughout Victoria. The locations were selected in regions of proximity to multi-day bushfire incidences which have occurred within Victoria over the last five years. Metropolitan sites were selected as personnel are frequently deployed from these areas during prolonged multi-day bushfire suppression. Participants in this study represented eight out of twenty regions throughout Victoria. Sessions were held within fire agencies premises including regional training colleges, brigades and state headquarters. Five sessions (23 men) completed the full task analysis and one session (8 men) did not evaluate the activity type, activity category or difficulty parts of the task analysis due to unforeseen time constraints.

A job inventory list of 51 core fire tasks was developed over two rounds of a consultation phase with two senior operational RTBS personnel and three training officers at the CFA training college. These personnel were only used in the consultation phase and new participants were recruited from the discussion session 40 phase. The inventory list was amended to incorporate feedback during each phase and aimed to comprehensively reflect all physical RTBS tasks which could realistically be performed by crew members deployed to a multi-day bushfire. Participants at each discussion session were invited to add to the task list if they believed additional tasks were required for evaluation. Four tasks were added to the task list during the first session and each subsequent group accepted the revised task list of 55 tasks (Table A.1) accurately depicted activities encountered during bushfire suppression.

Each discussion session required participants to collectively define a variety of variables for each task (Table A.2). Collective opinions were recorded on a master sheet kept by the same experimenter at every session. During analysis, each session was weighted by participant number when combining overall group means on each variable. Collective opinion was defined as agreement of more than 74 percent of participants on any variable (Table A.1). This value was attained by dividing the number of subjects required to agree in four out of the six discussion sessions by the total amount of subjects (23 out of 31 participants). As one session of 8 participants did not complete the activity type or activity category assessment, collective opinion in these areas of 65 percent was determined from a majority score in 3 out of five sessions (15 out of 23 participants). If variables did not exceed these cut off scores, no variable was identified for that task.

Activity categories of lift, carry and climb were adapted from Rayson (1998). Task selection, operational importance, and perceived level of difficulty were adapted from Sanchez & Levine (1989). Additional activity categories were included to reflect tasks included in SFF or WFF job tasks analysis, including dig/rake, loaded march, suppress, urban/structure encroachment and tanker mounted suppression (Gledhill & Jamnik, 1992a; Sharkey, 1997). Action categories of suppress, static hold, drag, and assemble were added to the methodology to describe the unique demands of this style of fire fighting.

Participants from three sessions within this study (N =15) were also asked about the number of hours in a shift, the amount of shifts per multi-day fire deployment and the amount of multi-day fire deployments per year to convert all task durations to a per season score. Seasonal frequency of each task was calculated from the participant agreed rate of two multi-day fires per season, 2.5 shifts per multi-day fire and 14 hours per shift.

Statistical Analysis: Data are expressed as mean ± standard deviation, unless otherwise mentioned. Raw data from individual session was combined into overall data using weighted averages which reflected the number of participants in each session.

RESULTS

Characterisation of Tasks

41

The duration of more physically demanding RTBS tasks ranged from 5.2 ± 4.2 to 21.9 ± 7.6 minutes (Table 3.1; Table 3.2; Table 3.3; Table 3.4; Table 3.5). Forty three per cent of tasks showed a perceived load per person in excess of 10 kg per operator (Table 3.1; Table 3.2; Table 3.3; Table 3.4; Table 3.5). The heaviest perceived weights involved three hose work tasks which were thought to generate between 26.6 ± 8.5 and 33.1 ± 11.9 kg per operator involved in the task (Table 3.1; Table 3.5). All tasks were carried out over distances ranging between 32.3 ± 30.8 and 95.8 ± 104.2m (Table 3.1; Table 3.2; Table 3.3; Table 3.4; Table 3.5). Carry work was identified in all seven physically demanding tasks and 40% of all 55 tasks (Figure3.5). The drag action featured in all hose work tasks, 43% of physically demanding tasks and 24% of all tasks (Figure3.5). When expressed by task frequency, drag and carry actions overlap in 90% physically demanding tasks completed per season. Dig/ rake action, characterizing all rake work, was present in 57% of physically demanding tasks (Table 3.2; Table 3.6) but represented only ten percent of physically demanding tasks completed per season (Figure 3.2; Figure 3.3; Figure 3.4), suggesting it is a small contributor to regular work performance over a season of RTBS work. Strength endurance activity type, the only activity type identified and representing 98% of the frequency of all physically demanding tasks (Table 3.6), was acknowledged in charged hose advance, lateral repositioning, full repositioning work and rake work during blacking out (Figure3.7). Of all fifty five tasks, 69% of these tasks had no clear activity type (Figure3.6). The largest agreed activity type was strength-endurance which was acknowledged in 16% of all tasks.

Physically demanding tasks represented 42% of the frequency of work tasks on the fire ground (Table 3.5; Figure 3.2; Figure 3.3; Figure 3.4). Tasks completed during RTBS work are characterized by carry/drag work and carry/dig/rake work in 90% and ten percent of physically demanding work respectively. Individual tasks last for between 5 and 20 minutes, requiring the operator to move loads between 4kg and 33 kg (Table 3.1; Table 3.2; Table 3.3; Table 3.4; Table 3.5). Operational working distances do not typically exceed 100m from the fire tanker. Physically demanding work is generally considered to involve a moderate or high difficulty of successful task completion. 98% of physically demanding RTBS work has a strength endurance activity type (Figure 3.6).

42

80

60

40

Mean Age of 20 Participants (yrs) Participants

0 1 2 3 4 5 6 Session

40 *

30

20

experience (yrs) 10 Overall Fire Industry

0 1 2 3 4 5 6

Session

Figure 3.1, Overall years of fire industry experience across six experimental conditions.

All data are means ± standard error of the mean. Yrs, years; session, discussion group for task analysis experts; sec, seconds; * P<0.05 4 vs 5 & 6.

43

Table 3.1, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as hose work. Relative Frequency Duration Load Distance Task Difficulty Importance (per season) (min) (kg) (m)

Uncharged 38mm hose advance onto fire break 3.8 ± 0.9 34.5 ± 47.7 1.4 ± 0.5 2.4 ± 1.5 8.2 ± 1.8 72.6 ± 15.1

Uncharged 38mm hose advance into terrain 4.0 ± 0.6 48.2 ± 72.4 1. 8 ± 0.4 4.0 ± 1.7 9.0 ± 1.2 72.6 ± 15.1

Lateral charged repositioning of 38mm hose 4.3 ± 0.7 523.9 ± 168.5 1.9 ± 0.6 5.2 ± 4.2 26.6 ± 8.5 53.2 ± 35.3

Full charged repositioning of 38mm hose 4.2 ± 0.7 269.0 ± 197.2 2.1 ± 0.3 9.1 ± 5.1 26.6 ± 8.5 94.8 ± 30.1

Operating 25mm rubber delivery hose 3.8 ± 0.7 93.7 ± 108.9 2.0 ± 0.0 7.6 ± 4.5 16.9 ± 4.0 25.8 ± 4.1

Hose work during blacking out activity 3.7 ± 0.8 70.3 ± 51.8 1.9 ± 0.3 25.2 ± 5.8 20.5 ± 10.1 89.0 ± 95.0

Charged 38mm hose advance 4.9 ± 0.4 23.4 ± 39.1 2.8 ± 0.4 6.6 ± 5.1 33.0 ± 11.9 55.2 ± 11.2

Pump operation on tanker 4.8 ± 0.4 79.4 ± 60.0 1.2 ± 0.4 31.3 ± 18.2 N/A 52.5 ± 50.1

Manual hose retraction 2.4 ± 0.5 12.4 ± 6.7 1.5 ± 0.7 2.9 ± 1.0 9.5 ± 4.6 46.7 ± 15.3

Hose bowling 3.0 ± 0.8 29.2 ± 49.7 1.0 ± 0.0 1.3 ± 1.7 10.2 ± 3.2 46.0 ± 57.5

Hose making up on the bite 3.4 ± 0.8 34.4 ± 47.9 1.0 ± 0.0 5.4 ± 6.6 10.2 ± 4.7 30.0

44

Table 3.2, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as rake work.

Relative Frequency Duration Load Distance Task Difficulty Importance (per season) (min) (kg) (m)

Solo rake work 3.4 ± 0.7 4.2 ± 3.4 1.9 ± 0.6 20.0 ± 8.6 4.1 ± 0.9 32.3 ± 30.8

Rapid rake work during spot fire containment 4.1 ± 0.6 4.2 ± 3.7 2.8 ± 0.4 9.8 ± 9.2 4.1 ± 0.9 58.2 ± 66.5

Rake during team line building 3.9 ± 0.8 7.8 ± 16.6 2.4 ± 0.5 20.5 ± 7.4 4.9 ± 2.4 83.2 ± 59.9

Chainsaw use in rakehoe crew 3.6 ± 0.8 7.8 ± 16.6 2.4 ± 0.5 16.9 ± 6.7 14.4 ± 6.2 170.4 ± 176.3

Chain saw use for vehicle access 4.4 ± 0.5 3.0 ± 3.1 2.2 ± 0.4 18.4 ± 7.7 11.9 ± 4.0 10.0 ± 0.0

Using an axe for vehicle access 3.2 ± 0.70 2.1 ± 3.6 2.1 ± 0.3 8.9 ± 5.3 3.8 ± 2.2 10.0 ± 0.0

Rake work during blacking out 3.1 ± 0.3 70.2 ± 54.9 1.9 ± 0.6 21.9 ± 7.6 4.1 ± 0.9 95.8 ± 104.2

Patrolling on foot whilst carrying hand tool 2.8 ± 0.6 43.7 ± 48.1 1.70 ± 0.70 39.4 ± 21.0 7.0 ± 8.1 748.4 ± 648.5

45

Table 3.3, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as miscellaneous work.

Relative Frequency Duration Load Distance Task Difficulty Importance (per season) (min) (kg) (m)

Preparation of individual equipment 4.8 ± 0.4 3.3 ± 1.6 1.0 ± 0.0 17.3 ± 8.5 11.8 ± 2.9 N/A Individually climb and dismount tanker 4.8 ± 0.4 433.6 ± 167.0 1.4 ± 0.5 0.3 ± 0.3 N/A 1.70 ± 0.0 Mobile patrolling: member of a tanker crew 2.9 ± 1.0 63.6 ± 99.1 1.0 ± 0.0 106.0 ± 112.1 N/A 6357.1 ± 6544.5 Mobile patrolling in a support vehicle 3.9 ± 0.8 16.4 ± 7.2 1.0 ± 0.0 135.1 ± 111.7 N/A 32609 ± 36633 Knapsack hiking 2.8 ± 0.8 0.0 ± 0.0 2.4 ± 0.5 29.8 ± 22.4 16.1 ± 3.7 395.7 ± 361.2 Knapsack spraying 3.1 ± 0.8 0.0 ± 0.0 2.4 ± 0.5 26.7 ± 20.7 14.0 ± 1.5 378.3 ± 377.3 Quick fill pump carry 3.0 ± 1.1 5.0 ± 2.9 1.1 ± 0.8 4.8 ± 2.8 18.3 ± 5.4 26.5 ± 18.1 Generator carry 2.5 ± 0.9 2.0 ± 2.0 0.9 ± 0.9 4.5 ± 3.4 18.3 ± 5.1 28.1 ± 18.6 Trailer mounted quick fill pump set up 2.9 ± 0.2 3.3 ± 1.8 1.7 ± 0.5 10.3 ± 2.6 16.0 ± 5.7 8.6 ± 2.3 Draughting set up 4.0 ± 0.0 5.8 ± 4.7 1.6 ± 0.5 8.1 ± 2.3 8.1 ± 3.6 10.2 ± 4.2 Quick fill pump set up 3.8 ± 0.4 5.0 ± 2.9 1.0 ± 0.7 8.5 ± 2.4 6.9 ± 4.0 10.9 ± 4.6 Water refuelling of truck 4.4 ± 0.8 109.6 ± 35.5 1.0 ± 0.0 6.3 ± 2.2 7.4 ± 5.1 11.9 ± 2.5 Adding class A foam 3.3 ± 0.5 5.0 ± 2.8 1.2 ± 0.8 9.0 ± 2.0 20.0 ± 0.0 3.1 ± 2.0 Hang hoses at station 2.5 ± 0.5 N/A 0.4 ± 0.5 37.5 ± 23.7 12.5 ± 7.9 N/A Fire line driving in support of crew work 4.1 ± 0.9 43.4 ± 44.9 1.7 ± 1.3 111.6 ± 91.2 N/A 25192 ± 37870

46

Transit driving: staging area and fire ground 4.0 ± 0.8 10.0 ± 0.0 1.0 ± 0.7 115.2 ± 62.0 N/A 45384± 17024 4 wheel drive driving in terrain 4.5 ± 0.9 26.9 ± 16.2 1.70 ± 1.2 98.2 ± 100.1 N/A 6923.1 ± 6939.3 Transit driving between home and staging area 3.2 ± 1.0 4.0 ± 2.2 1.4 ± 1.1 221.6 ± 110.3 15.0 ± 0.0 305769 ± 173624 Administer minor first aid 3.5 ± 1.1 1.7 ± 1.6 1.2 ± 0.4 8.2 ± 5.2 N/A 2.0 ± 0.0 Evacuate injured but assisting crew member 4.4 ± 0.8 0.1 ± 0.3 2.8 ± 0.4 21.0 ± 21.2 20.0 ± 0.0 56.9 ± 19.6 Evacuate seriously injured unassisting crew member 4.7 ± 0.5 0.0 ± 0.0 3.0 ± 0.0 30.2 ± 34.4 30.0 ± 0.0 56.9 ± 19.6 Vehicle repair 2.4 ± 0.9 0.4 ± 0.5 1.4 ± 0.5 22.4 ± 17.8 4.5 ± 2.6 10.0 ± 0.0 Vehicle recovery 2.8 ± 0.6 0.6 ± 0.5 1.5 ± 1.4 30.5 ± 17.3 13.9 ± 5.5 19.2 ± 7.6 Vehicle tire change 3.0 ± 0.7 0.4 ± 0.5 1.9 ± 1.1 33.4 ± 23.2 25.8 ± 16.6 6.4 ± 3.0 Burnout ignition 3.8 ± 0.7 2.8 ± 2.0 1.8 ± 0.4 46.9 ± 44.4 6.2 ± 1.8 677.4 ± 496.6 PLANT supervision 3.6 ± 1.2 4.8 ± 6.2 2.0 ± 0.7 159.2 ± 94.4 15.0 ± 0.0 3150.0 ± 3975.8

47

Table 3.4, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as integrated and involving a number of individual tasks.

Relative Frequency Duration Load Distance Task Difficulty Importance (per season) (min) (kg) (m)

Structure preparation for imminent ember attack 4.6 ± 0.5 6.5 ± 2.4 2.4 ± 0.8 20.3 ± 10.6 12.4 ± 3.7 56.7 ± 34.8

Rapid construction of a sheltering ditch 5.0 ± 0.0 N/A 2.8 ± 0.4 7.9 ± 6.3 N/A N/A

Rapid preparation of refuge site with tanker 4.7 ± 0.8 N/A 2.8 ± 0.4 5.5 ± 3.4 16.9 ± 17.1 49.1 ± 81.3

Preparation of tanker for burn over 5.0 ± 0.0 N/A 2.4 ± 0.8 2.7 ± 1.3 22.3 ± 23.0 14.1 ± 13.0

Preparation of support vehicles for burn over 4.8 ± 0.4 N/A 2.4 ± 0.8 2.6 ± 1.3 N/A 3.2 ± 1.5

Integrated suppression effort on structure fire 4.6 ± 0.5 2.7 ± 1.7 2.8 ± 0.4 15.8 ± 13.8 13.8 ± 7.7 236.2 ± 380.5

Integrated suppression effort to contain spot fires 4.6 ± 0.5 3.3 ± 2.1 2.8 ± 0.4 58.9 ± 102.0 21.6 ± 8.70 199.0 ± 357.3

Integrated crew suppression effort on a back burn 4.6 ± 0.8 9.4 ± 16.6 2.1 ± 0.3 177.4 ± 112.2 17.0 ± 8.5 323.9 ± 432.2

48

Table 3.5, Relative Importance, Perceived frequency per season, Difficulty of successful completion, Perceived task durations, Load encountered per operator and Operational distance of RTBS tasks identified as physically demanding.

Relative Frequency Duration Load Distance Task Difficulty Importance (per season) (min) (kg) (m)

Charged 38mm hose advance 4.8 ± 0.4 23.4 ± 39.1 2.8 ± 0.4 6.6 ± 5.1 33.1 ± 11.9 55.2 ± 11.2

Lateral charged repositioning of 38mm hose 4.3 ± 0.7 524 ± 169 1.9 ± 0.6 5.2 ± 4.2 26.6 ± 8.5 53.2 ± 35.3

Full charged repositioning of 38mm hose 4.2 ± 0.7 269 ± 197 2.1 ± 0.3 9.1 ± 5.1 26.6 ± 8.5 94.8 ± 30.1

Rapid rake work during spot fire containment 4.1 ± 0.6 4.2 ± 3.7 2.8 ± 0.4 9.8 ± 9.2 4.1 ± 0.9 58.2 ± 66.5

Rake work during team line building 3.9 ± 0.8 7.8 ± 16.6 2.4 ± 0.5 20.5 ± 7.4 4.9 ± 2.4 83.2 ± 59.9

Solo rake work 3.4 ± 0.7 4.2 ± 3.4 1.9± 0.6 20.0 ± 8.9 4.1 ± 0.9 32.3 ± 30.8

Rake work during blacking out 3.1 ± 0.3 70.2 ± 54.9 1.9 ± 0.6 21.9 ± 7.6 4.1 ± 0.9 95.8 ± 104

49

Identification of Critical Tasks.

Participants selected seven RTBS tasks as more physically demanding than other tasks (Table 3.5). These tasks are operationally performed by a single tanker crew consisting of three to five crew members. 43% of these tasks represented work done by a single operator and 57% represented integrated tasks requiring up to four operators. Of the tasks identified as more physically demanding, there was a substantial variation of relative importance, frequency and difficulty. Four of the seven physically demanding tasks were considered to have a relative importance above important, but not critical levels (Table 3.5). Five tasks were identified to involve a frequency higher than five occurrences per season (Figure 3.2; Figure 3.3), a value suggesting they are likely to occur at least once within any normal bushfire shift. Participants also considered four tasks required a higher difficulty of successful completion compared to other tasks (Table 3.1; Table 3.2; Table 3.3; Table 3.4; Table 3.5). Of importance, only two tasks, charged hose advance and full charged repositioning of a 38mm hose were considered to involve a relative importance greater than important, but not critical, frequency greater than 1 occurrence per fire shift and greater than moderate difficulty to successfully complete.

50

1.0 Lateral charged hose repositioning

Climb and dismount tanker 0.8

0.6

Charged hose repositioning

0.4 Relative percieved frequency/ season Rakehoe operation post fire 0.2 Pump operation

Charged hose advance 0.0 0 1 2 3 4 5

Operational Importance

Figure 3.2, Operational importance1 Vs. perceived relative frequency2 of all tasks.

550 500 Lateral charged repositioning 450 400 350 300 250 Charged repositioning 200 150 100 50 Manual tool work during black out 50 40 30 Charged hose advance 20

Perceived frequency/season 10 0 0 1 2 3 4 5

Operational Importance

Figure 3.3, Operational importance Vs. perceived frequency of perceived physically demanding tasks.

1 Operational importance was determined using a 5 point scale; 1. Not important, 2. Mildly important, 3. Moderately important, 4. Important, but not critical, 5. Critical 2 Relative frequency represented the perceived frequency of task performance per fire season. 51

1.0 Lateral charged hose repositioning

0.8

0.6

Charged hose repositioning

0.4 Relative percieved Relative frequency/ season 0.2 rakehoe use post fire work

Charged hose advance 0.0 0 1 2 3 4 5

Operational Importance

Figure 3.4, Operational importance3 Vs. perceived relative frequency4 of perceived physically demanding tasks.

3 Operational importance was determined using a 5 point scale; 1. Not important, 2. Mildly important, 3. Moderately important, 4. Important, but not critical, 5. Critical 4 Relative frequency represented the perceived frequency of task performance per fire season. 52

Assemble Carry Climb Dig/ Rake Drag Lift Loaded March Other Action Identified Pass Action Not Identified Patrol Push/ Pull Static Hold Strike Suppress Tanker- Mounted Urban/ Structure

0 20 40 60 80 100 0 20 40 60 80 100

Action identified in tasks (%) Percieved Action Frequency (%)

Figure 3.5, SME perceived action type and frequency of physically demanding RTBS tasks.

53

Strength

Endurance

Speed

Strength- Endurance

Strength- Speed

Endurance- Speed

Strength- Endurance- Speed

0 20 40 60 80 100 0 20 40 60 80 100

Activity Type Identified in Tasks (%) Percieved Activity Type Frequency (%)

Activity Type Identified Activity Type Not Identified

Figure 3.6, SME perceived activity type and frequency of physically demanding RTBS tasks.

54

DISCUSSION

The main aim of this study was to generate a list of critical criterion work tasks and characterize physically demanding fire ground work. These aims will serve as a first step in a job task analysis and development of PES for Australian RTBS work. All seven tasks identified as physically demanding were considered appropriate for further physical demands analysis. The seven critical tasks fall into two distinct work groups which can generally be differentiated according to integrated hose or rake work tasks.

Drag action is absent from all rake tasks but due to its universal involvement with hose work, and the high frequency of hose work, it is involved with eighty nine percent of physically demanding work performed on the fire ground. A high operational importance reported for tasks related to hose work is also likely to mirror the high reported frequency when compared to other tasks. The overlapping prevalence of carry action with this drag action is most likely reflective of the integrative nature of multiple operators conducting charged hose work in forested terrain. Multiple operators may be required to work together to carry the hose over multiple obstacles, or provide a drag force for forward momentum at different times in order to successfully reach the target deployment location. The inclusion of a drag action category appears to reveal hose work tasks for this occupational category in a way not comprehensively described in SFF work. Gledhill and Jamnik (1992a) reported SFF hose advance or repositioning work durations of between 21 and 65 seconds for hose deployments of between 15 and 61 meters respectively. BFFs estimated significantly longer durations in similar work tasks between 5.1 ± 4.2 and 9.1 ± 5.1 minutes over distances of 55.2 ± 11.2 to 94.8 ± 30.1 meters. Duration differences in hose work are most likely due to the terrain variations between a forest and cemented area for BFF and SFF fighting respectively. The prolonged nature of hose work in RTBS work was supported by consistent identification of the strength endurance activity type. This activity type was identified in 57% of physically demanding tasks but represented 98% the frequency of these tasks, which indicated tasks require sustained muscle load. Hose work tasks have traditionally only been studied in a content specific SFF setting (Gledhill & Jamnik, 1992a; Bilzon et al., 2001a; Sothmann et al., 2004). Given potential operational differences with SFF, it appears important to study hose work in a content specific setting.

A dig/rake action is the collective characteristic reported of all rake work conducted during RTBS work. Carry actions also reported during this work most likely refer to the requirement to carry a 2.5kg rake hoe tool for periods of up to 21.9 ± 7.6 minutes. Subject matter experts engaged in this study, reported all rake tasks, with the exception of patrolling on foot whilst carrying a rake hoe, as more physically demanding than other tasks. In Australian LMFF, rake work has shown to generate relative work rates of up to 88 ± 10 % O2max during solo rake work and 67 ± 16% O2max during rake work during team line building (Brotherhood et al., 1997). These values achieved RPE values of 16.4 ± 2.4 and 13.4 ± 2.1 respectively indicated these fire fighters found this work to range between somewhat hard and hard (heavy) on the Borg scale. It is not surprising Lusa (1994) reported SFFs of all ages report debris clearing with heavy manual tools was the most demanding 55 muscular task encountered during SFF work. Collectively these data indicate regardless of setting and rate on task frequency, rake work will be identified as physically demanding by fire fighters.

Lift action has been identified as the major action in up to 88% of military tasks (Rayson, 1998) and serving as a criterion measure in numerous job selection tests (Nottrodt & Celentano, 1987; Stevenson et al., 1988). Lift action was identified in 15% of all RTBS tasks, but as this action was not identified in any rake or hose work tasks, it was not identified in any physically demanding tasks. The physically demand and use of lift actions during manual handling work may be encountered differently during RTBS work when compared to other physically demanding occupations. The heaviest manual handling loads, hose work tasks, involve equipment transported by a fire tanker to within 55.2 ± 11.2 m of the desired deployment area. At this stage multi-operator crews would deploy the fire hose from the tanker using an integrated combination of carry and drag actions dependant on the operational situation. Over a fire season uncharged hose advance tasks occur 3.5 times more frequently than charged hose advance tasks. Charged hose advance, most likely due to the heavier hose weight, may only occur for safety reasons in response to the presence of live fire in the desired deployment location (Wildfire Firefighter learning manual, 2006). Much of the heavy lifting encountered in other physically demanding occupations may have been ergonomically removed by different work practices and much of the equipment weight being transported to within close proximity of a desired deployment area by the fire tanker.

Job selection tests must reflect musculoskeletal and metabolic load demands of an occupation in order to demonstrate genuine occupation requirements (Rayson, 2000b). Rake tasks encountered by BFFs may require a comparable metabolic demand to hose work tasks. Hose work tasks reported for SFF and military fire fighters require between 23 ± 2 and 38 ± 5 mL.kg-1.min-1 (Bilzon et al., 2001a; Gledhill & Jamnik, 1992a). During variably paced crew and solo rake hoe trials, Australian wild land fire fighters required oxygen consumption rates between 22 ± 3 and 41 ± 6 mL.kg-1.min-1 (Brotherhood et al., 1997a). Minimum satisfactory performance on criterion RTBS tasks is likely to reflect the lower end of these ranges as operational work is largely self paced and conducted by significant amounts of older individuals (McLellan, 2004). Due to higher relative importance and frequency of occurrence, it may be possible hose work tasks could be the most content valid tasks to include in a job selection test provided they are comparable with the maximal physical demand of fire fighting.

Experienced incumbents rate the frequency of the most demanding tasks less frequently than inexperienced incumbents (Landy &Vasey, 1991). Although there were potential differences in higher level operational experience with the subject matter experts it is likely perception differences were not present at the crew member level due to a similar range of expertise. Gender may also play a role in influencing perceived job frequencies of tasks, particularly physically demanding tasks which elicit higher relative fire fighting work rates in female subjects for a given absolute fire fighting task (Bilzon et al., 2001). This job task

56 analysis did not utilize any female subject matter experts and therefore may not be comprehensive of the physical demands of crew member female fire fighters. It may be possible additional tasks could be classified as more physically demanding by female subject matter experts in RTBS work. We were unable to recruit any females from the 7% in operational roles within the CFA (McLellan, 2004). The qualitative assessments by SME in this job task analysis provide content validity for selecting critical tasks for further physical demands analysis but not for determining criterion tasks (Rayson, 2000b). This analysis is instead a first step in the pathway to the development of a job selection test for RTBS work.

CONCLUSION

Experienced tanker-based BFFs identified seven tasks involving a greater physical demand than other tasks. Due to the relatively high levels of importance, frequency and difficulty to successfully complete, charged hose advance and full charged repositioning of a 38mm hose were considered critical tasks. RTBS work was characterized by hose related and rake tasks which were responsible for 90% and 10% of physically demanding task frequencies respectively. The majority of work on the fire ground utilizes drag or carry action with a combination of a strength endurance activity type. This job task analysis is intended as a content specific first step in the process of PES development to identify and characterize critical or physically demanding RTBS work.

57

CHAPTER 4: Aerobic demands of simulated bushfire suppression tasks performed by tanker-based fire fighters

INTRODUCTION

Rural fire authorities and land management agencies are primarily responsible for suppressing bushfires on private and public lands, respectively (CFA, 2006). Fire crews from these agencies use a combination of ‘wet’ and ‘dry’ fire suppression techniques to curtail the spread of the fire (CFA, 2006). The dry fire fighting tactics comprise clearing combustible fuel (e.g., small shrubs, plant litter) to create fire breaks of bare earth over which fire cannot pass. Fire breaks are created using earth moving machinery (e.g., bulldozer, grader) and teams of fire-fighters using rakes or other hand tools (e.g., rakes, chainsaws). The work demands for Australian and North American WFFs have been quantified in simulated (Brotherhood et al., 1997), controlled (Brotherhood et al., 1997), and wildfire settings (Ruby et al., 2002). In contrast, the work demands of wet bushfire suppression were fire fighters douse fires with water and other liquid suppressants delivered by hoses connected to fire trucks (called tankers) has never been measured. RTBS crews comprise the majority of Australia’s RTBS force each summer. At present, the fitness levels required of BFF to perform safe and productive RTBS work have not been determined.

Though the work demands of RTBS work are unknown, many fire ground tasks (e.g., hose work, miscellaneous equipment carry) share similarities with those performed by urban SFFs. Cardiovascular responses during simulated and ‘live-fire’ urban SFF work have been measured using North American (Manning & Griggs, 1983; Sothmann et al., 1992; Bennett et al., 1994; Louhevaara et al., 1994; Stewart et al., 2008), and European (Bilzon et al., 2001; Holmér & Gavhed, 2006) fire fighters. Heart rate responses for individual and integrated tasks range between 70% and 100% of age-predicted Maximal heart rate (HRmax). Similarly, aerobic energy demands span ~ 45 and 85%  Maximal oxygen uptake ( VO2 max). These work-rates would be considered ‘moderate’ to ‘very- hard’ by the physical activity intensity classifications endorsed by the American College of Sports Medicine (Balady et al., 1998). The high work intensities prompt many fire agencies to pre-screen fire fighters to ensure they have sufficient cardiovascular fitness to perform their duties safely and productively. At present, screening Australia’s BFFs cardiovascular fitness is not possible as the cardiovascular demands of RTBS have not been quantified. The aim of the present study was,

58 therefore, to quantify the aerobic energy demands and heart rate responses to a range of bushfire suppression tasks routinely performed by Australia’s BFFs. This work is a second step in the process of PES and follows on from a job task analysis.

