A gender-sensitised weight-loss and healthy living program for men with overweight and obesity in Australian Football League settings (Aussie-FIT): A pilot randomised controlled trial.

BACKGROUND: Recent evidence shows that sport settings can act as a powerful draw to engage men in weight loss. The primary objective of this pilot study was to test the feasibility of delivering and to evaluate preliminary efficacy of Aussie-FIT, a weight-loss program for men with overweight/obesity delivered in Australian Football League (AFL) settings, in preparation for a future definitive trial. METHODS AND
FINDINGS: This 6-month pilot trial took place in Perth, Australia. Participants were overweight/obese (Body Mass Index [BMI] ≥ 28 kg/m2), middle-aged (35-65 years old) men. Participants were recruited in May 2018, and the intervention took place between June and December 2018. The intervention involved 12 weekly 90-min face-to-face sessions, incorporating physical activity, nutrition, and behaviour change information and practical activities delivered by coaches at 2 clubs. Data were collected at baseline and immediately postintervention. For trial feasibility purposes, 6-month follow-ups were completed. Outcomes were differences in weight loss (primary outcome) and recruitment and retention rates, self-reported measures (for example, psychological well-being), device-measured physical activity, waist size, and blood pressure at 3 months. Within 3 days of advertising at each club, 426 men registered interest; 306 (72%) were eligible. Men were selected on a first-come first-served basis (n = 130; M age = 45.8, SD = 8; M BMI = 34.48 kg/m2, SD = 4.87) and randomised by a blinded researcher. Trial retention was 86% and 63% at 3- and 6-month follow-ups (respectively). No adverse events were reported. At 3 months, mean difference in weight between groups, adjusted for baseline weight and group, was 3.3 kg (95% CI 1.9, 4.8) in favour of the intervention group (p < 0.001). The intervention group's moderate-to-vigorous physical activity (MVPA) was higher than the control group by 8.54 min/day (95% CI 1.37, 15.71, p = 0.02). MVPA among men attracted to Aussie-FIT was high at baseline (intervention arm 35.61 min/day, control arm 38.38 min/day), which may have limited the scope for improvement.
CONCLUSIONS: Aussie-FIT was feasible to deliver; participants increased physical activity, decreased weight, and reported improvements in other outcomes. Issues with retention were a limitation of this trial. In a future, fully powered randomised controlled trial (RCT), retention could be improved by conducting assessments outside of holiday seasons. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry: ACTRN12617000515392.

Introduction

Obesity is a global public health issue, and prevalence is increasing; in 2016, over 1.9 billion people worldwide were overweight, and over 650 million were obese [1]. In Australia in 2016, approximately 60% of adults were estimated to be overweight, including 25% classified as obese [2]. Although in Australia, overweight and obesity are more prevalent in men than in women (71% versus 56%) [3], men are underrepresented in trials of weight loss, and current community programs have not appealed to them [4,5]. The prevalence of overweight and obesity in men and low participation in trials may be in part because cultural constructions of masculinity in Western societies often promote overeating and excessive alcohol consumption and stereotype dieting or healthy eating as a ‘female-only’ activity [6,7,8]. Weight-loss programs are typically marketed to females and are rarely tailored to attract and meet the needs of men [6,9].

Professional sports can be used as a powerful draw to attract men to weight-loss programs [10,11]. For example, the Football Fans in Training (FFIT) program was designed to appeal to soccer fans with overweight and obesity in Scotland and to support them in losing weight, by fostering changes in their dietary and physical activity behaviours [12,13,14]. The program was delivered over 12 weekly sessions in the context of professional soccer (for example, premier league stadia, training facilities) and was effective in reducing weight (baseline-adjusted mean difference in weight loss between intervention and control groups at 12-month follow-up was 4.94 kg). FFIT also improved health behaviours and psychological well-being [14], and these effects were maintained 3.5 years postbaseline [15].

FFIT has been successfully adapted for other professional sport settings and for different countries, including rugby in England [16] and New Zealand (Rugby Fans in Training [RuFIT] [17]) and hockey in Canada (Hockey FIT [18] and HAT TRICK [19]). The model has also been successfully used to inform healthy lifestyle programs for other outcomes, as in EuroFIT, which focussed on physical activity and sedentary time, delivered in 4 European countries [20]. To date, this program has not been evaluated in Australia, and programs that address unique masculinity constructs are still an exception in Australia [21].

Building on FFIT, we developed the Australian Fans in Training (Aussie-FIT) program, a culturally sensitised version of FFIT for men in Australia, for delivery in Australian Football League (AFL) settings. The key features of Aussie-FIT have been described elsewhere [22]. Key innovations included the program being further developed in the context of behaviour change theories summarised in a recent theory review [23] and on reviews of successful studies of weight-loss maintenance [24]. This included the integration of principles from Self-Determination Theory (SDT) as a means to strengthen the quality, quantity, and longevity of motivation to change health behaviours [25]. According to SDT, humans have 3 psychological needs. These include the need to feel that one is competent (i.e., has the skills, support, and resources to meet challenges or strive for goals), autonomous (i.e., that what one is doing reflects personal choice, values, and volition), and related to others (i.e., respected, supported, cared for, and connected). A plethora of studies have indicated health behaviour change to be more effective and well-being to be enhanced when these needs are satisfied. Need satisfaction is important to foster because it almost unequivocally predicts more autonomous (i.e., self-determined, valued, and relevant) motives [25] which, according to a recent systematic review [26], are strongly related to long-term weight maintenance and physical activity behaviour. Accordingly, principles of SDT were applied throughout the program via integration into coach training and program activities. In line with theoretical understanding [23] and previous studies [24], men need to effectively self-regulate their behaviour and successfully form healthy habits to maintain long-term behaviour change. Social support and environmental factors play a crucial role in health behaviour change and long-term maintenance [9,21]. Men need access to plentiful resources to maintain their behaviour long-term, including physical (access to healthy food, exercise equipment), psychological (energy levels, positive mood), and social resources to facilitate health behaviour change maintenance [27,28].

The overarching aim of the study was to test the feasibility of delivering and evaluating the 12-week Aussie-FIT program and to test preliminary efficacy in the context of AFL in Australia to determine the appropriateness of a future randomised controlled trial (RCT). Specifically, the primary objectives were to examine (1) the feasibility of recruiting the target sample and retaining them at 3- and 6-month follow-ups; (2) the appropriateness of the selected measures to examine effects and to report effects of the intervention at 3-month follow-up on body weight, waist circumference, blood pressure, device-measured time spent in moderate-to-vigorous physical activity (MVPA) and sedentary time, self-reported dietary screener, and a range of secondary psychological outcomes; (3) the feasibility of administering the data collection protocols and questionnaire administration procedures in the AFL club setting at all 3 time points; and (4) the feasibility of collecting data to develop and undertake preliminary tests of a model to determine the cost-effectiveness of the program. Additional feasibility outcomes will be examined in a full process evaluation, reported elsewhere.

Methods

Study design

The Curtin University Human Research Ethics Committee (HREC2018-0458) provided ethical approval; all participants provided written informed consent before participation. The study protocol has been published elsewhere [22]. We undertook a two-group pilot waitlist RCT from April–December 2018 in the Perth metropolitan area of Western Australia. The waitlist control treatment was chosen to align with FFIT RCT design and to minimise any ethical concerns related to withholding treatment in the control group. Participating groups were stratified by site (2), and measures were collected at baseline and 3 and 6 months postbaseline. The waitlist control group started participation in the program at 3 months; therefore, the 6-month measure represents a ‘postprogram’ assessment for the control arm and a short-term maintenance effect for the intervention arm. The Aussie-FIT pilot study was designed to inform the development of a definitive future trial. The study sample size followed guidelines for pilot RCTs [29,30].

Participants

Men were recruited through a variety of methods, including features in weekly fan emails from the participating AFL clubs, social media announcements from the clubs and directly from the Aussie-FIT project social media accounts, word of mouth, and via the Aussie-FIT website [31]. We invited potential participants to register their interest and complete an eligibility check by visiting the program website and providing their age, weight, height, contact details, and availability to attend Aussie-FIT sessions. The study inclusion criteria were men aged 35–65 years, with Body Mass Index (BMI) 28 kg/m2 or higher (eligibility based on baseline measures). Age and BMI cutoffs were based on the FFIT study [32]. The study exclusion criteria were men who were unable to comprehend English, men who were unable to provide informed consent and/or agree to the randomisation, and those who were already participating in a health promotion or weight-loss program.

