For Peer Review Only

BMJ Open BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from

Aqueous Versus Alcoholic Antisepsis with Chlorhexidine for Skin Excisions – The AVALANCHE Trial. Protocol for randomized controlled trial for the prevention of superficial surgical site infection after minor surgery in general practice. For peer review only

Journal: BMJ Open

Manuscript ID bmjopen-2016-011604

Article Type: Protocol

Date Submitted by the Author: 20-Feb-2016

Complete List of Authors: Heal, Clare; James Cook University, School of Medicine and Dentistry; Anton Breinl Research Centre for Health Systems Strengthening, Australian Institute of Tropical Health and Medicine Charles, Daniel; James Cook University, School of Medicine and Dentistry Hardy, Alexandra; James Cook University, School of Medicine and Dentistry Delpachitra, Meth; James Cook University, School of Medicine and Dentistry Banks, Jennifer; James Cook University, School of Medicine and Dentistry Wolfahrt, Michael; James Cook University, School of Medicine and Dentistry Buttner, Petra; James Cook University, School of Public Health, Tropical

Medicine and Rehabilitation Sciences http://bmjopen.bmj.com/

Primary Subject Public health Heading:

Secondary Subject Heading: Evidence based practice

Keywords: PRIMARY CARE, WOUND MANAGEMENT, SURGERY

on October 3, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 12 BMJ Open

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 TITLE 4 5 Aqueous Versus Alcoholic Antisepsis with Chlorhexidine for Skin Excisions – The 6 AVALANCHE Trial. Protocol for randomized controlled trial for the prevention of superficial 7 surgical site infection after minor surgery in general practice. 8 9 CF Heal1, 2, D Charles1, A Hardy1, M Delpachitra1, J Banks 1M Wohlfahrt1, P Buettner3 10 11 12 13 14 Corresponding Author 15 For peer review only 16 Professor Clare Heal 17 Promotional Chair 18 Discipline of General Practice and Rural Medicine 19 Mackay Clinical School 20 College of Medicine and Dentistry 21 James Cook University 22 Mackay Base Hospital, Bridge Rd 23 Mackay QLD 4740 Australia 24 [email protected] 25 26 Author Affiliations 27 28 1 Discipline of General Practice and Rural Medicine 29 Mackay Clinical School 30 College of Medicine and Dentistry 31 James Cook University 32

33 2 Anton Breinl Research Centre for Health Systems Strengthening, Australian Institute of http://bmjopen.bmj.com/ 34 35 Tropical Health and Medicine 36 3Tropical Health Solutions, Townsville, Qld and Centre for Chronic Disease Prevention, 37 James Cook University, Cairns, QLD 38 39 40 41 42 on October 3, 2021 by guest. Protected copyright.

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 12

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4 ABSTRACT 5

6 Introduction: 7 Surgical site infection (SSI) after minor skin excisions has a significant impact on patient 8 morbidity and healthcare resources. Skin antisepsis prior to surgical incision is used to 9 10 prevent surgical site infection, and is performed routinely worldwide. However, in spite of the 11 routine use of skin antisepsis, there is no consensus regarding which antiseptic agents are 12 most effective. 13 14 Methods and analysis: 15 The study designFor is a prospective peer randomised review controlled trial (RCT)only with the aim of 16 investigating the impact of two different antiseptic preparations upon the incidence of 17 superficial surgical site infection in patients undergoing minor skin excisions. The 18 intervention of 0.5% chlorhexidine gluconate (CHG) in 70% alcohol will be compared with 19 0.5% chlorhexidine gluconate in aqueous solution. The trial will be conducted in four 20 Australian General Practices over a 9-month period, with 920 participants to be recruited. 21 Consecutive patients presenting for minor skin excisions will be eligible to participate. 22 Randomisation will be on the level of the patient. The primary outcome is superficial surgical 23 site infection in the first 30 days following the excision. Secondary outcomes will be adverse 24 effects including anaphylaxis, skin irritation, contact dermatitis and rash and patterns of 25 antibiotic resistance. 26 27 Ethics and Dissemination: The study has been approved by the James Cook University 28 Human Research Ethics Committee (HREC). Findings will be disseminated in conference 29 presentations and journals and through online electronic media. 30 31 Registration Details: The trial is registered with the Australian New Zealand Clinical Trials 32 Registry, registration number: ACTRN12615001045505. 33 34 http://bmjopen.bmj.com/ Discussion: RCTs conducted in General Practice differ from hospital based projects in 35 terms of feasibility, pragmatism and funding. The success of this trial will be cemented in 36 the fact that the research question was established by a group of GPs who identified an 37 interesting question which is relevant to their clinical practice and not answered by current 38 evidence. 39 40

41 42 on October 3, 2021 by guest. Protected copyright. Study Strengths 43

44 45 • Practical clinical question relevant to clinical practice 46 • Recruitment of clinicians and participants will be very feasible 47 • Independent outcome assessor will assess photographs of all infections 48 • Conduct of trial in General Practice will provide clinically relevant results to end user 49 50 Study Limitations 51 • Diagnosis of infection has element of subjectivity 52 • Practice based outcome assessors will not be blinded 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 12 BMJ Open

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on October 3, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 12

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 BACKGROUND 4 5 It is routine practice prior to surgery to carry out preoperative cleansing of the skin with 6 antiseptic preparations at the site of surgical incisions (preoperative skin antisepsis). [1-3] 7 The purpose of preoperative skin antisepsis is to reduce the incidence of surgical site 8 infection (SSI) by removing microorganisms on the skin through a combination of 9 mechanical removal and chemical killing. [1, 3, 4] The incidence of SSI arising from surgery 10 in general practice is usually 1-3%, but has been recorded at rates as high as 10%. [5-8] 11 The consequences of SSI include pain and discomfort for patients, increased healthcare- 12 associated costs and temporarily reduced occupational and recreational productivity and 13 functionality. [1, 3, 4, 9] As a result, reducing the incidence of SSIs is in the interest of all 14 stakeholders in healthcare. 15 For peer review only 16 The most commonly used preoperative skin antiseptic preparations are povidone iodine 17 (PVI) and chlorhexidine gluconate (CHG). Both are available in aqueous and alcoholic 18 preparations and in multiple different concentrations. [1, 3, 4] Both PVI and CHG are 19 effective against a wide range of gram positive and negative bacteria, viruses and fungi, 20 though CHG has a more appreciable residual antiseptic activity on the skin after application. 21 [1, 2, 4] 22 23 Surprisingly, despite the routine use of preoperative skin antisepsis, there is no definitive 24 scientific consensus regarding which antiseptic preparation is most effective at preventing 25 SSI. [1, 4, 9] There is even less definitive data available for antisepsis prior to clean surgery, 26 which the Centre for Disease Control (CDC) defines as “an uninfected operation in which no 27 inflammation is encountered and the respiratory, alimentary, genital or uninfected urinary 28 tract is not entered” (CDC, 2011). [10] Prior studies of wound infection after minor surgery in 29 General Practice in the Mackay region have shown a SSI rate of approximately 9.5% - an 30 incidence which is much higher than expected based on published results of similar cohorts 31 in other regions of Australia and the world. [11-14] The reason for this high infection rate is 32 unclear, but may be related to the hot, humid environment or to patient behaviour in our rural 33 setting. On one hand, this infection rate is suboptimal, as the CDC suggests that an http://bmjopen.bmj.com/ 34 35 acceptable rate of SSI after clean minor surgery is less than 5% [15]. However, in settings 36 where there is a low risk of infection after clean surgery, studies of over 5,000 procedures 37 may be required to detect a clinically relevant difference in infection from an intervention with 38 statistical confidence, making such trials unfeasible [16]. Conversely, because of the high 39 rate of infection in our patient cohort and the high minor surgery workload in rural general 40 practice [17], a study of skin antisepsis for the prevention of SSI in in our setting is highly 41 feasible. 42 on October 3, 2021 by guest. Protected copyright. 43 In general, the evidence base guiding appropriate selection of antiseptic agents is poor. A 44 landmark study found 2% CHG in 70% isopropyl alcohol to be superior to an aqueous 45 solution of 10% PVI, however, as alcohol is known to have significant antimicrobial 46 properties this was likely to be an active treatment component in in this study [18]. Most 47 recent meta-analyses, including a Cochrane review, concur that it is difficult to make 48 conclusive statements about whether there are differences in the efficacy of CHG and PVI

49 [1, 2, 9] and the latest guideline on surgical skin antisepsis released by the Queensland 50 Department of Health agrees with this position. [4] The Queensland Department of Health 51 does however endorse the use of alcoholic solutions in preference to aqueous preparations, 52 [4] and two recent meta-analyses have asserted that there is some evidence to suggest that 53 alcoholic solutions may be more effective than aqueous solutions. [1, 2] Nonetheless, both 54 meta-analyses agree that the marked heterogeneity in type and application of antiseptic 55 preparations between available studies hinder direct comparisons between them. [1, 2] This, 56 combined with the overall low statistical power of almost all available studies makes it 57 difficult to draw firm conclusions about the proposed superiority of alcoholic solutions. [1, 9] 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 12 BMJ Open

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 Previous research in the Mackay region [6, 7, 19, 20] and recent interviews (pers. comm.) 4 have revealed that the majority of Mackay general practitioners (GPs) use CHG in 5 preference to PVI, one of the reasons behind this being perceptions of ‘messiness’ and skin 6 staining. Therefore for practical reasons, our team has chosen to examine the difference 7 between alcoholic and aqueous CHG, rather than comparing the relative efficacy of CHG 8 and PVI. 9 10 The aim of our study is to determine whether there is a difference in the incidence of SSI 11 after minor skin excisions in general practice (clean surgery) when alcoholic chlorhexidine 12 gluconate is used for preoperative skin antisepsis in comparison to aqueous chlorhexidine 13 gluconate. 14 15 We hope thisFor research peer will provide morereview authoritative direction only about skin antisepsis to 16 clinicians carrying out clean surgery. If our research demonstrates a difference in efficacy 17 between alcoholic and aqueous CHG, its dissemination may lead a change in behaviour 18 which may serve to reduce overall incidence of SSI after minor skin excisions in general 19 practice in Australia. 20 21 METHODS AND ANALYSIS 22 23 Study centre 24 25 This study will be conducted in four private general practices in Mackay, Queensland 26 (latitude 21E8S). Mackay is a rural centre with around 100 local GPs servicing a population 27 of 112,798. [21] The study team have previously carried out a number of successful 28 randomised controlled trials (RCTs) on minor skin excisions within the Mackay region. [6, 8, 29 19, 20] 30

31 Study design 32

33 This is a prospective randomised controlled trial (RCT) comparing the intervention of 0.5% http://bmjopen.bmj.com/ 34 35 CHG in 70% alcohol with 0.5% aqueous CHG surgical skin preparation for the prevention 36 of SSI following minor skin excisions conducted in general practice. Data will be collected 37 over a 9 month period. The study will be conducted in accordance with the CONSORT 38 statement. 39 40 Intervention 41 42 The intervention of surgical skin antisepsis with 0.5% CHG in 70% alcohol will be compared on October 3, 2021 by guest. Protected copyright. 43 with the control group of 0.5% CHG aqueous solution. The 0.5% concentration of CHG 44 aligns with guidelines released by the Queensland Centre for Healthcare Related Infection 45 Surveillance and Prevention. [22] The 70% alcoholic concentration of CHG is standard for