METHODS

Subjects. Twenty six volunteer BFF (20 men, 6 women, 43.1 ± 14.6 yrs, 1.7 ± 0.9 m, 81.1 ± 13.7 kg) were recruited for this study. All participants were operationally accredited for RTBS work and volunteer members of the Country Fire Authority, Victoria, Australia. During all trials, participants wore full personal protective equipment which consisted of a two piece probane treated pant and jacket ensemble or button up probane treated overalls (Stewart & Heaton, Australia), leather fire resistant gloves (Fire Rescue Safety Australia, Australia), treated leather work boots (Taipan, Australia) and a BR3 BFF hard hat (Pacific Helmets, Australia). Participants were instructed to wear similar clothes under their protective equipment as they would wear during a live fire emergency. Participants did not wear protective smoke goggles or a respiration filter mask during the trial work. The net weight of fire fighting ensemble was 5kg.

Photo: Approved Australian RTBS fire fighting apparel displaying one piece overall (left) and two piece jacket and pants ensemble (right), CFA, 2008.

Task simulations. Fire fighting subject matter experts (SME) identified a list of critical and physically demanding crew member tasks for this physical demands analysis after participating in a job task analysis. Fifteen tasks representative of the SME assessments in Chapter 3 and operational manager opinion (Bosua, 2006; Table A.1) were simulated at 2 regular training sites in the urban fringe of the Lerderderg State Park, Victoria, Australia. Sites were selected based upon similarity with operational fire terrain, access to water points close to the site, co-operation with local fire brigades and the ability to replicate multiple trials of the task on the same course. Due to time constraints, subjects completed three to four tasks in a session. Due to BFF requirements to minimise BFF

59 resources unavailable for RTBS duties, individual subjects did not perform every task in the study. When tasks required multiple operators, in different positional variations, additional operationally qualified BFF performed these roles and were not measured. In this instance, the subject completed one repetition of each positional variation in order to complete the task. Twenty minutes of rest was enforced between repetitions of each task. Single day experimental sessions consisted of one rake work task, one hose work task and one, or both, of either loaded marching or miscellaneous work. Selected operational distances and the amount of operators involved in each task are shown in Table 4.1. All standard course areas were marked with witches hats and measured using a 50m tape measure. Ambient temperature at the deployment site ranged from 10 – 24 degrees during experimental work.

Hose tasks. Uphill uncharged (UU), flat charged advance (FA), flat uncharged advance (FU), hose relocation (HRe) and hose blacking out (HBO) work tasks each had standard courses. All subjects self selected the rate of hose work but were instructed to “advance the fire hose as close to what you consider normal as possible”. The UU advance courses had angle of five to six percent and vertical elevation of ten to twelve meters. FC and FU courses were conducted on the same course through light terrain on the track edge. HBO tasks involved an integrated effort from a hose operator and rake operator to wet self selected “major” fuel sources in a 50 m X 5m marked zone from the back of the fire truck. Expired air was sampled from the hose and rake hoe operators simultaneously during blacking out work. HRe work involved a charged relocation of 60 meters of charged hose through terrain from a marked deployed position, back around a marker 20m from the fire truck and out to a newly deployed marked position. Dry weight of the hose rig used for experimentation was 26.5kg for the 80m hose advance tasks and included 1 length of extruded 38mm fire hose, 2 lengths of 38mm canvas fire hose and a branch nozzle (Pro 366 Branch, Protek, Australia). Charged weight of the hose rig was 111.1kg due to the 84.6L of water contained within three lengths of fire hose.

60

Photo: Australian RTBS personnel engaged in hose blacking out (HBO)(left) and hose relocation after initial deployment (HRe) (right) tasks.

Manual work tasks. Each solo (SLB) and team line build (TLB) task was conducted within an area 100 meters by 50 meters characterized by consistently foliaged terrain and gradient. An experienced fire fighter acted as a spotter for the lead rake hoe operator and selected a collective line through the bush for operators. In all trials, the spotter counted each stroke completed by the lead operator, instructing air collection to begin after 50 completed strokes and to continue for 75 strokes. The lead rake hoe operator rotated to the back of a four man team every 25 strokes during TLB trials. All subjects self selected their work rate during SLB or TLB tasks and were instructed to “complete this task at the pace you would normally use operationally”. Spot fire containment with rake hoe (SFR) tasks were conducted on “out and back” courses and involved the operator completing 75 strokes “as fast as possible”. An experienced fire fighter selected a line for the fire fighter through terrain during the trial and counted 37 strokes out on the course, one stroke on the turn and 37 strokes back to the point of origin.

Photo: Australian RTBS personnel engaged in Blacking out using rakehoe (BOR)(left) and Solo line build with rake hoe (SLB) (right) tasks.

Miscellaneous work tasks. Knapsack spray (KS) and hike (KH) tasks were conducted on an out and back 50m marked course through light terrain on a track edge. Subjects completed a 300m circuit with a 20kg water pack (Fire & Safety Solutions, Australia) during the trial, composed of a 125m loaded hike, 50m transition zone and 125m loaded spray/march. Each subject was instructed to “complete the trial at a pace you would in the field and deliver multiple pump sprays of water to major fuel sources en route during the spray/march condition”. Major fuel sources for the KS, HBO and BOR tasks were defined to subjects as standing or fallen fuel sources with a branch or debris diameter which could sustain prolonged fire activity (>0.2m) within a 2.5 meter radius either side of the marked course. Hose crank (HC) work involved manual cranking retraction of 80m of uncharged

61 hose onto a tanker-mounted hose reel from a deployed position. Additional qualified fire fighters were used for two support positions during HC work which involved either feeding the incoming hose smoothly onto the hose reel or lifting the final meter of the distal end of the hose of the ground to prevent excessive wear to the hose branch. The trailer pump set up (TPS) task involved working in a team of two operators to assemble a trailer mounted quickfill pump at a water source. This self paced task involved assembling filling hoses and filters, deploying the rig into a water source and then starting the pump so the unit was operational. Quickfill pump carry (QPC) involved two operators carrying a 20kg motorised quickfill pump and filling hoses 50m through flat lightly foliaged terrain to a water source. During TPS and QPC tasks, smaller tasks involved in completing both overall tasks were allocated to one of two operator roles. Operators completed both roles over two trials and expired air was collected from both operators during each trial. Static hose spraying (SHS) required participants to spray water at a target point twenty five meters away for two minutes at a standard operational hose pressure of 700kPa. Accuracy during this time was not recorded.

Photo: Australian RTBS personnel engaged in Quickfill pump carry (QPC)(left) and Knapsack spraying (KS) (right) tasks.

Equipment & data collection. Subjects were briefed prior to testing on the tasks they would be completing and were ask to complete the tasks to a standard they would perform operationally. Work was supervised by an experienced training officer. They were then fitted with a global positioning systems (SPI Elite, GPSports, AUS) and heart rate monitor (S410, Polar, FI). Heart rate and GPS data were taken from the GPS and sampled at one second epochs unless otherwise stated. Maximal heart rate and percentage of maximal heart rate (%HRmax ) were determined using the predictive formula, HRmax ൌ ʹͲͺ െ ͲǤ͹ ൈ ܣ݃݁ (Tanaka et al., 2001). Subjects were allocated a reusable 2 way valve mouthpiece (Hans Rudolph, USA) and familiarized with breathing through the mouthpiece before the start of testing. Subjects were then transported to the research sites in a

62 standard fire tanker. Expired air was collected directly via a mouthpiece and rubber tubing into a 120L reusable Douglas bag (Scholle, AUS). Douglas bags were pumped through a dry gas meter (Harvard, USA) and metabolic analyzer (True One 2400, Parvomedics, USA) after each testing session. Metabolic demand recorded from Douglas bag measurements will represent the average oxygen consumption measured across the duration of the task. As breath-by-breath analysis was unable to be performed during RTBS tasks, peak metabolic demand stated in this thesis will represent the highest average oxygen consumption across the entire duration of a simulated task. As such the peak oxygen demands stated during this thesis may conservative representations of the peak metabolic demands encountered during the RTBS task.

Photo: Australian RTBS personnel engaged in Spot fire containment with rake hoe (SFR)(left) and Trailer pump set up (TPS) (right) tasks.

All Douglas bags were analysed within 12 hours of their collection at the testing sites. Ethical and technical approval for the study was obtained from the University of Melbourne Human Research Ethics Committee and the Australasian Fire Authorities Council. Inclusion criteria as outlined by ACSM for intensive exercise (Baladay et al., 1998) was adopted and informed consent obtained was from each participant prior to the commencement of testing.

Statistical Analysis. An unpaired t-test was used to determine gender differences between subject characteristics and velocity differences between TLB and SLB. A paired t-test was used to measure differences between knapsack tasks. One way ANOVA was performed on all remaining group means and a Tukey test post hoc to determine significant differences between groups. ANOVA analysis was not conducted on miscellaneous work tasks due to the unrelated diversity of tasks represented. Graphs are shown as Mean ± SEM unless otherwise stated. Differences were considered statistically significant if P<0.05.

63

RESULTS

Oxygen consumption and duration.

The physical characteristics of the subjects are shown in Table 2. Women represented 23% of the sample population and had a significantly smaller height and mass when compared to male participants. No differences were reported in CFA BFFs experience between gender groups and all subjects were able to complete every task attempted at a self selected work rate until task completion. FC and UU tasks showed higher mean oxygen consumptions of 27.2 ± 7.1 and 26.7 ± 4.8 mL.kg-1.min-1 respectively, when compared to either FU or HBO tasks (Figure 4.1). Hose advance tasks, FU, FC, and UU; all shared similar durations of 52.0 ± 9.5, 46.2 ± 6.2 and 62.8 ± 12.6 seconds in contrast to HRe and HBO tasks which lasted for 95.7 ± 23.8 and 127.5 ± 30.4 seconds respectively (Figure 4.1). All manual work tasks required a similar oxygen consumption of between 28.4 ± 5.3 and 33.1 ± 5.1 mL.kg-1.min-1 (Figure 4.2). BOR lasted for 127.5± 30.4 seconds and took significantly longer than SFR task which lasted for 100.8 ± 16.5 seconds (Figure 4.2). KS required a higher mean oxygen consumption compared to KH of 29.5 ± 4.8 and 25.5 ± 5.0 mL.kg-1.min-1 respectively (Figure 4.3). No duration difference was seen for this higher oxygen consumption over the 125m course. The static hose spray task had the lowest mean oxygen consumption of 10.2 ± 2.9 mL.kg-1.min-1 (Figure 4.3) when compared to all tasks.

Velocity and Heart Rate.

Velocity and mean heart rate data were available for seven tasks; SLB, TLB, HC, HRe, UU, FC & FU. Manual work tasks displayed lower mean velocities compared with hose work tasks. FC & FU, both flat hose advance tasks, displayed higher mean velocities of 5.8 ± 0.8 and 5.5 ± 1.2 km.hr-1 when compared to 3.4 ± 0.6 and 2.8 ± 0.8 km/hr of UU and HRe tasks comparatively (Figure 4.4). TLB tasks were conducted at a faster mean speed of 1.2 ± 0.1 km/hr when compared to the 0.8 ± 0.2 km.hr-1 velocity recorded during SLB (Figure 4.4). Average heart rate data ranged from 128 ± 13.0 to 162 ± 15 b.minute-1 recorded during FU and SLB tasks respectively. Peak heart rate, also between FU and SLB tasks, ranged between from 140 ± 15.1 to 169 ± 12.2 b.minute-1. SLB tasks generated an average heart rate equivalent to 90.4 ± 7.4HR%max, a higher relative maximal heart rate than all tasks except

TLB which generated 83.5 ± 8.5HR%max (Figure 4.5). FU hose advance had a lower peak relative heart rate of 77.9 ± 7.2HR%max when compared to HRe, HC, UU, SLB (Figure 4.5). All other tasks represented comparable relative heart rates which ranged from 87.2 ± 8.0 to 94.8 ± 7.8 %HRmax (Figure 4.5).

Peak metabolic demand.

64

Mean peak metabolic demand encountered during tasks ranged from 25.5 ± 5.0 to 33.1 ± 5.1 mL.kg-1.min-1 observed during the KH and SLB tasks respectively (Figure 4.2; Figure 4.3). No oxygen consumption differences were found between the eight most aerobically demanding tasks; SLB, TLB, KS, BOR, SFR, FC, UU and KH (Figure 4.1; Figure 4.2; Figure 4.3). The energy consumption of HC and HRe of 24.0 ± 3.3 and 23.3 ± 3.7 mL.kg-1.min-1 tasks were similar to all aforementioned tasks except SLB and TLB tasks (Figure 4.2; Figure 4.3). Hose work tasks which generated peak metabolic demand lasted for 53.6 ± 5.1sec, a shorter duration than hose work and miscellaneous work tasks which lasted for 114.6 ± 27.3 and 112.3 ± 27.3 seconds respectively (Figure 4.6). A 95% confidence interval of the peak metabolically demanding tasks displayed a range of 25.3 to 28.7 mL.kg-1.min-1 for hose work tasks, 28.6 to 32.1 mL.kg-1.min-1 for rake work tasks and 25.3 to 29.7 mL.kg-1.min-1 for miscellaneous work tasks.

65

Table 4.1, Specifications of RTBS tasks simulations. Number of Group Tasks Operators Recording guidelines trials

Hose work Flat charged hose advance (FC) 30 (26M,4F) 4 80m advance Uphill uncharged hose advance (UU) 24 (22M,2F) 2 80m advance Hose relocation after initial deployment (HRe) 11 (11M) 2 120m relocation Hose blacking out (HBO) 13 (11M,2F) 2 50m into terrain Flat uncharged hose advance (FU) 22 (16M,6F) 2 80m advance Solo line build with rake hoe (SLB) 10 (9M,1F) 1 75 strokes Manual tool Team line build with rake hoe (TLB) 10 (9M,1F) 4 75 strokes work Spot fire containment with rake hoe (SFR) 14 (11M,3F) 1 75 strokes ASAP Blacking out using rakehoe (BOR) 13 (12M,1F) 1 50m into terrain Knapsack spraying (KS) 12 (11M,1F) 1 125m track edge hike Knapsack hiking (KH) 12 (11M,1F) 1 125m track edge hike Miscellaneous Hose crank (HC) 12 (8M,4F) 1 80m of hose work Trailer pump set up (TPS) 13 (11M,2F) 2 Time until operational Quickfill pump carry (QPC) 10 (8M,2F) 2 50m from track edge Static hose spraying (SHS) 7 (5M,2F) 1 120 seconds

66

Table 4.2, Physical Subject Characteristics. Age (yrs) Height (m) Weight (kg) CFA service (yrs)

Women 43.2 ± 13.7 1.61 ± 0.1a 70.3 ± 12.8b 9.9 ± 8.1

Men 44 ± 15.0 1.75 ± 0.1 84.4 ± 12.8 11.7 ± 9.0

Overall 43.1 ± 14.6 1.72 ± 0.1 81.1 ± 13.7 10.9 ± 8.7

All data are means ± standard deviation. yrs, years; m, metres; kg, kilogram body mass; a P<0.05 Women height vs Men height. b P<0.05 Women body mass vs Men body mass.

67

VO 2 e 35 Duration 140 a 30 b 120 )

d Duration (sec) -1 25 100 20 c 80 15 60 (ml.kg.min 2 10 40

VO 5 20 0 0 FC UU HRe HBO FU

Figure 4.1, Oxygen consumption and duration of hose work tasks. -1 -1 All data are means ± standard error of the mean. O2max, maximal oxygen uptake; mL·kg ·min , milliliters per kilogram body mass per minute; sec, seconds; FC, Flat charged hose advance; UU, Uncharged uphill hose advance; HRe, Hose relocation; HBO, Hose blacking out; FU, Flat uncharged hose advance; a P<0.05 FC O2 vs HBO & FU O2. b P<0.05 UU O2 vs HBO & FU O2. c P<0.05 UU duration vs FC & FU duration. d P<0.05 HRe duration vs FC, UU, HBO & FU duration. e P<0.05 HBO duration vs FC, UU & FU duration.

68

VO2 35 Duration 140 30 120 ) Duration (sec)

-1 * 25 100 20 80 15 60 (ml.kg.min 2 10 40

VO 5 20 0 0 SLB TLB SFR BOR

Figure 4.2, Oxygen consumption and duration of rakework tasks. -1 -1 All data are means ± standard error of the mean. O2max, maximal oxygen uptake; mL·kg ·min , milliliters per kilogram body mass per minute; sec, seconds; SLB, Solo line build with rakehoe; TLB, Team line build with rake hoe; SFR, Spot fire containment with rake hoe; BOR, Blacking out using rake hoe; *P<0.05 BOR duration vs SFR duration.

69

VO2 35 Duration 140 30 120 ) Duration (sec)

-1 * 25 100 20 80 15 60 (ml.kg.min 2 10 40

VO 5 20 0 0 KS KH HC TPS QPC SHS

Figure 4.3, Oxygen consumption and duration of miscellaneous work tasks. -1 -1 All data are means ± standard error of the mean. O2max, maximal oxygen uptake; mL·kg· min , milliliters per kilogram body mass per minute; sec, seconds; KS, Knapsack spraying; KH, Knapsack hiking; HC, Hose crank; TPS, Trailer pump set up; QPC, Quickfill pump carry; SHS, Static hose pray; * P<0.05 KS

O2 vs KH O2

70

8 ab )

-1 6

4

2 c Velocity (km.hr Velocity

0 FC FU UU HRe TLB SLB Figure 4.4, Velocity of bushfire suppression task simulations. All data are means ± standard error of the mean. km.hr-1, kilometres per hour; FC, Flat charged hose advance; FU, Flat uncharged hose advance; UU, Uncharged uncharged hose advance; HRe, Hose relocation; TLB, Team line build with rake hoe; SLB, Solo line build with rake hoe; a P<0.05 FC km.hr-1 vs UU & HRe km.hr-1. b P<0.05 FU km.hr-1 vs UU & HRe km.hr-1. c P<0.05 TLB km.hr-1 vs SLB km.hr-1.

71

average HR%max 100 a peak HR%max 90 c

max 80 b

70 % HR

60 20 0 SLB TLB HC HRe UU FC FU

Figure 4.5, Peak and average percentage of predicted Maximal heart rate encountered during tasks. All data are means ± standard error of the mean. SLB, Solo line build with rake hoe; TLB, Team line build with rake hoe; HC, Hose crank; HRe, Hose relocation; UU, Uncharged uncharged hose advance; FC, Flat charged hose advance; FU, Flat uncharged hose advance; a P<0.05 SLB HR%max vs UU, FU, FC,

HC & HRe HR%max. b P<0.05 FU HR%max vs TLB HR%max. c P<0.05 FU HR%max vs HRe, HC, UU & SLB HR%max.

72

VO2 35 Duration 140 30 120 ) Duration (sec) -1 25 100 20 80 15 * 60 (ml.kg.min 2 10 40

VO 5 20 0 0 Hose Manual tool Miscellaneous

Figure 4.6, Mean oxygen consumption and duration categorized by hose (N = 54), rake(N = 47) and miscellaneous work (N = 24) tasks found to elicit peak metabolic demand. -1 -1 All data are means ± standard error of the mean. O2max, maximal oxygen uptake; mL·kg ·min , milliliters per kilogram body mass per minute; sec, seconds; * P<0.05 Hose work duration vs rake & miscellaneous duration.

73

DISCUSSION

The main aim of this study was to quantify the aerobic demands and task durations of common RTBS tasks. Eight of fifteen common BFF tasks showed individual tasks may place a significant strain on the cardiovascular system generating up to 94.8 ± 7.8 %HRmax and 89% O2max for an aged matched member of the Australian population (Gore & Edwards, 1992). During individual task work, this study observed peak oxygen consumption can be generated within 46.1 ± 6.2 seconds and may be sustained for 127.5 ± 30.4 seconds. There is no metabolic difference between the most physically demanding individual tasks for hose, rake and miscellaneous work tasks; although hose work tasks have a shorter duration. Hose work tasks are commonly performed at mean velocities significantly higher than the observed rake tasks.

Performing traditional urban hose work tasks in varied terrain and without self-contained breathing apparatus (SCBA) appears to alter previously observed oxygen consumption rates for hose work tasks. Mean oxygen consumption rates were 4-7 mL.kg-1.min-1 higher than comparable tasks in Canadian urban fire fighters performing 64mm uncharged hose advances, charged hose advances, charged hose relocations or specialist four minute uncharged naval hose advancing tasks (Gledhill & Jamnik, 1992a; Bilzon et al., 2001a). Oxygen consumption increases significantly based solely on increased weight of fire equipment worn during tasks (Davis et al., 1982; O’Connell et al., 1986) suggesting the oxygen consumption differences observed between urban and RTBS hose tasks are most likely due to carrying the extra weight of SCBA and to a smaller extent the additional water weight in larger diameter fire hoses. Oxygen consumption has been shown to increase by around

30% or a 10% increase in relative percentage of O2 at a five percent gradient in subjects walking uphill in protective military clothing (Keren et al., 1981; Johnson et al., 2002). Similarly the UU task on a similar slop produces a 30% increase in oxygen consumption when compared to the FU task. Introduction of water weight to the hose has been shown to increase the oxygen consumption by around 17-25% (Gledhill & Jamnik, 1992a). The comparative 32% difference shown in this study between FU and FC tasks may be more representative of the increase in operators used to complete this task rather than the actual oxygen consumption change. Manning and Griggs (1983) opined heart rate is not dependent on SCBA and fire fighters worked at near maximal heart rate levels during all urban SFF hose work simulations. Heart rates recorded during hose work in this study are consistent with the near maximal values reported in previous simulated or real hose work of between 70 – 100% of maximal levels (Manning & Griggs, 1983; Sothmann et al., 1992b; Bennett et al., 1994; Louhevaara et al., 1994; Stewart et al., 2008).

74

Oxygen consumption and heart rate values of all observed rake work were comparable to normal paced solo or team raking tasks of LMFF (Brotherhood et al., 1997). The instruction to complete the SFR task “as fast as possible” generated no increase in oxygen consumption values and only decreased duration when compared to the BOR task. During the SLB and TLB tasks, average heart rates observed were equivalent to 90.4 ± 7.4 and 83.5 ± 7.4 %HRmax. Rural BFFs engaged in rake work would likely have been working close to maximal levels and would be unable to self select a harder work rate. There may be significant cardiovascular consequences of the absolute demands of rake work exceeding a self paced level in higher fat mass or less fit individuals (Brotherhood et al., 1997). At near maximal relative intensities, it is unlikely Australian BFFs would be able to conduct frequent or sustained rake work tasks as described in Australian LMFF or North American WFF environments (Brotherhood et al., 1997; Ruby et al., 2002). Alternatively, work of near maximal intensity is likely to increase the potential for adverse work outcomes such as; fatigue or injury.

The knapsack hike (KH) and spray (KS) tasks generated similar oxygen consumption to water backpacking tasks in urban fire fighters (Gledhill & Jamnik, 1992a). Surprisingly in two separate studies, Indian agricultural workers were capable of completing longer duration portable knapsack spray tasks with much lower oxygen consumption of 0.7 ± 0.1 and 0.5 ± 1.7 L.min-1 (Nag et al., 1980; Ghugare et al., 1991). Both rates substantially less than the oxygen consumption of 2.4 ± 1.7 L.min- 1and 2.1 ± 0.6 L.min-1 of the KS and KH tasks observed in this study. Substantial mean heart rates variations between aforementioned knapsack studies (Nag et al., 1980; Ghugare et al., 1991; Gledhill & Jamnik, 1992a) may indicate differences in oxygen demand are due to differing methodologies being used for this task, mechanical efficiency of movement, absolute body size of participants (Bilzon et al., 2001) or productivity (Brotherhood et al., 1997). The physical demand of most miscellaneous fire tasks such as pump carry, pump set up, hose reel cranking and hose spray are likely to reflect regional operational practices and differences in fire suppression techniques or equipment.

Occupational physiologists have traditionally indentified and indexed the mean peak physical demand of task related work in order to build content related PES (Gledhill & Jamnik, 1992a; Gledhill & Jamnik, 1992b; Bilzon et al., 2001a). Peak metabolic demand in this study reflects a mean oxygen consumption range between 25.5 and 33.1 mL.kg-1.min-1 with no oxygen consumption differences between the hardest hose, rake or miscellaneous work tasks. Individuals in this study were still capable of acceptable integrated task performance at energy consumption rates well below the group mean and therefore would classify as false positives should group means be used as a PES (Gebhardt, 2000). On the other hand caution must also be given to the development of PES

75

from the minimal oxygen consumption observed during satisfactory task performance in integrated job tasks. Significant variation has been shown in individual performance of maximal integrated tasks dependant on gender and number of participants involved in lifting tasks (Sharp et al., 1997; Lee, 2004). These data suggested Individuals may not perform integrated intensive work tasks as equal partners. PES for RTBS based off the lower 95% confidence intervals of 28.6 mL.kg-1.min-1 for rake work tasks and 25.3 mL.kg-1.min-1 for both hose work and miscellaneous work tasks, are likely to reflect the minimal peak metabolic demands of RTBS work.

Healthy male subjects between the average age of 24 and 36 years of age are generally used to characterize the physical demands of RTBS work (Manning & Griggs, 1983; Gledhill & Jamnik, 1992a; Bennett et al., 1994; Bilzon et al., 2001; Ruby et al., 2002). There may be little or no variation in the physical job demands of younger and older fire fighting populations during operational activities (Lusa et al., 1994) and fire ground work may produce absolute energy expenditure independent of aerobic power or fatness (Brotherhood et al., 1997). A higher relative work rate during peak fire ground work would be expected for older, smaller and less fit subjects in order to meet the same absolute physical demand (Bilzon et al., 2001). As body size characteristics of the participants are reasonably consistent with the literature and the number of female participants in peak metabolic demand tasks is small, higher relative heart rate data reported from this study may be elevated due solely to a well acknowledged general deterioration in work capacity with age. Potentially high relative work demands observed in this study highlight the need for undertaking a job task analysis and developing PES for RTBS work.

CONCLUSION

Synchronised video and physiological analysis of RTBS task and RTBS work bouts during single day controlled fires. Hose and rake work bouts comprise 87 and 13% of the relative duration of an average fire ground work bout. Analysis of 393 hose work bouts observed mean hose work bout was composed of 4.8 ± 7.2 tasks, had a duration of 192.0 ± 310.6 seconds, a heart rate equivalent to 67.0 ±

-1 0.1% HRmax and was conducted at a speed of 1.2 ± 1.3 km.hr . Five tasks accounted for 78% of all RTBS work by time including Support operator, blacking out work, lateral repositioning and operating RTBS tasks. Although the Patrol and carry hand tool task accounted for only 8% of all RTBS work, this task represented 62% of all rake hoe work.

Mean activity count and heart rate for all fire sites was observed as 295.7 ± 450.1 counts.min-1

-1 and 108.8± 19.7 b.min respectively, the later corresponding to 58.8± 11.4%HRmax. Mean peak heart rate was observed as 146.5 ± 25.3 b.min-1 and corresponded to an percentage of Maximal heart rate

76

of 78.2 ± 15.3 %HRmax. The large time spent in sedentary activity and lower intensity heart rate zones across numerous fire sites indicate a substantial portion of RTBS work is inactive in nature but characterised by tasks which potentially generate high heart rates and given the results of study two, relatively high oxygen uptake levels. The lower absolute and relative heart rates observed during Australian RTBS work when compared to SFF or LMFF are likely due to reductions in loaded bearing work and increased duration of operational deployment (CFA, 2006; Sharkey, 1997; Davis et al., 1982; Payne, 2005).

77

CHAPTER 5: Detailed work patterns and physiological responses of Australian tanker based bushfire fighters to rural bushfire fighting.

INTRODUCTION

Conducting a job inventory and subsequent job task analysis is considered a fundamental step toward developing content valid employment standards for physically demanding occupations (Rayson, 2000b; Rayson, 2000c). Subject matter expert (SME) qualitative assessment of work patterns has been used to successfully isolate critical and criterion tasks (Sanchez, 1989; Rayson, 2000b). Task analysis work in military and emergency services occupations has focused on identifying and characterizing the physical demands of distinct content related criterion tasks for content valid PES (Sothmann et al., 2004; Arvey et al., 1992; Nottrodt & Celentano 1987; Rayson, 1998; Gledhill & Jamnik, 1992a). Payne and Harvey (2010) stated a “job task analysis involves breaking down the requirements of the job into component tasks and analysing the mode, frequency, duration, intensity and work: rest ration of each task” (page 859) and suggested observation, survey, interviews and workshops could be used to collect this data. Not only does this approach satisfy an employee’s legislative obligations to the health and wellbeing of an employee (EEOC, 1979) but it also corresponds to beneficial outcomes such as increased work productivity, increased accuracy of job related training and reduced incidence of injury (Rayson, 2000b; Rayson, 2000c). Bos et al., (2004) observed operational time during SFF comprised of 47% preparation, 18% inside structure tasks, 16% SCBA tasks, 11% water-collection, 4% operation of equipment and 4% extinguishing tasks. The authors suggested up to four times as many personnel were involved in tasks not involving SCBA, operation of equipment or extinguishing tasks. In this case, 80% of the personnel supported a primary fire fighting effort conducted by 20% of personnel or 80% of fire fighters are capable of adequately meeting the physical demands of the task without a primary involvement. Similarly, Gledhill and Jamnik (1992a) observed 90% of the SFF

-1 -1 operations investigated required a mean O2 of 23.4 mL.kg .min , despite recommending a PES of 44.0 ± 1.5 mL.kg-1.min-1 determined during equipment carry up high rise stairs tasks. Viewed from another perspective, Gledhill and Jamnik’s SFFs who could satisfactorily work at 23.4 mL.kg-1.min-1 could complete 90% of work tasks. The context, or frequency and duration with which a physically demanding task occurs operationally were not investigated. Approaches such as the universality of service (National Defence and the Canadian Forces, 2006a), which determine each fire fighter is physically capable of performing each task, may over-

78

weigh infrequent but more physically demanding tasks. In order to create a reasonable job performance standard, PES must be proportional and balanced against the frequency and duration of all tasks. The context of fire ground work is vital to understanding whether the most physically demanding tasks are critical to successful job performance or whether the job can be completed satisfactorily without them. Gledhill and Jamnik’s PES methodology may have been improved if they documented the operational frequency and duration of tasks which require more than 23.4 mL.kg- 1.min-1.

