A peer-reviewed version of this preprint was published in PeerJ on 26 April 2017.

View the peer-reviewed version (peerj.com/articles/3218), which is the preferred citable publication unless you specifically need to cite this preprint.

Ifediora CO. 2017. The determinants and engagement patterns of chaperones and chauffeurs by Australian doctors in after-hours house-call services. PeerJ 5:e3218 https://doi.org/10.7717/peerj.3218 Doctor safety in Australian after-hours house-call medical services: the use of chaperones and chauffeurs

Objectives: The use of escorts (chauffeurs and chaperones) while on duty in after-hours- house-call (AHHC) is one key protective option available to doctors in the service and has been linked to low burnout and increased satisfaction in AHHC. This study aims to explore the patterns of engagement of escorts in Australian AHHC. Method: A questionnaire-based, electronic survey of all 300 doctors involved in AHHC through the National Home Doctor Service (NHDS), Australia’s largest providers of the service. Results: A total of 168 valid responses (56.0%) were received. 60.8% of the doctors engaged escorts. Of the doctors that engage chauffeurs, three-quarters do so “all or most times”, while only one-quarter engage chaperones to the same degree of frequency. Hiring escorts is very popular among Brisbane (91.7%) and Sydney-based (88.2%) practitioners, but is unpopular in the City of Gold Coast (26.1%). There were moderate patronages in Adelaide (52.9%) and Melbourne (46.4%). Compared to females, males were more likely to drive themselves (OR 5.34; P=0.001; CI 2.08 to 13.74) and less likely to use chauffeurs (OR 0.19; p<0.001; CI 0.07 to 0.51). Doctors in legally recognized social unions (OR 0.24; p=0.03; CI 0.07 to 0.83) and those that have attained the postgraduate fellowships (OR 0.43; p=0.02; CI 0.21 to 0.87) were both less likely to work without escorts. Conclusion: More needs to be done to increase the engagement of escorts by doctors involved in the Australian AHHC, particularly given their proven benefits in the service. Future studies may be needed to fully explore the real reasons behind the significant associations identified in this study.

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2581v1 | CC BY 4.0 Open Access | rec: 4 Nov 2016, publ: TITLE PAGE AND ACKNOWLEDGEMENT

TITLE: Doctor safety in Australian after-hours house-call medical services: the use of chaperones and chauffeurs.

ARTICLE CATEGORY: Quantitative Research/Survey.

AUTHOR: Chris O. Ifediora

AFFILIATIONS: School of Medicine, Griffith University, Gold Coast Campus, Australia

CORRESPONDENCE: Dr. C. O. Ifediora, Senior Lecturer, School of Medicine, Griffith University,

Gold Coast Campus, Parklands Drive, Southport, QLD 4215, Australia.

Email: [email protected].

ACKNOWLEDGMENT: Special thanks to Ben Keneally, The CEO, National Home Doctor service, for his co-operation during this study.

Page 1 of 18

ABSTRACT

Objectives: The use of escorts (chauffeurs and chaperones) while on duty in after-hours-house-call

(AHHC) is one key protective option available to doctors in the service and has been linked to low burnout and increased satisfaction in AHHC. This study aims to explore the patterns of engagement of escorts in Australian AHHC.

Method: A questionnaire-based, electronic survey of all 300 doctors involved in AHHC through the

National Home Doctor Service (NHDS), Australia’s largest providers of the service.

Results: A total of 168 valid responses (56.0%) were received. 60.8% of the doctors engaged escorts.

Of the doctors that engage chauffeurs, three-quarters do so “all or most times”, while only one- quarter engage chaperones to the same degree of frequency. Hiring escorts is very popular among

Brisbane (91.7%) and Sydney-based (88.2%) practitioners, but is unpopular in the City of Gold Coast

(26.1%). There were moderate patronages in Adelaide (52.9%) and Melbourne (46.4%).