Randomisation and blinding

The randomisation sequence was generated by a researcher not involved in the trial using SPSS with block sizes of 4, stratified by site and BMI category (with 4 BMI categories: <30, 30–34.9, 35–39.9, and >40 kg/m2), and concealed until conditions were assigned. After baseline measures were taken, participants were allocated to the intervention or waitlist control group according to the randomisation sequence and informed about their group allocation via email and phone call. At follow-up measurements, we asked participants not to disclose their group allocation to research assistants; the latter were trained to avoid engaging in extensive conversations with the men during the weight assessment. The weight assessments were taken in a separate area for participant privacy and also to help mask assessors from hearing men mention their experiences of the program at follow-ups. An independent researcher blinded to the study allocation analysed study data. The study is reported in line with Consolidated Standards of Reporting Trials (CONSORT) guidelines extension for randomised pilot and feasibility trials [33] and the Template for Intervention Description and Replication (TIDieR) checklist [34] (S1 Table).

Procedure

Full details of the procedure, intervention content, and measurement tools are provided in the study protocol [22]. Before delivery of the Aussie-FIT program, the research team received feedback on the program handbook from a purposive sample of consumers (i.e., male AFL fans; N = 5). Small adjustments were subsequently made based on this feedback (for example, some content was reworded or reordered). Six coaches (4 male and 2 female) with experience of coaching in community or recreational programs were recruited to deliver the program. The intention was for all coaches to be recruited directly from the participating clubs. In reality, 3 coaches were recruited via the AFL clubs, and 3 additional coaches were identified by the research team because of limited capacity within the clubs to provide additional coaches. The Aussie-FIT coaches had experience in coaching and also professional experience in teaching, exercise instruction, or sport science. The coaches attended 4 half-day program delivery training workshops. These were delivered by 2 female members of the research team (EQ, DK) who had expertise in health psychology and motivation and behaviour change principles and experience designing and delivering coach education workshops (EQ). The program was delivered at the 2 AFL clubs in Perth, Western Australia. There were 2 deliveries to groups of approximately 15 men at each club (on 2 different days of the week) in each group (i.e., intervention and control). In total, there were 4 groups of men in the intervention arm, and 4 groups of men in the waitlist control group received the intervention at a later date.

Aligned with the FFIT program, the Aussie-FIT intervention included 12 weekly 90-min sessions designed to promote physical activity, healthy eating, and weight loss. The program was delivered to groups of approximately 15 men by one coach and included classroom-based activities and physical activity sessions. In the early weeks, a little more than half of the session was dedicated to classroom activities, with less time allotted to the physical activity sessions. Over the course of 12 weeks, the balance shifted to expand the physical activity component to align with the men’s progress in fitness. The delivery style was informal, encouraging positive social interaction, humour, and ‘friendly banter’. The program was gender-sensitised with an emphasis on dietary and physical activity changes that are consistent with masculinised practices (for example, ordering salad with a steak, getting back to being able to play football with children or grandchildren, discussion of the role of alcohol in weight loss, and fostering group support). The program supported participants to make small sustainable changes to their eating through portion control; reduced consumption of sugary drinks, energy-dense foods, and alcohol; and a gradual increase in physical activity by choosing the activity that the men enjoy the most or could most easily incorporate into daily life. To make the program culturally appropriate, program content was adapted to reflect the Australian Guidelines for Healthy Eating [35], and resources from ‘LiveLighter’ (a healthy eating campaign in Australia) were built into the program to illustrate key principles (for example, a tool for ease of reading food labels). The program was designed to teach participants strategies for self-regulation, goal setting, and avoiding compensatory behaviours (for example, overeating after intense physical activity) and to prevent relapse.

Building on the FFIT program [14], Aussie-FIT added new content to both the coach training and intervention content. In particular, new coach training included information on what coaches can say and do to create more need supportive and less controlling environments and why. This innovation was added because need support is known to promote self-determined motivation for behaviour change [36], and increased focus on this aspect of the intervention could further contribute to behaviour change and maintenance. This content was delivered in coach training with discussion of the basic principles of the theory, interactive activities (for example, scenarios, role-playing), detailed descriptions of these environmental components, and specific planning activities in which coaches detailed in writing what they planned to do during each session to be need supportive and how.

The program also included specific activities in which participants reflected on their personal motives and identified sources of basic psychological need satisfaction. In-session activities were included that encouraged men to recognise their own experiences of need satisfaction and sources of autonomous motivation. Previous studies reported that focus on weight-loss maintenance should be incorporated from the beginning of the weight-loss program [23,37,38]; therefore, we placed greater emphasis from the start on relapse prevention and on long-term maintenance of behavioural changes. Building on the content of the original FFIT program, Aussie-FIT participants were supported in how to best form habits [14] and how to form specific action and coping plans (expanding on their initial SMART goals); these plans were revisited and revised during subsequent Aussie-FIT sessions.

Other innovative aspects of the Aussie-FIT intervention included participants and coaches being invited to join closed Facebook groups which comprised the approximately 15 men in their Aussie-FIT group and their coach. Automated text messages, written in language to promote feelings of autonomy, competence, and relatedness, were sent each week to encourage session attendance and included a brief description of the topic of the upcoming session (for example, “Today we will talk about junk food—it can be tough to cut it out altogether, but there are plenty of ways you can make it healthier”). In session 1, participants received an Aussie-FIT booklet with session summaries and space to complete in-session activities and to self-monitor their weight-loss progress and goals. Men also received activity monitors (Fitbit Zip), club t-shirts, and reusable ‘LiveLighter’ branded water bottles. Additional session information (for example, summaries of key points covered in the program) was available online, via a password-protected subsection of the program website.

There was no formally structured intervention content or contact provided after 12 weeks. Participants were free to communicate through the Facebook group with the coach and with each other. We encouraged the participants to keep in touch and to schedule to meet and to exercise together after the program had finished.

Data collection

Measures were collected at the 2 AFL club facilities where we also delivered the Aussie-FIT sessions (alternative arrangements were made for men who could not attend measurement sessions at the club [for example, meetings at university premises]). After a brief welcome and summary of what the assessment session involved, men provided informed consent and completed measures at supervised stations in the following order: (1) weight, height, and waist circumference; (2) questionnaires completed on iPads; (3) blood pressure and health screening; (4) ActiGraph fitted and instructions given for use; and (5) final questions answered. As a thank-you gesture, men received an AUD$20 team store voucher at the completion of each assessment session.

During health screening, men completed the Adult Pre-exercise Screening System form [39] to identify contraindications to participate in the physical activity sessions. Following the recommendations of Exercise and Sport Science Australia [40], participants whose responses on the screening form or blood pressure readings raised a concern had the opportunity to discuss with an allied health professional (i.e., university-qualified practitioner with expertise in preventing, diagnosing, and treating a range of conditions and illnesses) how to adapt the physical activity components of the program to meet their needs. Thirteen participants with a systolic blood pressure reading of over 160 mm/Hg were advised to seek an opinion from their General Practitioner (GP) prior to commencing the program; none were considered unfit to participate by their GP.

Measures

The primary outcome in this trial was mean difference in weight between groups at 3 months, adjusted for baseline weight. Secondary outcome measures were feasibility of recruitment and trial retention; device-measured weight, waist, and blood pressure; physical activity; self-reported diet and alcohol; well-being; quality of life; motivation for physical activity and other variables relevant to behaviour change; and self-reported sleep.

Feasibility of recruitment and trial retention

Recruitment rates were recorded as the number of men who registered interest in the trial either via a web-based form or over the phone. The online form included a self-administered screening tool to check men’s age and BMI met the inclusion criteria. We also recorded the number of men who were retained to the randomisation stage and those who attended follow-ups at 3 and 6 months. Feasibility-related outcomes are reported descriptively.

Device-measured weight, waist, and blood pressure

Weight was assessed using an electronic scale (Seca 813 Flat Scale, EMSE81; Birmingham, United Kingdom). Waist circumference was measured twice, or 3 times if 2 measures differed by more than 5 mm; at each time point, we took the average of these measures. Resting blood pressure was measured with a digital blood pressure monitor (Omron HEM-705CP; Milton Keynes, United Kingdom) after sitting for at least 5 min. If systolic blood pressure was over 139 mm/Hg and/or diastolic blood pressure was over 89 mm/Hg, 2 further measures were taken and recorded, and a mean was calculated from the second and third measures.

Physical activity

To provide a device-based measure of physical activity and sedentary behaviour/time; participants were fitted with a hip-worn ActiGraph GTX-9 (ActiGraph, Pensacola, FL, USA) [41] accelerometer that they were asked to wear continuously (24 h/day) for the following 9 days. The GTX-9 was programmed to record raw data at a frequency of 30 Hz, which were later reduced to vertical axis movement counts of 60-s epoch for the purpose of the current analyses. Participants were instructed to wear the accelerometer on the right hip continuously except during bathing or aquatic activities. Accelerometer data were downloaded using ActiLife version 6.5.4, and 1-min epoch data were processed using a validated algorithm in SAS (version 9.3) [42]. Common cut-points [43,44] were used to classify each minute as sedentary (<100 counts per minute, cpm), light intensity (100–1,951 cpm), moderate intensity (1,952–5,724 cpm), or vigorous intensity (>5,724 cpm). The uncensored step count (i.e., all steps counted) recorded per minute was also used. All participants with ≥4 valid days of data were included in the analyses of physical activity and sedentary behaviour.