Australian wildfire (known as bushfire) behaviour is unpredictable and can result in catastrophic loss of property or life. Physical demands analysis by direct observation of activities in military or emergency service incidents, such as bushfire suppression, is often problematic and considered impracticably dangerous to both participants and researchers (Rayson, 2000c; Sharkey & Davis, 1998). The accepted alternative to direct observation in real life situations is direct observation during job related task or work simulations (Rayson, 2000b). Simulation work involves the replication of the physical and job specific knowledge skill abilities (KSA) derived from a SME questionnaire in an environment with a more controlled risk to participants and researchers (Gledhill & Jamnik, 1992a; Kuruganti & Rickards, 2004; Sanchez & Levine, 1989).

Task simulations have been used to characterize the demands of urban SFF and rural LMFF work (Gledhill & Jamnik, 1992a; Budd et al., 1997a; Budd et al., 1997c). Individual tasks normally conducted as part of an integrated work shift are characterized in isolation to determine the peak physically demand encountered during the hardest tasks (Gledhill & Jamnik, 1992a; Rayson, 2000b). This work relies on the assumption that performance of the most physically demanding tasks will reasonably correlate with successful performance of the overall occupation and the hardest tasks faced by an incumbent are critical to successful performance of the overall occupation. In reality, it is likely isolated task simulations are limited in their accuracy as they represent only a small context or range of fire fighting work. Task simulations are unlikely to account for the cumulative physical demand or fatigue encountered during repetitive sequential task performance or tasks of significantly different duration.

An alternative approach to task simulations is to utilise a prolonged work simulation in order to observe intensity, frequency and duration of tasks conducted in an integrated work setting. Prolonged work simulations allow researchers to study the work demands of sequential individual tasks and have also been used to characterize the work demands of urban SFF and Australian LMFF (Budd et al., 1997c; Brotherhood et al., 1997a; Bennet et al., 1994; Smith et al., 2001). In addition to a more accurate characterisation of intensity, frequency and duration, it is likely this approach also

79

accounts for a workers or groups’ ability to perform sequential and integrated tasks with a targeted goal directly reflective of the job itself. The purpose of this investigation was to a) Characterise RTBS work in a more controlled setting, b) Characterise the frequency and intensity of each task performed and c) Characterise the frequency and intensity of tasks performed as part of a sequential work bout.

METHODS

Participants. Twenty eight operationally accredited volunteer fire fighters from the Tasmanian Fire Service (TFS), six females and 22 males (39.2 ± 14.7 yrs, 1.7 ± 0.1 m, 92.3 ± 17.5 kg, 8.6 ± 6.9 years or service), volunteered to participate in the study in Spring 2008. Participants in this study were all local to the regional fire sites and worked operationally within their normal fire turn out crews. No adaptations were made to the normal operational deployment practices of the Tasmanian Fire Service. During all shifts, participants wore full personal protective equipment consisting of a two piece probane treated pant and jacket ensemble (Stewart & Heaton, Australia), FirePro fire resistant gloves (Allglove Industries, Australia), treated leather work boots (Taipan, Australia) and a Tuffmaster Type 3 BFF hard hat (Protector Alsafe, Australia). Participants carried and may have used fire safety goggles (UVEX, GER) and disposable breathing masks (Dräger, GER). Clothing worn under personal protective equipment was not recorded. The net weight of BFF equipment is approximately 8kg. BFF utilised standard Isuzu pumper and tanker fire trucks fitted with 2-3000 litre tanker, 38mm hose nozzles and live 25mm hose reel. Ethical and technical approval for the study was obtained from the Deakin University Human Research Ethics Committee and the Australasian Fire Authorities Council. Informed consent was obtained from each participant prior to their involvement in the study.

Equipment & data collection. Participants were recruited during a one month period during four single day controlled perimeter fire incidents (Figure 5.1). Single day fire incidents occurred in the dry eucalyptus forests of the Huon Valley, Tasman Peninsula regions of Southern Tasmania and in of North Eastern Tasmania around the township of St Helens (Anson’s bay region). This work consisted of initiating, maintaining and controlling a fuel reduction burn in a bushfire prone region. Participant recruitment occurred at pre-shift briefings prior to participants commencing their deployment. Demographic data including, height, body mass, age and years of fire fighting experience was obtained via verbal interview at the time of informed consent. Body mass and height were measured using standard body mass scales and portable stadiometer (SECA, Birmingham, United Kingdom). Subjects were fitted with a GP Sports© SPI Elite global positioning system (GPSports, Canberra, Australia), worn in a harness mounted between the shoulder blades (Figure 5.2). Heart rate data were received through the GPS device from a HRM mounted on an elastic chest strap (Figure 5.2). The GPS units used in this study are a small (91 X 45 X 21mm3) water resistant monitor with a built in tri-axial

80

accelerometer. All devices recorded continuously for the duration of that shift and were removed by the experimenters as the subject returned to the pre-shift briefing area at the completion of their shift.

Figure 5.1. Map of Tasmanian fire data collection sites; 1, Sorell region (N=7); 2, Houn valley region (N=7); 3, Ansons Bay (N=7); 4, Ansons Bay (N=7). Google maps, Accessed November 2009. Differentially logged GPS co-ordinates downloaded from the devices at the end of each shift were converted into distance and velocity information using Team AMS software (GPSports, Canberra, Australia). For further analysis, distance and velocity information was exported into a Microsoft Excel® (Microsoft, Redmond, Washington) spreadsheet for statistical analysis. Heart rate and GPS data were taken from the GPS and sampled at one second epochs unless otherwise stated. Maximal heart rate and percentage of maximal heart rate (%HRmax) were determined using the predictive formula,

HRmax = 208 – 0.7 x Age (Tanaka et al., 2001). Cuddy et al (2007) and Heil et al., (2002) mounted activity monitors in a pocket on the upper torso. Australian BFFs are required to maintain free pocket space for storage of food, navigation tools and personal belongings. Australian BFFs are also able to

81

remove their jackets when away from operational areas or riding in the truck. During this study an Actical® (Mini-mitter, Bend, Oregon) activity monitor was attached to the most proximal non elastic segment of the chest mounted heart rate strap at the level of the upper torso (Figure 5.2). These monitors are a small (28 X 27 X 10mm3) uni-axial accelerometer which record one minute epochs for periods of up to 44 days and have previously been validated against accurate methods of determining energy expenditure free living conditions in adults (Welk et al., 2004).

Figure 5.2, Data collection techniques used in Tasmania fire sites . A) Heart rate GPS and Activity monitor positioning; B) Digital video camera synchronisation and analysis & C) Remote monitoring of Heart rate, GPS Speed and Distance.

Data obtained during the shift was downloaded using a Mini-Mitter Actireader® (Mini-Mitter, Bend, Oregon) onto Mini-Mitter Actical Software. For further statistical analysis, activity data was exported into a Microsoft Excel® (Microsoft, Redmond, WA) spreadsheet. Digital video was collected during work tasks using JVC model GZ-MG330 and GZ-MG555 Everio® HDD camcorders (JVC, Yokohama, Japan). One researcher was allocated to each participant and recorded each participant with a single

82

camera. Researchers were instructed “to record all movement and work bouts during the shift” by logging the start and stop times of all work and vehicle transit. After collection, digital files were burnt onto DVD using a JVC CU-VD3 share station direct DVD burner (JVC, Yokohama, Japan) for further analysis. Commencement and cessation periods of digital recording were manually logged by each researcher.

Analysis. Digital video was entered into Dartfish TeamPro® software (Dartfish, Fribourg, Switzerland). Tasks and work bouts were isolated using a custom made RTBS specific tagging profile. The custom made tagging profile evolved from a job task analysis and physical demands analysis conducted in Chapter 3 and 4. Table 5.1 displays hose work tasks included in the tagging profile derived from Chapter 3 and 4. Hose work tasks were consistent with hose work tasks described in Chapter 3. All tasks, rather than just physically demanding tasks, were included in the tagging profile. An individual experienced at identifying all RTBS tasks conducted all of the tagging work.

Work bouts comprised sequential tasks performed by participants until they left a task or placed equipment on the ground for a period longer than 15 seconds whilst not involved in the same or any other task. Rake and hose tasks (as defined by SME in Chapter 3) did not occur in the same work bouts and were consequently analysed separately. Physiological data from tasks and work bouts were synchronized with the GPS data manually in Microsoft Excel® (Microsoft, Redmond, WA) using commencement and cessation times logged in the field by researchers. Rest time between tasks was removed from the subsequent analysis. After synchronization, work bout data was averaged to obtain average heart rate (b.min-1), average peak heart rate (b.min-1), average speed (km.hr-1), and duration (seconds) for each given task and work bout. Heart rates were divided into American College of Sports

Medicine (ACSM) intensity classifications (Balady et al., 1998); very light (<30% HRmax), light (30 - 49%

HRmax), moderate (50 - 69% HRmax), hard (70 – 89% HRmax), very hard (>90% HRmax) and Maximal

(~100% HRmax). Time spent in ACSM classifications and absolute heart rate were analysed by real time and by two-hour blocks during each shift. Task frequency is expressed as an absolute measure, percentage of work bout as a relative percentage by time of the isolated work bout, percentage of hose work as a relative percentage of a task compared to the sum time of all hose work tasks.

Activity counts were divided into three ranges specified by Actical® software; sedentary (0-99 counts·min-1), light (100-1499 counts·min-1), and moderate/vigorous (1500 counts·min-1). The time spent in the three different intensities, total counts and average counts were compiled for overall shift time and for each two hour block during each shift. Average distance (meters) covered in each task was generated by multiplying speed generated (m/sec) by task duration (seconds). GPS speed data was also expressed as average for each two hour block during a shift and for each overall shift.

83

Additionally GPS distance data was expressed as the cumulative distance covered on foot and on tanker (speeds over km.hr-1) for each two hour block and overall during the shift. Operational regulations do not allow RTBS personnel to run on fire sites. 8 km.hr-1 was selected as a speed which would exceed all locomotion conducted by fire fighters on foot and would solely represent truck travel. It is acknowledged some truck travel will occur at slower speeds than 8 km.hr-1 but the volume of this travel is likely to be negligible.

Statistical Analysis. All statistics were derived using the statistical analysis program SPSS (IBM, Chicago, Illinois). One way ANOVA was performed on Figure 5.3, Figure 5.4, Figure 5.6, Figure 5.10, Figure 5.12, Figure 5.16 and on all Tables. A Tukey test was utilised post hoc to determine significant differences between groups. Mixed ANOVA was conducted on remaining Figures using between subjects for fire location, within-subjects for time, activity and intensity classifications where appropriate. A Tukey test was utilised post hoc to determine significant differences between groups. Graphs are shown as Mean ± SEM unless otherwise stated. Differences were considered statistically significant if P<0.05.

RESULTS

Participants

Participants in this study completed 10525 minutes of RTBS work at all fire sites (Table 5.3). Operationally active fire fighters used in this study were 39.0 ± 14.5 years of age (Table 5.2). Figure 5.3 demonstrates a significant difference in age of fire fighters recruited for the St Helens and Hobart fire sites of 44.9 ± 15.7 and 33.1 ± 10.6 respectively (P<0.05 ). A significant difference was also identified for the mean years of participant fire industry experience for St Helens and Hobart fire sites of 11.1 ± 8.2 and 6.0 ± 4.2 years respectively (Figure 5.3) (P<0.05).

Mean Activity Count

Mean activity count is displayed in Figure 5.4. Significantly higher mean average counts were recorded for the second Hobart fire shift of 431.2 ± 192.4 counts.min-1 compared to all St Helens fire sites of 237.9 ± 84.3 counts.min-1 (P<0.05). Figure 5.5 displays mean activity count for all fire sites across two hour segments. Main effects were observed for fire location (location; P = 0.04). St Helens fire sites produced a significantly lower mean average activity count of 188.1 ± 105.6 counts.min-1 in the 4-6 hours two hour time segment than for this period of the Hobart fire of 285.7 ± 185.6 counts.min-1 (P<0.05).

84

Characterisation of work demands

Cumulative Activity Count

Figure 5.6 displays over cumulative activity count at St Helens, Hobart and all fire sites. No difference in overall activity count was observed between fires (p>0.05). Cumulative activity count for all fire sites across two hour segments of fire shift is displayed in Figure 5.7. Main effects were observed for time (P <0.01). Significantly lower cumulative activity counts were observed in the 4- 6hours time segment when compared to the 0-2 hours time segment (P<0.05).

Activity Intensity Classification

Mean activity intensity classification as a time portion of the whole shift is displayed in Figure 5.8. Mean activity intensity classification for time spent in Sedentary, Light and Moderate intensity ranges was observed as 47.2 ± 13.2, 49.5 ± 11.2 and 3.3 ± 4.2% for all fire sites. Main effects were observed for intensity classification but not location (intensity classification; P = 0.0, location; P = 1.00). The portion of time spent in a moderate mean activity intensity classification occurred significantly less than time spent in both Light and Sedentary activity intensity classifications (P<0.05).The mean time spent in moderate activity classifications was significantly longer at the second Hobart fire when compared to combined St Helens fires (P<0.05). No significant differences were observed between fire locations for time spent in Sedentary or Light mean activity intensity classifications (p>0.05).

85

Table 5.1, Video analysis tasks categories for Single day fire sites as determined by SME in Chapter 4.

Task No. Hose Task Description Real life tasks CM-03 Uncharged 38mm hose advance onto fire break 38mm Uncharged onto fire break CM-04 Uncharged 38mm hose advance into terrain 38mm Uncharged advance CM-05 Lateral charged repositioning of 38mm hose 25mm Lateral repositioning 38mm Lateral repositioning CM-06 Full charged repositioning of 38mm hose 25mm Full repositioning 38mm Full repositioning CM-07 Operating 25mm rubber delivery hose 25mm Operating 25mm Support operator 38mm Operating 38mm Support operator CM-08 Hose work during blacking out 38mm Blacking out work 25mm Blacking out work CM-09 Charged 38mm hose advance 38mm Charge advance 25mm Charged advance CM-40 Rapid preparation of refuge site with tanker All CM-54 Hose bowling 38mm Hose bowling CM-55 Hose making up on the bite 38mm Carry coiled 38mm Making up on bite

86

Table 5.2. Physical Subject Characteristics N Value ± SD Range

Age (yrs) 27 39.0 ± 14.5 19 - 71

Height (m) 28 1.7 ± 0.1 1.6 – 1.9

Body mass (kg) 28 92.3 ± 17.5 61 - 137

Years of Service 28 8.6 ± 6.9 0.5 - 28.0

60 15 Years of Service (yrs) of Service Years a 40 b 10

Age (yrs) 20 5

0 0

St Helensage Hobartage St Helensexperience Hobartexperience

Figure 5.3. Age and experience differences between participants and Hobart and St Helens fire sites. All data are means ± standard error of the mean. Years of service (yrs), Years of cumulative experience in the fire industry; Age(yrs), Years of Age; Age, Average age of participants in St Helens and Hobart Fires; Experience, Years of cumulative experience in the fire industry; a P<0.05 St

Helensage vs Hobartage. b P<0.05 St Helensexperience vs Hobartexperience.

Table 5.3, Activity data recorded during simulated fire work of Hobart, St Helens and all fire sites.

Hobart St Helens Overall Duration of Simulation 5373 5152 10525 (min.weekend-1)

Sum (Counts-1) 1887504 1224426 3111930 Mean ± SD 351.3 ± 508.1 237.7 ± 371.6 295.7 ± 450.1

SD, Standard Deviation

87

1000

) 800 -1 600

400

(counts.min 200 Mean Activity Counts Activity Mean 0 St Helens Hobart Overall

Fire locations

Figure 5.4, Mean average activity data recorded during single day fire work of Hobart, St Helens and all fire sites. All data are means ± standard error of the mean. No significant differences observed between means.

St Helens 500 Hobart * Overall

) 400 -1 300

200

(counts.min 100 Mean Activity Counts Activity Mean 0 0-2 hours 2-4 hours 4-6 hours

Time (2hr.min-1)

Figure 5.5, Mean two hour activity data recorded during single day fire work of Hobart, St Helens and all fire sites.

All data are means ± standard error of the mean. * P<0.05 Mean activity counts Hobart4-6hours vs mean activity counts St Helens4-6hours.

88

250000

200000

150000

100000

50000 Whole Shift Activity Count Activity Shift Whole 0 St Helens Hobart Overall

Fire locations

Figure 5.6, Cumulative whole shift activity data recorded during single day fire work of Hobart, St Helens and all fire sites. All data are means ± standard error of the mean. No significant differences observed between means.

60000 St Helens 50000 Hobart Overall

) 40000 -1 30000 *

(2hr.min 20000 Activity Count Activity 2hr Cummulative 10000

0 0-2 hours 2-4 hours 4-6 hours

Time (2hr.min-1)

Figure 5.7, Mean cumulative 2 hr activity data recorded during single day fire work of Hobart, St Helens and all fire sites.

All data are means ± standard error of the mean. * P<0.05 Hobart0-2hours vs Hobart4-6hours.

89

Mean activity intensity ranges were observed over subsequent two hour segments at All fire sites displayed in Figure 5.9. Main effects were observed for activity intensity classification and the interaction between Time and activity intensity classification but not for time (intensity classification; P = 0.00, time*intensity classification; P = 0.04, time; P = 1.00). Similarly to the trend observed across the whole shift, the portion of time spent in a moderate mean activity intensity classification occurred significantly less than time spent in both Light and Sedentary activity intensity classifications across two hour time segments (P<0.05).

Heart rate

Mean heart rate is displayed in Figure 5.10. Mean heart rate for all fire sites, St Helens fire sites and Hobart fire sites was observed as 109 ± 20, 103 ± 16 and 116 ± 22 b.min-1 respectively. Mean relative heart rate for all fire sites, St Helens fire sites and Hobart fire sites was observed as 60

± 11, 58 ± 9 and 62 ± 62% HRmax respectively (Figure 5.12). No significant difference was observed between any fire shifts or location for absolute or relative heart rates (p>0.05). Mean heart rate, peak heart rate, % HRmax over subsequent two hour segments of all fire sites are displayed in Figure 5.11. Main effects for mean average peak heart rate were observed for Time (time; P = 0.03). Peak heart rate was significantly higher during the 2-4 hour time segment compared to the 4-6 hour time segment (P<0.05). Mean % HRmax at all fire sites, St Helens fire sites and Hobart fire sites over subsequent two hour segments are displayed in Table 5.13. No main effects or significant differences were observed for time or location (p>0.05).

ACSM heart rate classification

Mean proportion of time spent in ACSM classification during all fire sites is displayed in Figure 5.14. Mean proportion of time spent in ACSM classification during all fire sites was observed as 23 ± 29, 49 ± 22, 21± 20 and 3± 6% for Light, Moderate, Hard and Very Hard ACSM heart rate classifications respectively. Mean percentage of time spent in ACSM classification at all fire sites, St Helens fire sites and Hobart fire sites over subsequent two hour segments is displayed in Figure 5.15. Main effects were observed for ACSM classification but not location or the interaction between ACSM classification and location (ACSM classification; P = 0.00, Location; 1.00; ACSM classification*location; P = 1.00). For all fire sites more time was spent in Moderate when compared to Light, Hard or Very Hard ACSM classifications (P<0.05) and less time was spent in Very Hard when compared to Light, Hard or Moderate ACSM classifications (P<0.05).

90

100

) 75 -1

50 (counts.min 25 % Activity Count Intensity Count Activity %

0 St Helens Hobart Overall

Sedentary Fire locations Light Moderatea

Figure 5.8, Mean intensity distribution of activity data recorded during single day fire work of Hobart, St Helens and all fire sites.

All data are means ± standard error of the mean. a, P<0.05 % activity counts in moderate classifications Vs % activity counts in Sedentary and Light classifications.

91

100 ) -1

Sedentary 50 Light a Moderate (counts.min % Activity Count Intensity Count Activity % 0 0-2 hours 2-4 hours 4-6 hours

Time (2hr.min-1)

Figure 5.9, Mean intensity distribution of activity data recorded during subsequent 2 hour periods of single day fire work of all fire sites.

All data are means ± standard error of the mean. a, P<0.05 % activity counts in moderate classifications Vs % activity counts in Sedentary and Light classifications.

92

150 )

-1 135

120

105

90

75 Mean Heart Rate (b.min Rate Heart Mean 60 Hobart St Helens Overall

Fire locations

Figure 5.10, Mean average heart rate recorded during single day fire work of all fire sites. All data are means ± standard error of the mean. No significant difference was observed between the mean average heart rate between fire sites.

180 0.9 ) -1 160 0.8 % HR 140 0.7 max 120

0.6 100 Mean Heart Rate (b.min Rate Heart Mean 80 0.5 0 - 2 2 - 4 4 - 6 -1 Mean HR Time (2hr.min )

%HRmax Peak HR

Peak %HRmax

Figure 5.11, Mean heart rate data recorded during two hour segments during single day fire work of all fire sites. All data are means ± standard error of the mean. No significant difference was observed between the mean average heart rate between fire sites.

93

75

max 60 % HR 45

Hobart St Helens Overall

Fire locations

Figure 5.12, Mean Maximal heart rate related to age predicted Maximal recorded during single day fire work of all fire sites.

All data are means ± standard error of the mean. %HRmax, percentage heart rate of age predicted

Maximal (Tanaka et al., 2001). No significant difference was observed between the mean %HRmax between fire sites.

75 St Helens 70 Hobart Overall 65 60

max 55 50

% HR 45 40 35 30 0-2 hours 2-4 hours 4-6 hours

Time (2hr.min-1)

Figure 5.13, Mean heart rate related to age predicted Maximal recorded during single day fire work over two hour segments at St Helens, Hobart and all fire sites.

All data are means ± standard error of the mean. %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001)

94

Speed

Mean speed is shown in Figure 5.16. Mean speed for all fire sites, 2.3 ± 1.1 km.hr-1. No significant differences were observed in mean average speed between any individual fire shifts (P>0.05). Percentage of shift spent travelling in truck or speeds above 8 km.hr-1was observed as 0.1 ± 0.0, 0.1 ± 0.0 and 0.1 ± 0.0%; and speeds below 8 km.hr-1 as 1.0 ± 0.0, 1.0 ± 0.0 and 1.0 ± 0.0% for all fire sites, St Helens fire sites and Hobart fire sites. Main effects were observed for speed classification (location; P = 0.63, Speed classification; P = 0.00, location*Speed classification; P = 0.99). Time spent travelling below 8km/hr was higher than time spent travelling over 8 km.hr-1 in all fire locations.

Average percentage of shift time spent in selected 2 km.hr-1 speed increments was observed at St Helens, Hobart and All fire sites in shown in Figure 5.17. Main effects were observed for Speed classification but not location or interactions between these factors (location; P = 0.99, Speed classification; P = 0.00, location*Speed classification; P = 0.2,). More time was spent in 0-1.9 km.hr-1 speed classification when compared to 2 - 3.9, 4 – 5.9 and 6 – 7.9 km.hr-1 speed classifications (P<0.05).

95

100

75 max

50 % HR

25

0 St Helens Hobart Overall

Light (30 - 49%) a Fire locations Moderate (50 - 69%) Hard (70 - 89%) b Very Hard (90+%)

Figure 5.14, Average ACSM classification % Maximal heart rate data recorded during single day fire work of all fire sites.

All data are means ± standard error of the mean. %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); a, P<0.05 % time spent in Moderate ACSM heart rate classification Vs. Light, Hard and Very Hard ACSM heart rate classifications; b, Very Hard ACSM classification Vs. Light, Moderate and Hard ACSM classifications.

96

100 Light (30 - 49%) a Moderate (50 - 69%) 75 Hard (70 - 89%) b Very Hard (90+%)

50 % Time spent in

ACSM classification 25

0 Overall 0-2hr Overall 2-4hr Overall 4-6hr

Fire Locations

Figure 5.15, Mean average ACSM classification of heart rate data recorded during single day fire work of all fire sites.

All data are means ± standard error of the mean. 0-2hr, first two hour period of experimental work; 2-4hr, second two hour period of experimental work; 4- 6hr, third two hour period of experimental work; a, P<0.05 % Time spent in ACSM classification Moderate vs. Light, Hard, Very Hard 0-2, 2-4 & 4-6 hours; b, P<0.05 % Time spent in ACSM classification Very Hard vs. Light, Moderate, Hard 0-2, 2-4 & 4-6 hours; no significant difference were observed between 0- 2,2-4 and 4-6 hours respectively.

97

6

5 )

-1 4

3

2 Speed (km.hrSpeed

1

0 St Helens Hobart Overall

Fire locations

Figure 5.16, Mean speed recorded during single day fire work of all fire sites.

All data are means ± standard error of the mean. km.hr-1, kilometres per hour. No significant differences were observed between means.

98

100

75

50 % Time spent % Time 25

0 St Helens Hobart Overall a 0 - 1.9 km/hr Fire locations 2 - 3.9 km/hr 4 - 5.9 km/hr 6 - 7.9 km/hr

Figure 5.17, Mean time spent in speed classifications recorded during single day fire work of all fire sites.

All data are means ± standard error of the mean. km.hr-1, kilometres per hour. comb; fire sites combined; a P<0.05, Mean 0 - 1.9km/hr Vs mean 2 - 3.9km/hr, 4 - 5.9km/hr, 6 - 7.9km/hr;

99

Individual tasks task analysis

Analysis of task work isolated from total shift work for every identified task during St Helens and Hobart fire shifts is displayed in Table 5.4. 3096 tasks occurred during fire work at all fire sites, 1704 during 14 shifts at St Helens fire sites and 1392 during 12 shifts at Hobart fire sites. The average task in all fires lasted for 48.3 ± 101.7 seconds in duration, elicited an mean heart rate of 119 ± 25

-1 -1 b.min corresponding to 66 ± 13 %HRmax, elicited a peak heart rate of 124 ± 26 b.min and was performed at 1.4 ± 1.4 km.hr-1. No significant differences were observed for Task duration, mean

-1 Heart rate (b.min or %HRmax), mean Speed or peak heart rate between St Helens and Hobart fire sites (P>0.05).

Table 5.5 display the inter fire differences in duration, heart rate and speed observed during performance of 25mm hose work tasks. 25mm hose task performance was consistent across both fire sites with the exception of; task duration for full repositioning; heart rate for charged advance, full repositioning, Lateral repositioning; and peak heart rate for Lateral repositioning operating and Support Operator tasks (P<0.05). Table 5.6, Table 5.7 and Table 5.8 display inter fire differences observed during 38mm hose, miscellaneous and rake tasks. Significant differences between the Peak Heart rate (b.min-1) observed in the 38mm Lateral repositioning; 38mm Operating; 38mm Support Operator; Manual Handling equipment on the truck and Targeted walk tasks between Hobart and St Helens fires.

100

Table 5.4, Mean task values from all work tasks performed during RTBS shifts.

Heart Rate Task N Duration (sec) % HR Speed (km.hr-1) Peak HR (b.min-1) (b.min-1) max Hobart 1392 48.1 ± 107.1 131 ± 26 70 ± 13 1.6 ± 1.3 136 ± 26 St Helens 1704 48.3 ± 99.1 109 ± 21 62 ± 12 1.3 ± 1.5 114 ± 21 Combined 3096 47.3 ± 101.7 119 ± 26 66 ± 13 1.4 ± 1.4 124 ± 26

-1 All data are means ± standard error of the mean. N, Number of recorded work tasks; sec, seconds; km.hr , kilometres per hour; %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1); No Significant difference.

101

Table 5.5, Mean task values from 25mm hose work tasks performed during RTBS shifts.

Heart Rate Task N Duration (sec) % HR Speed (km.hr-1) Peak HR (b.min-1) (b.min-1) max 25 Blacking out work 81 130.4 ± 138.1 107 ± 18 59 ± 11 0.7 ± 0.6 116 ± 20 25 Charged advance 57 17.8 ± 10.3 104 ± 19 a 59 ± 10 1.8 ± 1.1 109 ± 19 25 Full repositioning 28 78.5 ± 41.5b 106 ± 19 c 59 ± 10 2.0 ± 1.0 115 ± 19 25 Lateral repositioning 245 17.4 ± 15.8 110 ± 26 d 62 ± 12 1.4 ± 1.1 114 ± 26 e 25 Manual hose retraction 29 54.2 ± 67.4 126 ± 27 68 ± 14 0.5 ± 0.5 133 ± 28 25 Operating 70 53.7 ± 58.0 125 ± 25 68 ± 11 f 1.1 ± 0.8 132 ± 25 g 25 Support operator 226 45.5 ± 58.5 107 ± 25 61 ± 12 h 0.7 ± 0.7 113 ± 26 i

All data are means ± standard error of the mean. 25, 25mm hose; 38, 38mm hose; N, Number of recorded work tasks; sec, seconds; km.hr-1, kilometres per

-1 hour; %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min ); a P<0.05 Hobart heart rate Vs St Helens heart rate; b P<0.05 Hobart duration Vs St Helens duration; c P<0.05, Hobart heart rate Vs St Helens heart rate; d P<0.05, Hobart heart rate Vs St

Helens heart rate; e P<0.05, Hobart peak heart rate Vs St Helens peak heart rate; f P<0.05, Hobart %HRmax Vs St Helens %HRmax; g P<0.05, Hobart peak heart rate Vs St Helens peak heart rate; h P<0.05, Hobart %HRmax Vs St Helens %HRmax; i P<0.05, Hobart peak heart rate Vs St Helens peak heart rate.