Compared to females, males were more likely to drive themselves (OR 5.34; P=0.001; CI 2.08 to

13.74) and less likely to use chauffeurs (OR 0.19; p<0.001; CI 0.07 to 0.51). Doctors in legally recognized social unions (OR 0.24; p=0.03; CI 0.07 to 0.83) and those that have attained the postgraduate fellowships (OR 0.43; p=0.02; CI 0.21 to 0.87) were both less likely to work without escorts.

Conclusion: More needs to be done to increase the engagement of escorts by doctors involved in the

Australian AHHC, particularly given their proven benefits in the service. Future studies may be needed to fully explore the real reasons behind the significant associations identified in this study.

Key Words: General practice; after-hours, safety, protection, home visits, doctors, family physicians.

Page 2 of 18

INTRODUCTION AND BACKGROUND

Also known as Medical Deputizing Service (MDS), the after-hours, house call (AHHC) services is becoming increasingly popular in Australia, with about 1.5 million patients benefitting from the service in 2013 (National Association for Medical Deputising Service 2014). This number represents nearly 38% of all “urgent” after-hours presentations seen at the time, with the Emergency

Departments (EDs) accounting for another 62% (National Association for Medical Deputising Service

2014). With this increase in popularity comes the need to ensure the welfare of the doctors are not ignored in the quest for quality service-delivery. As a matter of fact, this welfare has come into focus lately following a recent finding that over half of the doctors involved in AHHC have no protective measures in place while on duty (Ifediora 2015). The same study also reported that among those that have protective measures, only one-third employ chaperones while at work. The debate on this subject was further fueled by other online articles regarding use of chaperones and security measures in AHHC (Medical Observer 2015, Ozturk 2015).

These aforementioned discussions underline the need to evaluate the use of “escorts”, a key aspect of security measures available to doctors involved in AHHC. We define an escort as “an attendant employed to accompany someone” (Wordweb [Internet] 2015), in this case, a doctor. In AHHC, they are hired either as Chaperones or Chauffeurs. Basically, a chauffeur is “a person whose job is to drive people around in a car” (Merriam-Webster[Internet] 2015), while medical chaperones are “employed to accompany physicians during physical examinations, especially when the opposite gender is involved” (Farlex 2012). The roles of escorts in AHHC go beyond these primary roles, and may include working as security details, providing company for the doctor during drives between patients, carrying medical equipments, communications, locating patients’ addresses, and so on. Given that the use of these escorts is a key a protective measure for doctors in AHHC (Ifediora 2015), a knowledge of the doctors’ habits regarding their engagement is very important.

Page 3 of 18

This study, therefore, aims to identify the frequency and patterns of use of escorts among Australian doctors involved in AHHC. The independent doctor-variables associated with the use or otherwise of these escorts will also be explored. Not much exists in the literature regarding escorts in AHHC, and answering these research questions will help reduce this existing knowledge-gap, and help policy makers and the concerned doctors to consolidate, plan, improve and adjust as necessary regarding this important aspect of AHHC service.

Our findings will also have international relevance. For instance, as at 2011, 56% of the doctors in

Australia were either born overseas, or obtained their primary medical degree therefrom (Australian

Bureau of Statistics 2013). Given that this cosmopolitan trend is expected to continue, results from this survey will help both the overseas-trained doctors (OTDs) and the international recruitment agencies involved in their movements make better-informed decisions. In addition, various countries around the world (United Kingdom, France, Canada, the Netherlands, and so on) are at different levels of involvement and development of their AHHC industry (National Association for Medical

Deputising Service 2014), and healthcare managers in these countries may find the results from this study helpful as they design their own systems.

It should be noted that in Australian AHHC, the choice of whether to employ an escort or not rests with the doctor, but some service providers do facilitate these engagements. Some doctors engage the escorts as chauffeurs, others drive themselves but still have them as chaperones, while some do not engage them at all, and prefer to work alone. The escorts generally negotiate fees with the doctors that employ them, but our private enquiries reveal rates of between AU$20 to AU$40 per hour.