46 alcoholic preoperative skin preparations. [2] 47 48 Recruitment of study participants 49 50 Consecutive patients over the age of 18 with capacity to give informed consent, who present 51 to participating GP practices for minor skin excisions, will be invited to participate. The 52 practice nurses will be responsible for recruitment. A number of provisions have been 53 developed to assure informed consent. Firstly, all eligible participants will be provided with a 54 participant information sheet before giving written informed consent. In addition, practice 55 nurses, rather than practice GPs, will recruit patients. This is intended to minimise risk of 56 perceived coercion, as nurses are somewhat less responsible for direct decisions regarding 57 patient care than the patient’s GP. Further, all nurses involved in the trial will receive formal 58 training regarding appropriate consenting procedures. 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 12

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4 Randomisation 5 6 In this prospective RCT, randomisation will be performed at the level of the patient. The 7 random sequence will be generated from a computer-generated random number table. 8 Allocation concealment will be attained by using sealed, numbered, tamperproof opaque 9 envelopes such that neither the patient, nor the clinicians involved in their care, will be aware 10 of their allocation until after they have consented to be a part of the trial, thereby minimizing 11 selection and confounding bias. The research team involved in the assessment or treatment 12 of patients will have no role in the assignment process. The patients will be blinded to 13 treatment allocation, although there are differences in the alcoholic skin preparation which 14 are identifiable to the patient. Blinding of the operating doctors to the assigned skin 15 antiseptic is Fornot feasible peergiven the differing review smell of the two solutions. only 16 17 Inclusion criteria 18 19 • All patients over the age of 18 undergoing minor skin procedures at the participating 20 practices during the study period who: 21 o have capacity to give informed consent and; 22 o are able to return for removal of sutures 23 • Patients who are not presenting for: 24 o excision of sebaceous cyst 25 o 26 suturing of lacerations 27 o excisions not requiring sutures, such as shave biopsies 28 o punch biopsies 29 o excisions on body sites where adrenaline is contraindicated 30 31 Exclusion Criteria 32 33 • Allergy to alcohol or chlorhexidine http://bmjopen.bmj.com/ 34 • Evidence of infection at or adjacent to the operative site 35 • Current use of antibiotics 36 37 • Clinical indication for antibiotic treatment following excision (besides SSI) 38 • Peri-ocular excisions 39 • Patients with a primary language other than English for which certiﬁed translation 40 services for that language are not available 41 42 Surgical and Wound Protocol on October 3, 2021 by guest. Protected copyright. 43 44 A surgical and wound management protocol will standardise the management across both 45 study arms. The protocol is modelled on previous protocols used in similar trials, as well as 46 international guidelines, [6, 8, 10, 19, 20] and was developed in consultation with 47 participating doctors and nurses. As per this protocol, skin antisepsis will be applied in a 48 consistent manner for both study arms - drapes, gloves, sutures, local anaesthetic and 49 dressings will be the same across all sites and post-operative wound care processes will 50 be identical, with all patients receiving a standard set of verbal and written post-operative 51 wound care instructions. 52 53 Outcome Measures 54 55 Primary outcome measure 56

57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 12 BMJ Open

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 The primary outcome measure is the incidence of post-operative surgical site infection 4 occurring within 30 days of the procedure (deﬁned below). Patients’ wounds will be 5 assessed for evidence of SSI when they present for removal of sutures; at any other time if 6 they present for wound review due to signs and/or symptoms of SSI, or opportunistically if 7 they re-present for any other reason. Wound assessment will be carried out by doctors or 8 nurses at each general practice and the presence or absence of SSI recorded... There will 9 be standardised in-house training regarding the definition of infection. 10 All infections will be photographed and assessed for infection by a second blinded 11 independent outcome assessor to improve validity and reliability. 12 If patients are deemed to have an SSI, they will be treated with antibiotics as clinically 13 14 indicated, and as per standard practice, all wounds with a purulent discharge will be 15 swabbed. For peer review only 16 17 Secondary outcome measure 18 19 Secondary outcome measures will be 20 1. Adverse reactions to the preoperative skin antiseptic agent, manifesting as any one 21 of 22 a. anaphylaxis 23 b. skin irritation or contact dermatitis 24 c. rash 25 2. Microbiology of infected wounds with a purulent discharge, and any patterns of 26 antibiotic resistance 27 28 Definitions 29 30 Surgical site infection 31 32 Surgical site infection will be determined in accordance with a modified version of the CDC 33 definition for superficial surgical site infection: 34 • Infection occurs within 30 days after the excision, AND http://bmjopen.bmj.com/ 35 • Infection involves ONLY skin or subcutaneous tissue of the incision, AND 36 • At least ONE of the following: 37 o Purulent drainage with or without laboratory confirmation from the superficial 38 incision 39 o At least one of the following signs or symptoms: pain or tenderness, localised 40 swelling, redness or heat 41 o Diagnosis of superficial SSI by the GP on October 3, 2021 by guest. Protected copyright. 42 • Stitch abscesses, characterised by minimal inflammation and discharge confined to 43 the points of suture penetration, will not be included as SSIs [6, 8, 10] 44

45 Data Collection 46

47 Data will be primarily collected through use of a written spreadsheet which will be completed 48 by practice nurses. A member of the research team will visit practices on a fortnightly basis 49 50 to audit the data collection. 51 52 Baseline data will be collected regarding patient demographics, including age, sex, 53 occupation and smoking status, as well as co-morbidities such as diabetes mellitus or 54 peripheral vascular disease and current relevant medications, such as anticoagulants and 55 immunosuppressants. Data will also be recorded regarding the excision itself, such as the 56 incision length, the suture size and the type of excision performed (i.e. simple, flap, 2-layer 57 procedure). A body site map will be used to record excision site and the histology of the 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 12

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 lesion will also be recorded. Each item of data has been chosen based on data extracted 4 from other trials on risk factors for SSI. [5-8] 5 6 Sample Size Calculation 7 8 Our sample size was calculated on the basis of previous studies of surgical site infection in 9 the Mackay region - pooled analyses showed a weighted mean SSI rate of 9.35%, which has 10 been rounded up to 10% as our predicted baseline infection rate. [6, 19, 20] We decided that 11 an absolute reduction in the SSI rate of 5% (to 5%) would be clinically significant. To reach 12 this conclusion with statistical confidence, a power in excess of 80% and a significance level 13 of 0.05, a total of 435 patients would be required in the intervention group and 435 patients 14 in the control group, thus 870 in total. 15 For peer review only 16 Our previous similar trials have had a drop-out rate of less than 5%, so we will enrol an 17 additional 50 patients to counter potential attrition, providing a final sample size of 920. [6, 8, 18 19, 20] 19 20 21 22 Data Analysis 23 24 The primary analysis is an intention to treat analysis including all participants who undergo 25 randomisation. The analysis will be performed taking the individual person as the unit of 26 analysis. All reported p values will be two tailed and for each analysis p<0.05 will be 27 considered statistically significant. The main analysis will follow the intention-to-treat 28 principle. Baseline data across control and intervention groups will first be assessed for 29 marked differences. The incidence of SSI (the primary dependent variable) in each of the 30 two groups of the trial will then be compared using Pearson’s chi-square test. Multivariable 31 logistic regression analysis will be applied in case differences exist between intervention and 32 control groups at baseline and the analysis requires adjustment for confounders. We will 33 also carry out sensitivity testing for lost to follow-up patients and per protocol analysis for http://bmjopen.bmj.com/ 34 35 non-compliers to assess for the possible effects of systematic biases on results. 36 37 Potential problems 38 39 Based on our previous studies, we feel that recruitment of adequate patient numbers is 40 feasible, however if we fail to recruit patients, we will invite additional General Practices to 41 participate. 42 In our previous studies we have found that assessing for infection at time of removal of on October 3, 2021 by guest. Protected copyright. 43 sutures facilitates a high rate of follow-up. Any patients not followed-up will be analysed on 44 an intention to treat basis. 45 We have not planned to perform an interim analysis as we feel that variation of antisepsis is 46 a minor intervention, and the outcome of SSI is usually a minor medical issue which is 47 treated with a course of antibiotics. An interim analysis would further increase the required 48 sample size reducing the feasibility of our trial. 49 50 51 52 DISSEMINATION 53 54 This project is due for completion one year after commencement of data collection. The 55 translation of important ﬁndings to clinical practice will be facilitated through dissemination in 56 conference presentations and journals as well as electronic media. The researchers will also 57 aim to publish their findings on a range of Australian FOAM (Free Open Access Meducation) 58 websites to reach the next generation of technology-savvy health professionals. A written lay 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 12 BMJ Open

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 summary of the results will also be displayed at the participating general practices for the 4 information of study participants. 5 6 ETHICAL CONSIDERATIONS 7 8 This project has been reviewed and approved by the James Cook University Human 9 Research Ethic Committee; HREC approval ID H6065. We do not expect the interventions of 10 this study to place participants at any significant risk of harm, as we hypothesise a lower 11 incidence of SSI in the intervention group due to the use of alcoholic CHG. In any case, SSI 12 is a minor condition which can be easily treated without significant long term sequelae. To 13 assure privacy and confidentiality, all data spreadsheets and consent forms will be kept in a 14 locked cupboard throughout the trial, then transferred to a locked safe at the conclusion of 15 the trial, whereFor they willpeer be kept for 15review years. Patients willonly be de-identified in all data 16 collection. 17 18 DISCUSSION 19 20 21 Very few large RCTs are conducted in a primary care setting [23, 24]. Difficulties have been 22 reported in recruiting both patients and clinicians [25], and RCTs have been reported as 23 being methodologically and practically difficult to conduct in general practice. [11] However, 24 it is important that clinical practice be informed by adequate primary care evidence. 25 Otherwise GPs, the end-user of the research process, who attempt to practise evidence- 26 27 based medicine may have flawed tools and the guidelines they use may not be applicable to 28 the patients they see. [26, 27] 29

30 31 Funding for primary care research in Australia is very limited, particularly compared with UK 32 and Netherlands with only 2% of NHMRC grants awarded to primary care research between 33