102

Table 5.6, Mean average task values from 38mm hose work tasks performed during RTBS shifts.

Heart Rate Task N Duration (sec) % HR Speed (km.hr-1) Peak HR (b.min-1) (b.min-1) max 38 Blacking out work 164 75.7 ± 70.4 a 126 ± 24 71 ± 15 0.9 ± 0.70 131 ± 24 38 Carry coiled 25 48.7 ± 60.4 156 ± 29 83 ± 14 2.8 ± 1.5 161 ± 28 38 Charge advance 57 16.8 ± 10.7 115 ± 23 66 ± 13 2.0 ± 1.7 118 ± 23 38 Full repositioning 29 75.9 ± 64.1 113 ± 27 65 ± 18 1.4 ± 0.8 119 ± 28 38 Hose bowling 7 3.6 ± 1.8 130 ± 36 69 ± 19 1.9 ± 2.0 130 ± 36 38 Lateral repositioning 412 16.8 ± 14.2 b 127 ± 23 71 ± 13 1.4 ± 1.1 130 ± 23 d 38 Making up on bite 18 61.6 ± 47.2 155 ± 24 82 ± 13 1.5 ± 0.9 164 ± 25 38 Manual hose retraction 11 45.4 ± 34.6 141 ± 29 78 ± 16 0.5 ± 0.3 145 ± 29 38 Operating 162 40.2 ± 69.3 124 ± 19 69 ± 11 1.2 ± 1.3 129 ± 20 f 38 Support operator 262 49.6 ± 59.0 g 123 ± 26 h 68 ± 13 1.0 ± 1.3 128 ± 26 i 38 Uncharged onto fire break 11 19.7 ± 15.3 120 ± 15 66 ± 9 2.1 ± 1.6 126 ± 16 38 Uncharged advance 14 17.4 ± 12.6 132 ± 19 72 ± 8 1.70 ± 1.0 135 ± 20

All data are means ± standard error of the mean. 25, 25mm hose; 38, 38mm hose; N, Number of recorded work tasks; sec, seconds; km.hr-1, kilometres per

-1 hour; %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min ); a P<0.05 Hobart duration Vs St Helens duration; b P<0.05 Hobart duration Vs St Helens duration; c P<0.05, Hobart heart rate Vs St Helens heart rate; d P<0.05, Hobart peak heart rate Vs St Helens peak heart rate; e P<0.05, Hobart heart rate Vs St Helens heart rate; f P<0.05, Hobart peak heart rate Vs St Helens peak heart rate; g P<0.05, Hobart duration Vs St Helens duration; h P<0.05, Hobart heart rate Vs St Helens heart rate; i P<0.05, Hobart peak heart rate Vs St Helens peak heart rate.

103

Table 5.7, Mean average task values from miscellaneous work tasks performed during RTBS shifts. Heart Rate Task N Duration (sec) % HR Speed (km.hr-1) Peak HR (b.min-1) (b.min-1) max Burn out ignition (torch) 114 293.1 ± 367.4 119 ± 21 65 ± 12 1.6 ± 0.9 129 ± 24 Draughting 22 119.2 ± 112.3 97 ± 16 55 ± 8 0.6 ± 0.5 108 ± 20 Manual handling equipment 96 25.1 ± 30.7 118 ± 27 a 63 ± 14 1.0 ± 1.5 122 ± 26 b on truck Mobile patrol - tanker 1 176 88 47 4.8 104 Mount / dismount 114 5.4 ± 4.3 113 ± 25 61 ± 12 1.1 ± 1.0 115 ± 24 Pump operation 135 28.6 ± 35.1 111 ± 20 61 ± 11 1.1 ± 2.2 116 ± 20 Quickfill carry 2 25.0 ± 9.9 97 ± 16 53 ± 6 1.1 ± 0.70 110 ± 28 Quickfill set up 19 70.0 ± 72.5 107 ± 13 58 ± 8 0.8 ± 0.5 115 ± 13 Support crew on fire line 21 78.6 ± 51.3 101 ± 18 55 ± 10 2.8 ± 2.5 112 ± 21 Targeted walk 378 22.9 ± 29.4 117 ± 25 c 64 ± 13 2.7 ± 1.9 121 ± 25 d

-1 All data are means ± standard error of the mean. N, Number of recorded work tasks; sec, seconds; km.hr , kilometres per hour; %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1); a P<0.05 Hobart heart rate Vs St Helens heart rate; b P<0.05 Hobart peak heart rate Vs St Helens peak heart rate; c P<0.05, Hobart heart rate Vs St Helens heart rate; d P<0.05, Hobart peak heart rate Vs St Helens peak heart rate;

104

Table 5.8, Mean average task values from rake work tasks performed during RTBS shifts.

Heart Rate Task N Duration (sec) % HR Speed (km.hr-1) Peak HR (b.min-1) (b.min-1) max Rake hoe during blacking out 94 24.6 ± 25.3 130 ± 29 69 ± 15 0.8 ± 0.4 134± 31 Solo line building 1 123 167 96 0.3 169 Team line building 3 460.6 ± 386.6 157 ± 15 86 ± 11 0.5 ± 0.3 168 ± 10 Spot fire containment 25 51.9 ± 83.4 a 108 ± 15 62 ± 10 1.2 ± 1.0 113 ± 16 Patrol & carry hand tool 162 47.1 ± 59.2 b 127 ± 28 68 ± 14 1.4 ± 0.9 133 ± 28

-1 All data are means ± standard error of the mean. N, Number of recorded work tasks; sec, seconds; km.hr , kilometres per hour; %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1); a P<0.05 Hobart duration Vs St Helens duration; b P<0.05 Hobart duration Vs St Helens duration.

105

Characterisation of work bouts

The final analysis section of Chapter 5 details analysis of hose and rake work bouts, subsequent sequential tasks without rest, isolated from total shift work during St Helens and Hobart fire shifts. Variation between St Helens and Hobart fire sites in hose work bout duration, mean heart rate, %HRmax, peak heart rate, mean speed and number of tasks is displayed in Table 5.9. A number of differences were observed between hose work bouts at St Helens and Hobart fire sites. Duration of hose work bouts at Hobart fire sites was significantly shorter than at St Helens fire sites (P<0.05). Hose work bouts conducted at Hobart fire sites elicited a significantly higher mean heart rate, peak heart rate in both b.min-1 and relative to age predicted Maximal (Tanaka et al., 2001) (P<0.05). Hose work bouts at Hobart fires were conducted at a lower average speed and comprised fewer tasks in each work bout (P<0.05). Variation between St Helens and Hobart fire sites in rake work bout duration, mean heart rate, peak heart rate, %HRmax, mean speed and number of tasks is displayed in Table 5.10. Rake work bouts at Hobart fire sites elicited a higher mean heart rate in both absolute and relative terms but a lower mean average speed when compared to St Helens fire sites (P<0.05). Table 5.10 displays there were also less rake work bouts performed at St Helens fires when compared to the Hobart fire sites.

The descriptive compositions of tasks comprising an average hose work bout, all hose work and all fire ground work is displayed in Table 5.9, Table 5.11 and Figure 5.20 respectively. Individual hose tasks comprised between 83.6 ± 27.6 and 8.1 ± 7.7% of the duration of hose work bouts in which they occurred. Support operator work and blacking out work comprised high portions of hose work bouts and overall work. 25mm and 38mm support operator tasks comprised 36.7 ± 69.5 and 35.3 ± 34.9% of the duration of hose work bouts which they were involved in, 12.1 ± 0.0 and 15.4 ± 0.1% of all hose work and 11.8 ± 0.1 and 14.8 ± 0.1% of all fire ground work respectively. 25 and 38 blacking out work tasks comprised 31.5 ± 32.0 and 21.1 ± 24.2% of the duration of hose work bouts which they were involved in, 13.6 ± 0. 2 and 16.0 ± 0.1% of all hose work and 12.1 ± 0.2 and 14.2 ± 0.1% of all fire ground work respectively. The most frequently occurring task, 38 lateral repositioning work comprised only 12.9 ± 24.3% of individual hose work bouts, 2.9 ± 0.0% of all hose work and 8.0 ± 0.0% of all fire ground work.

The descriptive composition of tasks comprising an average individual rake work bout, average rake work and all fire ground work is displayed in Table 5.11. Individual rake hoe tasks comprised between 49.0 ± 34.1 and 10.0 ± 21.6% of the duration of rake work bouts in which they occurred. Patrol and carry rake tasks occurred with the highest frequency of all rake work tasks.

106

Although it represented only 12.8 ± 15.0% of the duration of individual rake work bouts, the Patrol and carry hand tool task was responsible for 49.0 ± 34.1% of all rake work and 7.8 ± 0.1% of all fire ground work. Similarly rake hoe during blacking out task represented only 10.8 ± 21.6% of the duration of individual rake work bouts but 19.7 ± 0.2% of all rake work and 2.5 ± 0.0% of all fire ground work. Tasks composing higher portions of individual work bouts, such as Team line building, spot fire containment and solo line building, comprised lower portions of rake hoe work and overall work suggesting these tasks occur as isolated tasks and do not require much integration with other work tasks.

The relative frequency by time of tasks comprising rake (A) and hose (B) work bouts is displayed in Figure 5.20. Hose and rake work bouts comprise 87 and 13% of the relative duration of an average fire ground work bout. Combined 25 and 38mm support operator, blacking out work, lateral repositioning and operating tasks account individually for 27, 26, 13 and 12% of all RTBS work respectively or 78% of all RTBS work. Patrol and carry hand tool accounted for 8% of all RTBS work and represented 62% of all rake hoe work. Full repositioning, Charged advance, Targeted walk, Uncharged advance, Spot fire containment, Pump operation, Rake hoe during blacking out, Team line building, Spot fire containment and Solo line building accounted for 5% or less individually or 14% cumulatively of RTBS work. The physiological characteristics of the average hose and rake work bouts encountered during RTBS work are described in Table 5.11, Figure 5.18 and Figure 5.19. 393 hose and 34 rake work bouts were analysed across Hobart and St Helens fires. The average hose work bout is composed of 4.8 ± 7.2 tasks, has a duration of 192.0 ± 310.6 seconds, elicits a heart rate

-1 equivalent to 67.0 ± 10% HRmax and is conducted at a speed of 1.2 ± 1.3 km.hr . The average rake work bout is composed of 7.9 ± 7.4 tasks, has a duration of 341.4 ± 306.1 seconds, elicits a heart rate

-1 equivalent to 68.0 ± 10%HRmax and is conducted at a speed of 1.1 ± 0.6 km.hr .

107

Table 5.9, Descriptive statistics between hose work bouts at Hobart and St Helens fires during single day fire work.

Number Fire Mean ± SD of values Duration (sec) Hobart 191 149.3 ± 245.9a St Helens 201 232.7 ± 357.4 Heart Rate (b.min-1) Hobart 192 130 ± 23.4b St Helens 184 110.4 ± 20.5 Peak HR Hobart 192 139.3 ± 24.1c (b.min-1) St Helens 184 120.6 ± 21.4

d % HRmax Hobart 192 70 ± 12 St Helens 184 63 ± 12 Speed (km.hr-1) Hobart 192 1.0 ± 0.7e St Helens 187 1.4 ± 1.6 Number of tasks Hobart 192 3.7 ± 4.70f St Helens 201 5.9 ± 8.8

All data are means ± standard error of the mean. sec, seconds; km/hr, kilometres per hour; %HRmax, percentage of age predicted Maximal heart rate (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1); a P<0.05 Mean duration Hobart vs mean duration St Helens ; b P<0.05 Mean heart rate Hobart vs mean heart rate St Helens; c P<0.05 Mean peak heart rate Hobart vs mean peak heart rate St Helens; d P<0.05 mean % HRmax Hobart vs mean % HR age predicted max St Helens; e P<0.05 Mean speed Hobart vs Mean speed St Helens; f P<0.05 Mean number of tasks Hobart vs Mean number of tasks St Helens.

108

Table 5.10, Descriptive statistics between rake work bouts at Hobart and St Helens fires during single day fire work.

Number Fire Mean ± SD of values Duration (sec) Hobart 30 337.7 ± 286.7 St Helens 4 368.5 ± 484.8 Heart Rate (b.min-1) Hobart 30 129.6 ± 25.0a St Helens 4 103.6 ± 6.2 Peak HR (b.min-1) Hobart 30 144.9 ± 27.4 St Helens 4 119.7 ± 4.1

% HRmax Hobart 30 70 ± 12 St Helens 4 57 ± 6 Speed (km.hr-1) Hobart 30 1.1 ± 0.5b St Helens 4 1.5 ± 1.1 Number of tasks Hobart 30 8.5 ± 7.6 St Helens 4 3.3 ± 2.6

-1 All data are means ± standard error of the mean. sec, seconds; km.hr , kilometres per hour; %HRmax, percentage of age predicted Maximal heart rate (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1); a P<0.05 Mean heart rate Hobart vs mean heart rate St Helens. b P<0.05 Mean speed Hobart vs Mean speed St Helens.

109

Table 5.11, Descriptive statistics of combined hose and rake work bouts during single day fire work.

Heart Rate Hose work Tasks Number of tasks Duration (sec) % HR Peak HR (b.min-1) Speed (km.hr-1) (b.min-1) max Mean ± SD 4.8 ± 7.2 192.0 ± 310.6 120 ± 24 67 ± 8 130 ± 25 1.2 ± 1.3 95% CI of mean 4.1 x 5.5 161.2 x 222.9 118.0 x 122.0 65 x 68 127.7 x 132.7 1.1 x 1.3 Heart Rate Rake work Tasks Number of tasks Duration (sec) % HR Peak HR (b.min-1) Speed (km.hr-1) (b.min-1) max Mean ± SD 7.9 ± 7.4 341.4 ± 306.1 126.6 ± 25.0 68 ± 13 141.9 ± 27.0 1.1 ± 0.6

95% CI of mean 5.3 x 10.5 234.6 x 448.2 117.8 x 135.3 64 x 73 132.5 x 151.3 0.9 x 1.4

-1 All data are means ± standard error of the mean. sec, seconds; km.hr , kilometres per hour; %HRmax, percentage of age predicted Maximal heart rate (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1);

110

100

80

60

40 Percentage of Percentage

20 individual work bout (%)

0

25 Operating38 Operating Targeted walk

38 Charge advance 25 Support 38operator Support25 operator Blacking 25out Charged work 38 advance Blacking out work 25 Full repositioning38 full repositioning 38 Uncharged advance 38 Lateral repositioning 25 Lateral repositioning

38 Uncharged onto fire break Task

Figure 5.18, Percentage of individual work bout comprising hose work RTBS tasks during single day fire work

All data are means ± standard error of the mean. 25, 25mm hose; 38, 38mm hose; %HRmax

111

80

60

40

Percentage of Percentage 20

individual work bout (%) 0

Team line buildingSolo line building Spot fire containment Patrol & carry hand tool

Rake hoe during blacking out Task

Figure 5.19, Percentage of individual work bout comprising rake work RTBS tasks during single day fire work

All data are means ± standard error of the mean. 25, 25mm hose; 38, 38mm hose; %HRmax

112

Patrol & carry hand Rake hoe during tool blacking out A 8% 3% Team line building 2% Spot fire containment 1%

Solo line building 0%

Hose Work 87%

Pump operation Spot fire 0% B containment 0% Support operator Rake Work 27% Uncharged advance 13% 1%

Targeted walk 1%

Charged advance 2%

Full repositioning 5%

Blacking out work Operating 26% 12% Lateral repositioning 13%

Figure 5.20, Percentage of rake (A) and hose (B) tasks by duration composing rake or hose work bouts during single day fire work respectively.

113

DISCUSSION

Characterisation of work demands

This study determined the work demands of single day RTBS work using activity monitors, GPS systems and heart rate monitors. The consistently moderate activity counts and relatively high ACSM heart rate intensities within this study provide a strong argument for the inclusion of PES into the BFF workforce. Despite the predominant period of work time spent in low intensity classifications, the presence of more intense work is consistent across all fire sites and individual fire shifts. It is reasonable for any rural fire agency to expect all fire fighters may be exposed to physical/ more arduous work throughout any RTBS shift 3.3 ± 4.2 % or 4.0 ± 9.0% by vigorous activity count or very hard ACSM heart rate classification respectively. Estimates of the amount of more physically demanding work using these data suggest RTBS equates to 27.7 ± 35.3 or 33.6 ± 75.6 minutes by vigorous activity count or very hard ACSM heart rate classification respectively when extrapolated for every 14 hour RTBS shift.

Bos et al., (2004) observed 85 real SFF shifts were composed of 1.5 sporadic, high intensity incidents per shift with an average operational time of 88 ± 76 min. SFF training drills and task simulations typically last between 64 seconds and 15 minutes (Sothmann et al., 1992b; Bilzon et al., 2002; Louhevaara et al., 1994; Smith et al., 1996; Gledhill & Jamnik, 1992a; Manning & Griggs, 1983) and may be conducted as higher intensities than BFF partly due to an individual’s ability to work at higher relative intensities over shorter working duration (Wu & Wang, 2002). Operational duration during SFF or related training drills is likely to be significantly shorter when compared to BFF (Chapter 4).

Mean heart rate for all fire sites was observed as 109 ± 20 b.min-1 respectively or 58.8±

11.4%HRmax. A multitude of studies have observed higher heart rates of between 122 to 158 -1 beats·min or 66 to 86%HRmax. Louhevaara et al., (1994) observed these values during 14.5 minutes of multi-task fire work; Sothmann et al., (1990)observed heart rates of 173 ± 9 b.min-1 for around 9

-1 minutes of work during a multi stage SFF work bout and 157 ± 8 b.min or 88 ± 6% HRmax over durations of 15 ± 7 minutes during actual SFF duties (Sothmann et al., 1992b). Holmér & Gavhed, (2006) similarly observed heart rates of 168 ± 11.7 b.min-1 on a multi stage SFF drill. 170 ± 2 beats·min-1 during charged hose advancing up two flights of stairs over 8 minutes of SFF training drills in an elevated temperature facility (Smith et al., 1996); Heart rate values of 137 ± 16, 161 ± 17, 180 ± 13 and 158 ± 17 b.min-1 or relative raking work rates of 47 ± 10, 70 ± 12, 88 ± 10 and 67 ± 16%

114

during simulated self paced 5 minute ‘slow’, ‘normal’, ‘fast’ and ‘crew’ rake hoe tasks respectively and between 158 to 180 beats·min-1 for 60% of their shift time (Brotherhood et al., 1997a).

Peak heart rate was observed as 147 ± 25 b.min-1 (Figure 5.11). Values lower than recorded by Sothmann et al., (1992b) of up to 88 ± 6% HRmax over durations of 15 ± 7 minutes during actual SFF duties; by Smith et al., (2001) during three repeated 7 minute strenuous live fire drills on recruit

-1 fire fighters 174.8 ± 8.3, 186.0 ± 9.1 and 189.3 ± 10.5 b.min ; by Williford et al., (1999) 92 %HRmax during simulated SFF drills; by Manning & Griggs (1983) 90 to 100 %HRmax during simulated SFF drills between 150 and 430 seconds in duration; peak mean heart rate of 181 ± 9 during a 64 second pitched roof ventilation exercise (Gledhill & Jamnik, 1992a). Budd et al., (1997c) observed average heart rates of 153.0 ± 2.4, 153.8 ± 2.9, 152.2 ± 2.1 and 151.7 ± 2.2 beat.min-1 in four crews of LMFFs suppressing fires over durations of 35-220 minutes. The intensity of LMFF was more comparable to BFF but still likely to by more intense due to shorter operational duration (Wu & Wang, 2002).

The lower mean and peak heart rates observed during this Australian BFF when compared to SFF or LMFF are likely due to reductions in loaded weight of personnel and the longer operational durations (CFA, 2006; Sharkey, 1997; Davis et al., 1982; Payne, 2005). Manning & Griggs, (1983) observed SFFs achieved near maximal work rates regardless of load carried during short duration simulated work tasks. Numerous authors observed increased work load caused by SCBA or hose weight during fire fighting work and increased physiological responses such as heart rate, oxygen consumption, blood lactate, ratings of perceived physical exertion and core body temperature during live fire work (Richmond et al., 2008; Ricciardi et al., 2008; Beekley et al., 2007; Keren et al., 1981; Skoldstrom, 1987; Hooper et al., 2001). White & Hodous (1987) observed decrease work tolerance during live fire work. It is likely Australian BFFs may have lower heart rates during operational work as they are not burdened with SCBA, hiking with a pack or large diameter SFF fire hoses.

Mean average activity count for all fire sites in this study was observed as 295.7 ± 450.1 counts.min-1. Cuddy et al., (2007) observed similar activity counts of 352 ± 252, 317 ± 183, 521 ± 421 and 366 ± 249 counts.min-1 in a number of American WFF lasting 12 hours. It is likely Australian BFFs self select a similar pace as American WFFs over a twelve hour shift when participating in crew work to maintain an average level of productivity relevant to the fire behaviour over 12-14 hours of a shift. Cuddy et al., (2007) also displayed a similar level of variability in a 12 hour fire shift. Given American Wildland fire fighting is composed of more intense rake work and hiking, it is likely this type of work also contains large periods of sedentary time and is composed of intermittent work similarly to Australian RTBS work over a twelve hour period.

115

Simulated work bouts performed by crews of Australian LMFFs reported 57.8 ± 15.1, 75.6 ± 15.9, 65.9 ± 15.2% of total shift time was spent involved in the primary task for the roles of raker, slasher and chainsaw operator respectively (Budd et al., 1997c). Cuddy et al., (2007) observed American WFFs spent 33% of fire ground time in light activity ranges and 66% in sedentary activity intensities or 79.1 ± 21.1 and 40.8 ± 21.0 min.2hr-1 respectively. Comparatively, Sedentary, Light and Moderate intensity ranges during Australian single day bushfire suppression were observed as 47.2 ± 13.2, 49.5 ± 11.2 and 3.3 ± 4.2% of overall shift or 56.6 ± 15.8, 59.4 ± 13.4 and 4.0 ± 5.0 min.2hr-1 respectively. When compared to Australian LMFFs (Budd et al., 1997a), lower portions of the shift were spent actively involved in task performance, particularly slasher and chainsaw operator roles, not core tasks of RTBS work. When compared to American WFFs, lower portions of the shift were spent in sedentary activity intensities and higher portions of the shift in light activity intensities. The large time spent in sedentary activity and lower intensity heart rate zones across numerous fire sites indicate a substantial portion of fire fighting work is inactive in nature but characterised by infrequent, sporadic tasks which potentially generate higher heart rates, aerobic demand and relative work rates.

Characterisation of work tasks and work bouts

This study also determined the physical characteristics of tasks performed during single day RTBS work measured using video analysis, activity monitors, GPS systems and heart rate monitors. The frequency and duration of tasks, encountered during intermittent industrial work was observed during a prolonged single day fire workday as individually or sequentially related tasks, a work bout. Cumulatively subjects in the current study completed 393 hose and 34 rake work bouts, comprising 87 and 13% respectively of the relative duration of an average fire ground work bout (Table 5.4; Figure 5.2). Combined 25mm and 38mm support operator, blacking out work, lateral repositioning and operating tasks account individually for 78% of all RTBS work (Figure 5.2). Patrol and carry hand tool accounted for 8% of all fire ground work and represented 62% of all rake hoe work on the fire ground. Full repositioning, Charged advance, Targeted walk, Uncharged advance, Spot fire containment, Pump operation, Rake hoe during blacking out, Team line building, Spot fire containment and Solo line building tasks demonstrated low frequency of occurrence during a single day bushfire suppression model (Figure 5.2).

This study demonstrated higher frequency tasks, such as lateral hose repositioning (Figure 5.2), did not comprise the highest portions of individual tasks (Figure 5.2). High frequency tasks tended to occur multiple times with short duration within individual work bouts. Tasks which generated higher individual aerobic energy demands (Chapter 4) occurred with less frequency than

116

other tasks and are more likely to occur in isolation (Figure 5.2). It is possible incumbents in this study self selected work which did not encompass physically demanding tasks and have potentially developed their work patterns to avoid this work were possible. Chapter 4 demonstrated Australian RTBS tasks can generate substantial demands on the cardiovascular system, potentially increasing relative work demand of older individuals. Self selection of less physically demanding work may be particularly important in reducing risk during fire fighting work were heart rate and demands on the cardiovascular system increase rapidly. It is also likely work practices in Australian BFF have evolved to favour less physically demanding tasks given the age of the workforce, unpaid status of workers and lack of pre-work health screening on RTBS personnel.

Brotherhood et al., (1997) demonstrated larger and higher fat mass Australian LMFFs had higher productivity in contrast to their predicted productivity from a stepping O2max assessment when performing prolonged rake hoe work. The authors suggested higher fat mass LMFF fire fighters may have an ability to self pace their effort against sub-maximal demands of the work although would be overwhelmed during sustained near maximal effort, such as those encountered working against steep terrain or in emergency circumstances. Such events would predispose higher fat mass fire fighters to adverse effects not seen in their study. Similarly Australian BFF observed in this study may also be exposed, albeit infrequently to higher relative work rates at or beyond normal self pacing behaviour in operational situations. Higher operational demands may occur when working with manual tools, working against gradient, or responding rapidly to emergencies (Chapter 4). For example, infrequent RTBS tasks such as solo line building, which occur less than once a shift, generate cardiovascular demands in excess of 95% HRmax. Analysis of single day RTBS work conducted in this study indicates these operational situations may occur less frequently than other BFF work.

Previous research has highlighted the importance of a thorough task analysis in developing content valid test but has not fully utilised the nature of the work encountered to develop test duration. Matching of video analysis to physiological variables has allowed researchers to understand the context by which the task is performed not just the frequency by which it occurs. Although some aspects of this process may be subjective, for example; when a threshold is determined for commencement and cessation of the task, it nonetheless enables a sensitive analysis of variation between fires. A higher level of sensitivity is particularly important given the requirements of Australian law.

Not only is it important to quantify the maximal work demands likely to be faced by a worker, but it is equally important to determine the work demands fire agencies or an employer can

117

reasonably expect a worker will face. This thesis has taken steps toward the development of a model for attributing a content valid duration to industrial work, in this instance a content valid RTBS test. This approach has been taken with specific regard to the OHS requirements of Australian workplace law and has been designed to achieve compliance in this domain. Confidence levels quoted in the work have been used to set a benchmark to determine what fire agencies or an employer may consider the reasonable operating range of Australian bushfire fighting.

CONCLUSION

Detailed video movement and physiological analysis of single day RTBS work potentially offers valuable insight into the composition and intensity of the work encountered during campaign bushfire suppression. When compared to SFF or LMFF, BFF involved lower average and Maximal heart rate. More intense work is however consistent across all fire sites and individual fire shifts. Arduous physical work occurred during all RTBS shifts of 3.3 ± 4.2 % or 4.0 ± 9.0% by activity count or ACSM heart rate classification respectively, equating to 27.7 ± 35.3 or 33.6 ± 75.6 minutes by activity count and heart rate respectively when extrapolated for a 14 hour shift (Figure 5.2). The large time spent in sedentary activity and lower intensity heart rate zones across numerous fire sites indicate a substantial portion of fire fighting work is inactive in nature but characterised by tasks or work bouts which potentially generate higher heart rates, aerobic demand and relative work rates (Figure 5.11; Figure 5.9; Figure 5.12). Fire ground work is likely to comprise of 87 and 13% Hose and rake work respectively (Figure 5.20). 78% of all RTBS work comprises of the hose tasks; support operator, blacking out work, and lateral repositioning tasks (Figure 5.20). This study will serve as a reference of operational intensity, duration and frequency of occurrence during further development of PES for RTBS work.

118

CHAPTER 6: Work demands and physiological responses of Australian tanker based bushfire fighters to campaign bushfire fighting.

INTRODUCTION

Training material produced by bushfire suppression agencies and controlled single day RTBS work (Chapter 5), use a combination of dry LMFF and SFF techniques whilst working from a fire fighting tanker truck (Bosua, 2006; Chapters 3& 4; CFA, 2006). Operational RTBS techniques commonly include mineral earth control line construction with hand tools, post back burning mopping up with hoses and direct fire attack with hoses (Bosua, 2006; Wildfire Firefighter Learning Manual, 2002). The work demands of fire suppression techniques used in a wildfire suppression context by Australian volunteer fire fighting agencies are yet to be investigated. Work demands of American wildfire suppression have been measured using doubly labelled water and activity monitors in wildland fire fighters (Heil, 2002; Ruby et al., 2002; Cuddy et al., 2007). Reported daily energy expenditure rates for these fire fighters were 4878 ± 716 and 4768 ± 478 kcal.d- 1g using doubly labelled water and activity monitor methods respectively. Doubly labelled water techniques for measurement of energy expenditure requires at least four days of recording to accurately calculate the changes in isotope level (Ruby et al., 2002). This technique is problematic for measurement of energy expenditure in Australian campaign fire suppression as the typical crew deployment of volunteer fire fighters is three days in duration. During multi-day recording periods, activity monitors have shown to predict similar daily energy expenditure to doubly labelled water (Heil et al., 2002). Due to their portability and relative accuracy, activity monitors are a viable operational alternative during a fire suppresion shift. Average operational and training heart rates have been observed by a number of

-1 researchers. Heart rate values of between 122 to 158 beats·min or 66 to 86%HRmax in SFF (Louhevaara et al., 1994; Sothmann et al., 1992b; Holmér & Gavhed, 2006; Smith et al., 1996) and 158 to 180 beats·min-1 for 60% of their shift time in Australian LMFF (Brotherhood et al., 1997a). Sothmann and colleagues (1992b) effectively monitored SFF during actual fire call outs and recorded similar heart rate values as observed during simulated testing. The use of heart rate monitoring in this context suggests BFF completing RTBS work can be similarly monitored using heart rate monitors.