Page 4 of 18

METHODS

Setting and Participants

The Participants included all Australian-based medical practitioners (GPs and others) who undertake

AHHC through the National Home Doctor Service (NHDS), Australia’s largest AHHC service provider

(National Home Doctor Service 2014). As at the time of this survey, the company rendered services in a number of locations, including Sydney, the Gold Coast, Adelaide, Brisbane Area (including Sunshine

Coast), and Melbourne Area (including Geelong and Canberra) (National Home Doctor Service 2014).

The terms “Melbourne and Brisbane Areas” reflect the NHDS administrative groupings, and were not based on geographical or political classifications. Official NHDS sources indicated that there were 300 doctors working for them at the time of this survey, and this represents the study population as the study reached out to all of them. Each NHDS-location is overseen by a Clinical or General Manager, and, given that the company had successfully annexed the largest after-hours general practice clinics in most Australian major towns and cities over the past few years prior to this study, it can be safely assumed that a study of NHDS-doctors reasonably represents the Australian after-hours doctor- population.

This study surveyed the experiences of the participants over a twelve-month period spanning from

October 2013 to September 2014. They were contacted through e-mails sent directly to them by the managers overseeing their respective locations. A total of two reminders were sent out at fortnightly intervals after the initial despatch. Data collection took approximately six weeks, from the end of

September 2014 to the middle of November 2014.

Questionnaire

The SurveyMonkeyR software was used in the designing and collation of the questionnaire, which was an 11-paged, electronic document divided into seven sections with a total of 25 questions designed to collect data for multiple studies. The aspect relating to this survey covered Pages 1 to 4,

Page 5 of 18 with a total of 14 questions. As no validated, off-the-shelf questionnaire existed to answer the key research questions of this survey, a suitable tool was devised and its validity tested by a pilot survey with 10 GPs in Australia who were not part of the study population. Recommendations and observations arising from the pilot study were used to modify the relevant section of the draft questionnaire, culminating in the final tool.

Analysis:

Analysis was with IBM SPSS Version 22, aiming to answer the two key research questions of “patterns of engagement of escorts among doctors involved in AHHC” and “associations between escort- engagements and various doctor-variables”. For the analysis, a respondent is considered to have used an escort if, at any time in the 12-month survey period, the doctor had engaged the services of an escort for work in AHHC. To give an idea of the frequency of usage, the responses included options on a 5-point Likert Scale: “not at all”, “rarely”, “sometimes”, “most times” and “all the time”.

The first research question identified the working habits of the doctors (driving self and working alone, driven by chauffeurs, and those that drive themselves but have an accompanying chaperone) and their frequencies.

The second research question explored possible associations between the three driving habits (which form the dependent variables) and nine independent doctor-variables, which were all presented as dichotomous variables (some were re-coded where necessary). These independent variables include gender (male and female), age (<40 years and ≥40 years), specialty (GPs and Non-GPs), postgraduate vocational status (fellowship attained and fellowship not attained), duration in after-hours service provision (≤2 years and >2 years), marriage status (in a legally-recognised social union or not), hours worked per week (< 24hours and ≥24 hours), living with kids (yes and no) and country of primary degree (Australian-trained and overseas-trained).

Given that the dependent variables were “ordinal categorical data”, a multi-staged Ordinal Logistics

Regression (OLR) was used to identify significant associations. Firstly, a univariate OLR analysis was

Page 6 of 18 performed for all the nine independent variables, and only those found to be statistically significant were included in the final multivariate OLR analysis. For each comparison, an odds ratio (OR) was generated, along with its corresponding 95% confidence interval (CI), with a significance level (p value) set at <0.05.

RESULTS

Basic Demographics

A total of 172 questionnaires were returned, out of which the “basic bio-data section” in four of them were not completed. These four were excluded from the analysis, leaving 168 valid responses out of the 300 that were dispatched. This gave a 56.0% response rate. The basic demographics of the respondents are summarized in Table 1.