Self-reported dietary screener and alcohol assessment

Nutrition relevant outcomes were assessed using the Dietary Instrument for Nutrition Education-based measures [45] adapted for the Australian population [46]. Briefly, the dietary screener asked participants to report how many times over the past 7 days they ate or drank specific foods. From the responses, we calculated the average change in fatty food score, fruit and vegetable score, sugary food score (all 3 on a 1–4 scale, with higher scores indicative of higher consumption). We used a 7-day recall method to measure alcohol consumption, and based on the calendar recall data, we calculated total alcohol consumption (reported as total average number of alcohol units consumed in a week; 1 unit is 10 ml of pure alcohol).

Well-being, quality of life, and motivation-related variables

We used the 10-item Rosenberg self-esteem scale (range 1–4), for which higher scores indicate higher self-esteem [47], and the 10-item Short Form of the Positive and Negative Affect Scale (PANAS) separating the 2 constructs [48] (range 1–5 for both), for which higher scores indicate higher positive or negative affect. The 12-item Interpersonal Behaviours Questionnaire (IBQ) was applied to assess men’s perceptions of the psychological need support in relation to weight loss that they received from family and friends (range 1–7), with a higher score indicating more support [49]. We used items from an adapted measure of the Treatment Self-Regulation Questionnaire of weight-loss motivation [50] to assess autonomous and controlled motivation regulating weight-loss behaviours (range 1–5), where higher scores indicate higher autonomous motivation [50]. We assessed autonomy, competence [51], and relatedness [52] psychological need satisfaction in relation to weight-loss behaviours and collapsed all needs into 1 score for the purposes of the analysis (range 1–5), with higher scores indicative of higher need satisfaction. We measured health-related quality of life using Quality-Adjusted Life Years (QALYs) derived from the EuroQol five-dimensional five-level version (EQ-5D-5L) [53] and EQ-VAS overall score (range 0–100), for which high scores indicate better overall health. We converted the EQ-5D-5L values into utility weights on the basis of the preference weights of a pilot sample from the Australian general population to calculate QALYs [54].

Variables relevant to behaviour change

We also assessed goal facilitation and competing goals in relation to weight-loss goals [55], barriers [56] and planning [57] (all scores range 1–5), and habits using the Self-Report Behavioural Automaticity Index (SRBAI) for physical activity and for healthy eating [58] (both scales range 1–7). For all these measures, higher scores are indicative of improvement on the construct assessed.

Self-reported sleep

We used the Pittsburgh Sleep Quality Index (PSQI) to assess 7 sleep components; a score of zero indicates the highest possible quality sleep, and 21 indicates the worst sleep quality [59].

Statistical and economic analyses

Statistical analyses included descriptive statistics and percentages for each group. We used a one-way random effects analysis of covariance in Mplus 8.2 [60] to estimate the treatment effect between groups [61,62], adjusting for baseline values of the dependent variables and clustering effects (teams of 15 men). We did not stratify by club because we only had 2 clubs participating, both based in Perth, Western Australia. We conducted an all-cases analysis that involved all participants who were randomised to the intervention or control group and provided at least baseline data; missing data were handled using full information maximum likelihood estimation [63]. We compared the baseline characteristics of the participants who dropped out with the characteristics of the participants who completed the full study to check for differences between the 2 groups. Secondary analyses focussed on maintenance effects for the intervention group and treatment effects for the control group; these analyses were conducted on each group separately. In separate, within-group analyses, we modelled the linear effect of time (3 to 6 months) as a predictor of dependent variables, adjusting for baseline values and the clustering effect of groups. In so doing, the effect can be interpreted as the average amount of change on the dependent variable between 3 and 6 months.

Economic analysis included developing and piloting an economic model to test procedures and measures that could be used to estimate the cost-effectiveness of the program in a future full-scale trial. Direct costs associated with the program included program set-up, promotion, and delivery material costs. As a gauge of preliminary cost-effectiveness, the cost per 5% weight reduction and cost per QALY were assessed using the EQ-5D-5L data. The values of these data were converted into utility weights on the basis of the preference weights of a pilot sample from the Australian general population [54].

Results

Recruitment and retention

We assessed 426 men who applied online to participate in the program for eligibility; 120 did not meet the inclusion criteria (Fig 1). Places were allocated on a first-come, first-served basis; 130 men were invited to undertake baseline measures, all of whom were assessed and considered eligible and randomised to the intervention (n = 64) or waitlist control group (n = 66). One hundred and seventy-six were not offered a place in the program because of limited capacity in this study. Study retention was moderate, with 50 participants (78%) completing 3-month follow-up measures in the intervention group and 62 (93%) in the waitlist control group. For 6-month measures, 35 participants (54%) in the intervention group and 47 participants (71%) in the control group were retained. Assessment of program attendance was not feasible because not all coaches followed the protocol and regularly checked attendance.

Fig 1

CONSORT flow diagram for the Aussie-FIT pilot trial.

Aussie-FIT, Australian Fans in Training; CONSORT, Consolidated Standards of Reporting Trials.

Sample characteristics

Demographics, device-derived outcome measures, and scores for self-reported outcomes for the intervention, control group, and total sample at baseline are provided in Table 1. Accelerometer data at all 3 time points for participants with 4+ days of valid (i.e., ≥10 h/day) wear time are presented in S2 Table. At baseline, 22 men were categorised as overweight and 108 men as obese (83% of total sample), with a mean weight of 111.4 kg (SD = 18.2) and mean waist circumference of 116.2 cm (SD = 13.7). Mean baseline blood pressure (137.8 mm/Hg [SD = 16.97] systolic and 87.3 mm/Hg [SD = 9.8] diastolic) was at the higher end of recommended levels. At baseline, the mean number of steps per day was high (10,928 [SD = 3,454]), with 35.3 (SD = 19.2) min of MVPA per day (uncensored, unless otherwise specified).

Table 1

Baseline characteristics of participants allocated to the intervention arm or waitlist control arm.

	Aussie-FIT Intervention Group (n = 64)	Aussie-FIT Waitlist Control Group (n = 66)	Total (N = 130)
Age (average years, SD)	44.2 (7.6)	47.2 (8)	45.8 (7.9)
Ethnic origin (%)
—White	61 (95)	63 (95)	124 (95)
—Mixed	2 (3)	2 (3)	4 (3)
—Other	1 (2)	1 (2)	2 (2)
Employment status (%)
—In paid employment/self-employed	60 (94)	61 (92)	121 (93)
—Retired	2 (3)	0 (0)	2 2 (2)
—Other	2 (3)	5 (8)	7 (5)
Average number of hours worked per week (SD)	42.01 (11.74)	40.16 (15.12)	41.07 (13.54)
Estimated number of days taken off because of illness in the last 6 months (SD)	1.32 (1.88)	2.46 (4.15)	1.90 (3.28)
Number of children (SD)	2.08 (1.65)	2.03 (1.22)	2.05 (1.44)
Years of full-time education (SD)	15.09 (3.12)	13.00 (2.81)	14.03 (3.14)
Housing tenure (%)
—Mortgage or loan	46 (71.90)	49 (74.24)	95 (73.07)
—Rent	9 (14.10)	6 (9.10)	15 (11.53)
—Own	6 (9.40)	9 (13.60)	15 (11.53)
—Shared ownership	2 (3.10)	0 (0)	2 (1.5)
—Live rent free	0 (0)	1 (1.57)	1 (0.76)
—Other	1 (1.60)	1 (1.57)	2 (1.5)
Marital status (%)
—Married and living together	49 (76.56)	50 (75.75)	99 (76.15)
—Living together but not married	4 (6.25)	6 (9.09)	10 (7.69)
—Single	3 (4.68)	5 (7.57)	8 (6.15)
—Separated	6 (9.37)	0 (0)	6 (4.61)
—Divorced	2 (3.05)	3 (4.54)	5 (3.84)
—Other	0 (0)	2 (3.03)	2 (1.5)

Abbreviations: Aussie-FIT, Australian Fans in Training.

Preliminary analysis

There were no significant differences in any baseline characteristics of the participants who dropped out compared to those who completed the study; all scales were reliable at all 3 time points (using Cronbach's alpha to measure internal consistency; S1 Text).