119

Global Positioning Systems (GPS) has been used to demonstrate work demands in Australian soldiers (Demczuk, 1998), Orienteering athletes (Larsson et al., 2001); Australian rules footballers (Edgecomb & Norton, 2006; Dawson et al., 2004; Wiseby et al., 2009) and American WFFs (Ruby et al., 2003; McKenzie, 2007). Edgecomb & Norton (2006) observed GPS overestimated actual values during a team sport (AFL football) simulation by 4.8% but was no less erroneous than video analysis methods of measuring distance. Wiseby and colleagues (2008) found AFL players ran 12.5 ± 1.7 km in under 120 minutes of game time, completed 240 moderate accelerations and 10 rapid accelerations per game at speeds exceeding 18km.hr-1. Although 65% of AFL competition time was spent at speeds below 8km.hr-1, the average motion speed of AFL footballers was 6.8 ± 0.9, 6.8 ± 0.9, 7.2 ± 0.8 and 7.3 ± 0.7km.hr-1 over seasons 2055, 2006, 2007 and 2008 respectively. Ruby and colleagues (2003) observed average velocities of 3.8 ± 0.5 and 2.9 ± 0.3 km.hr-1 in male and female American WFFs during a simulated loaded escape hike (Table 2.4). Average velocities of 2.3 ± 1.1 km.hr-1 at single day fire sites (Chapter 5) are likely to be representative of fire work without truck travel and conducted at significantly lower speeds than those observed during team sports, such as AFL football (Wiseby et al., 2009). With the exception of Chapter 5 of this thesis, GPS has not been used to observe the movement of BFFs during RTBS shifts.

Chapter 5 measured Australian RTBS work over a single day RTBS shift in a controlled burn or controlled edge of an emergency fire. The emergency nature of fire fighting work does not permit analysis of individual tasks or their context within sequential work bouts. Consequently no detailed comparison can be made between the frequency, duration and intensity of individual hose, rake and miscellaneous tasks between single day and campaign RTBS. The first aim of this chapter was to characterise campaign RTBS work. The second aim of this chapter was to compare intensity, activity data and speed data obtained during single day RTBS work (Chapter 5) and campaign RTBS work. Data obtained from GPS, activity and heart rate loggers were able to act as a control for comparison and extrapolation of observed behaviour in Chapter 5 to campaign shifts.

METHODS

Subjects. Thirty eight volunteer fire fighters, 5 females and 33 males, from the Country Fire Authority (CFA) and New South Wales Rural Fire Service (RFS) volunteered to participate in the study in conjunction with RTBS shifts during the 2006-07 bushfires. All participants were operationally accredited volunteer members of rural fire agencies. During all shifts, participants wore full personal protective equipment consisting of a two piece probane treated pant and jacket ensemble (Stewart & Heaton, Australia), leather fire resistant gloves (Fire Rescue Safety Australia, Australia), treated leather

120

work boots (Taipan, Melbourne, Australia) and a BR3 BFF hard hat (Pacific Helmets, Australia). Participants carried and may have used fire safety goggles (UVEX, Feurth, Germany) and disposable breathing masks (Dräger, Lubek, Germany). Clothing worn under personal protective equipment was not recorded. The net weight of BFF equipment is approximately 5kg. Ethical and technical approval for the study was obtained from the University of Melbourne Human Research Ethics Committee and the Australasian Fire Authorities Council. Informed consent was obtained from each participant prior to the involvement in the study.

Figure 6.1, Map of campaign fire data collection sites; 1, Mugdee (N=6); 2, Jamieson (N=12); 3, Mt Buller Ski Resort (N=6); 4, Swifts Creek (N=6); 5, Bairnsdale (N=4); 6, Mansfield (N=8) (DSE fire incident map, accessed July 2009).

Participants in Chapter 5 were local BFFs, deployed when needed to local bushfires. It is likely they are representative of local bushfire crews throughout Australia and probably represent a slightly different operational resource compared to normal campaign BFFs (Chapter 6). Local crews are the primary and rapid response used to contain local bushfires as they are the only resources within a practical proximity of the bushfires origin. In contrast, campaign bushfires have usually evolved from a local fire incident and become too large to be contained by local resources (Bosua,

121

2006; Hunter, 2003). In order to fight a campaign fire, surplus resources are usually redeployed from surrounding areas or states depending on the level of immediate community risk (Bosua, 2006).

Equipment & data collection. Participants were recruited at regular pre-shift briefings prior to participating in their normal work shift during the 2006-07 bushfires. All fire fighters were crew members working from standard bushfire tanker trucks during the second or third day of a standard three day RTBS deployment. Data in this study was collected at the 2006-07 bushfires which occurred over 70 days and burned over 1.1 million hectares of terrain. Significant variations in the terrain, weather conditions, operating equipment and operational priorities were noted between participants in this study. The collective experiences of the participants in this study likely represent the typical range of fire deployment of Australian campaign bushfire for BFF crews.

Demographic data was obtained via verbal interview at the time of informed consent. Body mass and height were measured using standard body mass scales and portable stadiometer (SECA, Birmingham, United Kingdom). All devices recorded continuously for the duration of a shift and were removed by the experimenters as the participant returned to the pre-shift briefing area. Subjects were fitted with a GP Sports© SPI Elite global positioning system (GPSports, Canberra, Australia), worn in a harness mounted on between the shoulder blades. The GPS units used in this study are a small (91 X 45 X 21mm3) water resistant monitor with a built in tri-axial accelerometer. Differentially logged GPS co-ordinates downloaded from the devices at the end of each shift were converted into distance and velocity information using Team AMS software (GPSports, Canberra, AUS). For further analysis, distance and velocity information was exported into an Excel®(Microsoft, Redmond, Washington) spreadsheet for statistical analysis. A heart rate monitor (S410, Polar, FI), comprising of a wrist unit and elastic chest strap was also fitted. Heart rate and GPS data were taken from the GPS and sampled at one second epochs unless otherwise stated. Polar heart rate data was used as a secondary logging device where GPS data was unavailable due to battery power limitations or insufficient GPS signal.

Maximal heart rate and percentage of maximal heart rate (%HRmax) were determined using the predictive formula, HRmax = 208 – 0.7 x Age (Tanaka et al., 2001). An Actical® (Mini-Mitter, Bend, Oregon) activity monitor was attached to the most proximal non elastic segment of the chest mounted heart rate strap (Same procedure followed in Chapter 5). Activity monitors were not mounted as in previous research (Heil, 2002) in pockets as fire agency consultants believed it was critical free pocket space was preserved for storage of food, navigation tools and personal belongings. These monitors are a small (28 X 27 X 10mm3) uni-axial accelerometer which recorded one minute epochs for periods of up to 44 days. They have been validated against accurate methods of determining energy expenditure free living conditions in adults (Welk et al., 2004). Data obtained

122

during the shift was downloaded using a Mini-Mitter Actireader® (Mini-mitter, Bend, Oregon) onto Actical Software (Mini-Mitter, Bend, Oregon). For further statistical analysis, activity data was exported into a Microsoft Excel® spreadsheet (Microsoft, Redmond, Washington).

Analysis. Twelve subjects declined to volunteer their years of service, height and body mass information which is subsequently reflected separately in the analysis. Two fire fighters withdrew during the study and their data was withdrawn from the study (one Mudgee and one Jamieson). GPS speed data was expressed as average for each two hour block during a shift and for each overall shift. GPS distance data was expressed as the cumulative distance covered on foot and on tanker (speeds over 8km.hr-1) for each two hour block and overall during the shift. 8 km.hr-1 was selected as cut off speed which would likely indicate participants were running rather than walking (Zamparo et al., 1992). Fire fighters are not allowed to run on fire sites and it was therefore considered speeds of this magnitude would solely represent truck travel. Heart rates were divided into American College of Sports Medicine (ACSM) intensity classifications (Balady et al., 1998); very light (<30% HR max), light (30 - 49% HR max), moderate (50 - 69% HR max), hard (70 – 89% HR max), very hard (>90% HR max) and Maximal (~100% HR max). Time spent in ACSM classifications and average raw HR were expressed relative to overall shift length and by two hour blocks during each shift. Activity counts were divided into three ranges specified by Actical® (Mini-Mitter, Bend, Oregon) software; sedentary (0-99 counts·min-1), light (100-1499 counts·min-1), and moderate/vigorous (1500 counts·min-1). The time spent in the three different intensities, total counts and average counts were compiled for overall shift time and for each two hour block during each shift.

Statistical Analysis. All statistics were derived using the statistical analysis program SPSS (IBM, Chicago, Illinois). One way ANOVA were performed on Figure 6.2, Figure 6.6 and Figure 6.1. A Tukey test was utilised post hoc to determine significant differences between groups. Comparative analysis between Chapter 5 and 6 was performed using three-way mixed ANOVA with the factors fire Location, Time, Intensity classification and for between or within participants. T-test with Bonferroni was used post hoc to determine difference between groups. Mixed ANOVA was conducted on remaining Figures using between subjects for fire location, within-subjects for time, activity and intensity classifications where appropriate. A Tukey test was utilised post hoc to determine significant differences between groups. Graphs are shown as Mean ± SEM unless otherwise stated. Differences were considered statistically significant if P<0.05.

RESULTS

Participants.

123

Age and years of service information is displayed in Table 6.1. Operational fire fighters averaged 37.5 ± 12.7 years of age and 11.2 ± 9.3 years of service. No significant difference was observed in single day participants regarding years of age and service (Chapter 5) (P>0.05). Campaign fire fighters were characterised by height, body mass and BMI values of 1.7 ± 0.1m, 76.1 ± 10.9kg and 25.1 ± 3.3 kg.m-2 respectively. Analogous characteristics in fire fighters working on single day fire sites (Chapter 5) for height and BMI of 1.7 ± 0.1m and 30.4 ± 4.8 kg.m-2 reflected no difference respectively (P>0.05). Fire fighters at single day fire sites (Chapter 5) were heavier than fire fighters at campaign fires (P<0.05).

Activity Count

Mean activity count, Peak activity count and Whole shift cumulative activity count are displayed in Figure 6.2 and Figure 6.3. Higher mean and lower peak activity counts were observed during single day fire work (Chapter 5) when compared to campaign fire sites (P<0.05). No differences was observed when compared to whole shift cumulative activity counts noted in single day fire sites (P>0.05). Mean activity count, Maximal activity count and Whole shift cumulative activity counts are displayed over two hour segments of fire shift in Table 6.3. No difference was observed between two hour mean activity count of campaign fire sites and single day fire sites. Cumulative counts in were lower in the 10-12 hour time period when compared to 2-4 hour period for campaign fire sites and the 0-2, 2-4, and 4-6 hour period of single day fire sites (P<0.05).

Activity Intensity Classification

Mean activity intensity classification as a time portion of the whole shift is displayed in Figure 6.4. Mean percentage of time spent in Sedentary, Light and Moderate activity zones at campaign fire incidents was observed as 64.0 ± 15.2, 32.6 ± 13.7 and 3.4 ± 3.8%; and at single day fire sites (Chapter 5) as 47.2 ± 13.2, 49.5 ± 11.2 and 3.3 ± 4.2% respectively. Main effects were observed for activity classification*fire location and activity classification but not fire location (activity classification*fire location; P=0.00, activity classification; P = 0.00, fire location; P = 0.99). During campaign fire suppression significantly more time is spent in Sedentary when compared to Light and Moderate activity ranges; and more time is spent in Light when compared to Moderate activity ranges (Figure 6.4) (P<0.05). When single day fire RTBS work (Chapter 5) is compared to campaign RTBS work, less time is spent working in sedentary activity ranges and more time is spent working in light activity ranges (P<0.05). The portion of time spent in Moderate activity ranges between single day (Chapter 5) and campaign RTBS work (Figure 6.4) is unchanged (p>0.05).

Mean activity intensity classification as a time portion of the whole shift is displayed over subsequent two hour segments in Table 6.5. Main effects were observed for time*activity

124

classification*fire location and activity classification but not fire location and time (time*activity classification*fire location; P=0.00, intensity classification; P = 0.0, fire location; P = 0.70, time; P = 0.98). Within a campaign fire shift longer portions of time were spent in Sedentary and Light activity ranges when compared to moderate activity ranges (P<0.05). In all 2hr time segments, time spent in sedentary ranges was longer than time spent in light ranges (P<0.05). Single day fire sites exhibited less time spent in sedentary activity ranges and more time spent in light activity ranges than campaign fire sites (P<0.05). There was no observed difference of time spent in moderate activity ranges between campaign and single day fire sites (P>0.05).

Heart rate

Mean and peak (absolute and relative to age predicted Maximal) are displayed in Figure 6.6. Mean heart rate during campaign RTBS work was observed as 99.0 ± 17.9 b.min-1 and was not different from heart rates observed at single day fire sites of 108.8 ± 19.7 b.min-1 (P>0.05). Similarly no significant difference in relative mean heart rate between single day and campaign fire sites as

53.6 ± 10.7 and 59.9 ± 10.7% HRmax was observed respectively (P>0.05). Peak heart rate at campaign -1 fire sites was observed as 167.8± 30.2 b.min which corresponded to 91.0± 19.1% HRmax. Similar peak heart rates were observed during single day RTBS work (Chapter 5).

Mean, peak, mean heart rate (absolute and relative to age predicted Maximal) were observed over subsequent two hour segments at single day fire sites in Figure 6.7. Main effects for mean heart rate were not observed for group*time*location and time (group*time*location; P=0.2, time; P = 0.6). Main effects for peak heart rate were not observed for group*time*location and time (group*time*location; P=0.2, time; P = 0.6). Main effects for mean relative heart rate were not observed for time*location and time (time*location; P=0.13, time; P = 0.9). Main effects for peak relative heart rate were not observed for time*location and time (time*location; P=0.64, time; P = 0.10). No differences were observed between mean or peak heart rate (in both absolute and relative terms) during single day fire sites (chapter 5) when compared to campaign fire sites (p>0.05).

ACSM heart rate classification

The mean proportion of time spent in ACSM intensity classifications during campaign and single day fire sites are displayed in Figure 6.8. Main effects were observed for ACSM classification and ACSM classification*fire location but not fire location (ACSM classification; P = 0.00, Location; P = 0.97; ACSM classification*location; P = 0.00). During campaign fires, less time was spent in Hard and Very hard ACSM classifications when compared to Light or Moderate ACSM classifications (P<0.05). No difference was observed in the portion of time spent in Light or Moderate ACSM

125

classifications (P>0.05). Campaign fire sites contained a significantly higher portion of time spent in light ACSM classifications than single day fire sites (P<0.05).

126

Table 6.1: Descriptive data for the sample of Australian campaign BFFs studied during 2006-07 fire season.

Fire Incident

Mudgee Jamieson Mt Buller Swifts Creek Bairnsdale Mansfield Overall (NSW) (Vic) (Vic) (Vic) (Vic) (Vic)

Experimental Period Dec 2nd - 3rd Dec 11th - 14th Dec 20th - 21st Jan 10th - 11th Jan 16th Jan 20th - 21st Dec 2006 to Jan 2007

Number 6 12 6 6 4 4 38

Males : Females 5 : 1 10 : 2 5 : 1 6 : 0 3 : 1 4 : 0 33 : 5

Age (yr) 37.0 ± 5.2 39.0 ± 3.5 43.8 ± 5.2 30.3 ± 5.5 41.5 ± 6.9 30.8 ± 5.9 37.5 ± 2.1

Length of N / A N / A 14.3 ± 5.5 9.0 ± 3.6 11.3 ± 3.4 9.5 ± 3.8 11.2 ± 2.1 Service (yr)

Height (m) N / A N / A 1.8 ± 0.0 1.7 ± 0.0 1.7 ± 0.0 1.8 ± 0.0 1.7 ± 0.0

Body Mass (kg) N / A N / A 78.3 ± 4.8 76.3 ± 3.7 67.5 ± 5.7 81.3 ± 5.2 76.1 ± 2.4

BMI (kg·m-2) N / A N / A 25.5 ± 1.4 25.4 ± 1.1 24.1 ± 0.2 25.0 ± 1.7 25.1 ± 0.8

All data presented as means ± SEM; NSW, New South Wales; Vic, Victoria; yr, years; m, metres; kg, kilograms; BMI, Body Mass Index; m, meters. Note: Length of Service, Height, Body Mass, and BMI means are based on 24 participants.

127

1000 7500 b 6250 800 Activity Counts Activity (counts.min )

-1 5000 600 a 3750 400 -1

2500 ) (counts.min Activity Counts

200 1250

0 0 AverageSD AverageC MaxSD MaxC

Fire locations

Figure 6.2. Mean Activity counts during campaign bushfire shifts.

All data are means ± standard error of the mean (N = 32C & 41SD). C, campaign fire observations; SD, Single day fire observations; a, P<0.05 average countSD vs. average countcampaign; b, P<0.05 max countSD vs. max countcampaign;

128

45000 37500 a 30000 22500 b

) 15000 -1 7500 3500 3000 2500 2000 1500 (counts.min Activity Counts Activity 400 300 200 100 0-2 2-4 4-6 6-8 8-10 10-12 12-14

Time (2hr.min -1)

Cummulative count SD Peak count SD Average counts SD Cummulative count C Peak count C Average counts C

Figure 6.3. Mean Activity intensity distribution over two hour periods during campaign and single day bushfire shifts.

All data are means ± standard error of the mean. C, campaign fire observations; SD, Single day fire observations; a P<0.05 Cumulative countSD 0-2 vs.

Cumulative countSD 4-6; b P<0.05 Cumulative countC 10-12 vs. Cumulative countC 2-4, Cumulative countSD 0-2, Cumulative countSD 2-4 and Cummulative countSD 4-6.

129

ab 100

75

Sedentaryc 50 Lightd Moderate

Percentage of Shift (%) Shift of Percentage 25

0 Single Day Campaign

Fire Type

Figure 6.4, Mean Activity intensity distribution during campaign bushfire shifts.

All data are means ± standard error of the mean (N = 32C & 41SD). a, P<0.05 Moderate vs. Sedentary & Light; b, P<0.05 Sedentary vs. Moderate & Light, Light vs. Moderate; c, P<0.05 Sedentary Single day vs. Sedentary Campaign; d, P<0.05 Light Single day vs. Light Campaign.

130

a b c 100

80

60

40

20 Percentage of Shift (%) Shift of Percentage 0 0-2C 0-2SD 2-4C 2-4SD 4-6C 4-6SD Sedentary Time (hrs) Light Moderate

Figure 6.5, Mean Activity intensity distribution over two hour periods during campaign bushfire shifts.

All data are means ± standard error of the mean (N = 42). a, P<0.05 0-2SD Sedentary vs. 0-2, and 4-6C Sedentary, 0-2SD Light vs. 0-2, 4-6C Light; b, P<0.05 2-4SD

Sedentary vs. 4-6C Sedentary, 2-4SD Light vs. 2-4 and 4-6C Light; c, P<0.05 4-6SD Light vs. 2-4 and 4-6C Light.

131

The mean percentage of time spent in ACSM intensity classifications at campaign and single day fire sites over consecutive two hour segments are displayed in Figure 6.9. Main effects were observed for ACSM classification and interactions between fire location*time*ACSM classification but not fire location and time (ACSM classification; P = 0.00, fire location*time*ACSM classification; P = 0.01, fire location; P = 0.9, time; P = 1.00). More time at campaign than single day fire sites was spent in Moderate when compared to Hard or Very Hard ACSM classifications (P<0.05) and less time was spent in Very Hard when compared to Light, Hard or Moderate ACSM classifications (P<0.05).

Speed

Mean speed of campaign and single day fire sited is shown in Figure 6.10. Mean time spent on truck determined by speed at campaign and single day fire sites is shown in Figure 6.11. Percentage of shift spent travelling in truck or speeds above 8 km.hr-1 at campaign fire sites was observed as 20.4 ± 15.6% and speeds below 8km/hr as 79.6 ± 15.6%. No main effects were observed for fire location, speed classification or fire location*speed classification (location; P = 0.2, speed classification; P = 0.89, location*Speed classification; P = 0.96).

Average percentage of shift time spent in selected 2km.hr-1 speed increments at campaign and single day fire sites in shown in Figure 6.12. Main effects were observed for fire location, speed classification and fire location*speed classification (location; P = 0.1, speed classification; P = 0.06, location*Speed classification; P = 0.05). During Campaign and single day fire shifts more time was spent in 0-1.9 km.hr-1 speed classification when compared to 2 - 3.9, 4 – 5.9 and 6 – 7.9 km.hr-1 speed classification; greater periods of the shift were also spent in the 2-3.9 and 4-5.9 km.hr-1 speed classification compared to the 6 – 7.9 km.hr-1 speed classification (P<0.05). Less time was spent in 0- 1.9 km.hr-1 speed classification at campaign fire sites when compared to single day fire sites (P<0.05). When compared to single day fire sites, more time at campaign fire sites was spent in a 6 – 7.9 km.hr-1 speed classification (P<0.05).

132

200 100

175 ) -1

150 % HR 75

125 max

100 50 Heart Rate (b.min Rate Heart

75

50 25

C C C C SD SD

Peak HR Average HR Average HR Peak %HRmax Average %HRmax Average %HRmax

Figure 6.6, Mean and peak (absolute and relative to age predicted Maximal) heart rate during campaign and single day bushfire shifts.

-1 All data are means ± standard error of the mean. C, campaign fire observations; SD, Single day fire observations; b.min , beats per minute, %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1);

133

200 90

175 80 ) -1 % HR 150 70 max

125 60 Heart Rate (b.min Rate Heart 50 100

40 0-2 2-4 4-6 6-8 8-10 10-12 12-14 Time (2hr.min -1)

Average HR C Peak HR C Average %HR C Peak %HR C

Average HR SD Peak HR SD Average %HR SD Peak %HR SD

Figure 6.7, Heart rate comparisons between single day (N= 26) and campaign fire sites (N= 32).

All data are means ± standard error of the mean. C, campaign fire observations; SD, Single day fire observations; HR, Heart rate; %HRmax, percentage heart rate of age predicted Maximal (Tanaka et al., 2001); Peak HR, Peak heart rate (b.min-1); No significant differences within heart rate groups.

134

a b 100 c Light (30 - 49%) 80 Moderate (50 - 69%) Hard (70 - 89%) 60 Very Hard (90+%)

40

Percentage of Shift (%) Shift of Percentage 20

Campaign Single Day

Fire Type

Figure 6.8, Mean ACSM classification distribution during campaign bushfire shifts.

All data are means ± standard error of the mean. ACSM Classification, Heart rate intensity classifications (Balady et al., 1998); a, P<0.05 Very Hard vs. Light & Moderate, Hard vs. Light & Moderate; b, P<0.05 Light vs. Moderate & Very Hard, Moderate vs. Very Hard, Hard vs. Moderate & Very Hard; c, P<0.05

Light Campaign vs. Light Single day.

135

bcd 100

80 Light (30 - 49%) a 60 Moderate (50 - 69%) Hard (70 - 89%) 40 Very Hard (90+%)

20 Percentage of Shift (%) Shift of Percentage

0-2C 0-2SD 2-4C 2-4SD 4-6C 4-6SD

Time (2hr.min-1)

Figure 6.9, Comparison of mean ACSM classification distribution during the first six hour period of campaign and single day bushfire shifts.

All data are means ± standard error of the mean. C, campaign fire observations; SD, Single day fire observations; L, Light; M, Moderate; H, Heavy; VH, Very Heavy; a P<0.05 Moderate vs. Hard & Very Hard; b P<0.05 0-2C L vs. 0-2SD L, 0-2C H, 0-2 SD H & 0-2 SD VH; 0-2 C M vs. 0-2 SD H & 0-2 SD VH; c P<0.05 2-4 C L vs. 2-4 SD L, 2-4 C H, 2-4 SD H, 2-4 C VH & 2-4 SD VH; 2-4 C M vs. 2-4 SD M, 2-4 SD H & 2-4 SD VH; d P<0.05 4-6 SD L vs 4-6 C VH, 4-6 SD VH, 4-6 SD M & 4-6 C M; 4-6 SD M vs 4-6 C H, 4-6 SD H, 4-6 SD VH & 4-6 C VH.

136

10 Likely truck travel

) 8 -1

6 4 *

Speed (km.hrSpeed 2

0 Campaign Single Day

Fire Type

Figure 6.10, Mean speed recorded during single day fire work of all fire sites.

All data are means ± standard error of the mean; km/hr, kilometres per hour; * P<0.05 Mean Speed Campaign vs mean Speed Single day.

137

a

100 x  8 km.hr-1 80 x  7.9 km.hr -1

60

40 Time spent in spent Time

Speed Zones (%) Zones Speed 20

Campaign Single Day

Fire Locations

Figure 6.11, Mean time spent on truck Vs. Time spent off truck recorded during campaign and single day fire work.

All data are means ± standard error of the mean. km.hr-1, kilometres per hour; a P<0.05 Mean time spent at speed 8 km.hr-1 vs. Mean time spent in speed 8 km.hr-1.

138

a 100

80 b 0 - 1.9 km.hr-1 -1 60 2 - 3.9 km.hr 4 - 5.9 km.hr-1 c 6 - 7.9 km.hr-1 40 Time spent in spent Time Speed Zones (%) 20

Campaign Single Day Fire Locations

Figure 6.12, Mean time spent in speed classifications recorded during single day fire work of all fire sites.

All data are means ± standard error of the mean. km.hr-1, kilometres per hour; a P<0.05 Mean time spent in speed 0-1.9 km.hr-1 Campaign vs. Mean time spent in speed 0-1.9 km.hr-1 Single day; b, P<0.05 Mean time spent in speed 0-1.9 km.hr-1 vs. Mean time spent in speeds 2-3.9, 4-5.9 and 6-7.9 km.hr-1, Mean time spent in speed 2-3.9 km.hr-1 vs. Mean time spent in speed 6-7.9 km.hr-1, Mean time spent in speed 4-5.9 km.hr-1 vs. Mean time spent in speed 6- 7.9 km.hr-1 ; c, P<0.05 Mean time spent in speed 6-7.9 km.hr-1 Campaign vs. Mean time spent in speed 6-7.9 km.hr-1 Single day.

139

DISCUSSION

Characterisation of work demands

The first aim of this study was to characterise the physical demands of campaign RTBS. Despite a low mean heart rate of 99.0 ± 17.9 b.min-1, campaign fire work regularly generated high relative peak cardiovascular loads equivalent to 91.0± 19.1% HRmax (Figure 6.7). Analysis of the shift suggested 85% of all time on the fire ground is spent in light or moderate ACSM heart rate classifications (Figure 6.8) suggesting campaign TBS work involves significant periods with low relative stress place on the cardiovascular system. Relative observations of heart rate similarly suggest campaign BFFs average between 50 and 55%HRmax for the majority of their bushfire suppression shift (Figure 6.8). Observations of mean activity count indicate campaign BFFs conduct around 95% of the shift in sedentary or light activity ranges (Cuddy et al., 2007). Observed mean activity count suggest campaign BFFs complete their work in activity ranges close to the sedentary intensity cut off (Figure 6.4). Peak heart rates in excess of 70% HRmax were observed throughout the campaign work shift.

Although campaign fire suppression is characterised by a large portion of low cardiovascular demand, low activity, this work is also characterised by small but consistent portions higher cardiovascular demand and activity. Heart rate values observed in campaign

BFFs regularly peaked over 70%HRmax in every two hour segment of a twelve hour shift (Figure 6.8). Around 10% of the shift was spent in hard or very ACSM heart rate intensities across the whole shift (Figure 6.9) or during sequential two hour portions of the shift (Figure 6.10). Given McLennan (2004) observed 55% of Australian operational BFF personnel are older than 40 years of age, an observation consistent with participants in this study, the regularly encountered higher intensity activity and relatively high ACSM heart rate classifications provide a strong argument for the inclusion of PES in this occupation.

Similarly to Lusa et al., (1994), this study observed consistent exposure to higher absolute and relative physical demands regardless of role, fire or age. It is reasonable for any rural fire agency to expect all fire fighters may be exposed to physical/ more arduous work throughout any RTBS shift, 3.4 ± 3.8 % or 1.0 ± 2.0% by moderate activity count or very hard ACSM heart rate classification respectively (Figure 6.5; Figure 6.9). Estimates of the amount of more physically demanding work using these data suggest fire fighting equates to 28.6 ± 31.9 or

140

8.4 ± 16.8 minutes by activity count and heart rate respectively for every 14 hour shift. When hard and very hard ACSM heart rate classifications are included in this calculation the estimate of time spent in higher relative heart rates throughout a 14 hour increases to 98.3 ± 141.1 minutes (Figure 6.9). Cardiovascular intensities of this level may explain in some part the occurrence of over exertion related injuries on Australian fire grounds (ABS, 2004). The inclusion of a realistic PES to insure a more physically prepared fire fighter may assist Australian fire agencies in reducing the individual risks of exposure to high cardiovascular or other work demands (Mangan, 2002; Payne & Harvey, 2010; Cady et al., 1979; Cady et al., 1985).