Patterns of Escort Engagements

Table 2 shows that, overall, about three out of five doctors (60.8%) engaged an escort when working in AHHC at the period under survey. There was a wide variation by location in this patronage, as an overwhelming majority of the doctors working in the Brisbane Area (91.7%) and Sydney (88.2%) employed them, while only 26.1% did the same in the Gold Coast. There were moderate usages in

Adelaide and the Melbourne Areas, where 52.9% and 46.4% engaged escorts respectively.

Table 3 presents the frequencies of escort-engagement on a 5-point scale. It shows that 39.2% of the entire respondents used escorts all the time while at work for the 12-month survey period, while

26.2% did not use them at all at any time over the 12-month period.

When those who never engage escorts are excluded (Table 4), we found that about three-quarters of the doctors who engage chauffeurs do so either “all the time” (53.1%), or “most times” (24.0%). A similar rate (73.1%) was found among those who drive themselves without an accompanying escort, comprising of 51.6% for those who drive alone “all the time” and another 21.5% who do so “most times”. Conversely, among those who chose to drive themselves but still have an accompanying chaperon, only about a quarter (25.9%) do so at the same frequency as the other two groups (14.8% Page 7 of 18 for “all the time” and 11.1% for “most times”). The significance of these are explored in the

“Discussion” Section.

Associations of escort-usage with independent doctor-variables

Out of the nine independent variables, significant associations were found with three. These include

Gender, Marital Status and Registration Status. As shown in Table 5, males were over five times more likely to drive themselves without an escort compared to females (OR 5.34; P=0.001; CI 2.08 to

13.74). They were also 81% less likely to use chauffeurs (OR 0.19; p<0.001; CI 0.07 to 0.51).

Doctors in legally-recognized social unions were also 76% less likely to work without an escort (OR

0.24; p=0.03; CI 0.07 to 0.83).

Finally, we found that doctors who have attained the postgraduate fellowships were 57% less likely to engage escorts while working in AHHC (OR 0.43; p=0.02; CI 0.21 to 0.87).

No statistical associations were found between escort-engagement behaviours of the doctors and their age, specialty, duration in AHHC, hours worked per week, and whether they live with kids or not. The country where they obtained their primary medical degree also had no influence on the attitudes on whether to engage escorts or not.

DISCUSSION

Male doctors comprised 80.4% of the respondents, while the remaining 19.6% were females, indicating that female doctors were represented less in AHHC compared to their proportion in the

Australian general practice population (where 43 per cent of doctors are females) (Australian Bureau of Statistics 2013). This may be related to the higher apprehension female doctors express regarding the safety of AHHC, as was found in another study(Tolhurst, Baker et al. 2003). Interestingly, 71.9% of the respondents to this work were trained overseas, re-enforcing the international relevance of our study.

Page 8 of 18

It is interesting to find that about three out of five doctors involved in AHHC engaged an escort while on duty. Even though this represents a majority, it is not clear why the remaining 39.2% do not patronize escorts despite their proven benefits (Ifediora 2015, Ifediora 2016). One line of thinking is that the fees involved may limit the patronage of these escorts as the doctors have to pay them from their own pockets. Another possibility is the need for privacy, as some doctors may feel that having these individuals as company may impact on their privacies. Unfortunately, there is no existing study to allow a comparison to this finding, and our survey did not explore the real reasons behind the non-patronage for those who do not do so. It may be important to have future studies look into this, given that the use of chaperones in AHHC (and possibly chauffeurs) have been identified as a protective measure in the service (Ifediora 2015), and that the availability of protective measures have been significantly associated with reduced burnout (Ifediora 2015) and increased satisfaction

(Ifediora 2016).

The wide variation in patronage by location is another puzzling finding from this work. It is not clear why roughly 9-in-10 doctors working in Sydney and Brisbane engage escorts, while about half do so in each of Adelaide and the Melbourne. Even worse is the Gold Coast, where only one-quarter engage them. There is a chance that there might be peculiarities in the characteristics of the doctors in these locations, whether in terms of ideological differences or in other areas. Again, future studies, perhaps with a qualitative approach, may be the best way to explore these findings.