2000 and 2008. [28] General practice based research differs in many ways from hospital http://bmjopen.bmj.com/ 34 based research in issues of funding, feasibility and pragmatism, and we have used our 35 36 experience from conducting previous successful trials in general practice to inform the 37 design and methods of the present study[6, 8, 19, 20]. Our study will be conducted for a total 38 cost of $20,000 which is similar to our previous trials. 39 40 41 Skin excisions form a large proportion of the workload of Australian GPs [17] and this is even 42 greater in Queensland, the state with the highest incidence of skin cancer in the world. [29] on October 3, 2021 by guest. Protected copyright. 43 This effect is magnified in regional towns such as Mackay, where there are no permanent 44 dermatologists or plastic surgeons. Using a research question which is relevant to our 45 clinical situation increases the feasibility of recruitment of patients because of our high case 46 load of patients presenting for skin excisions. 47 Our research question is practical and clinically relevant. Local clinicians do not use betadine 48 antisepsis because of perceptions of ‘messiness’ and skin staining. Therefore, in order to be 49 pragmatic our team has chosen to examine the difference between alcoholic and aqueous 50 CHG, rather than comparing the relative efficacy of CHG and PVI. We have also found that 51 using a clinically relevant research question engages general practitioners and practice 52 nurses and facilitates practitioner recruitment, as well as increasing the potential for 53 translation into clinical practice[30]. 54 55 Our surgical and wound management protocol was developed in consultation with 56 participating doctors and practice nurses, which again increases the ownership and 57 practicality of the project. Occasionally, scientific rigor may be compromised at the expense 58 of pragmatism. For instance in contrast with hospital based research, it is simply not 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 12

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 practically or financially feasible to have an independent outcome assessor assessing each 4 wound at each of the three geographically dispersed practices. To compensate for this, 5 infections will be photographed and assessed for infection by a second independent and 6 blinded outcome assessor, who will re-assess every wound. 7 It is also not feasible to have an independent researcher to recruit patients. In our trial 8 practice nurses will be responsible for recruitment. A number of provisions have been 9 10 developed to assure informed consent. This is intended to minimise risk of perceived 11 coercion, as nurses are somewhat less responsible for direct decisions regarding patient 12 care than the patient’s GP. The practice nurses are also responsible for data collection and 13 are paid on a fee per service basis that compensates them for their time. The study involves 14 very little extra work for the participating GPs – they are not responsible for any data 15 collection, andFor were only peer required to havereview knowledge of the only process involved in order to 16 answer any possible queries. 17 The trial will provide guidance to GPs regarding skin antisepsis, and will inform current 18 clinical guidelines and healthcare worker education. If we detect a measurable decrease in 19 incidence of SSI with alcoholic CHG, this may result in a change in clinical practice which 20 could reduce SSI rates following clean surgery. As this is a pragmatic trial, the findings can 21 potentially be immediately translated into clinical practice. 22 23 24 25 26 ACKNOWLEDGEMENTS 27 The authors thank Ms Debbie Kimber, Ms Julie O’Sullivan, Dr Sheldon Browning, Dr Luke 28 Notely, Dr Andrew O’Neil, Dr Andrea Cosgrove. 29 30 Authors Contributions 31 CH conceived the study idea and oversaw the development of the study design and 32 protocol. DC led the development of the study protocol. AH, MD, and JB assisted with the 33

development of the study design and protocol. PB assisted with the sample size calculation http://bmjopen.bmj.com/ 34 and statistics. All authors contributed to the drafting of the manuscript 35

36 COMPETING INTERESTS 37 None declared 38

39 40 FUNDING 41 James Cook University Honours program grant: $1000 42 Royal Australian College of General Practitioners PWH Grieve Award: $2 500 on October 3, 2021 by guest. Protected copyright. 43 Royal Australian College of General Practitioners Family Medical Care Education and 44 Research Grant: $13 500 45 46 47 48 REFERENCES 49 50 1. Dumville, J.C., et al., Preoperative skin antiseptics for preventing surgical wound infections 51 after clean surgery. Cochrane Database Syst Rev, 2015. 4: p. CD003949. 52 2. Maiwald, M. and E.S. Chan, The forgotten role of alcohol: a systematic review and meta- 53 analysis of the clinical efficacy and perceived role of chlorhexidine in skin antisepsis. PLoS 54 One, 2012. 7(9): p. e44277. 55 3. Lee, I., et al., Systematic review and cost analysis comparing use of chlorhexidine with use of 56 iodine for preoperative skin antisepsis to prevent surgical site infection. Infect Control Hosp 57 Epidemiol, 2010. 31(12): p. 1219-29. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 12 BMJ Open

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4. Health, Q.G.D.o., Surgical Skin Disinfection Guideline. 2015. 4 5. Heal, C., P. Buettner, and S. Browning, Risk factors for wound infection after minor surgery in 5 general practice. Med J Aust, 2006. 185(5): p. 255-8. 6 6. Heal, C., et al., Comparing non-sterile to sterile gloves for minor surgery: a prospective 7 randomised controlled non-inferiority trial. Med J Aust, 2015. 202(1): p. 27-31. 8 7. Heal, C.F., P.G. Buettner, and H. Drobetz, Risk factors for surgical site infection after 9 dermatological surgery. Int J Dermatol, 2012. 51(7): p. 796-803. 10 8. Smith, S.C., C.F. Heal, and P.G. Buttner, Prevention of surgical site infection in lower limb skin 11 12 lesion excisions with single dose oral antibiotic prophylaxis: a prospective randomised 13 placebo-controlled double-blind trial. BMJ Open, 2014. 4(7): p. e005270. 14 9. Kamel, C., et al., Preoperative Skin Antiseptic Preparations and Application Techniques for 15 PreventingFor Surgical peer Site Infections: review A Systematic Review only of the Clinical Evidence and 16 Guidelines, in Rapid Response Reports. 2011: Ottawa (ON). 17 10. Mangram, A.J., et al., Guideline for prevention of surgical site infection, 1999. Hospital 18 Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol, 1999. 20(4): 19 p. 250-78; quiz 279-80. 20 11. Amici, J.M., et al., A prospective study of the incidence of complications associated with 21 dermatological surgery. Br J Dermatol, 2005. 153(5): p. 967-71. 22 12. Dixon, A.J., M.P. Dixon, and J.B. Dixon, Randomized clinical trial of the effect of applying 23 ointment to surgical wounds before occlusive dressing. Br J Surg, 2006. 93(8): p. 937-43. 24 25 13. Futoryan, T. and D. Grande, Postoperative wound infection rates in dermatologic surgery. 26 Dermatol Surg, 1995. 21(6): p. 509-14. 27 14. Lathlean, S., Skin cancer in general practice in South Australia. A five year study. Aust Fam 28 Physician, 1999. 28 Suppl 1: p. S28-31. 29 15. Culver, D.H., et al., Surgical wound infection rates by wound class, operative procedure, and 30 patient risk index. National Nosocomial Infections Surveillance System. Am J Med, 1991. 31 91(3B): p. 152S-157S. 32 16. Platt, R., Antibiotic prophylaxis in clean surgery: does it work? Should it be used if it does? 33

New Horiz, 1998. 6(2 Suppl): p. S53-7. http://bmjopen.bmj.com/ 34 17. Askew, D.A., et al., Skin cancer surgery in Australia 2001-2005: the changing role of the 35 general practitioner. Med J Aust, 2007. 187(4): p. 210-4. 36 18. Darouiche, R.O., et al., Chlorhexidine-Alcohol versus Povidone-Iodine for Surgical-Site 37 38 Antisepsis. N Engl J Med, 2010. 362(1): p. 18-26. 39 19. Heal, C., et al., Does single application of topical chloramphenicol to high risk sutured 40 wounds reduce incidence of wound infection after minor surgery? Prospective randomised 41 placebo controlled double blind trial. BMJ, 2009. 338(338): p. a2812. 42 20. Heal, C., et al., Can sutures get wet? Prospective randomised controlled trial of wound on October 3, 2021 by guest. Protected copyright. 43 management in general practice. BMJ, 2006. 332(7549): p. 1053-6. 44 21. Australian Bureau of Statistics, 2011 Census QuickStats: Mackay, Commonwealth Bureau of 45 Census and Statistics, Editor. 2013: Canberra, . 46 22. Queensland Centre for Healthcare Related Infection Surveillance and Prevention and 47 Tuberculosis Prevention, Guideline for Surgical Skin Disinfection. 2012. 48 23. Mendis, K., et al., A bibliometric analysis of Australian general practice publications from 49 1980 to 2007 using PubMed. Inform Prim Care, 2010. 18(4): p. 223-33. 50 51 24. Mendis, K. and I. Solangaarachchi, PubMed perspective of family medicine research: where 52 does it stand? Fam Pract, 2005. 22(5): p. 570-5. 53 25. Rendell, J.M., R.D. Merritt, and J.R. Geddes, Incentives and disincentives to participation by 54 clinicians in randomised controlled trials. Cochrane Database Syst Rev, 2007(2): p. 55 MR000021. 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 12

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 26. Steel, N., et al., A review of clinical practice guidelines found that they were often based on 4 evidence of uncertain relevance to primary care patients. J Clin Epidemiol, 2014. 67(11): p. 5 1251-7. 6 27. Scullard, P., et al., Does the evidence referenced in NICE guidelines reflect a primary care 7 population? Br J Gen Pract, 2011. 61(584): p. e112-7. 8 28. McIntyre, E.L., D. Mazza, and N.P. Harris, NHMRC funding for primary health care research, 9 2000-2008. Med J Aust, 2011. 195(4): p. 230. 10 29. Buettner, P.G. and B.A. Raasch, Incidence rates of skin cancer in Townsville, Australia. Int J 11 12 Cancer, 1998. 78(5): p. 587-93. 13 30. Heal, C.F., C. Veitch, and R. Preston, Practice based research - lessons from the field. Aust 14 Fam Physician, 2008. 37(5): p. 381-4. 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on October 3, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from

Protocol for randomized controlled trial comparing aqueous with alcoholic chlorhexidine antisepsis for the prevention of superficial surgical site infection after minor surgery in general practice – the AVALANCHE trial.

For peer review only Journal: BMJ Open

Manuscript ID bmjopen-2016-011604.R1

Article Type: Protocol

Date Submitted by the Author: 03-May-2016

Secondary Subject Heading: Evidence based practice

Keywords: PRIMARY CARE, WOUND MANAGEMENT, SURGERY

on October 3, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 25 BMJ Open

1 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 TITLE 4 5 Protocol for randomized controlled trial comparing aqueous with alcoholic chlorhexidine 6 antisepsis for the prevention of superficial surgical site infection after minor surgery in 7 general practice – the AVALANCHE trial. 8 9 CF Heal1, 2, D Charles1, A Hardy1, M Delpachitra1, J Banks 1M Wohlfahrt1, P Buettner3 10 11 12 13 14 Corresponding Author 15 For peer review only 16 Professor Clare Heal 17 Promotional Chair 18 Discipline of General Practice and Rural Medicine 19 Mackay Clinical School 20 College of Medicine and Dentistry 21 James Cook University 22 Mackay Base Hospital, Bridge Rd 23 Mackay QLD 4740 Australia 24 [email protected] 25 26 Author Affiliations 27 28 1 Discipline of General Practice and Rural Medicine 29 Mackay Clinical School 30 College of Medicine and Dentistry 31 James Cook University 32