Naughton (2006) suggested GPS provided accurate real time position and velocity of motion well in linear pathways of AFL footballers but not necessarily in circular pathways characteristic of team sports. Video analysis of RTBS work indicated non linear movement may provide a potential source of error in the data. Although movement pathways were outside the scope of this thesis, researchers did observe fire fighters may move non-linearly within a 180 degree range from the hose deployment area on the back of the truck and their movement can vary depending on the terrain they are operating on or task they are completing. When used in an Australian bushfire suppression environment GPS devices offer potentially accurate insights into the amount of time spent travelling in the truck, commuting between briefing areas, water sources and the fire ground work sites. GPS devices also offer potentially accurate insights into the speed of tasks and change in acceleration during task based simulations. As expected, GPS measurement accurately determined GPS may however be of limited use from a physiological standpoint during high sedentary work encountered during single day or campaign bushfire suppression. In assessing high risk, physically demanding occupations, which have high levels of sedentary movement and rely operationally on vehicles, a GPS mounted in an operational or tactical vehicle may be as useful as and far less obtrusive than mounted on the participant. The exception to this may occur during critically identified, higher movement tasks such as simulated escape from an approaching fire front (Ruby et al., 2003) in American hot shot WFFs and some Australian LMFFs.

Comparison of Single day and Campaign bushfire suppression

The second aim of this study was to compare the physiological demands of campaign RTBS to single day RTBS work using physiological data obtained from unobtrusive physiological monitoring devices; GPS, activity monitors and heart rate monitors. The emergency nature of RTBS work does not permit analysis of individual tasks or their context within sequential work

141

bouts. Consequently no detailed comparison can be made between the frequency, duration and intensity of individual hose, rake and miscellaneous tasks between single day and campaign RTBS. Developing a control group from unobtrusive physiological devices within high risk- physically demanding work environments, is important in order to evaluate the content validity of task based simulations, field simulations or SME ratings of the job demands. To the authors’ knowledge, the standard methodology of developing PES does not regularly incorporate evaluations of the content validity of work based simulations.

The mean absolute and relative heart rate values are comparable and exhibited few differences in single day and campaign fire suppression work (Figure 6.7). The mean absolute and relative heart rate values were also comparable between single day and campaign fire suppression work over subsequent two hour segments (Figure 6.8). Longer portions of the shift were spent in light ACSM heart rate classifications during campaign fire suppression (Figure 6. 9). Despite this difference, comparable portions of shift were spent in moderate, hard and very hard ACSM classifications between single day and campaign bushfire suppression. It is likely single day RTBS work generates absolute and relative heart rate values representative of the cardiovascular intensities encountered during campaign RTBS work. In this regard it is likely single day fire suppression work is an appropriate content valid simulation of the cardiovascular demands of campaign fire suppression work.

Campaign fire suppression work exhibited a lower average activity count than single day fire suppression (Figure 6.3). In contrast, campaign fire suppression generated higher peak activity counts (Figure 6.3). The difference in mean activity counts observed in single day and campaign of 304.3 ± 152.1 and 209.2 ± 104.3 counts.min-1 respectively is likely to have little practical physiological relevance. Both values recorded represent mean activity rates comparable to very light, near sedentary work (Cuddy, 2007; Minimitter© Actical software) over a two hour period. In addition, single day fire suppression work exhibited no difference in moderate activity counts, a lower portion of time spent in sedentary activity counts and longer time spent in light activity counts, over a full shift and two hour segments, when compared with campaign fire suppression work. In combination, these findings are likely to indicate single day fire suppression is consistently composed of more light activity work rather than more moderate activity work. Interestingly, despite some whole shift differences in mean and peak heart rate, campaign and single day fire suppression work generated similar average, peak and cumulative activity count during two hour segments. As a similar duration of time was spent in moderate activity counts during single day and campaign RTBS work (Figure 6.8), although higher peak

142

activity counts observed during campaign RTBS work (Figure 6.3). Some bouts of campaign RTBS work may generate higher activity but it is likely similarly active work bouts are faced less often, hence the lower mean activity count, during campaign RTBS work when compared to single day RTBS.

Observations in this study indicated a single day RTBS model accurately reflects the absolute and relative cardiovascular demands of campaign bushfire suppression. A single day fire suppression model may however over-predict the intensity of activity data at lower intensities when compared with campaign bushfire suppression. Given the substantial increase in mean average speed and higher portion of time spent over 8 km.hr-1 and in the 6-7 km.hr-1 range observed during campaign fire suppression work, it is likely higher sedentary levels observed at campaign fire sites are due to higher amounts of truck travel and the lack of physical movement encountered during this travel.

CONCLUSION

The current findings indicate campaign fire suppression exhibited a work profile similar to single day RTBS, predominantly sedentary activity, low relative cardiovascular demand work combined with brief periods of vigorous activity/ high relative cardiovascular demand work. Although no task based observations were carried out it is likely campaign RTBS shares an intermittent work profile comparable with single day RTBS. As expected, this study displayed similar mean maximal relative and absolute work intensities, represented by activity count and heart rate, which were comparable between single day and campaign RTBS. These observations support the use of a single day controlled perimeter burns model as a representative simulation of campaign bushfire work. In the absence of the ability to conduct detailed video and physiological response analysis in a campaign RTBS setting, single day controlled perimeter burns (Chapter 5) is likely to be the most content valid simulation. The single day fire model potentially over-distributes work from sedentary into light and moderate intensities when compared with campaign fire suppression. The higher volume of time spent riding on the fire truck in campaign fire shifts is likely to be the cause of lower average relative work rates and high portion of time spent in sedentary or inactive intensities.

Large scale, live fire work simulations have previously been used effectively to determine the work demands of Australian LMFF work (Budd et al., 1997a), WFF simulated escape (Ruby et al., 2003) and SFF work (Smith et al., 2001; Williams et al., 1996). From a methodological perspective, using advances in unobtrusive physiological monitoring devices such as GPS, activity monitors and heart rate monitors in high risk or emergency situations

143

(Chapter 6) as a basis for validation of a work simulation model (Chapter 5) may be useful in a number of other high risk, physically demanding occupations.

144

CHAPTER 7: DISCUSSION

The studies described in this thesis characterise the physical demands of Australian RTBS work. Additionally this thesis attempted to progress traditional job task analysis methodology by using unobtrusive physiological monitoring of a high risk, physically demanding occupation for comparison with SME ratings, task based simulations and a single day RTBS work simulations. This chapter brings together the major concepts of this thesis and discusses the practical considerations in further development of PES for Australian RTBS work.

Characterising Australian tanker based bushfire fighting.

Using a job task analysis this thesis quantified the work demands of simulated, single day and campaign RTBS work. Chapter 5 and 6 observed both types of work generated, despite small differences in time spent in activity intensity classifications, comparable cardiovascular stress. All fire ground work measured in this thesis displayed a work profile indicating the vast majority of RTBS work is conducted at low relative cardiovascular intensities and sedentary to light activity ranges. A smaller minority of work generates relatively higher cardiovascular stress and higher activity count in BFFs. Rather than occurring during specific portions of the shift, work of a higher relative demand appears to occur throughout a work shift. This thesis also collected physiological data and video analysis of work conducted on fuel reduction burning activities and controlled perimeters of bushfire incidents. Using this methodology it was not only possible to study the frequency, intensity and duration individual tasks but also to understand the context by which tasks occur on the fire ground. For the purpose of this thesis, tasks occurring in a sequential order, collectively aimed at achieving a uniform goal were termed a ‘work bout’.

Operational considerations, such as fire behaviour and the natural work patterns of their fire crews dictated participants were required to complete an average 111 tasks during their shift. Given the mean hose work bout is composed of 4.8 ± 7.2 tasks, an average RTBS shift is only composed of between 15 and 20 work bouts of RTBS work or 130 to 160 minutes of task based work distributed throughout a 14-hour shift. Over the same 14-hour shift, fire fighters are likely to be exposed to a cumulative duration between 30 and 90 minutes of higher relative activity count or ACSM heart rate. Given the low number of work bouts likely completed by a BFF it is understandable both single day and campaign RTBS work were characterised by observed low activity and low relative heart rate values. This finding also highlights the intermittent work profile of RTBS. Short duration, higher relative intensity work bouts have the

145

potential to occur at any time throughout the shift and are likely to have been the primary factor generating relatively higher cardiovascular stress and higher activity count in participants.

Video analysis of single day RTBS (Chapter 5) identified fire ground work comprises 87 and 13% hose and rake work respectively. 78% of all hose work involves the performance of support operator, blacking out work, and lateral repositioning tasks. In contrast, the most frequent rake task, Patrol and carry hand tool, accounted for only 8% of all fire ground work or 62% of all rake hoe work. Physiological measurement of task based simulations (Chapter 4) identified a number of tasks which demonstrated high relative aerobic energy and cardiovascular demands including; Full repositioning, Charged advance, Spot fire containment, Rake hoe during blacking out, Team line building and Solo line building tasks. Video analysis of Chapter 5 task performance and work bouts observed the majority of aforementioned tasks exhibited low frequency of occurrence, under one occurrence per shift, during a single day bushfire suppression model. A low frequency of occurrence during RTBS indicated incumbents may self selected work patterns or adapt operational practices to avoid physically demanding work.

Payne and Harvey (2010) suggested a representative job task analysis included both commonly performed job tasks and potentially less commonly performed but critical job tasks. SME in this thesis rated many of the highest work demands to occur so infrequently that employers may be unreasonable in assigning PES matched to less infrequent more critical tasks. Data recorded in chapter 5 and 6 confirmed these assessments, many rake work and emergency tasks are represented in the analysis but occur so infrequently that they are unlikely to be critical to the successful performance of the majority of work shifts. It is reasonable to expect an approach using infrequently occurring but physically demanding bushfire tasks to develop PES, risks over-weighting these tasks and potentially generating higher false negatives. In this occurrence, failed applicants could perform satisfactorily in the vast majority of fire work and only perform as false positives very infrequently. Reducing the staffing levels of fire agencies is likely to lead to increased operational occurrence of low frequency, high demand tasks for the remaining BFF or reduce the overall work capacity of the fire agency unnecessarily.

McFadyen et al., (1996) observed increased job performance in high fit wildland fire fighters when compared to low fit wildland fire fighters, each with O2max values of 55.7 ± 3.0 and 48.8 ± 3.0 mL.kg-1.min-1 respectively. High fit fire fighters completed more handline raked per hour (27%), hose carry (9%), hose deployment (14%) and worked at a higher initial and all day rake hoe suppression pace. Lusa and colleagues (1994) had previously observed the work

146

demands of fire ground work are consistent regardless of role, gender or age. Operational practice in Australian RTBS work may deliberately avoid physically demanding work due to high relative age of Australian operational BFFs (McLennan, 2004), who may have decreased fitness or work capacity (Swank et al., 2000; Horowitz and Montgomery, 1991) and greater risk of over- exertion related injury (Rodriguez and Eldridge, 2003; Fahy & LeBlanc, 2006; Fahy, 2005). Work rates at near maximal levels are unsustainable for 14 hour shift periods (Wu & Wang, 2002) and are likely to be significantly beyond their normal self pacing behaviour of BFFs. It is likely all Australian BFFs, regardless of role, gender or age, are exposed to intermittent short duration, higher relative intensity work bouts throughout a shift. It is also likely fitness plays a substantial role in mediating the magnitude of physiological stress imposed on the fire fighter (Budd et al., 1997a).

Due to the large population of BFF in Australia and disparity in living location, measurements of BFF fitness and force capabilities were outside the scope of this review. Observations from this thesis indicate experience incumbents high relative, near maximal physiological work rates when conducting commonly performed individual tasks or work bouts involved with RTBS (Chapter 4,5,6). Near-maximal relative heart rates during single day and campaign (Chapter 5 and 6) are likely to indicate BFF fitness levels are only slightly higher than the 25-30 mL.kg-1.min-1 aerobic levels observed during simulated work tasks work (Chapter 4).

Comparison of SME expert opinion, task based simulations, single day and campaign fire activity as a basis for a job task analysis

The standard methodology of job task analysis does not incorporate evaluations of the content validity provided by SME raters or of work based simulations used to further develop PES (Williford et al., 1999; Rhea et al., 2004; Davis et al., 1982; Sothmann et al., 2004; Arvey et al., 1992; Hughes et al., 1989; Nottrodt & Celentano, 1987; DeLorenzo-Green & Sharkey, 1995; Rayson, 1998; Rayson, 2000c; Sharkey, Rothwell & DeLorenzo-Green, 1994; Sharkey & Rothwell, 1996; Gledhill & Jamnik, 1992a). Practical factors such as the additional time and expense involved in data collection and processing compared with current methodology have likely limited methodological developments in accurately quantifying intensity, duration and composition of work bouts in physically demanding occupations.

This thesis has expanded on traditional content validation methodology to include novel evaluation of task frequency, duration and develop an understanding of how the sequential performance of tasks in a work bout achieves a targeted outcome. Direct observation, although

147

the most accurate means of conducting a job task analysis, is inaccurate and dangerous if attempted during high risk, physically demanding occupations. A large number of PES reflected in scientific literature rely heavily on levels of acceptable job performance determined by SME (Sothmann et al., 2004; Rayson, 1998; Arvey et al., 1992; Hughes et al., 1989; Nottrodt & Celentano, 1987).

This thesis attempted to improve on task analysis methodology by incorporating measurable content valid work simulations and evaluations of SME ratings in more controllable fire environments as a practical alternative to direct observation methods. The whole shift accuracy of the task analysis derived from a single day, relatively controlled RTBS environment was compared to a control group of BFFs at campaign RTBS work through a number of less qualitative portable physiological measuring devices worn at both fire sites. The accuracy of task based simulations was also compared against performance of comparable tasks in a controlled fire setting and physiological measurement of a control group of fire fighters at campaign fire sites. This methodology should provide additional validation of SME ratings and work simulations which often form the output of a job task analysis and are used to develop criterion measures in the development of a PES. The establishment of a PES using more stringent evaluation of content validation is an approach consistent with the Uniform Guidelines on Employee Selection Procedures (EEOC, 1978), a best practice guide which recommended utilising a variety of methods in combination to validate the development of a PES (Chapter 2).

Table 7.1 displayed a comparison of seasonal task frequency estimated by SME (Chapter 3) and extrapolated from task frequency at single day fire sites (Chapter 5). Although statistical comparison between these groups is problematic given measures are extrapolated, SME often indicated higher frequency in the occurrence of a full charge hose repositioning and to a lesser extent uncharged hose advance. In contrast, SME probably underestimated the seasonal frequency of tasks including hand or hose tool work during blacking out work, patrol whilst carrying a hand tool and charged hose advance. The author concedes the tasks likely to have been underestimated by SME in Table 7.1 are frequently performed during controlled perimeter work and fuel reduction burning but potential differences in seasonal frequency are unlikely to be operational between single day and campaign fires. All shift deployments in campaign fire incorporated blacking out work (Chapter 6) and this work is considered fundamental to a bushfire suppression strategy (Bosua, 2006; CFA, 2006).

148

Table 7.1, Subject matter expert (SME) perceived frequency per season of RTBS tasks V. Predicted frequency per season RTBS tasks extrapolated from single day (SD) fire observations.

Frequency estimateSME Frequency estimateSD Task (per season) (per season)

Uncharged 38mm hose advance onto fire break 34.5 ± 47.7 15.4

Uncharged 38mm hose advance into terrain 48.2 ± 72.4 16.8

Lateral charged repositioning of 38mm hose 523.9 ± 168.5 576.8

Full charged repositioning of 38mm hose 269.0 ± 197.2 40.6

Operating 25mm rubber delivery hose 25mm 93.7 ± 108.9 98

Hose work during blacking out activity 70.3 ± 51.8 229.6

Charged 38mm hose advance 23.4 ± 39.1 79.8

Manual hose retraction 12.4 ± 6.7 15.4

Solo rake work 4.2 ± 3.4 1.4

Rake during team line building 7.8 ± 16.6 4.2 70.2 ± 54.9 131.6 Rake work during blacking out 43.7 ± 48.1 226.8 Patrolling on foot whilst carrying hand tool

149

Average %HRmax task simulation 100 a Average %HRmax single day fire 90 b 80

max 70

%HR 60 50 20 0 SLB* TLB HC H Re FC FU

Figure 7.1, Comparison of average %HRmax observed during simulated tasks and at single day fires for the same tasks. All data are means ± standard error of the mean. SLB, solo line building with rake hoe; TLB, Team line building with rake hoe; HC, 38 mm Hose cranking; H

Re, 38mm Hose relocation; FC, Flat charged hose advance; FU, Flat uncharged hose advance; %HRmax (Tanaka et al., 2001); a P<0.05 Mean SLB average

%HRmax on the task simulation vs. mean average SLB %HRmax during single day fires; b P<0.05 Mean FC average %HRmax on the task simulation vs. mean average FC %HRmax during single day fires; SLB*, Average %HRmax single day fires SLB has been replaced with spot fire containment work from single day fires. Only one incidence of SLB was recorded at single day fires as fire fighters used spot fire containment rake work in preference to SLB work.

150

Figure 7.1 shows a comparison between observed relative heart rate intensity during the same tasks performed in a task simulation (Chapter 4) and single day RTBS (Chapter 5). Interestingly, the relative heart rate values were largely consistent between task simulations and task performance in operational single day fire work. This finding likely showed task simulations are representative of campaign fire ground work. Significantly higher relative heart rate values were recorded during flat charge advance and solo line building work when compared to operational comparisons. The average flat charged hose advance velocities of 5.8± 0.8 and 1.8± 1.1 observed during performance of task simulation (Chapter 4) and single day (Chapter 5) fire work respectively, indicated operational work is conducted at significantly lower speeds than observed in the task simulation (P<0.05). Task based video analysis observed hose work speeds of 1.0 ± 0.7 and 1.4 ± 1.6 km.hr-1 or rake work 1.1 ± 0.5 and 1.5 ± 1.1 km.hr-1 at Hobart and St Helens fire sites respectively (Chapter 5 and 6). No difference in velocity was observed during rake work tasks or hose work tasks during single day bushfire work.

Task based simulations derived higher GPS velocities of between 5.8 ± 0.8 and 2.8 ± 0.8 km.hr- 1 during hose tasks compared to 1.2 ± 0.1 and 0.8 ± 0.2 km.hr-1 during rake tasks. Task duration was observed as 192.0 ± 310.6 and 53.6 ± 12.6 seconds at single day fire sites and task based simulations respectively. Distance was not measured in this thesis due to the impractical and dangerous environment of single day fires but task distances can be estimated using speed and duration. Given the average hose work velocities at task based simulations and single day fire sites of 4.7 ± 1.5 and 1.2 ± 1.3 km.hr-1 respectively, hose tasks are likely to be conducted over distances of around 70 ± 5.3 and 64.0 ± 112.2 metres respectively. The 1Hz sampling rate of the GPSports GPS (GPSports, Canberra, Australia) is likely to be fast enough to accurately measure short duration, higher velocity task work common of team sport work (Wiseby et al., 2009) but not as accurate during the large amount of near sedentary, low distance, low velocity bushfire ground movement (Chapter 5 and 6). The task simulations from this thesis are only representative of the most physically demanding, shortest duration hose advance tasks and did not accurately represent the average or longer duration hose work. Further work needs to be conducted to characterise an accurate task based hose work simulation.

Similarly significantly higher velocities were observed in task simulations of flat uncharged hose advance and hose relocation tasks (Chapter 4) when compared to operational performance of these tasks (Chapter 5). These tasks produced similar average heart rates which showed the task simulation matched the physical demand of operational task performance more closely. Interestingly, rake hoe work produced consistent and comparable velocities throughout simulated and operational observations. The differences in heart rate in the performance of rake work likely

151

reflect differences in work productivity, i.e. area raked by the operator was only partially measured by velocity. Given the variability in fire ground performance of each task due to operational demands, simulations of isolated RTBS tasks are valid physical representations of RTBS work. Flat charged hose advance work is likely to represent the exception to this rule as task simulations (Chapter 4) were performed at an unrealistically high velocity and intensity. Operational performance of flat charged hose advance activities may require lower oxygen consumption rates than recorded during task simulations (Chapter 4).

Figure 7.2 displays the perceived SME task durations (Chapter 3) compared with task durations observed during task based simulations (Chapter 4) and single day bushfires (Chapter 5). SME experts overestimated task durations of all tasks when compared with observed durations during task based simulations or single day fires (Figure 7.2). The duration of tasks observed during work based simulations and single day bushfire suppression was comparable with the exception of 38mm Hose relocation task, which was shorter by 40 seconds in an operational context. These data are likely to indicate task based simulations were a more accurate reflection of operational task duration than SME ratings. This inaccuracy may occur as SME may be referring to task performance as part of a larger work bout. The average hose work bout has an operational duration of 288 ± 432 seconds (Figure 5.3), substantially closer in duration to the estimations of the SME than observed bushfire ground values.

Even if practical factors limits further development of research methodology developed used in this thesis, best practice for the PES needs to evaluate the duration of work intensity faced during high intensity tasks and how numerous sequential work tasks interact to produce the cumulative and sustained work bouts. In Australia, the same legislative arguments theoretically apply to the development of PES in terms of composition and duration of work, as well as intensity. Best practice from a scientific and legal perspective should focus on accurately and reasonably reflecting all critical content of the work conducted, including work intensity, duration and work bout composition. Process advancements in the job task analysis process (as trialled in this thesis) which enable more accurate quantification of job demands, coupled with a greater understanding of legislative differences, are likely to lead to a lower rate of false positives during job fitness testing. Consequently, when used by fire agencies, lower false positives are likely to direct translate into the operational advantage of more fit personnel and greater overall operational capacity.

152

2750 2500 2250 2000 Single day RTBS 1750 a 1500 SME rating 1250 Simulated task 1000 750 500 250

Duration (sec) Duration 100

50 b

0

38 BOut 38 FC 38 FU 38 HRe R BOut 38 HC P& C 25 OP

Figure 7.2, SME rated task duration of RTBS tasks (Chapter 3) V. task duration observed during task based simulations (Chapter 4) & single day fire observations (Chapter 5).

All data are means ± standard error of the mean. 38mm BOut, Blacking out with 38mm hose; FC, Flat charged hose advance; FU, Flat uncharged hose advance; H Re, 38mm full and lateral Hose relocation; R Bout, Blacking out with rake hoe tool; HC, 38 mm Hose cranking; P& C, patrol on foot and carry rake hoe; 25mmOP, Operating 25mm hose; a P<0.05 Mean SME vs. SD & SIM duration; b P<0.05 Mean SD 38mm HRe vs. SIM 38mm HRe duration.

153

OHS considerations with further development of PES for RTBS work

Chapter 1 discussed the implementation of PES as a practical compromise between OHS law and employment discrimination law. Discrimination is exempt from Australian employment law if an individual has impairment or physical features deemed reasonable necessary to protect the health or safety of any individual, member of the public or their property (NSW Anti-Discrimination Act, 1977; QLD Anti-Discrimination Act, 1991; Victorian Equal Opportunity Act, 1995; Tasmanian Anti- Discrimination Act, 1998; WA Equal Opportunity Act, 1984; SA Equal Opportunity Act, 1984). The development of PES in Australia must therefore target the practicable and necessary protection of an individual’s wellbeing at a work site, and additionally in the case of emergency services personnel, the practicable and necessary protection of members of the public or their property. This thesis has generated a number of important considerations for the development of PES in Australian RTBS personnel.

Hughes and colleagues (1989) stated it was “realistically not possible to expect a worker to be fully prepared to meet all physical emergency situations” but instead suggested workers who successfully meet PES have “adequate capacity to meet most of the strenuous physical demands experienced, as well as have a substantial reserve capacity when meeting routine physical requirements of the job”. Australian authors have reflected these statements, defining a job analysis as identification of “the physical domains which exist within the trade tasks, and which may reasonably be expected to be performed by personnel within the trade” (Taylor & Groeller, 2003). These statements highlight the balance between scientific validity or best practice and the reasonable workplace viability of PES which ultimately determines how employers develop PES for their workplace in Australia.

As discussed earlier in this chapter, SME are used as a gold standard in the development of levels of acceptable job performance. The rating of SME experts includes a component of subjective assessed productivity, a level in which the experienced incumbent finds the level of productivity acceptable. The assessment has primarily been made to date in career incumbent samples (Sothmann et al., 2004; Rayson, 1998; Arvey et al., 1992; Hughes et al., 1989; Nottrodt & Celentano, 1987) and is likely to incorporate a level of productivity given expected standards for ongoing employment (Payne & Harvey, 2010). In Australia, the implementation of PES enables employers to set a level of productivity based on “reasonable” expectations of employee work behaviour, beyond those required to simply maintain safety and indicative of economic considerations (Commonwealth v HR&EO Commission, 1998).

154

When using volunteer incumbents there appears to be no scientific or legislative justification to developing PES at productivity levels higher than the rate at which individuals’ have a matched physical job capability and exhibit lower rates of job related injury. The degree to which volunteer populations can be held to a paid career PES is questionable and as yet contains no precedence in judicial rulings to this author’s knowledge. Legislative and operational forces are likely to dictate a volunteer, who is not remunerated for their services in any way, may not necessarily be held to any productivity based PES beyond a required minimum physical capability (Figure 7.3).

Figure 7.3, Speculative acceptable development range for Australian PES showing the range between a minimum PES (a decreased risk of job related injury) compared with reasonably practicable PES (subjectively determined productivity or profitability).

The degree of productivity generated by a volunteer BFF is unlikely to be as productive as a paid individual as they perform general fire suppression work less frequent over a season (Swank et al., 2000; Bosua, 2006), may avoid many physically demanding tasks (Chapter 5), and have not passed a traditional PES set for career incumbents (Bosua, 2006). A PES designed specifically for volunteer populations is likely to be reflective of lower expected productivity and lower enforced physical capabilities. In contrast to current PES where productivity is a key driver, OHS concerns will be the primary factor motivating the development of volunteer PES.

155

Operational issues for fire agencies which can be informed by science

Australian volunteer fire fighters are an older sample within the community of personnel performing high risk, physically demanding occupations (McLennan, 2004). Although the sampling in this study did not facilitate direct comparison between older and younger participants, fire tasks appear to have produced similar maximal relative strain when compared to SFF, LMFF, military and fire suppression studies (Louhevaara et al., et al., 1986; Manning & Griggs, 1983; Gledhill & Jamnik, 1992a; Bennett et al., 1994; Bilzon et al., 2001a; Richmond et al., 2008; Smith et al., 2005). Despite similar relative work intensity faced during fire ground work, relatively few Australian BFF are killed on duty during operational activities compared to American WFFs (Burton, 2007; Washburn, 1996; Mangan, 1999; Fahy & LeBlanc, 2006; CFA annual report, 2008; TFS annual report, 2008). Australian fire fighting mortality rate is estimated at only 1.5 fire fighters per year over the past 15 years or 0.7 per 100,000 Australian BFF despite injury data from a single state suggesting seven fire fighters were killed and up to 400 fire fighters injured through heat stress, dehydration, smoke inhalation, cuts, strains and sprains during the 2002-2003 fire season (Aisbett, 2005; ABS, 2004).

Table 7.2, Pilot testing for the aerobic power of Western Australian fire fighters (Phillips et al., 2011). Variable SFF LMFF BFF P value N 22 20 18 Age (yrs) 35.4 ± 6.6* 39.9 ± 11.8 43.0 ± 12.4 -1 -1 O2 (mL.kg .min ) 53.4 ± 4.8* 43.4 ± 7.1 40.5 ± 6.3 <0.0 -1 O2 (L.min ) 4.4 ± 0.5* 3.7 ± 0.6 3.5 ± 0.7 <0.0

McFadyen et al., (1996) observed significantly improved job performance, including more handline raked per hour completed, hose deployed, hose carried and a higher sustained initial and all day sustained rake hoe pace between high and low fit LMFFs. Brotherhood and colleagues (1997b) hypothesized LMFFs self pace themselves at sustainable work rates which balance fire line productivity against relative physiological cost. This study recorded aerobic power levels of 47.0 ± 7.8 and 46.6 ± 6.7 mL.kg-1.min-1 for maximal raking and step test oxygen uptake respectively (Brotherhood et al., 1997a). Pilot testing displayed in Figure 7.2, despite a restrictively small sample shows LMFFs may have comparable aerobic power to volunteer BFF and LMFF observed in Brotherhood and colleagues (1997b) work. It is also likely Australian BFFs would also undertake physical work commensurate with their physical fitness.

Lower overall physical activity and injury rates encountered during RTBS work when compared with other styles of fire suppression may be due primarily to the operational use of a fire truck. Using

156

a truck to carry water and equipment likely results in significant reductions to many load bearing tasks, which are normally completed manually by LMFF (Cuddy et al., 2007). Conventional methods of fire suppression utilised by LMFF and WFF provide a less water intensive substitute for RTBS work such as knapsack sprays, hiking to a containment perimeter with gear and manual rake tool crews (Budd et al., 1997a; DSE, 2008). Preliminary evidence has been observed in this thesis which indicated the cyclic use of a fire truck, in addition to the operational advantageous of water carriage, may be a substantial mechanism by which BFF achieve rest during bushfire suppression shifts (Figure 6.12 and 6.13). Any decrease in physical work rate caused by the use of a fire truck as a piece of operational equipment is particularly important from an OHS and operational planning perspective.

Rural Fire agencies have consequently explored the prospect of adopting more dry LMFF techniques such as back-burning and rake hoe due to the poor water availability in many regions (Bosua, 2006). The CFA Training manual (2006) highlights Australian BFFs carried a finite source of water which when depleted required geographic relocation to a water source to replenish. Fire conditions will dictate the type of work conducted and consequently the amount of water used. Harder fire conditions will deplete the water storage capacity of the truck at a faster rate and will require more refills. SFF, by contrast historically have used a constant supply of water and do not frequently require geographic relocation to replenish water stocks. This thesis observed truck travel time of 20.4 ± 15.6 and 5.0 ± 4.0% during Campaign (Chapter 6) and Single Day (Chapter 5) fires. The refill time observed in this thesis is likely to create significant operational rest times within the shift, which reduce the level of prolonged physical strain to which any given fire fighter is exposed.