Outcomes at 3 months: All-cases analysis

The mean difference at 3 months between groups was 3.3 kg (95% CI 1.9–4.8) or 2.88% of total body weight lost (1.48–4.28) from baseline, both in favour of the intervention arm (p < 0.001) as per all-cases analysis (N = 130); see Table 2. The mean difference in BMI between groups (changes from baseline) was 1.48 kg/m2 (95% CI 0.57–1.73) in favour of the intervention (p < 0.001). Mean differences in changes in waist circumference (cm), blood pressure (mm/Hg), step counts, and time spent in sedentary or light activity were not significant between groups. However, the intervention group’s MVPA at 3 months was significantly higher than that of the control group by 8.5 min/day (95 CI 1.3, 15.7; p = 0.02).

Table 2

Group differences in 3-month changes in primary and secondary outcome (per protocol and all-cases analysis).

	Aussie-FIT Intervention Group (Baseline n = 64)*	Aussie-FIT Waitlist Control Group (Baseline n = 66)	Per Protocol (n = 112)	All-Cases Analysis (N = 130)
	Group mean	Group mean
Weight (kg)**	110.15 (16.37)	112.64 (19.91)
—3 months	107.81 (16.58)	111.21 (20.60)
—6 months	107.21 (18.70)	108.22 (20.93)	−3.92 (−6.10, −1.73), p < 0.001	−3.33 (−4.77, −1.89), p < 0.001
Weight (% of total body weight lost since baseline)
—3 months (from T0 to T1)	3.41 (4.39)	0.53 (3.19)
—6 months (from T0 to T2)	4.51 (6.26)	3.67 (4.58)	−3.47 (−5.43, −1.51), p = 0.001	2.88 (−4.28, −1.48), p < 0.001
Waist circumference (cm)	115.23 (11.79)	117.24 (15.42)
—3 months	113.42 (15.87)	116.08 (15.51)
—6 months	108.73 (13.68)	110.84 (15.46)	−2.08 (−6.60, 1.91), p = 0.32	−2.66 (−5.53, 0.21), p = 0.07
BMI (kg/m2)	34.66 (4.63)	35.32 (6.53)
—3 months	33.81 (5.03)	35.02 (6.99)
—6 months	33.71 (5.46)	34.33 (6.88)	−1.36 (−2.06, −0.66), p < 0.001	−1.48 (−1.73, −0.57), p < 0.001
Blood pressure (mm/Hg)
Systolic	139.80 (18.81)	135.95 (14.88)
—3 months	136.58 (14.73)	136.56 (13.13)
—6 months	138.29 (14.75)	136.62 (14.97)	−4.42 (−9.11, 0.26), p = 0.06	−2.16 (−5.72, 1.41), p = 0.24
Diastolic	87.95 (10.31)	86.67 (9.46)
—3 months	87.22 (10.17)	87.77 (8.64)
—6 months	88.69 (7.82)	88.13 (9.37)	−0.72 (−4.12, 2.67), p = 0.68	−1.09 (−4.51, 2.33), p = 0.53
Physical activity measured with ActiGraph
Sedentary time (min/day)	590.74 (94.55)	571.9 (101.76)
—3 months	650.84 (192.32)	644.22 (171.81)
—6 months	536.32 (96.13)	561.96 (76.19)	−1.52 (−38.25, 35.20), p = 0.94	−0.95 (−35.18, 33.27), p = 0.96
LPA time (min/day)	261.80 (70.21)	268.2 (71.7)
—3 months	264.72 (67.83)	276.47 (73.01)
—6 months	277.37 (79.24)	298.59 (75.82)	2.40 (−29.71, 34.50), p = 0.88	0.89 (−21.54, 23.31), p = 0.94
MVPA time (min/day)	35.61 (19.57)	38.38 (21.5)
—3 months	45.14 (22.50)	37.63 (20.45)
—6 months	39.90 (22.50)	51.00 (31.77)	7.86 (−1.25, 16.97), p = 0.09	8.54 (1.37, 15.71), p = 0.02
Steps (uncensored)	12,872 (3,770)	13,097 (3,754)
—3 months	13,717 (4,322)	13,421 (4,047)
—6 months	13,762 (4,395)	15,736 (4,620)	849 (−972, 2,670), p = 0.36	445 (−660, 1,551), p = 0.43
Steps (censored)	10,787 (3,359)	11,053 (3,557)
—3 months	11,654 (4,018)	11,356 (3,828)
—6 months	11,718 (4,049)	13,439 (4,407)	777 (−561, 2,115), p = 0.26	550 (−552, 1,652), p = 0.33
Self-reported outcome measures
Fatty food score (DINE)	1.91 (0.34)	1.91 (0.31)
—3 months	1.70 (0.29)	1.89 (0.33)
—6 months	1.74 (0.29)	1.61 (0.29)	−0.19 (−0.32, −0.06), p < 0.001	−0.20 (−0.28, −0.12), p < 0.001
Fruit and vegetable (DINE)	1.87 (0.37)	1.85 (0.53)
—3 months	2.24 (0.56)	2.03 (0.56)
—6 months	2.20 (0.47)	2.28 (0.64)	0.18 (−0.06, 0.42), p = 0.15	0.20 (0.01, 0.39), p = 0.036
Sugary food score (DINE)	1.39 (0.36)	1.44 (0.47)
—3 months	1.10 (0.16)	1.51 (0.56)
—6 months	1.16 (0.29)	1.16 (0.39)	−0.39 (−0.61, −0.18), p < 0.001	−0.39 (−0.66, −0.12), p = 0.005
Total alcohol consumption (units per week)	8.70 (11.34)	8.44 (10.88)
—3 months	11.50 (8.54)	8.54 (6.09)
—6 months	7.04 (8.95)	5.81 (5.43)	0.42 (−0.82, 1.66), p = 0.50	1.06 (−0.99, 3.11), p = 0.31
Self-esteem (Rosenberg scale, range 1–4)	3.02 (0.51)	2.98 (0.37)
—3 months	3.23 (0.51)	3.03 (0.44)
—6 months	3.21 (0.54)	3.25 (0.42)	0.13 (0.02, 0.24), p = 0.02	0.15 (0.05, 0.25), p = 0.003
Positive affect (PANAS, range 1–5)	2.92 (0.70)	2.89 (0.66)
—3 months	3.56 (0.64)	2.97 (0.66)
—6 months	3.35 (0.58)	3.59 (0.63)	0.58 (0.37, 0.79), p < 0.001	0.58 (0.35, 0.80), p < 0.001
Negative affect (PANAS, range 1–5)	1.71 (0.60)	1.75 (0.56)
—3 months	1.53 (0.51)	1.69 (0.58)
—6 months	1.59 (0.54)	1.48 (0.59)	−0.11 (−0.29, 0.07), p = 0.25	−0.14 (−0.30, 0.02), p = 0.09
Need support (IBQ scale, range 1–7)	5.07 (1.18)	5.33 (1.07)
—3 months	5.35 (1.27)	5.43 (1.13)
—6 months	5.33 (1.09)	5.64 (1.21)	0.18 (−0.69, 1.06), p = 0.68	0.15 (−0.43, 0.73), p = 0.61
Basic need satisfaction in relation to weight-loss behaviours (range 1–5)	3.23 (0.59)	3.40 (0.65)
—3 months	3.92 (0.61)	3.40 (0.72)
—6 months	3.73 (0.75)	3.97 (0.66)	0.72 (0.50, 0.94), p < 0.001	0.60 (0.35, 0.84), p < 0.001
Weight-loss motivation (range 1–5)
Autonomous	4.32 (0.53)	4.22 (0.55)
—3 months	6.32 (0.49)	6.23 (0.55)
—6 months	6.33 (0.67)	6.50 (0.45)	0.14 (−0.05, 0.32), p = 0.14	0.07 (−0.12, 0.24), p = 0.48
Controlled	2.79 (0.65)	2.91 (0.58)
—3 months	4.26 (1.22)	4.28 (0.97)
—6 months	3.97 (1.07)	4.10 (1.21)	0.04 (−0.46, 0.54), p = 0.88	0.11 (−1.22, 1.44), p = 0.87
Health-related quality of life (health utility score)	0.86 (0.20)	0.82 (0.16)
—3 months	0.89 (0.13)	0.84 (0.16)
—6 months	0.86 (0.16)	0.86 (0.19)	0.02 (−0.02, 0.07), p = 0.36
Total for health today (range 0–100)	59.28 (18.63)	54.92 (16.65)
—3 months	74.15 (15.10)	57.66 (15.82)
—6 months	75.37 (13.28)	69.50 (14.62)	16.10 (13.22, 18.97), p < 0.001	14.57 (9.77, 19.34), p < 0.001
Goal facilitation (range 1–5)	2.64 (0.76)	2.80 (0.75)
—3 months	3.17(0.91)	2.81 (0.85)
—6 months	3.19 (0.75)	3.41 0.79)	0.54 (−0.09, 1.16), p = 0.09	0.47 (0.27, 0.68), p < 0.001
Competing goals (range 1–5)	3.80 (0.65)	3.67 (0.72)
—3 months	3.38 (0.68)	3.52 (0.71)
—6 months	3.39 (0.75)	3.04 (0.92)	−0.20 (−0.48, 0.07), p = 0.15	−0.16 (−0.45, 0.13), p = 0.27
Habits for PA (range 1–7)	3.18 (1.14)	3.44 (1.27)
—3 months	3.96 (1.31)	3.30 (1.30)
—6 months	4.01 (1.30)	4.18 (1.40)	1.07 (0.70, 1.45), p < 0.001	0.89 (0.73, 1.05), p < 0.001
Habits for eating (range 1–7)	3.02 (1.22)	3.01 (1.07)
—3 months	4.01 (1.30)	2.99 (1.24)
—6 months	3.78 1.35)	4.05 (1.21)	1.17 (0.66, 1.57), p < 0.001	1.08 (0.74, 1.42), p < 0.001
Barriers (range 1–5)	3.16 (0.90)	3.28 (0.69)
—3 months	3.16 (0.78)	3.14 (0.70)
—6 months	3.09 (0.81)	3.36 (0.83)	0.21 (−0.07, 0.48), p = 0.14	0.07 (−0.08, 0.21), p = 0.38
Planning (range 1–5)	3.00 (0.83)	3.37 (0.86)
—3 months	3.70 (0.68)	3.39 (0.71)
—6 months	3.40 (0.92)	3.82 (0.82)	0.52 (0.26, 0.78), p < 0.001	0.39 (0.20, 0.58), p < 0.001
Sleep quality (PSQI, range 0–21; lower score indicates a better outcome)	5.76 (3.05)	6.59 (3.59)
—3 months	4.11 (2.30)	5.66 (3.02)	−1.29 (−2.03, −0.55), p = 0.001	−1.18 (−2.16, −0.20), p = 0.02
—6 months	4.77 (3.30)	4.14 (3.03)