Very interestingly, there is a big difference in the frequencies with which doctors engage escorts, work alone or combine both. As shown in Table 4, nearly four out of five doctors who engage chaperones do so either “all the time” (53.1%) or “most of the time” (24.0%). This shows that there is a strong commitment to the use of escorts among those who patronize them. Conversely, those who chose to drive themselves but have an accompanying chaperon, do so either rarely (25.9%) or only

“sometimes” (48.1%). This may indicate that doctors who bother driving themselves generally prefer Page 9 of 18 to go alone anyway, and would therefore have little need for chaperones most of the times. These are important findings for policy developments, as a knowledge of the real reasons for why doctors chose not to employ the services of chauffeurs would help in the campaign to get more of them interested in the service if these reasons are adequately tackled.

It is not a complete surprise that, compared to females, male doctors were significantly less likely to use escorts, while being more likely to drive themselves. Given that females have been found to be more apprehensive with AHHC services,(Tolhurst, Baker et al. 2003, Tolhurst, Talbot et al. 2003) one would expect that they would be more likely to cherish “company” when involved in the service. In fact, this theory has recently been confirmed by a recent publication (Ifediora 2015) that found that male doctors involved in AHHC were less likely to adopt protective measures (including escorts) than their female colleagues.

The finding that doctors in legal unions (coupled) were less likely to drive themselves compared to

“single” doctors is also not totally a surprise. It is generally believed that those without legal commitments to their partners or kids (through marriage or other forms of social unions) generally take more risks and are less cautious (Forrest and Hay 2011, DePaulo 2015). By extrapolation therefore, one can argue that they are less likely to engage companions wile on AHHC.

Finally, our observation that doctors that have attained postgraduate fellowships were less likely to engage chauffeurs is a bit of a surprise. However, given the confidence that comes with the attainment of their status, we are tempted to conclude that this group of practitioners are less likely to be bothered with having company while on duty. Of course, it would be worthwhile for future studies to look into this interesting finding.

Page 10 of 18

Study limitations

The major limitation though, is the fact that the study was not designed to explore the real reasons behind the escort-engagement behaviours of the AHHC doctors. As indicated in the survey, future surveys may do well to explore these. Another limitation is that locations in other Australian states and territories like Tasmania, Western Australia and Northern Territory were not represented in the study. However, the AHHC services in these areas were are few and not well developed at the time of this survey, and therefore, the study outcome is unlikely to have been significantly affected.

Conclusions

We conclude that 60.8% of doctors involved in after-hours house call services engaged escorts

(chauffeurs or chaperones) while at work, leaving about 2-in-5 of the practitioners without this support. There was a wide variation by location in the usage of escort services in Australia, with most of the patronage in the Brisbane Area and Sydney (nearly 90% patronage), while only 1-in-4 of those in the Gold Coast do the same. About half of those working in Adelaide and Melbourne engage them as well.

We also conclude that doctors are very likely to be committed to their behaviours regarding the engagement of escorts or otherwise, since those who patronize them are likely to do so often, while those who prefer to work by themselves without escort support are very likely to only engage them sparingly.

Finally, we conclude males are significantly less likely to employ escorts but likely to likely to drive out alone while on AHHC duty. Also, practitioners in legally recognized social unions are less likely to work without escorts, while those that have attained postgraduate fellowships are less likely to engage them while on duty.

Recommendations

We recommend that more needs to be done to encourage a higher number of doctors involved in

AHHC to engage escorts, particularly given the protective and psychological benefits they offer these Page 11 of 18 doctors. It is equally important that future studies fully explore the real reasons behind the patterns of escort-engagements displayed by doctors, as identified in this study.

Competing Interests

The author declares that he has no competing interests (financially or non-financial). It may be pointed out that the author does undertake AHHC services on a part-time basis as an independent

General Practitioner with the NHDS, in addition to his normal, regular-hour, day-time general practice work in a different Surgery. This study was independently organized, planned and executed solely by the researcher, and he solely bore the entire costs for the work (which involved only the amount spent to subscribe to the SurveyMonkeyR software, and any possible article-processing charges that may arise).