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 25

2 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4 ABSTRACT 5

6 Introduction: 7 Surgical site infection (SSI) after minor skin excisions has a significant impact on patient 8 morbidity and healthcare resources. Skin antisepsis prior to surgical incision is used to 9 10 prevent surgical site infection, and is performed routinely worldwide. However, in spite of the 11 routine use of skin antisepsis, there is no consensus regarding which antiseptic agents are 12 most effective. The AVALANCHE trial will compare Aqueous Versus Alcoholic Antisepsis 13 with Chlorhexidine for Skin Excisions. 14 15 Methods andFor analysis : peer review only 16 The study design is a prospective randomised controlled trial (RCT) with the aim of 17 investigating the impact of two different antiseptic preparations upon the incidence of 18 superficial surgical site infection in patients undergoing minor skin excisions. The 19 intervention of 0.5% chlorhexidine gluconate (CHG) in 70% alcohol will be compared with 20 0.5% chlorhexidine gluconate in aqueous solution. The trial will be conducted in four 21 Australian General Practices over a 9-month period, with 920 participants to be recruited. 22 Consecutive patients presenting for minor skin excisions will be eligible to participate. 23 Randomisation will be on the level of the patient. The primary outcome is superficial surgical 24 site infection in the first 30 days following the excision. Secondary outcomes will be adverse 25 effects including anaphylaxis, skin irritation, contact dermatitis and rash and patterns of 26 antibiotic resistance. 27 28 Ethics and Dissemination: The study has been approved by the James Cook University 29 Human Research Ethics Committee (HREC). Findings will be disseminated in conference 30 presentations and journals and through online electronic media. 31 32 Registration Details: The trial is registered with the Australian New Zealand Clinical Trials 33 Registry, registration number: ACTRN12615001045505. http://bmjopen.bmj.com/ 34 35 Discussion: RCTs conducted in General Practice differ from hospital based projects in 36 terms of feasibility, pragmatism and funding. The success of this trial will be cemented in 37 the fact that the research question was established by a group of GPs who identified an 38 interesting question which is relevant to their clinical practice and not answered by current 39 40 evidence. 41

on October 3, 2021 by guest. Protected copyright. 42 43 Study Strengths 44

45 46 • Practical clinical question relevant to clinical practice 47 • Recruitment of clinicians and participants will be very feasible 48 • Independent outcome assessor will assess photographs of all infections 49 • Conduct of trial in General Practice will provide clinically relevant results to end user 50 51 Study Limitations 52 • Diagnosis of infection has element of subjectivity 53 • Practice based outcome assessors will not be blinded 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 25 BMJ Open

3 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on October 3, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 25

4 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 BACKGROUND 4 5 It is routine practice prior to surgery to carry out preoperative cleansing of the skin with 6 antiseptic preparations at the site of surgical incisions (preoperative skin antisepsis). [1-3] 7 The purpose of preoperative skin antisepsis is to reduce the incidence of surgical site 8 infection (SSI) by removing microorganisms on the skin through a combination of 9 mechanical removal and chemical killing. [1, 3, 4] The incidence of SSI arising from surgery 10 in general practice is usually 1-3%, but has been recorded at rates as high as 10%. [5-8] 11 The consequences of SSI include pain and discomfort for patients, increased healthcare- 12 associated costs and temporarily reduced occupational and recreational productivity and 13 functionality. [1, 3, 4, 9] As a result, reducing the incidence of SSIs is in the interest of all 14 stakeholders in healthcare. 15 For peer review only 16 The most commonly used preoperative skin antiseptic preparations are povidone iodine 17 (PVI) and chlorhexidine gluconate (CHG). Both are available in aqueous and alcoholic 18 preparations and in multiple different concentrations. [1, 3, 4] Both PVI and CHG are 19 effective against a wide range of gram positive and negative bacteria, viruses and fungi, 20 though CHG has a more appreciable residual antiseptic activity on the skin after application. 21 [1, 2, 4] 22 23 Surprisingly, despite the routine use of preoperative skin antisepsis, there is no definitive 24 scientific consensus regarding which antiseptic preparation is most effective at preventing 25 SSI. [1, 4, 9] There is even less definitive data available for antisepsis prior to clean surgery, 26 which the Centre for Disease Control (CDC) defines as “an uninfected operation in which no 27 inflammation is encountered and the respiratory, alimentary, genital or uninfected urinary 28 tract is not entered” (CDC, 2011). [10] Prior studies of wound infection after minor surgery in 29 General Practice in the Mackay region have shown a SSI rate of approximately 9.5% - an 30 incidence which is much higher than expected based on published results of similar cohorts 31 in other regions of Australia and the world. [11-14] The reason for this high infection rate is 32 unclear, but may be related to the hot, humid environment or to patient behaviour in our rural 33 setting. On one hand, this infection rate is suboptimal, as the CDC suggests that an http://bmjopen.bmj.com/ 34 35 acceptable rate of SSI after clean minor surgery is less than 5% [15]. However, in settings 36 where there is a low risk of infection after clean surgery, studies of over 5,000 procedures 37 may be required to detect a clinically relevant difference in infection from an intervention with 38 statistical confidence, making such trials unfeasible [16]. Conversely, because of the high 39 rate of infection in our patient cohort and the high minor surgery workload in rural general 40 practice [17], a study of skin antisepsis for the prevention of SSI in our setting is highly 41 feasible. 42 on October 3, 2021 by guest. Protected copyright. 43 In general, the evidence base guiding appropriate selection of antiseptic agents is poor. A 44 landmark study found 2% CHG in 70% isopropyl alcohol to be superior to an aqueous 45 solution of 10% PVI, however, as alcohol is known to have significant antimicrobial 46 properties this was likely to be an active treatment component in this study [18]. Most recent 47 meta-analyses, including a Cochrane review, concur that it is difficult to make conclusive

48 statements about whether there are differences in the efficacy of CHG and PVI [1, 2, 9] and 49 the latest guideline on surgical skin antisepsis released by the Queensland Department of 50 Health agrees with this position. [4] The Queensland Department of Health does however 51 endorse the use of alcoholic solutions in preference to aqueous preparations, [4] and two 52 recent meta-analyses have asserted that there is some evidence to suggest that alcoholic 53 solutions may be more effective than aqueous solutions. [1, 2] Nonetheless, both meta- 54 analyses agree that the marked heterogeneity in type and application of antiseptic 55 preparations between available studies hinder direct comparisons between them. [1, 2] This, 56 combined with the overall low statistical power of almost all available studies makes it 57 difficult to draw firm conclusions about the proposed superiority of alcoholic solutions. [1, 9] 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 25 BMJ Open

5 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 Previous research in the Mackay region [6, 7, 19, 20] and recent interviews (pers. comm.) 4 have revealed that the majority of Mackay general practitioners (GPs) use CHG in 5 preference to PVI, one of the reasons behind this being perceptions of ‘messiness’ and skin 6 staining. Therefore for practical reasons, our team has chosen to examine the difference 7 between alcoholic and aqueous CHG, rather than comparing the relative efficacy of CHG 8 and PVI. 9 10 The aim of our study is to determine whether there is a difference in the incidence of SSI 11 after minor skin excisions in general practice (clean surgery) when alcoholic chlorhexidine 12 gluconate is used for preoperative skin antisepsis in comparison to aqueous chlorhexidine 13 gluconate. 14 15 We hope thisFor research peer will provide morereview authoritative direction only about skin antisepsis to 16 clinicians carrying out clean surgery. If our research demonstrates a difference in efficacy 17 between alcoholic and aqueous CHG, its dissemination may lead a change in behaviour 18 which may serve to reduce overall incidence of SSI after minor skin excisions in general 19 practice in Australia. 20 21 METHODS AND ANALYSIS 22 23 Study centre 24 25 This study will be conducted in four private general practices in Mackay, Queensland 26 (latitude 21E8S). Mackay is a rural centre with around 100 local GPs servicing a population 27 of 112,798. [21] The study team have previously carried out a number of successful 28 randomised controlled trials (RCTs) on minor skin excisions within the Mackay region. [6, 8, 29 19, 20] 30

31 Study design 32

33 This is a prospective randomised controlled trial (RCT) comparing the intervention of 0.5% http://bmjopen.bmj.com/ 34 35 CHG in 70% alcohol with 0.5% aqueous CHG surgical skin preparation for the prevention 36 of SSI following ‘minor skin excisions’ – benign or malignant skin lesions excised under 37 local anaesthetic - conducted in general practice. Data will be collected over a 9 month 38 period. The study will be conducted in accordance with the CONSORT statement. 39 40 Intervention 41 42 The intervention of surgical skin antisepsis with 0.5% CHG in 70% alcohol will be compared on October 3, 2021 by guest. Protected copyright. 43 with the control group of 0.5% CHG aqueous solution. The 0.5% concentration of CHG 44 aligns with guidelines released by the Queensland Centre for Healthcare Related Infection 45 Surveillance and Prevention. [22] The 70% alcoholic concentration of CHG is standard for 46 alcoholic preoperative skin preparations. [2] The antiseptic solutions will be purchased from

47 an independent supplier with research funding. 48 49 Recruitment of study participants 50 51 Consecutive patients over the age of 18 with capacity to give informed consent, who present 52 to participating GP practices for minor skin excisions, will be invited to participate. The 53 practice nurses will be responsible for recruitment. A number of provisions have been 54 developed to assure informed consent. Firstly, all eligible participants will be provided with a 55 participant information sheet before giving written informed consent. In addition, practice 56 nurses, rather than practice GPs, will recruit patients. This is intended to minimise risk of 57 perceived coercion, as nurses are somewhat less responsible for direct decisions regarding 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 25

6 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 patient care than the patient’s GP. Further, all nurses involved in the trial will receive formal 4 training regarding appropriate consenting procedures. 5 6 Randomisation 7 8 In this prospective RCT, randomisation will be performed at the level of the patient with an 9 allocation ratio of 1:1. The random sequence will be generated from a computer-generated 10 random number table. Allocation concealment will be attained by using sealed, numbered, 11 tamperproof opaque envelopes such that neither the patient, nor the clinicians involved in 12 their care, will be aware of their allocation until after they have consented to be a part of the 13 trial, thereby minimizing selection and confounding bias. The research team involved in the 14 assessment or treatment of patients will have no role in the assignment process. The 15 patients will Forbe blinded topeer treatment allocation, review although there areonly differences in the alcoholic 16 skin preparation which are identifiable to the patient. Blinding of the operating doctors to the 17 assigned skin antiseptic is not feasible given the differing smell of the two solutions. The 18 practice nurse or doctor assessing outcomes will be blinded to the treatment allocation, as 19 will the practice nurse collecting data and the investigator team. 20 21 Inclusion criteria 22 23 • All patients over the age of 18 undergoing minor skin procedures at the participating 24 practices during the study period who: 25 o have capacity to give informed consent and; 26 o are able to return for removal of sutures 27 • Patients who are not presenting for: 28 o excision of sebaceous cyst 29 o 30 suturing of lacerations 31 o excisions not requiring sutures, such as shave biopsies 32 o punch biopsies 33 o excisions on body sites where adrenaline is contraindicated 34 http://bmjopen.bmj.com/ 35 Exclusion Criteria 36 37 • Allergy to alcohol or chlorhexidine 38 • Evidence of infection at or adjacent to the operative site 39 • Current use of antibiotics 40 41 • Clinical indication for antibiotic treatment following excision (besides SSI) 42 • Peri-ocular excisions on October 3, 2021 by guest. Protected copyright. 43 • Patients with a primary language other than English for which certiﬁed translation 44 services for that language are not available 45 46 Surgical and Wound Protocol 47 48 A surgical and wound management protocol will standardise the management across both 49 study arms. The protocol is modelled on previous protocols used in similar trials, as well as 50 international guidelines, [6, 8, 10, 19, 20] and was developed in consultation with 51 participating doctors and nurses. As per this protocol, skin antisepsis will be applied in a 52 consistent manner for both study arms - drapes, gloves, sutures, local anaesthetic and 53 dressings will be the same across all sites and post-operative wound care processes will 54 be identical, with all patients receiving a standard set of verbal and written post-operative 55 wound care instructions. 56