*n = 50 for 3-month measure for intervention group and n = 62 for 3-month measure control group; n = 35 for 6-month measure for intervention group and n = 47 for control group at T2.

**Top line is a baseline measure. Group means are completer’s means. Censored step counts (n/day) are any steps taken during minutes in which the ActiGraph cpm were >100; variance for step count data was larger than the maximum permitted in Mplus (1,000,000), so the dependent variable and dependent variable at baseline (which we included in the model as a covariate) were rescaled by 10 (i.e., linear transformation) for censored and uncensored steps. Abbreviations: Aussie-FIT, Australian Fans in Training; BMI, Body Mass Index; cpm, counts per minute; DINE, dietary instrument for nutritional education; IBQ, Interpersonal Behaviours Questionnaire; LPA, Leisure Time Physical Activity; MVPA, moderate-to-vigorous physical activity; PANAS, Positive and Negative Affect Scale; PSQI, Pittsburg Sleep Quality Index.

At 3 months, the intervention group reported lower scores for high fat foods (−0.20; 95% CI −0.28, −0.12; p < 0.001) and sugar (−0.39; 95% CI −0.66, −0.12; p = 0.005) than the control group, but there was no significant improvement in fruit and vegetable score or alcohol intake. In terms of psychosocial outcomes, at 3 months the intervention group, as compared with the control group, reported significantly higher levels of self-esteem (0.23; 95% CI 0.15, 0.29; p < 0.001), positive affect (0.46; 95% CI 0.24, 0.69; p < 0.001), basic need satisfaction in relation to weight-loss behaviours (0.60; 95% CI, 0.35, 0.84; p < 0.001), overall health as per the EuroQol Visual Analogue Scale (14.5; 95% CI 9.7, 19.3; p < 0.001), goal facilitation (0.47; 95% CI 0.27, 0.68; p < 0.001), habits for physical activity (0.89; 95% CI 0.73, 1.05; p < 0.001) and healthy eating (1.08; 95% CI 0.74, 1.42; p < 0.001), and planning (0.39; 95% CI 0.20, 0.58; p < 0.001). Sleep quality improved from baseline to a greater extent in the intervention than in the control group (−1.18; 95% CI −2.16, −0.20; p = 0.02). All other differences on secondary outcomes were small and statistically nonsignificant (Table 2). Intervention (S3 Table) and control (S4 Table) group effects were analysed separately for the 3- and 6-month data.

Preliminary cost-effectiveness

The cost-effectiveness data were feasible to collect and analyse. The total direct costs associated with the Aussie-FIT program (i.e., program set-up, promotion, and delivery material costs) amounted to AUD$35,215 (i.e., AUD$270 per participant), compared with no costs incurred for the comparison group. Thus, the incremental cost per individual was AUD$270. The preliminary cost-effectiveness of the Aussie-FIT program was assessed using 2 outcome variables: (1) the number of men achieving a 5% weight reduction at 3 months and (2) QALYs. Thus, we calculated a preliminary indication of the incremental cost-effectiveness (ICER) per each additional man achieving a 5% weight loss at 3 months and the incremental cost-effectiveness per QALY.

At 3 months, 19 out of 64 men (approximately 29%) in the intervention group had achieved a 5% weight loss, while only 6 out of 66 men (approximately 9%) in the comparison group had achieved a 5% weight loss. The incremental effect of the program concerning a 5% weight loss at 3 months is estimated to be 0.206 (i.e., 0.297–0.091). Thus, ICER per each additional man achieving a 5% weight loss at 3 months was estimated to be AUD$1,315 (i.e., ). The EQ-5D-5L values were converted into utility weights on the basis of the preference weights of a pilot sample from the Australian general population to compute QALYs. At 3 months, the average QALYs for the intervention group were 0.216 (SD = 0.041), while for the comparison group, it was 0.209 (SD = 0.037). The incremental QALY gains at 3 months were estimated to be 0.007 (i.e., 0.216–0.209). Thus, the cost-effectiveness of the Aussie-FIT was estimated to be AUD$39,756 (i.e., ) per an additional QALY gained after 3 months.

Discussion

Summary of findings

This pilot RCT aimed to determine the feasibility of delivering and evaluating the 12-week Aussie-FIT program in the context of AFL in Australia to determine the appropriateness of a future RCT. Findings support the feasibility of delivering this intervention in this context and its preliminary efficacy. Men who participated in the Aussie-FIT program lost a significant amount of weight and sustained this weight loss 3 months postintervention. Pilot data also indicated that the program had positive effects on a range of behavioural and psychosocial outcomes. The recruitment process was very effective, with many more men applying than we had capacity to offer the program to as part of this study.

Similar to the experiences of FFIT in Scotland, capitalising on the high popularity of the 2 participating AFL clubs in Perth proved to be an effective way to attract men to the program. Trial retention at the 3-month follow-up was excellent and comparable with other FFIT trials. However, there was quite high loss at 6 months, with only 63% of the original sample retained. Overall, these findings support the appropriateness of a future fully powered RCT to more rigorously test the effectiveness and cost-effectiveness of the Aussie-FIT intervention as a means to address overweight and obesity among men in Australia.

Relationship to other literature

Our findings for weight loss at 3 months are comparable to most other gender-targeted interventions delivered in sports settings internationally. The results of our pilot study are comparable to those from the HockeyFIT pilot trial in Canada [18] (N = 80) and the RuFIT pilot in New Zealand [17] (N = 96), which reported mean differences in weight loss between groups at 3 months to be 3.6 kg (95% CI 1.9–5.3) and 2.5 kg (95% CI −0.4 to 5.4), respectively, in favour of the intervention. Compared to the FFIT (N = 748) trial, Aussie-FIT participants’ weight loss immediately post-program was lower; FFIT participants lost 5.2 kgs (−6 to −4.3) at 3 months [14]. However, any comparisons with FFIT must be interpreted with caution, given that the FFIT evaluation was not a pilot.

In relation to physical activity outcomes, we found only a small increase in step counts following participation in the program for the intervention arm; however, this warrants more rigorous testing via a fully powered RCT, given the relatively large proportion of participants with step counts meeting general PA guidelines at baseline. The FFIT study used a self-report measure of physical activity, whereas the recent EuroFIT intervention reported device-measured physical activity (steps) as the main outcome; intention to treat analyses in EuroFIT showed a baseline-adjusted mean difference of 1,208 steps per day (95% CI, 869–1,546) in favour of the intervention group at 3 months. It is difficult to explain the high baseline in step count in this study. Anecdotally, we are aware that many participants were in manual professions (for example, construction work); hence, it is possible that there was a sampling bias, However, we did not collect data on occupations, so we cannot draw conclusions as to a potential bias. We also note that the ActiGraph algorithm classifies intermittent stepping more readily than the ActivPAL algorithm (used in the EuroFIT trial),which is shown to result in a nontrivial discrepancy between devices [64].