Ethical Considerations:

Ethical clearance was obtained from the Human Research Ethics Committee of the Griffiths

University, Australia (GU Ref No: MED/47/14/HREC), prior to commencing the study.

Page 12 of 18

REFERENCES

Australian Bureau of Statistics. (2013, 2013-04-10). "Main Features - Doctors and Nurses." Retrieved July 7, 2016, from http://www.abs.gov.au/AUSSTATS/[email protected]/Lookup/4102.0Main+Features20April+2013. DePaulo, B. (2015, 25th April 2013). "Are Married People Less Likely to Kill Themselves?" Retrieved September 26, 2015, from http://www.psychologytoday.com/blog/living-single/201304/are-married- people-less-likely-kill-themselves. Farlex. (2012). "Farlex Partner Medical Dictionary." Retrieved September 28, 2015, from http://medical-dictionary.thefreedictionary.com/chaperone. Forrest, W. and C. Hay (2011). "Life-course transitions, self-control and desistance from crime." Criminology and Criminal Justice 11(5): 487-513. Ifediora, C. (2015). "Associations of stress and burnout among Australian-based doctors involved in after-hours home visits." The Australasian medical journal 8(11): 345-356. Ifediora, C. (2015). "Exploring the safety measures by doctors on after-hours house call services." The Australasian medical journal 8(7): 239–246. Ifediora, C. O. (2016). "Assessing the satisfaction levels among doctors who embark on after-hours home visits in Australia." Family Practice 33(1): 82. Medical Observer. (2015). "Doctors shun safety on home calls." Retrieved August 23, 2016, from http://www.medicalobserver.com.au/professional-news/doctors-shun-safety-on-home-calls. Merriam-Webster[Internet]. (2015). Retrieved September 21, 2015, from http://www.merriam- webster.com/dictionary/chauffeur. National Association for Medical Deputising Service. (2014). "After hours medical care in Australia, NAMDS After Hours Primary Care Summary Paper." Retrieved April 16, 2016, from http://www.namds.com/assets/files/After%20Hours%20Medical%20Care%20in%20Australia%20FIN AL.pdf. National Home Doctor Service. (2014). "Our Structure | National Home Doctor Service." Retrieved July 2, 2015, from http://www.homedoctor.com.au/section/About_us/Our_structure. Ozturk, S. (2015, 24 August 2015). "After-hours GPs using chaperones for home visits." Retrieved August 23, 2016, from http://www.australiandoctor.com.au/News/Latest-News/After-hours-GPs- using-chaperones-for-home-visits. Tolhurst, H., L. Baker, G. Murray, P. Bell, A. Sutton and S. Dean (2003). "Rural general practitioner experience of work-related violence in australia." The Australian Journal Of Rural Health 11(5): 231- 236. Tolhurst, H., J. Talbot, L. Baker, P. Bell, G. Murray, A. Sutton, S. Dean, C. Treloar and G. Harris (2003). "Rural general practitioner apprehension about work related violence in Australia." Australian Journal of Rural Health 11(5): 237-241. Wordweb [Internet]. (2015). Wordweb online Retrieved September 28, 2015, from http://www.wordwebonline.com/search.pl?w=escort.

Page 13 of 18

Table 1: Summary of the basic demographics of the respondents involved in Australian after-hours doctor house calls