57 Outcome Measures 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 25 BMJ Open

7 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4 Primary outcome measure 5 6 The primary outcome measure is the incidence of post-operative surgical site infection 7 occurring within 30 days of the procedure (deﬁned below). Patients’ wounds will be 8 assessed for evidence of SSI when they present for removal of sutures; at any other time if 9 they present for wound review due to signs and/or symptoms of SSI, or opportunistically if 10 they re-present for any other reason. Wound assessment will be carried out by doctors or 11 nurses at each general practice and the presence or absence of SSI recorded. There will be 12 standardised in-house training regarding the definition of infection. 13 All infections will be photographed and assessed for infection by a second blinded 14 independent outcome assessor to improve validity and reliability. 15 For peer review only 16 If patients are deemed to have an SSI, they will be treated with antibiotics as clinically 17 indicated, and as per standard practice, all wounds with a purulent discharge will be 18 swabbed. 19 20 Secondary outcome measure 21 22 Secondary outcome measures will be 23 1. Adverse reactions to the preoperative skin antiseptic agent, manifesting as any one 24 of 25 a. anaphylaxis 26 b. skin irritation or contact dermatitis 27 c. rash 28 2. Microbiology of infected wounds with a purulent discharge, and any patterns of 29 antibiotic resistance 30 31 Definitions 32 33 Surgical site infection 34 http://bmjopen.bmj.com/ 35 Surgical site infection will be determined in accordance with a modified version of the CDC 36 definition for superficial surgical site infection: 37 • Infection occurs within 30 days after the excision, AND 38 • Infection involves ONLY skin or subcutaneous tissue of the incision, AND 39 • At least ONE of the following: 40 o Purulent drainage with or without laboratory confirmation from the superficial 41 incision on October 3, 2021 by guest. Protected copyright. 42 o At least one of the following signs or symptoms: pain or tenderness, localised 43 swelling, redness or heat 44 o Diagnosis of superficial SSI by the GP 45 46 • Stitch abscesses, characterised by minimal inflammation and discharge confined to 47 the points of suture penetration, will not be included as SSIs [6, 8, 10] 48 49 Data Collection 50 51 Data will be primarily collected through use of a written spreadsheet which will be completed 52 by practice nurses. A member of the research team will visit practices on a fortnightly basis 53 to audit the data collection. 54 55 Baseline data will be collected regarding patient demographics, including age, sex, 56 occupation and smoking status, as well as co-morbidities such as diabetes mellitus or 57 peripheral vascular disease and current relevant medications, such as anticoagulants and 58 immunosuppressants. Data will also be recorded regarding the excision itself, such as the 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 25

8 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 incision length, the suture size and the type of excision performed (i.e. simple, flap, 2-layer 4 procedure). A body site map will be used to record excision site and the histology of the 5 lesion will also be recorded. Each item of data has been chosen based on data extracted 6 from other trials on risk factors for SSI. [5-8] 7 8 Sample Size Calculation 9 10 Our sample size was calculated on the basis of three previous studies of surgical site 11 infection in the Mackay region [6, 19, 20] . These studies used mostly aqueous chlorhexidine 12 as surgical antisepsis. Pooled analyses showed a weighted mean SSI rate of 9.35%, which 13 has been rounded up to 10% as our predicted baseline infection rate. [6, 19, 20] We decided 14 that an absolute reduction in the SSI rate of 5% (to 5%) would be clinically significant. To 15 reach this For conclusion withpeer statistical review confidence, a power only in excess of 80% and a 16 significance level of 0.05, a total of 435 patients would be required in the intervention group 17 and 435 patients in the control group, thus 870 in total. 18 19 Our previous similar trials [6,8,19,20[ have had a drop-out rate of less than 5%, so we will 20 enrol an additional 50 patients to counter potential attrition, providing a final sample size of 21 920. [6, 8, 19, 20] 22 23 24 25 Data Analysis 26 27 The primary analysis is an intention to treat analysis including all participants who undergo 28 randomisation. The analysis will be performed taking the individual person as the unit of 29 analysis. All reported p values will be two tailed and for each analysis p<0.05 will be 30 considered statistically significant. The main analysis will follow the intention-to-treat 31 principle. Baseline data across control and intervention groups will first be assessed for 32 marked differences. The incidence of SSI (the primary dependent variable) in each of the 33 two groups of the trial will then be compared using Pearson’s chi-square test. Multivariable http://bmjopen.bmj.com/ 34 35 logistic regression analysis will be applied in case differences exist between intervention and 36 control groups at baseline and the analysis requires adjustment for confounders. We will 37 also carry out sensitivity testing for lost to follow-up patients and per protocol analysis for 38 non-compliers to assess for the possible effects of systematic biases on results. 39 40 Potential problems 41 42 Based on our previous studies, we feel that recruitment of adequate patient numbers is on October 3, 2021 by guest. Protected copyright. 43 feasible, however if we fail to recruit patients, we will invite additional General Practices to 44 participate. 45 In our previous studies we have found that assessing for infection at time of removal of 46 sutures facilitates a high rate of follow-up. Any patients not followed-up will be analysed on 47 an intention to treat basis. 48 We have not planned to perform an interim analysis as we feel that variation of antisepsis is 49 a minor intervention, and the outcome of SSI is usually a minor medical issue which is 50 treated with a course of antibiotics. An interim analysis would further increase the required 51 sample size reducing the feasibility of our trial. 52 53 54 55 DISSEMINATION 56 57 This project is due for completion one year after commencement of data collection. The 58 translation of important ﬁndings to clinical practice will be facilitated through dissemination in 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 25 BMJ Open

9 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 conference presentations and journals as well as electronic media. The researchers will also 4 aim to publish their findings on a range of Australian FOAM (Free Open Access Meducation) 5 websites to reach the next generation of technology-savvy health professionals. A written lay 6 summary of the results will also be displayed at the participating general practices for the 7 information of study participants. 8 9 ETHICAL CONSIDERATIONS 10 11 This project has been reviewed and approved by the James Cook University Human 12 Research Ethic Committee; HREC approval ID H6065. We do not expect the interventions of 13 this study to place participants at any significant risk of harm, as we hypothesise a lower 14 incidence of SSI in the intervention group due to the use of alcoholic CHG. In any case, SSI 15 is a minor conditionFor which peer can be easily review treated without significant only long term sequelae. To 16 assure privacy and confidentiality, all data spreadsheets and consent forms will be kept in a 17 locked cupboard throughout the trial, then transferred to a locked safe at the conclusion of 18 the trial, where they will be kept for 15 years. Patients will be de-identified in all data 19 collection. 20 21 DISCUSSION 22 23 24 Very few large RCTs are conducted in a primary care setting [23, 24]. Difficulties have been 25 reported in recruiting both patients and clinicians [25], and RCTs have been reported as 26 being methodologically and practically difficult to conduct in general practice. [11] However, 27 it is important that clinical practice be informed by adequate primary care evidence. 28 29 Otherwise GPs, the end-user of the research process, who attempt to practise evidence- 30 based medicine may have flawed tools and the guidelines they use may not be applicable to 31 the patients they see. [26, 27] 32

33 http://bmjopen.bmj.com/ 34 Funding for primary care research in Australia is very limited, particularly compared with UK 35 and Netherlands with only 2% of NHMRC grants awarded to primary care research between 36 2000 and 2008. [28] General practice based research differs in many ways from hospital 37 38 based research in issues of funding, feasibility and pragmatism, and we have used our 39 experience from conducting previous successful trials in general practice to inform the 40 design and methods of the present study[6, 8, 19, 20]. Our study will be conducted for a total 41 cost of $20,000 which is similar to our previous trials. 42 on October 3, 2021 by guest. Protected copyright. 43 44 Skin excisions form a large proportion of the workload of Australian GPs [17] and this is even 45 greater in Queensland, the state with the highest incidence of skin cancer in the world. [29] 46 This effect is magnified in regional towns such as Mackay, where there are no permanent 47 dermatologists or plastic surgeons. Using a research question which is relevant to our 48 clinical situation increases the feasibility of recruitment of patients because of our high case 49 load of patients presenting for skin excisions. 50 Our research question is practical and clinically relevant. Local clinicians do not use betadine 51 antisepsis because of perceptions of ‘messiness’ and skin staining. Therefore, in order to be 52 pragmatic our team has chosen to examine the difference between alcoholic and aqueous 53 CHG, rather than comparing the relative efficacy of CHG and PVI. We have also found that 54 using a clinically relevant research question engages general practitioners and practice 55 nurses and facilitates practitioner recruitment, as well as increasing the potential for 56 translation into clinical practice[30]. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 25

10 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 Our surgical and wound management protocol was developed in consultation with 4 participating doctors and practice nurses, which again increases the ownership and 5 practicality of the project. Occasionally, scientific rigor may be compromised at the expense 6 of pragmatism. For instance in contrast with hospital based research, it is simply not 7 practically or financially feasible to have an independent outcome assessor assessing each 8 wound at each of the three geographically dispersed practices. To compensate for this, 9 infections will be photographed and assessed for infection by a second independent and 10 blinded outcome assessor, who will re-assess every wound. 11 12 It is also not feasible to have an independent researcher to recruit patients. In our trial 13 practice nurses will be responsible for recruitment. A number of provisions have been 14 developed to assure informed consent. This is intended to minimise risk of perceived 15 coercion, asFor nurses are peer somewhat less review responsible for direct only decisions regarding patient 16 care than the patient’s GP. The practice nurses are also responsible for data collection and 17 are paid on a fee per service basis that compensates them for their time. The study involves 18 very little extra work for the participating GPs – they are not responsible for any data 19 collection, and were only required to have knowledge of the process involved in order to 20 answer any possible queries. 21 The trial will provide guidance to GPs regarding skin antisepsis, and will inform current 22 clinical guidelines and healthcare worker education. Although our study is conducted in a 23 tropical rural setting and we are aware that our baseline infection rate is comparably high 24 (6,7) we have no reason to believe that any relative risk reduction detected will not be 25 generalizable to other settings. If we detect a measurable decrease in incidence of SSI with 26 alcoholic CHG, this may result in a change in clinical practice, with an increase in use of 27 alcoholic CHG, which could reduce SSI rates following clean surgery. As this is a pragmatic 28 trial, the findings can potentially be immediately translated into clinical practice. 29 30 31 32 33

ACKNOWLEDGEMENTS http://bmjopen.bmj.com/ 34 The authors thank Ms Debbie Kimber, Ms Julie O’Sullivan, Dr Sheldon Browning, Dr Luke 35 Notely, Dr Andrew O’Neil, Dr Andrea Cosgrove. 36