For the first time, the work demands of Australian RTBS work have been both characterised and quantified. A baseline record of physiological responses to campaign RTBS work has also been produced. The single day controlled RTBS shift combined with the range of deployed physiological monitoring devices provide a viable alternative for the development and validation of a range of OHS strategies, not achievable during campaign fire shifts. This work will assist not only in the development of PES but also in evaluating; effectiveness of operational sustenance, hydration strategies, fire clothing variation, ergonomic adaptation to fire trucks, operational deployment strategies, the effect of support appliances (bulldozers, aerial assests, interagency assistance) and the ongoing effects of ageing on the BFF workforce.

157

LIMITATIONS

A significant limitation throughout this thesis was the inability of fire agencies to organise advanced access to participants. A reduced access time with participants prior to any fire activity meant methodology was not able to be developed which could determine the resting heart rate or

O2max of participants involved in this thesis. Without resting heart rate or a O2max measurement it is difficult to predict individual relative oxygen demands (Bilzon et al., 2001), maximal acceptable work time (Wu & Wang, 2002; Bos et al., 2004) or measure a weighted cardiovascular load (Taylor &

Groeller, 2003) as has been completed for other industrial occupations. This study used O2max normative values derived from the Australian population (Gore & Edwards, 1992) as age matched means for comparison. The often remote location and large geographical distances of rural Australia made follow up fitness testing of participants, after participation, logistically impractical and was likely to have been perceived by politically -friendly elements within fire agencies as too socially obtrusive by a research team. Phillips and colleagues (2011) measured O2max in 21 Western

Australian volunteer fire fighters with an age of 44.5 ± 11.8. During a treadmill O2max test,

-1 -1 -1 volunteer fire fighters recorded a O2max of 40.5 ± 6.3 mL.kg .min or 3.2 ± 0.9 L.min . Gore and

Edwards (1992) observed the O2max of average 40-49 year old Australian male was 37.9 ± 8.5 mL.kg-1.min-1. For this purposes of this thesis, the Australian volunteer BFF population was assumed to be representative of age matched O2max values recorded in the broader Australian population. The mean relative aerobic demand of tasks performed by BFFs in this thesis was extrapolated from the O2max values observed in age matched members of the Australian population (Gore & Edwards, 1992).

Portable metabolic analysers were not utilised in field research due to the restrictions they place on BFF physical movement during tasks and a prioritisation to use resources on accessing greater numbers of participants. The Douglas Bag method for measuring oxygen consumption was used in preference as the majority of task based simulations employed in this study were comparable to those with similar methodology used in Budd et al., (1997c) or involved hose task simulations less than two minutes in duration.

The decision to conduct a validation of a single day controlled fire RTBS model (Chapter 5) came at the expense of a biomechanical analysis of the musculoskeletal demands of critical RTBS tasks or work bouts. A biomechanical analysis should be conducted to determine the physical demand of a number of high frequency critical tasks including support operator, blacking out work,

158

lateral repositioning and operating and patrol and carry hand tool. Additionally the same biomechanical analysis should be conducted on infrequent but physically demanding hose tasks including; team line building, spot fire containment and solo line building tasks. It is recommended the biomechanical analysis account for the positional and operational variations used in this thesis and preferentially should be performed where appropriate in a single day RTBS work shift.

Table 7.3, Number of participants sampled by gender and percentage of overall sample across Chapters 3 to 6.

Mean Age Male Female (yrs) Chapter 3 31 (100%) 0 (0%) 50.1 ± 11.6 Chapter 4 20 (77%) 6 (23%) 43.1 ± 14.6 Chapter 5 22 (79%) 6 (21%) 39.2 ± 14.7 Chapter 6 33 (86%) 5 (14%) 37.5 ± 12.7 Overall 106 (86%) 17 (14%)

Some observers may consider the low relative sampling of female subjects a significant limitation of the findings from this thesis. The author would argue the opposite is true. Australian courts have instead tended to rule toward content valid assessment of occupations which are considered objectively measurable and essential to performance of the employee (Cosma v Qantas Airways Limited, 2002; Qantas Airways Ltd v Christie, 1998), and should include analysis of normal operating conditions and less frequent “foreseeable circumstances” within an occupation (X v Commonwealth of Australia, 1999). Legislative compliance for the Australian fire agencies, in particular for volunteer fire fighters, is likely driven by ‘reasonably practicable’ and ‘reasonably necessary’ OHS measures aimed at the protection of a person on a work site rather than discrimination law compliance as has been the case in North America (EEOC, 1978; Jamnik et al., 2010c). As such the methodology for PES development from North America may have developed a focus on gender or age to avoid adverse impact litigation or demonstrate occupational requirement.

The methodology in this thesis has been designed to avoid adverse impact considerations and instead provide an objective, content valid assessment of RTBS work. Women represent roughly 17% or 35, 704 of the 206, 954 Australian RTBS personnel (McLennan, 2004). This study was only able to recruit women at a rate of 14% overall participation (Table 7.3), a small underrepresentation which can likely be mitigated through participant selection in the development of criterion tasks and minimum scores. Similarly the mean age reflected in Table 7.3, demonstrates the mean age of participants are comparable to McLennan (2004) estimate 55% of Australian operational BFF

159

personnel are older than 40 years of age. The author believes the representative global sample of this thesis as represented is a more content valid assessment of RTBS work than if further development was focused on gender or age.

FUTURE DIRECTIONS

Legislative compliance in Australian PES development is dependent on a number of factors beyond the identification and quantification of critical tasks. There seems to be little legislative protection for employers developing a PES based on critical tasks which are impractically infrequent in their occurrence. Australian PES instead may rely on accounting for the physical content of events considered to have a reasonable frequency of occurrence. As previously eluded to in Chapter 1, there is a substantial lack of judicial guidance concerning the acceptable level of appropriate physical demands analysis and subsequent PES development. Determination of acceptable performance standards based off critical tasks is scientifically problematic without practicable judicial guidance on what factors determine a critical task. In addition, guidance is required to determine best practice for developing PES as a reasonable balance between:

- infrequent but critically rated and less arduous regularly occurring bushfire suppression tasks

- safety and productivity during performance of bushfire suppression tasks

- the job performance requirements of volunteer and career fire fighters

A high degree of work variability observed in both campaign and single day fire sites is one of the principal problems facing the development of PES for Australian RTBS work. In many cases, particularly when using heart rate and activity count as physiological measures of work demands, the standard deviation of observed values is almost the same magnitude as average values obtained. SME (Chapter 3) provided similar insight into acceptable performance standards of work conducted on the fire ground by indicated acceptable work was a variable dependent on the task or work to be completed. Given the substantial variance in the physiological responses of tasks and work, identifying the effects of efficiency or experience on individual task performance, gender differences and age factors may be unobtainable using this methodology.

Variation in the tasks may also be due to a significant self pacing of performance, high terrain differences, low supervisor expectations of job productivity and the relative infrequency of highly important or critical tasks (Payne & Harvey, 2010). These factors would become substantially more important in emergency situations where work demands may exceed less capable fire fighters self selected work pace and require higher levels of time critical productivity (Brotherhood et al., 1997b; McFadyen et al., 1996). Additionally large portions of RTBS tasks involving hose work require

160

sustained input from individual fire fighter within an integrated team performance. This thesis did not focus on individual rotation through job roles or factors which may determine the regularity by which tasks were performed within a fire crew.

Risk of fire ground injury data has been collected from SME during chapter 3. The data fell outside the scope of this thesis but will be important in future PES development. When referenced against OHS data from fire agency this data will provide a valuable insight into the accuracy of fire fighter or SME perceptions toward their risk of injury whilst performing individual fire ground tasks. Quantification of musculoskeletal load encountered during bushfire suppression tasks also fell outside the scope of this thesis but will be of critical importance in the future development of a RTBS PES.

The characterisation of hose tasks and hose work bouts should be used utilised as a reference of operational intensity, duration and frequency of occurrence during further development of PES for RTBS work. Further work in developing a PES should make reference of the aerobic demands associated with hose work (Chapter 4), the relative composition of hose tasks and work bouts (Chapter 5), and the intermittent work profile of operational work which elicits near- maximal physiological demand in incumbent BFFs (Chapter 5 & 6).

REFERENCES

1. Aasa, U., Barnekow-Bergkvist, M., Angquist, K. and Brulin, C. (2005). "Relationships between work-related factors and disorders in the neck-shoulder and low-back region among female and male ambulance personnel." Journal of Occupational Health 47(6): 481-489

2. Aasa, U., Angquist, K., & Barnekow-Bergkvist, M. (2008). The effects of a 1-year physical exercise programme on development of fatigue during a simulated ambulance work task. Ergonomics, 51(8), 1179-1194.

3. ABS (2004). Bushfires. Year Book Australia - Environment. Canberra, Australian Capital Territory, Australian Bureau of Statistics: 7.

4. ACT RFS (2007). ACT Rural Fire Service: Operating guideline A.4; RFS fitness assessment. Canberra, Australian Capital Territory.

5. AFAC (2002). AFAC guidelines for health and fitness monitoring of Australian fire and emergency service workers. East Melbourne, Australia, Australasian Fire Authorities Council.

161

6. Aisbett, B., Phillips, M., Sargeant, M., Gilbert, B. and Nichols, D. (2007). "Fighting with fire: How bushfire supression can impact on fire fighters' health". Australian Family Physician 38(12): 994-997.

7. Anti-Discrimination Act (1977) 204, New South Wales Consolidated Acts

8. Anti-Discrimination Act (1991) 204, Queensland Consolidated Acts

9. Anti-Discrimination Act (1998) 204, Tasmanian Consolidated Acts

10. Anton, D., Cook, T.M., Rosecrance, J.C. and Merlino, L.A. (2003). "Method for quantitavely assessing physical risk factors during variable noncyclic work". Scandinavian Journal of Work, Environment and Health 29(5): 354-362.

11. Arthur, W., Jr., Doverspike, D. & Barrett, G.V. (1996). "Development of a job-analysis based procedure for weighing content-related tests into a single test battery score". Personnel Psychology 49: 971-985.

12. Arvey, R.J. and Renz, G.L. (1992). "Fairness in the Selection of Employees". Journal of Business Ethics 11: 331-340.

13. Astrand, P., K. Rodahl, H. Dahl and S. Stromme (2003). Applied work physiology. Textbook of work physiology. P. Astrand, K. Rodahl, H. Dahl and S. Stromme. Champaign, Illinois, Human Kinetics: 525-535.

14. Australasian Council of State Emergency Services (ACSES) (2007) The value of volunteers in State Emergency Services. Accessed March 2009.

15. Balady, G. J., B. Chaitman, D. Driscoll, C. Foster, E. Froelicher, N. Gordon, R. Pate, J. Rippe and T. Bazzarre (1998). "Recommendations for cardiovascular screening, staffing, and emergency policies at health/fitness facilities." Circulation 97(22): 2283-2293.

16. Bouchard, D.R. and Trudeau, F. (2007). "Reliability of the assessment of the oxygen/heart rate relationship during a workday." Applied Ergonomics 38(5): 491-497.

17. Barnard, R. and H. Duncan (1975). "Heart rate and ecg responses of fire fighters." Journal of Occupational Medicine 17(4): 247-250.

18. Beekley, M.D. (2007). "Effects of heavy load carriage during constant-speed, simulated, road marching". Military Medicine 172(6): 592-595.

19. Bell, N.S., Mangione, T.W., Hemenway, D., Amoroso, P.J. and Jones, B.H. (2000). "High injury rates among female army trainees: A function of gender?" American Journal of Preventative Medicine 18(3S): 141-146.

162

20. Bennett, B. L., R. D. Hagan, G. Banta and F. Williams (1994). "Physiological responses during shipboard fire fighting." Aviat Space Environ Med 66(3): 225-31.

21. Berkowitz, P.M. & Muller-Bonanni, T. (Ed) (2006). International labor and employment law: A practical guide. ABA Publishing, American Bar Association.

22. Bilzon, J., E. Scarpello, C. Smith, N. Ravenhill and M. Rayson (2001a). "Characterization of the metabolic demands of simulated shipboard royal navy fire-fighting tasks." Ergonomics 44(8): 776-780.

23. Bilzon, J. L. J., A. J. Allsopp and M. J. Tipton (2001b). "Assessment of physical fitness for occupations encompassing load-carriage tasks." Occup Med (Lond) 51(5): 357-361.

24. Bos, J., E. Mol, B. Visser and M. Frings-Dresen (2004). "The physical demands upon (dutch) fire-fighters in relation to the Maximal acceptable energetic workload." Ergonomics 47(4): 446-460.

25. Bosua, V. Personal Interview with the State Operations Controller for the Country Fire Authority: Country Fire Authority Headquarters, Burwood, Australia. August, 2006

26. Bowers, C.A., Baker, D.P. and Salas, E. (1994). "Measuring the importance of teamwork: The reliability and validity of job/task analysis indices for team-training design". Military Psychology 6(4): 205-214.

27. Brage, S., Brage, N., Franks, P., Ekelund, U., Wong, M., Andersen, L., Froberg, K., & Wareham, N. (2004). Branched equation modeling of simultaneous accelerometry and heart rate monitoring improves estimate of directly measured physical activity energy expenditure. Journal of Applied Physiology, 96, 343-351.

28. British Columbia (Public Service Employee Relations Commission) v. British Columbia (Government and Services Employees Union) [1999]. 3 S.C.R. 3, referred to as the Meiorin case.

29. British Columbia (Superintendant of motor vehicles) v. British Columbia (Council of human rights) [1999]. 3 S.C.R. 868, referred to as the Grismer case.

30. Brotherhood, J., G. Budd, A. Hendrie, S. Jeffrey, F. Beasley, B. Costin, W. Zhien, M. Baker, N. Cheney and M. Dawson (1997a). "Project aquarius 3. Effects of work rate on the productivity, energy expenditure, and physiological responses of men building fireline with a rakehoe in dry eucalypt forest." International Journal of Wildland Fire 7(2): 87-98.

163

31. Brotherhood, J., G. Budd, A. Hendrie, S. Jeffrey, F. Beasley, B. Costin, W. Zhien, M. Baker, N. Cheney and M. Dawson (1997b). "Project aquarius 11. Effects of fitness, fatness, body size, and age on the energy expenditure, strain, and productivity of men suppressing wildland fires." International Journal of Wildland Fire 7(2): 181-199.

32. Brown, J., Derchak, A. and Coleman, R. (2007). A detailed physiologic analysis of the willand firefighter pack test: Improving safety through science. Elk Grove, California: Vivometrics Government services.

33. Budd, G., J. Brotherhood, A. Hendrie, N. Cheney and M. Dawson (1996). Safe and productive bushfire fighting with hand tools. Canberra, ACT, Australian Government Publishing Service.

34. Budd, G., J. Brotherhood, A. Hendrie, S. Jeffrey, F. Beasley, B. Costin, W. Zhien, M. Baker, N. Cheney and M. Dawson (1997a). "Project aquarius 1. Stress, strain, and productivity in men suppressing australian summer bushfires with hand tools: Background, objectives, and methods." International Journal of Wildland Fire 7(2): 69-76.

35. Budd, G., J. Brotherhood, A. Hendrie, S. Jeffrey, F. Beasley, B. Costin, W. Zhien, M. Baker, N. Cheney and M. Dawson (1997b). "Project aquarius 5. Activity distribution, energy expenditure, and productivity of men suppressing free-running wildland fires with hand tools." International Journal of Wildland Fire 7(2): 105-118.

36. Budd, G., J. Brotherhood, A. Hendrie, S. Jeffrey, F. Beasley, B. Costin, W. Zhien, M. Baker, N. Cheney and M. Dawson (1997c). "Project aquarius 7. Physiological and subjective responses of men suppressing wildland fires." International Journal of Wildland Fire 7(2): 133-144.

37. Budd, G., J. Brotherhood, A. Hendrie, S. JEffrey, F. Beasley, B. Costin, W. Zhien, M. Baker, N. Cheney and M. Dawson (1997d). "Project aquarius 9. Relative influence of job demands and personal factors on the energy expenditure, strain, and productivity of men suppressing wildland fires." International Journal of Wildland Fire 7(2): 159-166.

38. Burton, G. (2007). "How the United States is reducing its firefighter fatalities". The Australian Journal of Emergency Management. 22(2): 37-43.

39. Byrd, R. and M. Collins (1980). "Physiologic characteristics of fire fighters." American Corrective Therapy Journal 34(4): 106-9.

40. Cady, L.D., Bishoff, D.P., O'Connell, E.R., Thomas, P.C. and Allan J.H. (1979). "Strength and fitness and subsequent back injuries in firefighters". Journal of Occupational Medicine 21(4): 269-272.

164

41. Cady, L.D., Thomas, P.C. and Karwasky, R.J. (1985). "Program for increasing health and physical fitness of fire fighters". Journal of Occupational Medicine 27(2): 110-114.

42. Canadian Human Rights Act of 1977, c. 33, s. 1. Department of Justice, Canada.

43. Canadian Human Rights Commission (2007). Bona fide occupational requirements and bona fide justification under the Canadian Human Rights Act: The implications of Meiorin and Grismer. Ottawa, Ontario.

44. National Defence and the Canadian Forces (2006a). Universality of Service: DAOD 5023-0. Assistant Deputy Minister, Defence Administration Orders and Directives.

45. National Defence and the Canadian Forces (2006b). Minimum Operational Standards related to Universality of Service: DAOD 5023-1. Assistant Deputy Minister, Defence Administration Orders and Directives.

46. Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C. 2000(e).

47. Country Fire Authority (2002). Wildfire firefighter - learning manual. Burwood East, Victoria, Australia, Learning Systems, Human Resource Development and Planning, CFA Headquarters.

48. CFA (2006). Wildfire Firefighter: Learning Manual. 2nd edition. Burwood East, Victoria, Australia, Learning Systems, Human Resource Development and Planning, CFA Headquarters.

49. Country Fire Authority (2008) Annual Report 2008. Accessed March 2009.

50. Country Fire Service (2008) South Australian Country Fire Service 2007-2008 Annual Report. Accessed March 2009.

51. Chan. G., Tan, V. and Koh, D. (2000). "Ageing and fitness to work". Occupational Medicine 50(7): 483-491.

52. Commonwealth v HR&EO Commission (1998) 152 ALR 182.

53. Conley, P.R. & Sackett, P.R. (1987). " Effects of using high- versus low performing incumbents as sources of job-analysis information". Journal of Applied Psychology 72(3): 434-437.

54. Coris, E., A. Ramirez and D. Van Durme (2004). "Heat illness in athletes: The dangerous combination of heat, humidity and exercise." Sports Medicine 34(1): 9-16.

55. Cosma v Qantas Airways Ltd (2002). 124 FCR 504.

56. Crouter S., Foster C., Esten P., Brice G., & Porcarl J. (2001). Comparison of incremental treadmill exercise and free range running. Medicine & Science in Sport & Exercise, 33, 332- 338.

165

57. Cuddy, J.S., Gaskill, S.E., Sharkey, B.J., Harger, S.G. and Ruby, B.C. (2007). Supplemental feedings increase self-selected work output during wildfire suppression. Medicine and Science in Sports and Exercise 39(6): 1004-1012.

58. Davis, P., C. Dotson and D. Santa Maria (1982). "Relationship between simulated fire fighting tasks and physical performance measures." Medicine and Science in Sports and Exercise 14(1): 65-71.

59. Dawson, B., Hopkinson, R., Appleby, B. & Stewart, G. (2004). " Comparison of training activities and game demands in the Australian Football League". Journal of Science and Medicine in Sport 7(3): 292-301.

60. Dawson, B., Randall, N. and Cowper,K. (2000). "Brief report: Minimum aerobic fitness standards and changes to the current physical performance test battery". Perth, Western Australia: The University of Western Australia, School of Human Movement and Exercise Science.

61. Deakin, J.M., Pelot, R., Smith, J.T., Stevenson, J.M. & Wolfe, L.A. (1997) Development of a Bona Fide Physical Maintenance Standard for CF and DND Fire Fighters. (Contract# 003SV.W8477-3-SC02) Kingston, Ontario: Queens University, Ergonomics Research Group.

62. Deakin, J.M., Pelot, R., Smith, J.T. and Weber, C.L. (1999). Bona fide occupational requirements for SARTech PMFS (Contract# 003SV.W8477-3-SC02) Kingston, Ontario: Queens University, Ergonomics Research Group.

63. The Department of Environment and Conservation (DEC) (2006) Fire fitness: Guidelines for managing The Department of Environment and Conservation fire fitness program. Perth, Western Australia; Management Services Branch, Fire Management Services Branch, The Department of Environment and Conservation.

64. Department of Environment and Conservation (DEC) (2007) Department of Environment and Conservation Annual Report 2006-2007. Accessed March 2009.

65. Department of Sustainability and the Environment (DSE) (2008). Living with Fire report: Victoria's Bushfire Strategy. Accessed March 2009.

66. Demczuk, V. (1998). "Field validation of an energy expenditure model for walking soldiers". International Journal of Industrial Ergonomics 22: 381-387.

67. Disability Discrimination Act (1992) 264, Commonwealth Consolidated Acts

166

68. Dreger, R.W. and Petersen, S.R. (2007). "Oxygen cost of the CF-DND fire fit test in males and females". Applied Physiology, Nutrition and Metabolism 32(3): 454-462.

69. Duncan, H. W., G. W. Gardner and R. J. Barnard (1979). "Physiological responses of men working in fire fighting equipment in the heat." Ergonomics 22(5): 521-7.

70. Edgecomb, S.J. and Norton, K.I. (2006). "Comparison of global positioning and computer- based tracking systems for measuring player movement distance during Australian Football". Journal of Science and Medicine in Sport 9: 25-32.

71. Ellis, S. and L. Gilbert (1997). Job related taskbased assessments. AFAC Guidelines: National Approach to Physical Performance Assessment for Firefighters. East Melbourne, Australia, Australasian Fire Authorities Council: 9.

72. Equal Employment Opportunity Commission (EEOC), the Department of Labor, the Department of Justice, and the Civil Service Commission (1978). Uniform Guidelines on Employee Selection Procedure; 43 FR 38295; 29 CFR Part 1607. Section 60-3.

73. Equal Opportunity Act (1995) 250, Victorian Consolidated Acts

74. Equal Opportunity Act, (1984) 250, South Australian Consolidated Acts.

75. Equal Opportunity Act, (1984) 250, Western Australian Consolidated Acts

76. Fahy, R. (2005). US firefighter fatalities due to sudden cardiac death, 1995 - 2004. Quincy, MA, National Fire Protection Association: 9.

77. Fahy, R. and LeBlanc, P. (2000). US firefighter fatalities. Quincy, MA, National Fire Protection Association: July/August.

78. Fahy, R. and LeBlanc, P. (2006). Firefighter fatalities in the United States - 2005. Quincy, MA, National Fire Protection Association: June.

79. Febbraio, M. (2003). Exercise in the heat. Physiological bases of sports performance. M. Hargreaves and J. Hawley. Sydney, McGraw-Hill: 254-276.

80. Fleishman, E.A. (1989). "Some New Frontiers in Personnel Selection Research". Personnel Psychology 41: 679-701.

81. Fleishman, E.A. and Mumford, M.D. (1991). "Evaluating classifications of job behaviour: A construct validation of the ability requirement scales". Personnel Psychology 44: 523-575.

167

82. Gaskill, S., B. Ruby, A. Walker, O. Sanchez, R. Serfass and A. Leon (2001). "Validity and reliability of combining three methods to determine ventilatory threshold." Medicine and Science in Sports and Exercise 33(11): 1841-1848.

83. Gebhart, D.L., (2000). Chapter 6: Establishing Performance Standards. In Constable, S. and Palmer, B. (Ed.), The Process of Physical Fitness Standards Development (pp. 179-201). State of the Art Report, Human Systems Information Analysis Center, US Department of Defence.

84. Ghugare, B. D., S. H. Adhaoo, L. P. Gite, A. C. Pandya and S. L. Patel (1991). "Ergonomics evaluation of a lever-operated knapsack sprayer." Applied Ergonomics 22(4): 241-50.

85. Gledhill, N., & Jamnik, V.K. (1992a). Characterization of the Physical Demands of Fire fighting. Canadian Journal of Sports Science. 17(3): 207-213

86. Gledhill, N. and Jamnik, V. K. (1992b). "Development and validation of a fitness screening protocol for firefighter applicants." Canadian Journal of Sport Science 17(3): 199-206.

87. Gledhill, J. (2001). "Sustaining volunteerism in the emergency services: Funding. Emergency Management Australia, National Summit on Volunteers in Emergency Management: Canberra, ACT.

88. Gore, C.J. & Edwards, D. (1992) Australian Fitness Norms: A Manual for Fitness Assessors. North Adelaide, The Health Development Foundation.

89. Green, S.B. & Veres, J.G. (1990). "Evaluation of an index to detect inaccurate respondents to a task analysis inventory". Journal of Business and Psychology 5(1): 47-61.

90. Guidotti, T.L. & Clough, V.M. (1992). "Occupational health concerns of fire fighting". Annual Review of Public Health. 13: 151 - 171.

91. Guidotti, T.L. (1995). "Occupational mortality among firefighters: Assessing the association". Journal of Occupational and Environmental Medicine. 37(12): 1348 - 1356.

92. Hahn, D.C. & Dipboye, R.L. (1988). " Effects of training and information on the accuracy and reliability of job evaluations". Journal of Applied Psychology73(2): 146-153.

93. Harvey, D.G., Kraemer, J.L., Sharratt, M.T. and Hughson, R.L. (2008). "Respiratory gas exchange and physiological demands during a fire fighter evaluation circuit in men and women". European Journal of Applied Physiology 103(1): 89-98.

94. Haynes, K., Tibbits, A., Coates, L., Ganewatta, G., Handmer, J. and McAneney, J. (2008). "100 years of Australian civilian bushfire fatalities: exploring the trends in relation to the 'stay or go policy'". Melbourne, Victoria: Bushfire Cooperative Research Centre.

168

95. Heil, D. P. (2002). "Estimating energy expenditure in wildland fire fighters using a physical activity monitor." Applied Ergonomics 33(5): 405-413.

96. Hendrie, A., J. Brotherhood, G. Budd, S. Jeffrey, F. Beasley, B. Costin, W. Zhien, M. Baker, N. Cheney and M. Dawson (1997). "Project aquarius 8. Sweating, drinking, and dehydration in men suppressing wildland fires." International Journal of Wildland Fire 7(2): 145-158.

97. Hogan, J. & Quigley, A.M. (1982) "Physical standards for employment and the courts". American Psychologist, 41(11): 1193-1217

98. Holmér, I. & Gavhed (2007) "Classification of metabolic and respiratory demands in fire fighting activity with extreme workloads". Applied Ergonomics.4(1): 25-52

99. Hooper, A.J., Crawford, J.O. & Thomas, D. (2001). "An evaluation of physiological demands and comfort between the use of conventional and lightweight self-contained breathing apparatus". Applied Ergonomics 32(4): 399-406

100. Hughes, M.A., Ratliff, R.A., Purswell, J.L. & Hadwiger, J. (1989). "A content validiation methodology for job related physical performance tests". Public Personnel Management 18(4): 487-504.

101. Hunter, L. (Ed.)(2003) The Campaign Fires: North-East/ East Gippsland Fires 2003. Country Fire Authority.

102. Hunter, S. K., A. Critchlow and R. M. Enoka (2005). "Muscle endurance is greater for old men compared with strength-matched young men." Journal of Applied Physiology 99(3): 890-897.

103. Hyde, A.C., Blazer, A., Caudle, S., Clevette, R.E., Shelly, J.R. & Szayna, T. (2008) 2007 U.S. Forest Service & Department of Interior Large Wildfire Cost Review: Assessing Progress Toward An Integrated Risk And Cost Fire Management Strategy. Brookings Institution, U.S. Secretary of Agriculture.

104. Jackson, A.S. (1994). "Preemployment physical evaluation". Exercise and Sports Science Reviews 22(1): 53- 90.

105. Jamnik, V. K., Thomas, S. G., Shaw, J. A., & Gledhill, N. (2010a). Identification and characterization of the critical physically demanding tasks encountered by correctional officers. Applied Physiology, Nutrition & Metabolism, 35(1), 45-58.

106. Jamnik, V. K., Thomas, S. G., Burr, J. F., & Gledhill, N. (2010b). Construction, validation, and derivation of performance standards for a fitness test for correctional officer applicants. Applied Physiology, Nutrition & Metabolism, 35(1), 59-70.

169

107. Jamnik, V. K., Thomas, S. G., & Gledhill, N. (2010c). "Applying the Meiorin Decision requirements to the fitness test for correctional officer applicants; examining adverse impact and accommodation". Applied Physiology, Nutrition & Metabolism, 35(1), 71-81.

108. Johnson, A.T., Benhur Benjamin, M. & Silverman, N. (2002) "Oxygen consumption, heat production, and muscular efficiency during uphill and downhill walking". Applied Ergonomics. 33(5): 485-491

109. Kallinen, M. (2005). "Cardiovascular benefits and potential hazards of physical exercise in elderley people." Journal of Sports Science and Medicine 4(Suppl. 7): 1 - 51.

110. Kazmierczak, K., Mathiassen, S.E., Neumann, P. & Winkel, J. (2006) " Observer reliability of industrial activity analysis based on video recordings." International Journal of Industrial Ergonomics, 36(3): 275-282.

111. Knapik, J.J. (1987)."The influence of physical fitness training on the manual material handling capability of women". Applied Ergonomics 28(5): 339-345.

112. Knapik, J.J., Harper, W. and Crowell, H.P. (1999)."Physiological factors in stretcher carriage performance". European Journal of Applied Physiology 79:409-413.

113. Knapik, J.J., Sharp, M.A., Canham-Chervak, M., Hauret, K., Patton, J.F. and Jones, B.H. (2001) "Risk factors for training-related injuries among men and women in basic combat training". Medicine and Science in Sports and Exercise 33(6): 946-954.