In line with other FIT studies [14,20], Aussie-FIT participants successfully decreased their consumption of fatty and sugary foods following the program, and point estimates indicated small increases in fruit and vegetable intake; however, it is noteworthy that scores were already relatively high at the start of the intervention. Excessive alcohol consumption is often perceived to be central to the culture of football spectatorship in Australia, and the program specifically aimed to reduce it [65]. In the current sample, men reported low alcohol intake on average at baseline, which was below the alcohol guideline limits for Australian males. It is possible that men underreported alcohol consumption or that this study attracted a sample of men who were moderate drinkers.

The inclusion of motivational components in the coach training and program content of Aussie-FIT were a key adaptation and extension of previous trials. There is preliminary evidence to suggest that participating in the Aussie-FIT program led participants to experience increased feelings of autonomy, competence, and relatedness in relation to weight-loss behaviours. These findings are important because, according to SDT [25], feeling that one is competent, autonomous, and related to others are critical determinants of adaptive behaviours, cognitions, and emotions [66]. In a future RCT, longer-term follow-ups will help to determine whether participation in Aussie-FIT leads to longer-term experience of need satisfaction and/or whether experiences of need support offered during Aussie-FIT are associated with better outcomes for participants.

We also assessed constructs that were targeted in the intervention as relevant for weight loss but were not previously addressed in FIT-based studies. For instance, there was a significant improvement in goal facilitation. This tentatively suggests that Aussie-FIT participants learnt how to prioritise their weight loss. However, there were no important/significant effects for dealing with competing goals. This may be due to a low emphasis in the program content on dealing with other competing goals [67]. The improvements in self-reported behavioural automaticity for physical activity and healthy eating were consistent with findings from other studies promoting habit formation [68]. There was also improvement in planning at 3 months in favour of the intervention group, consistent with previous research on forming action and coping plans [56,57].

Finally, there was improved self-reported sleep quality at 3 months in favour of the intervention group. Given the burgeoning concerns with the ill-health effects of bad quality of sleep [69], this result is promising. This finding is based on well-validated self-report instrument but warrants replication with device-measured monitoring of sleep quality. Other FIT studies did not examine sleep change as an outcome of the intervention. Sleep duration and quality have an important influence on physical activity levels and consequently on weight. Better sleep is correlated with higher physical activity and lower weight [70].

Our findings support progression to a fully powered RCT to further test the effectiveness of the Aussie-FIT program; a future trial should be powered with data from this pilot rather than the FFIT study and also be powered to detect changes in secondary outcomes. The long-term goal of the Aussie-FIT program is to promote long-term maintenance of behaviour changes that have led to weight loss; therefore, longer-term follow-up studies are required to determine whether specific additions included in this program have an impact on long-term outcomes. For instance, FFIT program participants maintained intervention effects at 3.5 years postbaseline (2.9 kg [95% CI 1.7, 4 kg], p < 0.001) [15]; further exploration is needed to assess if the inclusion of specific SDT techniques and behaviour change techniques adds value that corresponds with further weight reduction or weight maintenance at a desired level.

We intended to recruit coaches via the AFL clubs; however, the clubs did not have coaches readily available to fulfil this role. As a result, 3 coaches were recruited via recommendations from the clubs, and 3 coaches were independently identified by the research team. All Aussie-FIT coaches’ professional backgrounds (which included teaching, coaching, exercise instruction, and sport science) meant they were likely to be equipped to create an environment that is seen as central to the success of FFIT, i.e., a nondidactic, encouraging, and interactive delivery style that incorporates appropriate banter to support vicarious learning and team spirit [71]. However, those coaches who were not directly associated with the club may have been less able to integrate into their coaching style other characteristics seen as contributing to FFIT’s success, such as the ‘behind-the-scenes’ stories and tacit knowledge of the inner workings of the club into program delivery [71]. To overcome this, within the coach training, the coaches were encouraged to think about how to incorporate the sport and club ‘feel’ within their delivery style. Based on qualitative data, the level of connection between the coach and the club did not seem to contribute to any variability in quality of experience any more than other relevant variables such as the coaches’ personality and motivation or differences in facilities available to deliver the program. However, the extent to which the personal and situational characteristics that shape program delivery impact participants’ experiences of the program and observed outcomes could be further explored in future research to help inform the most appropriate implementation model for Australia.

The preliminary cost-effectiveness evaluation suggests that the Aussie-FIT program was relatively inexpensive to deliver and potentially cost-effective, with estimated ICERs of $1,315 per each additional man achieving a 5% weight loss and $39,756 per QALY gained after 3 months. Moreover, our calculated ICER per QALY lies within the commonly acceptable range considered to be good value for money in Australia [72,73]. However, the reported ICER in this pilot study should be interpreted with caution and may be overestimated because it does not include several relevant costs, such as direct medical costs and future health system costs due to the intervention. A complete and comprehensive cost-effectiveness evaluation using the model piloted in this study is warranted in a fully powered trial.

Strengths, limitations, and future directions

This study had several strengths. It was the first trial to assess an FIT intervention in Australia in the AFL setting. Key strengths also include the evidence of effectiveness of strategies to recruit men, feasible trial procedures such as the use of new questionnaires and device-based assessments, and feasible incorporation of principles of effective motivation into the intervention content. The limitations of the study include the inability to collect attendance data at the Aussie-FIT weekly sessions and low retention at final follow-up. One possible explanation for this loss could be the timing of data collections. Because of unavoidable time restrictions on when the program could be delivered, final follow-ups were scheduled the week before Christmas, which is usually a busy time of personal commitments and festivities and may have prevented men from being able to attend. Unfortunately, we were unable to collect reliable data on men’s attendance at the program as coaches only sporadically recorded attendance. This was likely because coaches were busy at the start of the program, talking to men and preparing for the session. In a future trial, it may be more advantageous to use a ‘self-check–in’ system on an iPad, on which men can register when arriving at the venue. This study also lacked objective measurement of sleep quality, body composition, and cardiorespiratory or neuromuscular fitness. A fully powered future trial could include these measurements to further improve quality of the outcome assessment. Also, some of the measured outcomes did not improve significantly in the intervention group at 3 months (for example, waist circumference, blood pressure, sedentary time, need support). Nevertheless, findings were consistently in favour of the intervention group. To fully understand whether there are no important/significant differences, a fully powered trial needs to be conducted.

Another limitation of the current study is the generalisability of findings with respect to ethnicity because the sample was not representative of the Australian population (95% were White). Future formative work should include feedback on resources from a diverse sample of men. Future trials should investigate whether findings can be replicated in other ethnic groups or whether the program requires additional tailoring to appeal to a more diverse sample of men in Australia (for example, men from culturally and linguistically diverse backgrounds, indigenous men, men from across the socioeconomic spectrum). Because there are only a limited number of AFL clubs in Australia, a fully powered RCT may need to also rely on lower-level clubs (for example, state-based leagues) to reach target numbers and attract men from a wider geographical reach.

Conclusion

The pilot findings may be generalizable to delivery of the intervention in other professional sport settings in Australia, and broader implementation may be better achieved by capitalising on rollouts of the program in lower-level state-based leagues and via adaptations to other sports. The Aussie-FIT model also has potential for expansion via delivery to specific population segments who are not necessarily overweight or obese but are insufficiently active to benefit health and for whom existing interventions may not appeal (for example, new dads, prostate cancer survivors). Pilot studies are needed to determine whether tailoring the program to engage these specific population groups can offer an alternative or superior opportunity to ‘usual care’. Overall, this study achieved the objectives of testing the feasibility of delivering and evaluating Aussie-FIT in the AFL context in Australia. Recommendations to use these methods in a future definitive RCT are warranted. The Aussie-FIT program should be tested and implemented on a larger scale with maintenance effects tested over a longer timeframe (for example, 1- and 2-year follow-ups).

CONSORT guidelines extension for randomised pilot and feasibility trials and TIDieR checklists.

(S1A) CONSORT 2010 checklist of information to include when reporting a pilot or feasibility trial. (S1B) The TIDieR Checklist. CONSORT, Consolidated Standards of Reporting Trials; TIDieR, Template for Intervention Description and Replication.

(DOCX)

Click here for additional data file.

Accelerometer data at all time points for participants with 4+ days of valid (i.e., ≥10 h/day) wear time; data are means (SD).

(DOCX)

Click here for additional data file.

Changes from 3 months to 6 months for the intervention group only.

(DOCX)

Click here for additional data file.

Program effects (i.e., change from 3 months to 6 months) for the waitlist control group.

(DOCX)

Click here for additional data file.

All scales reliability checks.

(SPSS file, here: https://osf.io/4vsng/files/).

(DOCX)

Click here for additional data file.