37 Authors Contributions 38 CH conceived the study idea and oversaw the development of the study design and 39 40 protocol. DC led the development of the study protocol. AH, MD, and JB assisted with the 41 development of the study design and protocol. PB assisted with the sample size calculation 42 and statistics. All authors contributed to the drafting of the manuscript on October 3, 2021 by guest. Protected copyright. 43 44 COMPETING INTERESTS 45 None declared 46 47 FUNDING 48 James Cook University Honours program grant: $1000 49 Royal Australian College of General Practitioners PWH Grieve Award: $2 500 50 Royal Australian College of General Practitioners Family Medical Care Education and 51 Research Grant: $13 500 52 The authors’ work is independent of this funding. 53 54 55 REFERENCES 56 57 1. Dumville, J.C., et al., Preoperative skin antiseptics for preventing surgical wound infections 58 after clean surgery. Cochrane Database Syst Rev, 2015. 4: p. CD003949. 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 25 BMJ Open

11 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 2. Maiwald, M. and E.S. Chan, The forgotten role of alcohol: a systematic review and meta- 4 analysis of the clinical efficacy and perceived role of chlorhexidine in skin antisepsis. PLoS 5 One, 2012. 7(9): p. e44277. 6 3. Lee, I., et al., Systematic review and cost analysis comparing use of chlorhexidine with use of 7 iodine for preoperative skin antisepsis to prevent surgical site infection. Infect Control Hosp 8 Epidemiol, 2010. 31(12): p. 1219-29. 9 4. Health, Q.G.D.o., Surgical Skin Disinfection Guideline. 2015. 10 5. Heal, C., P. Buettner, and S. Browning, Risk factors for wound infection after minor surgery in 11 12 general practice. Med J Aust, 2006. 185(5): p. 255-8. 13 6. Heal, C., et al., Comparing non-sterile to sterile gloves for minor surgery: a prospective 14 randomised controlled non-inferiority trial. Med J Aust, 2015. 202(1): p. 27-31. 15 7. Heal, For C.F., P.G. Buettner,peer and H. review Drobetz, Risk factors only for surgical site infection after 16 dermatological surgery. Int J Dermatol, 2012. 51(7): p. 796-803. 17 8. Smith, S.C., C.F. Heal, and P.G. Buttner, Prevention of surgical site infection in lower limb skin 18 lesion excisions with single dose oral antibiotic prophylaxis: a prospective randomised 19 placebo-controlled double-blind trial. BMJ Open, 2014. 4(7): p. e005270. 20 9. Kamel, C., et al., Preoperative Skin Antiseptic Preparations and Application Techniques for 21 Preventing Surgical Site Infections: A Systematic Review of the Clinical Evidence and 22 Guidelines, in Rapid Response Reports. 2011: Ottawa (ON). 23 10. Mangram, A.J., et al., Guideline for prevention of surgical site infection, 1999. Hospital 24 25 Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol, 1999. 20(4): 26 p. 250-78; quiz 279-80. 27 11. Amici, J.M., et al., A prospective study of the incidence of complications associated with 28 dermatological surgery. Br J Dermatol, 2005. 153(5): p. 967-71. 29 12. Dixon, A.J., M.P. Dixon, and J.B. Dixon, Randomized clinical trial of the effect of applying 30 ointment to surgical wounds before occlusive dressing. Br J Surg, 2006. 93(8): p. 937-43. 31 13. Futoryan, T. and D. Grande, Postoperative wound infection rates in dermatologic surgery. 32 Dermatol Surg, 1995. 21(6): p. 509-14. 33

14. Lathlean, S., Skin cancer in general practice in South Australia. A five year study. Aust Fam http://bmjopen.bmj.com/ 34 Physician, 1999. 28 Suppl 1: p. S28-31. 35 15. Culver, D.H., et al., Surgical wound infection rates by wound class, operative procedure, and 36 patient risk index. National Nosocomial Infections Surveillance System. Am J Med, 1991. 37 38 91(3B): p. 152S-157S. 39 16. Platt, R., Antibiotic prophylaxis in clean surgery: does it work? Should it be used if it does? 40 New Horiz, 1998. 6(2 Suppl): p. S53-7. 41 17. Askew, D.A., et al., Skin cancer surgery in Australia 2001-2005: the changing role of the 42 general practitioner. Med J Aust, 2007. 187(4): p. 210-4. on October 3, 2021 by guest. Protected copyright. 43 18. Darouiche, R.O., et al., Chlorhexidine-Alcohol versus Povidone-Iodine for Surgical-Site 44 Antisepsis. N Engl J Med, 2010. 362(1): p. 18-26. 45 19. Heal, C., et al., Does single application of topical chloramphenicol to high risk sutured 46 wounds reduce incidence of wound infection after minor surgery? Prospective randomised 47 placebo controlled double blind trial. BMJ, 2009. 338(338): p. a2812. 48 20. Heal, C., et al., Can sutures get wet? Prospective randomised controlled trial of wound 49 management in general practice. BMJ, 2006. 332(7549): p. 1053-6. 50 51 21. Australian Bureau of Statistics, 2011 Census QuickStats: Mackay, Commonwealth Bureau of 52 Census and Statistics, Editor. 2013: Canberra, . 53 22. Queensland Centre for Healthcare Related Infection Surveillance and Prevention and 54 Tuberculosis Prevention, Guideline for Surgical Skin Disinfection. 2012. 55 23. Mendis, K., et al., A bibliometric analysis of Australian general practice publications from 56 1980 to 2007 using PubMed. Inform Prim Care, 2010. 18(4): p. 223-33. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 25

12 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 24. Mendis, K. and I. Solangaarachchi, PubMed perspective of family medicine research: where 4 does it stand? Fam Pract, 2005. 22(5): p. 570-5. 5 25. Rendell, J.M., R.D. Merritt, and J.R. Geddes, Incentives and disincentives to participation by 6 clinicians in randomised controlled trials. Cochrane Database Syst Rev, 2007(2): p. 7 MR000021. 8 26. Steel, N., et al., A review of clinical practice guidelines found that they were often based on 9 evidence of uncertain relevance to primary care patients. J Clin Epidemiol, 2014. 67(11): p. 10 1251-7. 11 12 27. Scullard, P., et al., Does the evidence referenced in NICE guidelines reflect a primary care 13 population? Br J Gen Pract, 2011. 61(584): p. e112-7. 14 28. McIntyre, E.L., D. Mazza, and N.P. Harris, NHMRC funding for primary health care research, 15 2000-2008.For Med J Aust,peer 2011. 195 (4):review p. 230. only 16 29. Buettner, P.G. and B.A. Raasch, Incidence rates of skin cancer in Townsville, Australia. Int J 17 Cancer, 1998. 78(5): p. 587-93. 18 30. Heal, C.F., C. Veitch, and R. Preston, Practice based research - lessons from the field. Aust 19 Fam Physician, 2008. 37(5): p. 381-4. 20 21 22 23 24 25 26 27 28 29

30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on October 3, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from Docs Docs Addressed on on Addressed number page

_____2______2______1______N/A______10______10______1,10______10______10______Supplementary ___Supplementary BMJ Open http://bmjopen.bmj.com/ on October 3, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Description Description interpretation of data; writing of the report; and the decision to submit the report for publication, including including publication, for report the submit to decision the and report; of the writing ofdata; interpretation activities any ofthese over authority ultimate have will they whether

5c 5c and analysis, management, collection, design; study in ifany, funders, sponsorand study of Role 1 1 acronym trial if applicable, and, interventions, population, design, study the identifying title Descriptive _____1______5b 5b sponsor trial forthe information contact and Name 5a 5a contributors protocol of roles and affiliations, Names, 2b 2b Set Data Registration Trial Organization WorldHealth fromthe items All No No 4 4 support other and material, financial, of types and Sources Protocol version version Protocol 3 identifier version and Date Roles and and Roles responsibilities Title Title Funding Trial registration registration Trial 2a registry intended of name registered, yet Ifname. not registry and identifier Trial Administrative information Administrative documents* related and protocol trial clinical a in address to items Recommended Checklist: 2013 SPIRIT Section/item Item Page 13 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from Page 14 of 25 ______6______6______

______6______6______6,7,10______N/A______5,6______5,6______5______5______5______5______N/A______N/A______4,5,9______5______4,5,9,10______4,5,9,10__ BMJ Open http://bmjopen.bmj.com/ on October 3, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only be collected. Reference to where list of study sites can be obtained. obtained. be can sites ofstudy list where to Reference collected. be allocation ratio, and framework (eg, superiority, equivalence, noninferiority, exploratory) exploratory) noninferiority, equivalence, superiority, (eg, framework and ratio, allocation Relevant concomitant care and interventions that are permitted or prohibited during the trial trial the during prohibited or permitted are that interventions and care concomitant Relevant (eg, drug tablet return, laboratory tests) tests) laboratory return, tablet drug (eg, Strategies to improve adherence to intervention protocols, and any procedures for monitoring adherence adherence monitoring for procedures any and protocols, intervention to adherence improve to Strategies change in response to harms, participant request, or improving/worsening disease) disease) improving/worsening or request, participant harms, to response in change Criteria for discontinuing or modifying allocated interventions for a given trial participant (eg, drug dose drug (eg, participant trial given for a interventions allocated modifying or fordiscontinuing Criteria administered administered Interventions for each group with sufficient detail to allow replication, including how and when they will be be they will and when how including replication, allow to detail sufficient with group each for Interventions individuals who will perform the interventions (eg, surgeons, psychotherapists) psychotherapists) surgeons, (eg, interventions the perform will who individuals Inclusion and exclusion criteria for participants. If applicable, eligibility criteria for study centres and centres and criteriastudy for eligibility applicable, If for participants. criteria exclusion and Inclusion adjudication committee, data management team, and other individuals or groups overseeing the trial, if trial, the overseeing groups or individuals other and managementteam, data committee, adjudication committee) monitoring data for 21a (seeItem applicable studies (published and unpublished) examining benefits and harms for each intervention intervention each for harms and benefits examining unpublished) and (published studies 5d 5d endpoint committee, steering centre, coordinating of the responsibilities and roles, Composition, 6b 6b comparators of forchoice Explanation 11d 11d 11c 11c 11b 11b 11a 11a 10 10

6a 6a of relevant summary including trial, the forundertaking justification and question of research Description Interventions Interventions Eligibility criteria criteria Eligibility Study setting Study 9 will data where ofcountries list and hospital) academic clinic, community (eg, settings of study Description Methods: Participants, interventions, and outcomes outcomes and interventions, Participants, Methods: Objectives design Trial 7 8 hypotheses or objectives Specific group), single factorial, crossover, group, parallel (eg, oftrial type including design of trial Description Introduction Introduction Background and Background rationale 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from