Statistic Parameters N % Gender Male 135 80.4 Valid=168 Female 33 19.6 Age Range (Yrs) 39 or less 69 41.1 Valid=168 40-60 90 53.6 Over 60 9 5.4 Vocational/Registration Vocationally registered (Fellows) 61 44.5 status Valid=137 Non-vocationally registered (Non- 76 55.5 fellows) Primary degree Australian-trained 45 28.1 Valid=160 Overseas: New Zealand 6 3.8 Overseas: other 109 68.1 Specialty General Practice 135 84.4 Valid=160 Medical 7 4.4 Surgical 2 1.3 Emergency Department 6 3.8 Othersa 10 6.3 Location of Service Adelaide 51 31.9 Valid=160 Brisbane Areab 36 22.6 Gold Coast 23 14.4 Melbourne Areac 31 19.4 Sydney 17 10.6 Other (unfixed location) 2 1.3 Duration In After-Hours <= 2 yrs 80 50.0 Valid=160 2-10 yrs 54 33.8 >10 yrs 26 16.3 Hours worked/week <24 hrs/week 62 38.8 Valid=160 24 to 37.5 hrs/week 47 29.4 >37.5 hrs/week 51 31.9 Marital status Married 140 83.3 Valid=168 Single 12 2.4 De factod 10 6.0 Separated 4 2.4 Widowed 2 1.2 Whether Protective Yes 65 43.0 Measures used or not Valid=151 No 29 19.2 Have never thought about it 9 6.0

Have thought about it, but unsure of 48 31.8 what to do aOthers = Occupational physicians, Paediatricians, Public Health, etc; bBrisbane Area = Brisbane and Sunshine Coast; cMelbourne Area = Melbourne, Geelong and Canberra dDe facto = Co-habitation and civil partnership

Page 14 of 18

Table 2: Frequencies of escort-engagements by location among doctors involved in after-hours house calls

Location of Total Total number engaging Percentage use of service Respondents escorts escorts by location

Adelaide 51 27 52.9 Brisbane Area 36 33 91.7 Gold Coast 23 6 26.1 Melbourne Area 31 15 46.4 Sydney 17 15 88.2

TOTALS 158 96 (60.8%)

Page 15 of 18

Table 3: Patterns of escort-engagements among doctors involved in after-hours house calls

Frequency of Self-drive Chauffeur- Self-drive but accompanied use Only (%) driven only (%) by Chaperone (%)

Not at all 38 (26.8) 34 (26.2) 61 (69.3)

Rarely 8 (6.3) 3 (2.3) 7 (8.0)

Sometimes 17 (13.4) 19 (14.6) 13 (14.8)

Most times 20 (15.7) 23 (17.7) 3 (3.4)

All the time 48 (37.8) 51 (39.2) 4 (4.5)

TOTALS 131 (100.0) 130 (100.0%) 88 (100.0)

Page 16 of 18

Table 4: Frequencies of engagement of escorts among doctors who patronize their services at any rate for after-hours house calls (excludes those who never patronize them)

Frequency Self-drive only Chauffeur-driven Self-drive but accompanied by of use (%) (%) chaperone (%)

Rarely 8 (8.6) 3 (3.1) 7 (25.9)

Sometimes 17 (18.3) 19 (19.8) 13 (48.1) Most times 20 (21.5) 23 (24.0) 3 (11.1) All the time 48 (51.6) 51 (53.1) 4 (14.8)

TOTALS 93 (100.0) 96 (100.0%) 27 (99.9)

Page 17 of 18

TABLE 5: Ordinal Logistics Regression showing the associations between the escort- engagement habits of doctors in after-hours call services and independent doctor-variables

Estimate Odds ratio Significance 95% CI of OR Driving Habit Independent doctor-variables (b) OR=Exp(b) (p-value) Lower Upper

Gender Male (Vs Female) 1.675 5.34 0.001 2.08 13.74

Self-drive (N = 127)

Marriage -1.442 0.24 0.03 0.07 0.83 In a union (Vs Not in a union)

Gender -1.673 0.19 0.001 0.07 0.51 Use of Male (Vs Female) Chauffeur (N = 110)

Registration Status -0.854 0.43 0.02 0.21 0.87 VR (Vs Non VR)

NB:

 Only results from the stage 2 of the OLR analysis is shown. Most independent variables had no significant associations after the Stage 1 of the analysis and these include Age, Specialty, Duration in AHHC, and Country where Primary Degree was obtained)  Responses in parentheses represent the reference responses.  VR = Vocationally trained (attained postgraduate fellowship)

Page 18 of 18