114. Katch, F., V. Katch and W. McArdle (2001). Appendix c: Energy expenditure in housefhold, occupation, recreational, and sports activities. Exercise physiology: Energy, nutrition and human performance. W. McArdle, F. Katch and V. Katch. Baltimore, Maryland, Lippincott, Williams & Wilkins: 1103-1115.

115. Keren, G., Epstein, Y., Magazanik I, A. and Sohar, E. (1981). The Energy Cost of Walking and Running with and without a Backpack Load. European journal of applied Physiology. 46(3): 317-324

116. Kuruganti, U. and Rickards, J. (2004). "The role of human factors engineering in establishing occupational fitness standards". International Journal of Industrial Medicine. 34(6): 451-457.

117. Landy, F.J. and Vasey, J. (1991). "Job analysis: The composition of SME samples". Personnel Psychology 44: 27-50.

118. Larsson, P. (2003). "Global positioning system and sport-specific testing". Sports Medicine 33(15): 1093-1101.

170

119. Larsson, P. and Henriksson-Larsen, K. (2001). "The use of dGPS and simultaneous metabolic measurements during orienteering". Medicine & Science in Sports & Exercise 33(11): 1919- 1924.

120. Lee, T.H. (2004). "Maximal isometric lifting strengths of men in teamwork". Human Factors 46(4): 686-696.

121. Lemon, P. and R. Hermiston (1977). "The human energy cost of fire fighting." Journal of Occupational Medicine 19(8): 558-562.

122. LeBlanc, P.R. and Fahy, R.F. (2005). Full Report: Firefighter fatalities in the United States, 2004. Quincy, MA, National Fire Protection Association.

123. Lindell, M.K., Clause, C.S., Brandt, C.J. & Landis, R.L. (1998). "Relationship between organisational context and job analysis task ratings". Journal of Applied Psychology 83(5): 769- 776.

124. Louhevaara, V., Smolander, J., Korhonen, O. and Toumi, T. (1986). "Maximal working times with self contained breathing apparatus". Ergonomics 29(1): 77-85.

125. Louhevaara, V., Soukainen, J., Lusa, S., Tulppo, M., Toumi, T. and Kajaste, T. (1994). "Development and evaluation of a test drill for assessing physical work capacity of fire- fighters". International Journal of Industrial Ergonomics 13: 139-146.

126. Louhevaara, V., Toumi, T., Smolander, J., Korhonen, O., Tossanainen, A. and Jaakkola, J. (1985). "Cardiorespiratory strain in jobs that require respiratory protection". International Archives of Occupational and Environmental Health 55(3): 195-206.

127. Lusa, S., V. Louhevaara, J. Smolander, M. Kivimaki and O. Korhonen (1993). "Physiological responses of fire fighting students during simulated smoke-diving in the heat." American Industrial Hygiene Association Journal 54(5): 228-31.

128. Lusa, S., Louhevaara, V. & Kinnunen, K. (1994). Are the Job Demands on Physical Work Capacity Equal for Young and Aging Firefighters? Journal of Occupational Medicine, 36(1): 70- 74

129. Lyons, J., A. Allsopp and J. Bilzon (2005). "Influences of body composition upon the relative metabolic and cardiovascular demands of load-carriage." Occupational Medicine (London) 55(5): 380-384.

171

130. Maniscalco, P., Lane, R., Welke, M., Mitchell, J.H. and Husting, L. (1999). "Decreased rate of back injuries through a wellness program for offshore petroleum employees". Journal of Occupational and Environmental Medicine 41(9): 813-820.

131. Mangan, R. (1999). Wildland fire fatalities in the united states: 1990 to 1998. Missoula, MT, US Department of Agriculture, Forest Service, Missoula Technology and Development Centre: 14.

132. Mangan, R. (2002). Towards a healthier and safer wildland firefighter workforce. Missoula, Montana, Missoula Technology and Development Centre: 4.

133. Manning, J.E. & Griggs, T.R. (1983). Heart rates in fire fighters using light and heavy breathing equipment: Similar near-maximal exertion in response to multiple work load conditions. Journal of Occupational Medicine, 25(3): 215-218

134. Matticks, C.A., Westwater, J.J., Himel, H.N., Morgan, R.F. & Edlich, R.F. (1992). "Health risks to fire fighters". Journal of Burn Care and Rehabilitation. 13(2): 223- 235.

135. Maughan, R. and S. Shirreffs (2004). "Exercise in the heat: Challenges and opportunities." Journal of Sports Sciences 22(10): 917-927.

136. Maurer, T.J. & Tross, S.A. (2000). "SME committee Vs. field job analysis ratings: Convergence, cautions, and a call". Journal of Business and Psychology 14(3): 489-499.

137. McCallum, A. (2004). Meiorin: Five years later. Cheryl V. Rovis. Toronto: Ontario. Filion Wakely Thorup Angeletti LLP.

138. McKenzie, J.M. (2007). The use of GPS to predict energy expenditure for outdoor walking. Master of Science Thesis, Montana State University, Bozeman, Montana.

139. McFadyen, P., Wenger, H. A., Pethick, W & Sleivert, G. G.(1996) “The Influence of Fitness on the Physiological Cost of Wildland Firefighting” Medicine & Science in Sports & Exercise 28(5): 199

140. McLennan, J. (2004). Profiles of australia’s volunteer firefighters. Bundoora, Victoria, Bushfire CRC & Latrobe University: 24.

141. Metropolitan Fire and Emergency Services Board (MFB) (2008) Metropolitan Fire and Emergency Services Board Annual Report 2008. Accessed March 2009.

142. Michaelides, M.A., Parpa, K.M., Thompson, J. and Brown, B. (2008). "Predicting performance on a firefighter's ability test from fitness parameters". Research Quarterly for Exercise and Sport 79(4): 468-475.

172

143. Misner, J. E., R. A. Boileau and S. A. Plowman (1989a). "Development of placement tests for fire fighting ; a long-term analysis by race and sex." Applied Ergonomics 20(3): 218-24.

144. Misner, J. E., R. A. Boileau, S. A. Plowman, B. Joyce, S. Huorovitz, B. Elmore, M. Gates, J. Gilbert and C. Horswill (1989b). "Physical performance and physical fitness of a select group of female firefighter applicants." Journal of Applied Sport Science Research 3(3): 62-67.

145. Misner, J. E., S. A. Plowman and R. A. Boileau (1987). "Performance differences between males and females on simulated fire fighting tasks." Journal of Occupational Medicince 29(10): 801- 5.

146. Montoye, H., H. Kemper, W. Saris and R. Washburn (1996). Appendix c: Classification by energy cost of human physical activities. Measuring physical activity and energy expenditure. H. Montoye, H. Kemper, W. Saris and R. Washburn. Champaign, Illinois, Human Kinetics: 123- 133.

147. Morgeson, F.P. & Campion, M.A. (1997). "Social and cognitive sources of potential inaccuracy in job analysis". Journal of Applied Psychology 82(5): 627-655.

148. Mueller, M. & Belcher, G. (2000). "Observed divergence in the attitudes of incumbents and supervisors as subject matter experts in job analysis: A study of the fire captain rank". Public Personnel Management 29(4): 529-556.

149. Musich, S., Napier, D. and Edington, D.W. (2001). "The association of health risks with workers' compensation costs". Journal of Occupational and Environmental Medicine 43(6): 534-541.

150. Musk, A.W., Peters, J.M., Bernstein, L. Rubin, C. and Monroe, C.B. (1982). "Pulmonary function in firefighters: A six-year follow-up in the Boston Fire Department". American Journal of Industrial Medicine 3(1): 3-9.

151. Nag, P., N. Sebastian and M. Mavlankar (1980). "Occupational workload of indian agricultural workers." Ergonomics 23(2): 91-102.

152. Naughton, G. (2006). "A question of precision". Journal of Science and Medicine in Sport 9(1): 33-34.

153. Noble, B. J. and R. J. Robertson (1996). Perceived exertion. Campaign, IL, Human Kinetics.

154. Nottrodt, J.W. & Celentano, E.J. (1987) Development of predictive selection and placement tests for personnel evaluation. Applied Egronomics 18(4): 279-288

173

155. NSW Fire Brigades (2007) NSW Fire Brigades Annual Statistical Report 2006/07. Accessed March 2009.

156. NSW Rural Fire Service (NSWRFS) (2008) Together- for our community: NSW Rural Fire Service Annual Report 2008. Accessed March 2009.

157. Nuwayhid, I.A., Stewart, W. and Johnson, J.V. (1993). "Work activities and the onset of first- time low back pain among New York City fire fighters". American Journal of Epidemiology 137(5): 539-548.

158. O'Connell, E., P. Thomas, L. Cady and R. Karwasky (1986). "Energy costs of simulated stair climbing as a job-related task in fire fighting." Journal of Occupational Medicine 28(4): 282- 284.

159. Oldham, J.A., Schofield, S., Callaghan, M.J. & Winstanley, J. (2000). An investigation of the validity of ‘simulated’ work related tasks in relation to ‘real-life’ situations in the fire service training environment. Occupational Medicine, 50(8): 599-607

160. Ontario Human Rights Commision (OHRC) (2000). Policy and guidelines on disability and the duty to accomodate. Ontario, Canada.

161. Payne, W., B. Portier, I. Fairweather, S. Zhi and R. Snow (1988). Physiological and psychological responses to wearing fully encapsulated protective clothing during simulated work in various thermal environments, Department of Physical Education and Recreation, Footscray Institute of Technology and Department of Human Movement Studies, Ballarat College of Advanced Education: 68.

162. Payne, W.R., Payne, K.A. and Harvey, J.T. (2005). "Review of the firefighter fitness assessment". (For the Victorian Department of Sustainability and Environment) Ballarat, Victoria: University of Ballarat, School of Human Movement and Sport Sciences.

163. Payne, W. and Harvey, J. (2009). A framework for the design and development of physical employment tests and standards. Ergonomics 53(7):858-871.

164. Peterson, A., Payne, W., Phillips, M., Netto., K., Nichols. & Aisbett, B. (2010) Validity and relevance of the pack hike wildland firefighter work capacity test: a review. Ergonomics 53(10): 1276-1285.

165. Petersen, S., R. Dreger, B. Williams and W. McGarvey (2000). "The effects of hyperoxia on performance during simulated fire fighting work." Ergonomics 43(2): 210-222.

174

166. Phillips, M., Petersen, A., Abbiss, C.R., Netto, K., Payne, W., Nichols, D. & Aisbett, B. (2011). Pack Hike Test finishing time for Australian firefighters: pass rates and correlates of performance. Applied Ergonomics 42(3): 411-418.

167. Pine, D.E. (1995). "Assessing the validity of job ratings: An empirical study of false reporting in task inventories". Public Personnel Management 24(4): 451-460.

168. Plasqui, G. and Westerterp, K.R. (2007). "Physical activity assessment with accelerometers: An evaluation against doubly labeled water". Obesity 15: 2371-2379.

169. Poole, P. and B. Ross (1983). "The energy cost of sheep shearing." Search 14(3-4): 103-105.

170. Qantas Airways Ltd v Christie (1998). 152 ALR 365

171. Rayson, M.P. (1998). The development of physical selection procedures for the British Army. Phase 1: job analysis and identification of criterion tasks. In M.A. Hanson (Ed.), Contemporary Ergonomics (pp. 393-397). London: Taylor and Francis.

172. Rayson, M., Holliman, D. and Belyavin, A.(2000a). "Development of physical selection procedures for the British Army. Phase 2: Relationship between physical performance tests and citerion tasks". Ergonomics, 43(1), 73- 105.

173. Rayson, M.P. (2000b). Fitness for work: the need for conducting a job analysis. Occupational Medicine. 50(6): 434-436.

174. Rayson, M.P., (2000c). Chapter 3: Job Analysis. In Constable, S. and Palmer, B. (Ed.), The Process of Physical Fitness Standards Development (pp. 67-101). State of the Art Report, Human Systems Information Analysis Center, US Department of Defence.

175. Rayson, M.P., (2003). "Operational physiological capabilities of firefighters: literature review and research recommendations". Technical Report to the ODPM under contract number CI 71/4/12. Farnham, UK: Optimal Performance Limited.

176. Reaburn, P. (2003). Physical fitness of queensland firefighters. Rockhampton, Queensland, Central Queensland University: 10.

177. Reilly, T., Wooler, A. and Tipton, M. (2006a). "Occupational fitness standards for beach lifeguards. Phase 1: the physiological demands of beach lifeguarding". Occupational Medicine 56(1): 6-11.

178. Reilly, T., Iggleden, C., Gennser, M. and Tipton, M. (2006b). "Occupational fitness standards for beach lifeguards. Phase 2: the development of an easily administered fitness test". Occupational Medicine 56(1): 12-17.

175

179. Rhea, M. R., B. A. Alvar and R. Gray (2004). "Physical fitness and job performance of firefighters." Journal of Strength and Conditioning Ressearch 18(2): 348-52.

180. Richman, W.L. and Quinones, M.A. (1996). "Task frequency rating accuracy: The effect of task engagement and experience". Journal of Applied Psychology 81(5): 512-524.

181. Richmond, V.L., Rayson, M.P., Wilkinson, D.M., Carter, J.M. and Blacker, S.D. (2008). "Physical demands of firefighter search and rescue in ambient environment conditions". Ergonomics 51(7): 1023-1031.

182. Rodriguez, J.L. and Eldridge, J.A. (2003). "The function of aerobic capacity and strength in predicting the probability of injury in fire fighters". Medicine and Science in Sports and Exercise 35(5): Supplement 1, S209.

183. Ronalds, C. (2008). Discrimination Law and Practice: 3rd Edition. The Federation Press, Sydney, Australia.

184. Rorke, S.C. (2002). Physically and psychologically demanding occupations: International reference guidelines for occupational health and job readiness assessment. In: Thatcher, A, Fisher, J., and Miller, K. (Editors). CD-ROM Proceedings of CybErg’2002, The Third International Cyberspace Conference on Ergonomics, 764-773. ISBN: 0-9577485-1-5

185. Ruby, B. (1999). Energy expenditure and energy intake during wildfire suppression in male and female firefighters. Wildland Fire Fighter Health and Safety Conference, Missoula, Montana, Missoula Technology and Development Centre.

186. Ruby, B. and S. Gaskill (2003). Wildland firefighter research summaries - 2001 and 2002 fire seasons. Missoula, Human Performance Laboratory, Department of Health and Human Performance, University of Montana: 51.

187. Ruby, B., D. Schoeller, B. Sharkey, C. Burks and S. Tysk (2003). "Water turnover and changes in body composition during arduous wildfire suppression." Medicine and Science in Sports and Exercise 35(10): 1760-1765.

188. Ruby, B., T. Schriver, T. Zderic, B. Sharkey, C. Burks and S. Tysk (2002). "Total energy expenditure during arduous wildfire suppression." Medicine and Science in Sports and Exercise 34(6): 1048-1054.

189. Sanchez, J.I. & Levine, E.L. (1989). "Determining important tasks within jobs: A policy- capturing approach". Journal of Applied Psychology 74(2): 336-342.

176

190. Sanchez, J.I. & Fraser, S.L. (1992). "On the choice of scales for task analysis". Journal of Applied Psychology 77(4): 545-553.

191. Saupe, K., M. Sothmann and D. Jasenof (1991). "Aging and the fitness of fire fighters: The complex issues involved in abolishing mandatory retirement ages." American Journal of Public Health 81(9): 1192-4.

192. Schmuck, G., San-Miguel-Ayanz, J., Barbosa, P., Camia, A., Kucera, J., Liberta, G., Bucella, P., Schulte, E., Flies, R. & Colletti, L. (2004) Forest Fires in Europe: 2003 Fire Campaign. European Commission, Joint Research Center, Institute for Environment and Sustainability.

193. Selkirk, G. A. and T. M. McLellan (2004). "Physical work limits for toronto firefighters in warm environments." Journal of Occupational and Environmental Hygiene 1(4): 199-212.

194. Sharkey, B. (1999). The development and validation of a job-related work capacity test for wildland fire fighting. Wildland Fire Fighter Health and Safety Conference, Missoula, Montana, Missoula Technology and Development Centre.

195. Sharkey, B.J. (1997) Chapter 7: Work Capacity Tests and Training. In, Fitness and Work Capacity, Second Edition. (pp 47-55). National Wildfire Coordinating Group, US Department of Agriculture.

196. Sharkey, B.J. and Davis, P.O. (2008) Hard Work: Defining Physical Work Performance Requirements. Human Kinetics, Champaign, Illinois.

197. Sharp, M.A., Rice, V.J., Nindl, B.C. and Williamson, T.L. (1997). "Effects of team size on the Maximal weight bar lifting strength of military personnel". Human Factors 39(3): 481-488.

198. Shephard, R.J. (1991). "Occupational demand and human rights: Public safety officers and cardiorespiratory fitness". Sports Medicine 12(2): 94-109.

199. Shephard, R.J. and Bonneau, J. (2002). "Assuring gender equity in recruit standards for police officers". Canadian Journal of Applied Physiology 27(3): 263-295.

200. Shephard, R.J. and Bonneau, J. (2003). "Supervision of occupational fitness assessments". Canadian Journal of Applied Physiology 28(2): 225-239.

201. Skoldstrom, B. (1987). "Physiological responses of fire fighters to workload and thermal stress". Ergonomics 30(11): 1589-1597.

202. Sluiter, J.K. (2006). "High-demand jobs: Age-related diversity in work ability?". Applied Ergonomics 37: 429-440.

177

203. Smith, D., T. Manning and S. Petruzzello (2001). "Effect of strenuous live-fire drills on cardiovascular and psychological responses of recruit firefighters." Ergonomics 44(3): 244-254.

204. Smith, D. and S. Petruzzello (1998). "Selected physiological and psychological responses to live-fire drills in different configurations of fire fighting gear." Ergonomics 41(8): 1141-1154.

205. Smith, D., S. Petruzzello, M. Chludzinski, J. Reed and J. Woods (2005). "Selected hormonal and immunological responses to strenuous live-fire fire fighting drills." Ergonomics 48(1): 55-65.

206. Smith, D., S. Petruzzello, J. Kramer and J. Misner (1997). "The effects of different thermal environments on the physiological and psychological responses of firefighters to a training drill." Ergonomics 40(4): 500-510.

207. Smith, D. L., S. J. Petruzzello, J. M. Kramer and J. E. Misner (1996). "Physiological, psychophysical, and psychological responses of firefighters to fire fighting training drills." Aviation Space and Environmental Medicine 67(11): 1063-8.

208. Snook, S.H., Campanelli, R.A. and Hart, J.W. (1978). "A study of three preventive approaches to low back injury". Journal of Occupational Medicine 20(7): 478-481.

209. Sothmann, M., D. Gebhardt, T. Baker, G. Kastello and V. Sheppard (2004). "Performance requirements of physically strenuous occupations: Validating minimum standards for muscular strength and endurance." Ergonomics 47(8): 864-875.

210. Sothmann, M., K. Saupe, D. Jasenof, J. Blaney, S. Donahue Fuhrman, T. Woulfe, P. Raven, J. Pawelczyk, C. Dotson, F. Landy, J. Smith and P. Davis (1990). "Advancing age and the cardiorespiratory stress of fire suppression: Determining a minimum standard for aerobic fitness." Human Performance 3(4): 217-236.

211. Sothmann, M., K. Saupe, P. Raven, J. Pawelczyk, P. Davis, C. Dotson, F. Landy and M. Siliunas (1991). "Oxygen consumption during fire suppression: Error of heart rate estimation." Ergonomics 34(12): 1469-74.

212. Sothmann, M. S., F. Landy and K. Saupe (1992a). "Age as a bona fide occupational qualification for fire fighting. A review on the importance of measuring aerobic power." Journal of Occupational Medicine 34(1): 26-33.

213. Sothmann, M. S., K. Saupe, D. Jasenof and J. Blaney (1992b). "Heart rate response of firefighters to actual emergencies. Implications for cardiorespiratory fitness." Journal of Occupational Medicine 34(8): 797-800.

178

214. Sparrow, W. A., K. M. Hughes, A. P. Russell and P. F. Le Rossignol (2000). Movement economy, preferred modes, and pacing. Energetics of human activity. W. A. Sparrow. Champaign, Illinois, Human Kinetics. 1: 96 - 124.

215. Spurr GB, Prentice AM, & Murgatroyd PR. (1988). Energy expenditure from minute-by-minute heart-rate recordings: comparison with indirect calorimetery. American Journal of Clinical Nutrition, 48, 522-559.

216. Stammers., R.B., Carey, M.S. & Astley, J.A. (1990). Chapter 6: Task Analysis. In J.R. Wilson & Corlett E.N. (Ed.), Evaluation of Human Work: A practical ergonomics methodology (pp. 134- 160). London: Taylor and Francis.

217. Stevenson, J.M., Andrew, G.M., Bryant, J.T. & Thomson, J.M. (1988) Development of minimum physical fitness standards for the Canadian Armed Forces. Phase III. Kingston, Ontario: Queens University, Ergonomics Research Group.

218. Stewart, I.B., McDonald, M.D., Hunt, A.P. & Parker, T.W. (2008) Physical capacity of rescue personnel in the mining industry. Journal of occupational medicine and toxicology, 3 (1): 22-28

219. Strath SJ, Bassett Jr DR, Swartz AM, & Thompson DL. (2001). Simultaneous heart rate-motion sensor technique to estimate energy expenditure. Medicine & Science in Sports & Exercise, 33, 2118-2123.

220. Swank, A., K. Adams, K. Barnard, J. Berning and B. Stamford (2000). "Age-related aerobic power in volunteer firefighters, a comparative analysis." Journal of Strength and Conditioning Research 14(2): 170-174.

221. Tasmanian Fire Service (TFS) (2008) Annual Report 2007/08. Accessed March 2009.

222. Taylor, N.A.S. and Groeller, H. (2003). "Work-based physiological assessment of physically- demanding trades: a methodological overview". Journal of Physiological Anthropology and Applied Human Science 22(2): 73-81.

223. Tanaka, H., Monohan, K.D. & Seals, D.R. (2001) Age-predicted maximal heart rate revisited. Journal of the American College of Cardiology. 37: 153-156

224. Thornton, R., Handmer, J. and Leonard, M. (2008). "Fire in the future: 2020 Scenario". Fire Australia. Summer 2008-09: 14-17.

225. Trost, S.G. McIver, K.L. and Pate, R.R. (2005). "Conducting accelerometer-based activity assessments in field-based research". Medicine & Science in Sports & Exercise 37(11): S531- S543.

179

226. Walton, S.M., Conrad, K.M., Furner, S.E. and Samo, D.G. (2003). "Cause, type, and workers' compensation costs of injury to fire fighters". American Journal of Industrial Medicine 43(4): 454-458

227. Washburn, A.E. (1996). "1995 Firefighter Fatalities". NFPA Journal. 90(4): 63-5.

228. Welk, G.J., Schaben, J.A. and Morrow, J.R. (2004). "Reliability of accelerometry-based activity monitors: A generalizability study". Medicine & Science in Sports & Exercise 36(9): 1637-1645.

229. White, M.K. and Hodous, T.K. (1987). "Reduced work tolerance associated with wearing protective clothing and respirators". American Industrial Hygiene Association Journal 48(4): 304-310.

230. Williams, B., S. R. Petersen and K. Douglas (1996). "Physiological responses of training officers during live fire training." Fire Engineering September: 45-52.

231. Williams, A.G. (2006). "Can simply anthropometric and physical performance tests track training-induced changes in load-carriage ability". Military Medicine 171(8): 742-748.

232. Williams, A.G. (2007). "Material handling ability of regular and reserve British army soldiers". Military Medicine 172(2): 220-223.

233. Williams-Bell, M.F., Villar, R., Sharratt, M.T. and Hughson, R.L. (2009). " Physiological Demands of the Firefighter Candidate Physical Ability Test". Medicine & Science in Sports & Exercise 41(3): 653-662.

234. Williford, H., W. Duey, M. Olson, R. Howard and N. Wang (1999). "Relationship between fire fighting suppression tasks and physical fitness." Ergonomics 42(9): 1179-1186.

235. Wilmore, J. and D. Costill (1994). Physiology of sport and exercise. Champaign, Illinois, Human Kinetics.

236. Wilson, M.A. (1997). "The validity of task coverage ratings by incumbents and supervisors: Bad news". Journal of Business and Psychology 12(1): 85-95.

237. Wiseby, B., Montgomery, P.G., Pyne, D.B. & Rattray, B. (2009). "Quantifying movement demands of AFL football using GPS tracking". Journal of Science and Medicine in Sport 13(5): 531-536.

238. U.S. Department of Labor (1999). Testing and Assessment: An employer's guide to good practices. Washington, D.C

239. Vanderburgh, P.M. (2008). "Occupational relevance and body mass bias in military physical fitness tests." Medicine and Science in Sports and Exercise 40(8): 1538-1545.

180

240. Viswesvaran, C., Ones, D.S. & Schmidt, S.L. (1996). "Comparative analysis of thereliability of job performance ratings". Journal of Applied Psychology 81(5): 557-574.

241. von Heimburg, E.D., Rasmussen, AK. and Medbø, J.I. (2006). "Physiological responses of firefighters and performance predictors during a simulated rescue of hospital patients". Ergonomics 49(2): 111-126.

242. X v Commonwealth (1999) 200 CLR 177.

243. Zamparo, P., Perini, R., Orizio, C., Sacher, M. & Ferretti, G. (1992). The energy cost of walking and running on sand. European Journal of Applied Physiology 65: 183-187.

181

APPENDICES

Table A.1, Task list of RTBS work.

Task No. Task Description CM-01 Preparation of individual equipment CM-02 Individually climb and dismount tanker CM-03 Uncharged 38mm hose advance onto fire break CM-04 Uncharged 38mm hose advance into terrain CM-05 Lateral charged repositioning of 38mm hose CM-06 Full charged repositioning of 38mm hose CM-07 Operating 25mm rubber delivery hose CM-08 Hose work during blacking out activity CM-09 Charged 38mm hose advance CM-10 Pump operation at tanker CM-11 Manual hose retraction CM-12 Solo rake work CM-13 Rapid rake work during spot fire containment CM-14 Rake work during team line building CM-15 Chainsaw use in rakehoe crew CM-16 Chain saw use for vehicle access CM-17 Using an axe for vehicle access CM-18 Rake work during blacking out CM-19 Patrolling on foot whilst carrying hand tool CM-20 Mobile patrolling as a member of a tanker crew CM-21 Mobile patrolling in a support vehicle CM-22 Knapsack hiking CM-23 Knapsack spraying CM-25 Quick fill pump carry CM-26 Generator carry CM-27 Trailer mounted quick fill pump set up CM-28 Draughting set up

182

CM-29 Quick fill pump set up CM-30 Structure preparation for imminent ember attack CM-31 Water refuelling of truck CM-32 Adding class A foam CM-33 Hang hoses at station CM-34 Fire line driving in support of crew work CM-35 Transit driving between the staging area and fire ground CM-36 4 wheel drive driving in terrain CM-38 Transit driving between home and staging area CM-39 Rapid construction of a sheltering ditch CM-40 Rapid preparation of refuge site with tanker CM-41 Preparation of tanker for burn over CM-42 Preparation of support vehicles for burn over CM-43 Administer minor first aid CM-44 Evacuate injured but assisting crew member CM-45 Evacuate seriously injured crew member CM-46 Integrated suppression effort on structure fire CM-47 Integrated suppression effort to contain spot fires CM-48 Integrated crew suppression effort on a back burn CM-49 Vehicle repair CM-50 Vehicle recovery CM-51 Vehicle tire change CM-52 Burnout ignition CM-53 PLANT machinery supervision CM-54 Hose bowling CM-55 Hose making up on the bite

183

Table A.2, Potential options and definitions of variables discussed by participants for each task on the task list. Participants could select multiple options for variable seven, action category. For all other variables, participants could only select one variable. Variable Options/ Definitions

1. Task Selection “Is this task more physically demanding than other fire ground tasks?”

2. Operational 1. Not important Importance 2. Mildly important

3. Moderately important

4. Important, but not critical

5. Critical

3. Duration How long (in minutes) is the task expected to take to complete? 4. Load How much weight (kg) is used or moved when doing this task per person? 5. Distance Over what distance (m) is the task conducted? 6. Frequency With what frequency are task repetitions completed? 7. Action Category Lift - The action of picking up an item to one of three levels: low, medium, high Carry - Carrying an item from one location to another. Includes initial lift if constant height Pass - The action of moving an item from one side of the body to the other or diagonally vertical Climb - Climbing Action/ Ascent & Descent on/off vehicle Push/pull - The action such as manually priming a pump with a fore-aft or side-to- side movements.

184

Strike - The action typically used with a rake and includes striking actions with object or hand. Dig/ Rake - The action used in digging soil, specifically targeted line creation or suppression. Loaded hike - The action of walking at relatively sustained pace carrying load. Patrol - The action of monitoring tactically to maintain fire line perimeters. Suppress - The action of directly attacking a fire edge or live flame. Urban/ Structure encroachment - The action of defending, suppressing or monitoring structures affected by or in the path of a bushfire. Tanker mounted suppression - The action of following and suppressing a running fire whilst riding on a fire tanker Static Hole - The action of holding a load or resisting against a force without moving. Drag - The action of moving an object from one point to another without lifting the bulk of that objects weight off the ground Assemble - The action of building equipment from parts. Other - Any other action which does not fit easily within the criteria. 8. Activity Type Strength - The ability to exert maximal muscular force for short periods of time, ie. Weight lifting Endurance - The ability to maintain work for sustained periods of time, ie. Marathon Speed - The ability to move at a high rate of motion, ie. 100m dash

185

Speed- strength – Combining speed and strength, ie. Shot put Strength – endurance - Combining strength and endurance, ie. wrestling Speed – endurance - Combining speed and endurance, ie. 400m dash Power – endurance - Combining strength, speed and endurance, ie. volleyball 9. Perceived level 3. High of Difficulty 2. Moderate 1. Low

186