___N/A______6______6______5,6,7______6______6______6______6______8______8______8______8______7______7______7_____ ins and washouts), assessments, and visits for visits and assessments, washouts), and ins BMJ Open http://bmjopen.bmj.com/

on October 3, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only allocated intervention during the the trial during intervention allocated If blinded, circumstances under which unblinding is permissible, and procedure for revealing a participant’s participant’s a forrevealing procedure and is permissible, unblinding which under circumstances blinded, If assessors, data analysts), and how how and analysts), data assessors, Who will be blinded after assignment to interventions (eg, trial participants, care providers, outcome outcome careproviders, participants, trial (eg, interventions to afterassignment blinded be will Who interventions interventions Who will generate the allocation sequence, who will enrol participants, and who will assign participants to to participants assign will andwho participants, enrol will who sequence, allocation the generate will Who opaque, sealed envelopes), describing any steps to conceal the sequence until interventions are assigned assigned are interventions until sequence the conceal to steps any describing envelopes), sealed opaque, Mechanism of implementing the allocation sequence (eg, central telephone; sequentially numbered, numbered, sequentially telephone; central (eg, sequence allocation the implementing of Mechanism or assign interventions interventions assign or (eg, blocking) should be provided in a separate document that is unavailable to those who enrol participants participants enrol who those to unavailable is that document separate a in provided be should blocking) (eg, factors for stratification. To reduce predictability of a random sequence, details of any planned restriction restriction planned ofany details sequence, random a of predictability reduce To forstratification. factors Method of generating the allocation sequence (eg, computer-generated random numbers), and list of any of list any and numbers), random computer-generated (eg, sequence allocation the of generating Method

Strategies for achieving adequate participant enrolment to reach target sample size size sample target reach to enrolment participant adequate forachieving Strategies clinical and statistical assumptions supporting any sample size calculations calculations samplesize any supporting assumptions statistical and clinical Estimated number of participants needed to achieve study objectives and how it was determined, including including determined, it was how and objectives study achieve to needed participants of number Estimated participants. A schematic diagram is highly recommended (see Figure) Figure) (see recommended highly is diagram schematic A participants. Time schedule of enrolment, interventions (including any run- any (including interventions enrolment, of schedule Time efficacy and harm outcomes is strongly recommended recommended strongly is andharm outcomes efficacy median, proportion), and time point for each outcome. Explanation of the clinical relevance of chosen chosen of relevance clinical of the Explanation foroutcome. each point timeand proportion), median, pressure), analysis metric (eg, change from baseline, final value, time to event), method of aggregation (eg, aggregation methodof event), to time value, final baseline, from change (eg, metric analysis pressure), Primary, secondary, and other outcomes, including the specific measurement variable (eg, systolic blood blood (eg,systolic variable measurement specific the including outcomes, other and secondary, Primary, 17b 17b 17a 17a 16c 16c 16b 16b 16a 16a

15 15 14 14 13 13 12 12 Implementation Implementation mechanism mechanism concealment concealment generation generation Sequence Sequence Allocation Allocation Methods: Data collection, management, and analysis and management, collection, Data Methods:

Blinding (masking) (masking) Blinding Methods: Assignment of interventions (for controlled trials) controlled (for interventions of Assignment Methods: Recruitment Recruitment Sample size size Sample Participant timeline timeline Participant Outcomes Outcomes

Allocation: Allocation: Page 15 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from Page 16 of 25

______10______10______7,8______7,8______8______8______8______8______8______8______8______8______8______8______7,8,9______8______8______7,8______7,8_____ BMJ Open http://bmjopen.bmj.com/ on October 3, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only from investigators and the sponsor sponsor the and frominvestigators Frequency and procedures for auditing trial conduct, if any, and whether the process will be independent independent be will process the and whether if any, conduct, trial forauditing procedures and Frequency events and other unintended effects of trial interventions or trial conduct conduct trial or interventions trial of effects unintended other and events Plans for collecting, assessing, reporting, and managing solicited and spontaneously reported adverse reported spontaneously and solicited managing and reporting, assessing, collecting, for Plans results and make the final decision to terminate the trial trial the terminate to decision final makethe and results Description of any interim analyses and stopping guidelines, including who will have access to these interim access to have will who including guidelines, stopping and analyses interim of any Description needed needed about its charter can be found, if not in the protocol. Alternatively, an explanation of why a DMC is not is not DMC a ofwhy explanation an Alternatively, theprotocol. in ifnot found, be can itscharter about whether it is independent from the sponsor and competing interests; and reference to where further details furtherdetails where to reference and interests; competing and sponsor from the isindependent it whether Composition of data monitoring committee (DMC); summary of its role and reporting structure; statement of statement structure; reporting and role ofits summary (DMC); committee monitoring data of Composition statistical methods to handle missing data (eg, multiple imputation) imputation) multiple (eg, data missing handle to methods statistical Definition of analysis population relating to protocol non-adherence (eg, as randomised analysis), and any any and analysis), randomised as (eg, non-adherence protocol to relating population ofanalysis Definition Methods for any additional analyses (eg, subgroup and adjusted analyses) analyses) adjusted and subgroup (eg, analyses additional any for Methods statistical analysis plan can be found, if not in the protocol protocol the in iffound,not be can plan analysis statistical Statistical methods for analysing primary and secondary outcomes. Reference to where other details of the the of details other where to Reference outcomes. secondary and primary foranalysing methods Statistical procedures can be found, if not in the protocol protocol the in if not found, be can procedures (eg, double data entry; range checks for data values). Reference to where details of data management management data of details where to Reference values). data for checks range entry; data double (eg, Plans for data entry, coding, security, and storage, including any related processes to promote data quality quality data promote to processes related any including storage, and security, coding, entry, fordata Plans collected for participants who discontinue or deviate from intervention protocols protocols from intervention deviate or discontinue who forparticipants collected Plans to promote participant retention and complete follow-up, including list of any outcome data to to be data outcome of list any including follow-up, complete and retention participant promote to Plans Reference to where data collection forms can be found, if not in the protocol protocol inthe ifnot found, be can forms collection data where to Reference study instruments (eg, questionnaires, laboratory tests) along with their reliability and validity, if known. if known. validity, and reliability their with along tests) laboratory questionnaires, (eg, instruments study processes to promote data quality (eg, duplicate measurements, training of assessors) and a description of description a and assessors) of training measurements, duplicate (eg, quality data promote to processes Plans for assessment and collection of outcome, baseline, and other trial data, including any related related any including data, other trial and baseline, ofoutcome, collection and forassessment Plans

23 23 22 22 21b 21b 21a 21a 20c 20c 20b 20b 20a 20a 19 19 18b 18b 18a 18a Ethics and dissemination dissemination and Ethics Harms Harms

Data monitoring monitoring Data Methods: Monitoring Methods:

Statistical methods methods Statistical Data management management Data

methods methods Data collection collection Data

Auditing Auditing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

5 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from

______7______7______N/A______5,10______5,10______N/A______7, 8______8______7, ______8,9______8,9______7,9______7,9______N/A______9______9______BMJ Open http://bmjopen.bmj.com/ on October 3, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only how (see Item 32) 32) (seeItem how if applicable studies, analyses) to relevant parties (eg, investigators, REC/IRBs, trial participants, trial registries, journals, journals, registries, trial participants, trial REC/IRBs, investigators, (eg, parties relevant to analyses) regulators) participation participation Provisions, if any, for ancillary and post-trial care, and for compensation to those who suffer harm from trial trial from harmsuffer who those to compensation for care,and post-trial and for ancillary ifany, Provisions, limit such access for investigators investigators for access such limit Statement of who will have access to the final trial dataset, and disclosure of contractual agreements that that agreements contractual of disclosure and dataset, trial final the to access have will of who Statement the public, and other relevant groups (eg, via publication, reporting in results databases, or other data data other or results databases, in reporting publication, via groups(eg, relevant other and public, the restrictions publication any including arrangements), sharing in order to protect confidentiality before, during, and after the trial trial afterthe and during, before, confidentiality protect to order in 26b 26b ancillary in specimens and biological data ofuseparticipant and collection for provisions consent Additional 31b 31b writers professional of use intended any and guidelines eligibility Authorship 31c 31c code andstatistical dataset, participant-level protocol, full the to access public for granting any, if Plans, ______N/A_____ 31a 31a professionals, healthcare participants, to results trial communicate sponsorto and forinvestigators Plans 30 30 29 29

32 32 surrogates authorised and participants to given documentation related other formand consent Model ______appendix__ 25 25 outcomes, criteria, eligibility to changes (eg, modifications protocol important communicating for Plans 24 24 approval (REC/IRB) board review committee/institutional ethics research seeking for Plans 28 28 site study each and trial overall forthe investigators principal for interests competing other and Financial ______10______

Consent or assent assent or Consent 26a and surrogates, authorised or participants trial potential from assent or consent informed obtain will Who

Dissemination policy Dissemination Appendices Informed consent consent Informed materials trial care trial Protocol Protocol amendments Research ethics ethics Research approval Confidentiality 27 maintained and shared, be collected, will participants andenrolled potential about information personal How Declaration of Declaration interests Ancillary and post- and Ancillary Access to data data to Access Page 17 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

6 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from Page 18 of 25 ______N/A_____ BMJ Open http://bmjopen.bmj.com/ ” license. license. ” on October 3, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only analysis in the current trial and for future use in ancillary studies, if applicable ifapplicable studies, ancillary in use future for and trial current the in analysis 33 33 molecular or forgenetic specimens biological of storage and evaluation, laboratory collection, for Plans Attribution-NonCommercial-NoDerivs 3.0 Unported 3.0 Attribution-NonCommercial-NoDerivs

*It is strongly recommended that this checklist be read in conjunction with the SPIRIT 2013 Explanation & Elaboration for important clarification on the items. items. on the clarification for important Elaboration & Explanation 2013 SPIRIT the with conjunction in read be thischecklist that recommended is strongly *It “ Biological Biological specimens Amendments to the protocol should be tracked and dated. The SPIRIT checklist is copyrighted by the SPIRIT Group under the Creative Commons Commons Creative the under Group SPIRIT by the iscopyrighted checklist SPIRIT The dated. and tracked be should protocol the to Amendments 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 7 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from BMJ Open http://bmjopen.bmj.com/ on October 3, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only

Page 19 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 20 of 25

Appendix 1: WHO trial registration dataset 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 Note: All of this information is included in the ANZCTR registry 3 4 5 6 Primary registry and trial identifying number Australian and New Zealand Clinical Trials 7 Registry, registration number: 8 ACTRN12615001045505. 9 Date of registration in primary registry 07/10/2015 10 11 Secondary identifying numbers N/A 12 13 Sources of monetary or material support Royal Australian College of General Practitioners 14 James Cook University 15 For peer review only 16 Primary Sponsor James Cook University 17 18 Secondary Sponsor Dr Daniel Charles 19 Contact for public queries Dr Daniel Charles 20 21 C/- Mackay Clinical School 22 College of Medicine and Dentistry 23 James Cook University 24 Level 1, Building K 25 Mackay Base Hospital, 475 Bridge Rd 26 Mackay QLD 4740 Australia 27 [email protected]