24 Feb 2020

Dear Dr. Quested,

Thank you very much for submitting your manuscript "A gender-sensitised weight loss and healthy living program for overweight and obese men in Australian Football League settings (Aussie-FIT): A pilot randomised controlled trial." (PMEDICINE-D-19-03524) for consideration at PLOS Medicine. I sincerely apologise for the delay in getting this decision to you.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Mar 09 2020 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

Abstract- please use an alternative to “hook”

Abstract- please use specific language instead of “meaningful differences”. If you choose to report your secondary outcomes here, we suggest you include 95% CI and p values.

Abstract methods and findings- the last sentence of this section should highlight a limitation of your study design

At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

Please include the study protocol document and analysis plan, with any amendments, as Supporting Information to be published with the manuscript if accepted. If this is published, please provide clear details in the methods section.

References must be within square brackets throughout

Please provide the primary and secondary outcome measures within the methods section. A brief summary of the trial outcomes can be presented in the introduction section

Questionnaires -please provide the questionnaires used in this study as SI files or if previously published, please provide a citation

Please provide 95% confidence intervals along with all p-values, all p-values should be in 0.xx format

Discussion- I recommend that the results outlined at line 448 are reported in the results section

The primary outcomes on your trial registry do not fully match the primary outcomes reported in the manuscript, For instance- acceptability to participants and coaches. If a clarification is needed in the submission, please revise.

Please include sections and paragraph numbers instead of page numbers in the CONSORT checklist

The Data Availability Statement (DAS) requires revision. For each data source used in your study:

a) If the data are freely or publicly available, note this and state the location of the data: within the paper, in Supporting Information files, or in a public repository (include the DOI or accession number).

b) If the data are owned by a third party but freely available upon request, please note this and state the owner of the data set and contact information for data requests (web or email address). Note that a study author cannot be the contact person for the data.

c) If the data are not freely available, please describe briefly the ethical, legal, or contractual restriction that prevents you from sharing it. Please also include an appropriate contact (web or email address) for inquiries (again, this cannot be a study author).

Comments from the reviewers:

Reviewer #1: This manuscript reports the results from a pilot of the Aussie-FIT program. The manuscript is well written and descriptive and I only have a couple of minor comments.

Line 105: Please change to "...healthy habits to maintain long-term behaviour change."

Lines 115-124: You do not need to list all the outcomes here - just a brief summary is sufficient for the introduction.

Lines 173-175: Was feedback on the program only from 5 men? This is very small, and I would consider adding to the discussion that more formative work may have helped make the program more culturally appropriate for Australian men.

Line 181: Please state who delivered the training workshops.

Lines 349-350: Please add the proportions of men (i.e. in %) for follow-up measures.

Lines 372-373: When reporting the results, please provide more information as it is unclear exactly what you are measuring here i.e. what does the 2.88% refer to? I see this is more clear in the table but still needs to be written out in full in the text.

Line 390: Are you referring to sleep quality here, because as it stands I initially interpreted it as sleep duration (in which case decreasing sleep duration would not be a positive outcome!) Given that the sentence opens with "...reported significantly higher levels of..." this still refers to the sleep outcome (which decreased). This is very confusing. To aid understanding, I suggest including the sleep quality measure as its own sentence i.e. Sleep quality improved from baseline to a greater extent in the intervention versus control group. And I would update the table reporting sleep outcomes to indicate the a lower score indicates a better outcome.

Discussion: I think you need to discuss the generalisability of findings with respect to ethnicity - given that over 95% of participants were Caucasian, the sample isn't representative of the Australian population - and future work would need to investigate whether findings can be replicated in other groups - formative work may be needed to establish how the program could be tailored appropriately.

Reviewer #2: Colin Greaves review: Accept with Minor Revision

Overall, this is a very clearly written and high quality report of an important pilot study around the adaptation /piloting of the highly successful FFIT intervention for implementation in an Aussie Rules context in Western Australia. I have only some minor, mainly presentational comments (and a more substantive comment about health economics, which is made based on exposure to health economic evaluations in several full-scale trials, although I am not an expert in this area)

Abstract: Could mention 3-6 mth results (see below comments on presentation of these). One decimal place might be sufficiently precise?

BACKGROUND:

"Men need access to plentiful resources to maintain their behaviour long term, including both physical (access to healthy food, exercise 108 equipment) and psychological 109 (energy levels, positive mood) resources to facilitate health behaviour change maintenance 110 (28,29)." ... and social resources?

METHODS

Sample size:

"The study sample size followed guidelines for pilot RCTs (30,31) and approximated the FFIT pilot study sample size (32). The Aussie-FIT pilot study was designed to inform the development of a definitive future trial.

- Given the additional /non-standard pilot study aim to 'assess preliminary efficacy', you should present a formal sample size calculation (based on detecting a difference in weight loss). It is not sufficient to simply refer to other pilot studies as a justification for the sample size selected. Even if the aims were only to estimate retention rates, a sample size calculation can be presented (e.g. calculate N needed to estimate retention rates with CIs of +/- 8%)

Maybe say something about the background experience /qualifications of the facilitators? (in Discussion, how does this compare with FFIT and how might this have affected performance /interactions with participants?)

Intervention: what content /contact (if any) beyond 12 weeks? important to know given the emphasis on maintenance here.

"We measured health-related quality of life using the EuroQol five-dimensional five level version (EQ-5D-5L, range 1-5), with lower scores indicative of higher health-related quality of life."

- How was a single /overall QoL score computed?

Analysis:

How were missing data imputed for the ITT analysis? e.g. baseline carried forward, LOCF?

I am a bit confused by the logic behind the secondary analysis which seems to combine "maintenance effects for the intervention group and treatment effects for the control group". So, by combining, it measures neither of these things? It might be better to present the 3-6 month within-group changes separately? Not as a comparison - just mean differences with 95%CIs?

I also don't follow the following "We modelled the linear effect of time (3 to 6 months) as a predictor of dependent variables, adjusted for baseline values and the clustering effect of group. In so doing, the effect can be interpreted as the average amount of change on the dependent variable between 3 and 6 months." I'm not a statistician, but I cant see how that interpretation arises. It might be better just to lose this analysis (as above)?

RESULTS:

Baseline steps and MVPA are high? It is a bit hard to say anything definitive about MVPA as unbouted MVPA is usually 5-10 times higher than MVPA in ten-minute bouts, depending on the sample. Steps do look high though, suggesting there may be some sample bias here - so it may be worth picking this up in the limitations - and say how you might seek to ameliorate this in the full-scale trial?

3 month outcomes: Pease make clear that the analyses presented in the first and second paragraphs (3.33 kg (95% CI 1.89-4.77) etc) are the ITT (all participants with imputed values) figures. You could also summarise the per protocol data here?

Is "overall health the EQ-5D-5L score (and if so, which one - the thermometer or other summary score)?

3-6 month outcomes: As above, the rationale for this analysis does not make sense - it combines initial weight loss in the control group with post-intervention maintenance in the intervention group and uses a rather odd (to me) statistical approach (estimating 3-6 month change for the combined groups, whilst controlling for baseline values). Why not simply report the 3-6 month changes within each group (mean change = ***kg, 95%CI: *** to ***). The best estimate of 3-6 maintenance would be the within intervention group change from 3 to 6 months, where weight appears to further decrease by around 0.6kg (Table 2). Also need to report this figure with and without imputation of missing values.

Cost-effectiveness: The calculations here seem to be more back-of-the-envelope than based on professional health economic analytic methods? I am not a health economist, but I do know that a) you have to include health-care usage and other societal costs (depending on the perspective taken) in the "cost" side of the equation (for both groups) and b) you don't just subtract utility values and divide the cost by this difference - there is a statistical analysis of individual level data involved which yields 95%CIs around the estimate. NB: I wouldn't expect to see a good ICER for within-trial analysis of EQ-5D data in a weight loss trial. The sensitivity of EQ-5D is too low and the main QoL benefits (not getting diabetes, avoiding CV events, cancer etc) that will actually influence EQ-5D tend to accrue in the longer term. A decision-analytic modelling approach would therefore be more appropriate for the main trial.

Furthermore, it doesn't make sense to try to construct a full-blown health economic analysis based on only 110 participants and a 3-month timeframe. I would remove this health economic analysis and just report intervention costs, any between group difference in EQ-5D at 3 months (as part of Table 2) and the measures-completion rate - which indicates whether or not a health ec. analysis would be feasible for the main trial.

Similarly, I would remove the probabilistic sensitivity analysis - the data is much too thin (low N) to bear this type of analysis?

DISCUSSION

This is quiet long /rambling. It could benefit form some sub-headings to structure the text: Summary of findings; Relationship to other literature, Strengths and Limitations; Future research directions; Conclusion

I would cut /substantially edit the health economics text (as per comments above)

TABLES

In general, please consider the appropriate number of significant figures /decimal places. Are your estimates really so accurate as to justify reporting of age, weight etc to 2 decimal places?