28 Contact for scientific queries Professor Clare Heal 29 30 Contact details as per ‘corresponding author’ 31 section 32 33 Public title Aqueous vs Alcoholic Antisepsis with 34 Chlorhexidine for skin Excisions – the http://bmjopen.bmj.com/ 35 AVALACHE trial. 36 Scientific title Preoperative skin antisepsis with aqueous vs 37 alcoholic chlorhexidine for the prevention of 38 surgical site infection after minor skin excisions in 39 general practice – a randomised controlled trial. 40 41 Countries of recruitment Australia 42 on October 3, 2021 by guest. Protected copyright. 43 Health conditions or problems studied Surgical Site Infection 44 45 Interventions Intervention: 46 Preoperative skin antisepsis with 0.5% 47 chlorhexidine dissolved in 70% alcohol applied to 48 the surgical site immediately prior to minor skin 49 surgery in general practice. 50 51 Control: 52 53 Preoperative skin antisepsis with 0.5% chlorhexidine dissolved in water applied to the 54 surgical site immediately prior to minor skin 55 surgery in general practice. 56 57 Application protocol for intervention and control: 58 59 Approximately 30mL of antiseptic solution will be 60 applied over skin surface at the site of the For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 8 Page 21 of 25 BMJ Open

planned excision using a soaked gauze so as to 1 cover an area 1cm greater at all borders than the BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 area of skin exposed by the sterile drape which 3 will be placed over the site of the excision. 4 5 Key inclusion and exclusion criteria Inclusion criteria 6 7 • All patients over the age of 18 8 undergoing minor skin procedures at the participating practices during the study 9 period who: 10 o have capacity to give informed 11 consent and; 12 o are able to return for removal of 13 sutures 14 • Patients who are not presenting for: 15 For peer reviewo excision only of sebaceous cyst 16 17 o suturing of lacerations 18 19 o excisions not requiring sutures, 20 such as shave biopsies 21 22 o punch biopsies 23 o 24 excisions on body sites where 25 adrenaline is contraindicated 26 27 Exclusion Criteria 28 • Allergy to alcohol or chlorhexidine 29 30 • Evidence of infection at or adjacent to the 31 operative site 32 33 • Current use of antibiotics 34 http://bmjopen.bmj.com/ 35 • Clinical indication for antibiotic treatment 36 following excision (besides SSI) 37 38 • Peri-ocular excisions 39 40 • Patients with a primary language other 41 than English for which certiﬁed 42 translation services for that language are on October 3, 2021 by guest. Protected copyright. 43 not available 44 45 Study type Randomised Controlled Trial 46 47 Date of first enrolment 16/10/2015 48 Target sample size 920 49 50 Recruitment status Recruiting 51 52 Primary Outcome(s) Surgical Site Infection (SSI) diagnosed in 53 accordance with a modified version of the CDC 54 definition for superficial surgical site infection: 55 56 • Infection occurs within 30 days after the excision, AND 57 • Infection involves ONLY skin or 58 subcutaneous tissue of the incision, AND 59 • At least ONE of the following: 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 9 BMJ Open Page 22 of 25

o Purulent drainage with or without 1 laboratory confirmation from the BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 superficial incision 3 o At least one of the following signs or 4 symptoms: pain or tenderness, 5 localised swelling, redness or heat o 6 Diagnosis of superficial SSI by the GP 7 Stitch abscesses, characterised by minimal 8 inflammation and discharge confined to the points 9 of suture penetration, will not be included as 10 SSIs. 11 12 Key secondary outcomes 1. Adverse reactions to the preoperative 13 skin antiseptic agent, manifesting as any 14 one of 15 For peer reviewa. anaphylaxis only 16 b. skin irritation or contact 17 dermatitis 18 c. rash 19 2. Microbiology of infected wounds with a 20 purulent discharge, and any patterns of antibiotic resistance 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on October 3, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 10 Page 23 of 25 BMJ Open

Appendix 2: Consent Form and Patient Information Sheet 1 BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from

2 3 Participant Informed Consent Form 4 5 Project: Aqueous Vs Alcoholic Antisepsis with Chlorhexidine for Skin Excisions – The AVALANCHE Trial 6 7 Principal Investigator: Dr Daniel Charles 8 Study supervisor: Professor Clare Heal 9 10 School: James Cook University School of Medicine and Dentistry 11 12 13 I understand that the aim of this research study is to compare alcoholic chlorhexidine and aqueous chlorhexidine to see if 14 there are any differences between them in terms of how well they prevent wound infections when they are applied to 15 For peer review only 16 surgical sites before skin excisions. 17 18 I consent to participate in this project; the details of which have been explained to me by the practice nurse, and I have 19 been provided with a written information sheet to keep. 20 21 22 I understand that my participation in this study involves being randomly allocated to one of two groups. Depending upon 23 the group I am allocated to, the site where my skin excision will take place will be cleansed using either alcoholic 24 chlorhexidine or aqueous chlorhexidine. I also acknowledge that prior to giving consent to participate in this study, I will 25 have no knowledge of which of these preparations will be used for my surgery. 26 27 28 I understand that aside from the surgical site cleansing, the doctors and nurses involved in my care will follow standard 29 surgical and wound care protocols. I understand that, as per standard practice, my wound will be assessed for evidence of 30 wound infection when I come back to my GP for removal of sutures, and if a wound infection is noted, I will be treated as 31 per standard practice. 32 33 34 I also agree that the researcher may use the results from my participation as described in the information sheet. http://bmjopen.bmj.com/ 35 36 I acknowledge that: 37

38 39 - any risks and possible effects of participating in this study have been explained to my satisfaction; 40 - taking part in this study is voluntary and I am aware that I can stop taking part in it at any time without explanation or 41 prejudice and withdraw any unprocessed data I have provided; 42 - any information I give will be kept strictly confidential and that no names will be used to identify me with this study on October 3, 2021 by guest. Protected copyright. 43 without my approval; 44 - my details will be kept in storage and will be accessible to the researchers only 45 - after I sign this consent form it will be retained by the researcher 46 47 48 Name: (printed) 49

50 Signature: Date: 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 11 BMJ Open Page 24 of 25

1 Participant Information Sheet BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 2 3 4 5 Project: Aqueous Vs Alcoholic Antisepsis with Chlorhexidine for Skin Excisions – The AVALANCHE Trial 6 7 Principal Investigator: Dr Daniel Charles 8 Study supervisor: Professor Clare Heal 9 School: James Cook University School of Medicine and Dentistry 10

11 12 We would like to invite you to take part in a research study that your doctor is participating in. The study is being carried 13 out by an honours student studying the Bachelor of Medicine and Surgery course at James Cook University. 14 15 For peerDO YOU review NEED TO TAKE PART? only 16 We would like to stress that you do not need to take part in this study unless you wish to. Your decision to participate or 17 18 not will not impact upon the care you receive at this practice, nor will any additional cost be incurred regardless of your 19 decision. If you choose to participate, you can stop taking part at any time without explanation. 20 21 WHAT ARE THE AIMS OF THE STUDY? 22 23 Before a surgical procedure is carried out, the area on the patient’s body where the procedure takes place (the surgical 24 site) is routinely cleansed with a solution which kills bacteria on the skin (an antiseptic preparation). The aim of this 25 process is to reduce the likelihood of infection at the surgical site (wound infection). As it stands, however, not enough 26 research has been done to determine which antiseptic preparations are best at preventing wound infections. As such, this 27 study aims to compare two different antiseptic preparations - alcoholic chlorhexidine, and aqueous chlorhexidine - to see 28 29 which is most effective. 30 31 WHAT WILL HAPPEN IF YOU CHOOSE TO TAKE PART? 32 If you choose to take part in this study, the practice nurse will first take down some basic information from you, such as 33

your height and weight, your basic medical history and the names of any medications you are taking. You will then be http://bmjopen.bmj.com/ 34 35 randomly allocated to either the alcoholic chlorhexidine group, or the aqueous chlorhexidine group. 36 37 Before your skin excision, the surgical site will be cleansed with whichever of these two solutions you have been randomly 38 assigned to. The doctors and nurses involved in your procedure will then be using standard surgical techniques and 39 following routine surgical protocols. They will thoroughly wash their hands, wear gloves, and use routine instruments and 40 41 stitch materials to complete the surgery. After the skin excision, a dressing will be put over the wound and you will be 42 given instructions obout how to care for your wound. on October 3, 2021 by guest. Protected copyright. 43 44 WHAT WILL HAPPEN IF YOU GET A WOUND INFECTION? 45 46 As is routine, when you come back for removal of stitches, your doctor or nurse will look carefully for signs of wound 47 infection, however if you believe that you have developed a wound infection before this, please feel free to come back 48 earlier. There is always a small chance of developing a wound infection; they happen 9% of the time after skin excisions, 49 and many factors effect whether or not patients develop wound infections besides the type of antiseptic preparation 50 used. However, if you do develop a wound infection, it is unlikely to be a serious health issue; you may experience some 51 52 mild pain and discomfort, but the infection is likely to be able to be simply and effectively treated. 53 54 WHAT ARE THE RISKS AND BENEFITS OF THIS STUDY? 55 The risks of this study are minimal, as both of the solutions being used in this study are already routinely used in medical 56 practices in the area. There is a risk of developing a wound infection, though, as mentioned above, this risk is always 57 58 present after a skin excision, and your doctor will look carefully for wound infection and treat it if one is found. The only 59 other major risk of the study is the risk of a skin reaction to one of the antiseptic preparations. These reactions are 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 12 Page 25 of 25 BMJ Open

uncommon, but if you have a known allergy to either chlorhexidine or to alcohol, we would ask that you inform the BMJ Open: first published as 10.1136/bmjopen-2016-011604 on 7 July 2016. Downloaded from 1 practice nurse, and we would request that you DO NOT take part in this study. 2 3 4 The major benefit of this study will be to hopefully clarify which antiseptic preparations are best at preventing wound 5 infections and thereby help health workers to decide upon the best way to manage skin excisions in future. 6 7

8 9 WILL YOUR INFORMATION BE CONFIDENTIAL? 10 Yes, all information collected for the purposes of the study will be completely confidential. Only your GP, the practice 11 nurse, and the main study investigators will have access to your information. The data from this study will be used in 12 research publications and reports, but you will not be identified in any way in these publications. 13

14 15 ForWHAT peer SHOULD YOU DO review IF YOU DON’T WANT TO only TAKE PART? 16 If you do not wish to take part in this study, simply inform the practice nurse looking after you. If this is the case, your 17 excision will take place as normal. 18 19 20 If you have any questions about the study, please contact the principal investigator, Dr Daniel Charles, or the supervisor 21 for this project, Prof Clare Heal. Contact details are listed below. 22 23 24 Principal Investigator: Study Supervisor: 25 26 Dr Daniel Charles Professor Clare Heal 27 School of Medicine and Dentistry School of Medicine and Dentistry 28 James Cook University James Cook University 29 Phone: C/- JCU Mackay Clinical School: 07 4885 7100 Phone: 07 4885 7142 30 Email: [email protected] Email: [email protected] 31 32 33 34 If you have any concerns regarding the ethical conduct of the study, please contact: http://bmjopen.bmj.com/ 35 Human Ethics, Research Office 36 James Cook University, Townsville, Qld, 4811 37 Phone: (07) 4781 5011 ([email protected]) 38

39 40 41 42 on October 3, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 13