Table 2: Are the means reported at 3 and 6 months imputed means (for all 64/66 participants) or per protocol means (for responders only) - please indicate

Overall, this is a very worthy paper and a pleasure to read. Good luck with the revisions

Reviewer #3: Alex McConnachie, Statistical Review

Kwasnicka and colleagues present a report of a pilot RCT of a weight loss and healthy lifestyle intervention delivered in Australian Football League clubs. This review considers the statistical aspects of the paper.

I note that the analysis is described as intention to treat, including everyone who provided baseline data. It is stated that ITT analysis "was used to deal with missing data". In my opinion, ITT analysis has nothing to do with missing data. As the name suggests, I would say that ITT is about which intervention it was intended to deliver. In other words, ITT means to analyse according to randomised group, regardless of whether the intervention was received or not. Missing data is a separate issue. Given the description of the statistical methods, I assume that the analysis involved analysing data from all time points, including baseline, in a mixed effects regression model. This accounts for missing follow-up data to some extent, and is a perfectly reasonable way to carry out the analysis. My only concern is the description of this being "an ITT analysis"; I would separate the methods used to deal with missing data from the concept of ITT.

The 5 items of the EQ-5D questionnaire appear to have been averaged to give a score between 1 and 5. This is not usual; the responses to the 5 questions should be mapped to a health utility score at each time point. It is these health utilities that are used to derive QALYs.

Some of the data has been reported over-precisely. E.g. from line 355 onwards, we have summaries of a number of measures (weight, BP, MVPA) reported to 2 decimal places.

Perhaps a little too much emphasis is put onto the statistical significance (or lack of) for between-group comparisons of some of the outcomes. More emphasis should be put on the point estimates of intervention effects (and their CIs). As the authors note at one point, in a pilot trial, the power will be limited to detect important differences between groups. It appears that the majority of measures have point estimates for the between-group difference at 3 months that are in favour of the intervention group. Some of the estimates (e.g. waist circumference, estimated mean difference -2.66cm) are quite sizeable, so whilst they are not statistically significant, I would say that these estimates add to the overall picture of clinically important intervention effects.

Similar comments apply to the changes between 3 and 6 months - a point estimate of an additional 0.98kg weight loss after the end of the intervention is more than just a lack of weight regain - to me, this suggests that the 3-month intervention may continue to have beneficial effects after the end of intervention delivery. This is what would be hoped of a lifestyle intervention, and though the change was not statistically significant in this pilot trial, it is promising, and lends support to the plan to do a larger trial.

This may be a matter for the journal editors, but I prefer to read "0.xx", rather than ".xx".

Any attachments provided with reviews can be seen via the following link:

[LINK]

16 Mar 2020

Submitted filename: R1 response_Plos Medicine_AussieFIT_16.03.2020_FINAL.docx

Click here for additional data file.

16 Apr 2020

Dear Dr. Quested,

Thank you very much for re-submitting your manuscript "A gender-sensitised weight loss and healthy living program for overweight and obese men in Australian Football League settings (Aussie-FIT): A pilot randomised controlled trial." (PMEDICINE-D-19-03524R1) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by xxx reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

Our publications team (plosmedicine@plos.org) will be in touch shortly about the production requirements for your paper, and the link and deadline for resubmission. DO NOT RESUBMIT BEFORE YOU'VE RECEIVED THE PRODUCTION REQUIREMENTS.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Apr 23 2020 11:59PM.

Sincerely,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

Abstract

Last section of the methods and findings section must highlight a limitation of your study methodology

Please include dates of recruitment and intervention in the abstract

Data availability statement- please remove “option C applies”.

CONSORT checklist-please remove page numbers and use sections and paragraphs. The page numbers are likely to change during the publication process.

Some of the references towards the end of the introduction are in rounded brackets. Please change to square brackets

Please use Vancouver style for references

Please address the comments from the statistical reviewer regarding power analysis and EQ-5D. This involves removing the post-hoc power analysis and also remove the summary score for EQ-5D. We will be unable to accept your manuscript without this.

Please change all p-values to 0.xx instead of .xx

We would be grateful if you could provide a clean version of your manuscript and one containing track changes as a companion file.

Suggest rephrasing “meaningful” in the abstract and the discussion as it is too vague for a research article

Line 581 in discussion, please rephrase “footy fandom”

Some of the information in the Acknowledgements regarding grants should be provided in the funding disclosure

Comments from Reviewers:

Reviewer #2: I am happy with the responses provided. No further comments.

Reviewer #3: Alex McConnachie, Statistical Review

I would like to thank the authors for considering my earlier points. However, I feel that the changes they have made have not all been for the better.

In response to another reviewer, a sample size calculation has been added. Previously, as in the protocol paper, the sample size was said to be based on guidelines for pilot trials. This was fine, and was appropriately referenced. The new, post-hoc sample size calculation is not required, and is incorrect. To have 90% power, at 5% significance, to detect a difference of 4.71, assuming a SD of 3.95, requires 2 x 10.51 x (3.95)^2 / (4.71)^2 = 15 per group. If 12 per cluster are followed up (80% of 15), and the ICC is 0.05, the design effect is 1 + 0.05 x (12-1) = 1.55, increasing the sample size to 24 per group. Dividing by 0.8, to allow for loss to follow-up, gives a total of 30 per group randomised.

In my opinion, the sample size justification should be reported in the way the study was designed. If this was based on published guidelines for studies of this type, so be it. A post-hoc power calculation is not required, and is generally not recommended.

One area where I agree with the other reviewer is the number of decimal places used to summarise many of the results. The authors state that they have followed the journal style, though I could find no reference to the number of decimal places to use on the journal website, and "APA guidelines". I have looked into the American Psychological Association guidelines, and was shocked at how bad some of them are. To report two decimal places for everything, regardless of context, is simply wrong. Summaries for age, weight, waist circumference, blood pressure, minutes of activity, and EQ-5D VAS, should all be given to one decimal place. Given that the denominators are less than 100, percentages can be presented as whole numbers, or to one decimal place at most. The cost data could also be reported to the nearest AUS$.

Another area where the APA has given some rather odd advice is on the use of leading zeros. The authors stated in their response to my original comments, that this has been changed throughout, but it has not. The APA guildeline is to omit the leading zero if the quantity cannot exceed 1 (or -1). In my opinion, based on over 30 years studying and practising statistics, this is not good advice. Aesthetically, numbers between +/- 1 without the leading zero just look bad.

I still have reservations about the presentation of the EQ-5D data. In my experience, it is unusual to average the five items to produce a summary score for each individual. It is far more common to report the derived health utility score at each time point.

Minor point: in the section "What do these finding mean?", under the second point, I would add that studies with longer follow-up are needed.

Another minor point (which may be ignored) is that I would still say that the analysis was done according to the Intention To Treat principle, because individuals were (I assume) included in the analysis according to their randomised group allocation, regardless of their participation in the intervention. My earlier comment was to avoid connecting ITT with the issue of missing data. The new text to describe the analysis in this respect is good.

Any attachments provided with reviews can be seen via the following link:

[LINK]

16 Jun 2020

Dear Dr. Quested,

On behalf of my colleagues and the academic editor, Dr. Colin J Greaves, I am delighted to inform you that your manuscript entitled "A gender-sensitised weight loss and healthy living program for men with overweight and obesity in Australian Football League settings (Aussie-FIT): A pilot randomised controlled trial." (PMEDICINE-D-19-03524R2) has been accepted for publication in PLOS Medicine.

PRODUCTION PROCESS

Before publication you will see the copyedited word document (in around 1-2 weeks from now) and a PDF galley proof shortly after that. The copyeditor will be in touch shortly before sending you the copyedited Word document. We will make some revisions at the copyediting stage to conform to our general style, and for clarification. When you receive this version you should check and revise it very carefully, including figures, tables, references, and supporting information, because corrections at the next stage (proofs) will be strictly limited to (1) errors in author names or affiliations, (2) errors of scientific fact that would cause misunderstandings to readers, and (3) printer's (introduced) errors.

If you are likely to be away when either this document or the proof is sent, please ensure we have contact information of a second person, as we will need you to respond quickly at each point.

PRESS

A selection of our articles each week are press released by the journal. You will be contacted nearer the time if we are press releasing your article in order to approve the content and check the contact information for journalists is correct. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact.

PROFILE INFORMATION

Now that your manuscript has been accepted, please log into EM and update your profile. Go to https://www.editorialmanager.com/pmedicine, log in, and click on the "Update My Information" link at the top of the page. Please update your user information to ensure an efficient production and billing process.

Thank you again for submitting the manuscript to PLOS Medicine. We look forward to publishing it.

Best wishes,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org