This study used a 2-arm, randomized pre-post intervention design. The study protocol was reported in accordance with the CONSORT (Consolidated Standards of Reporting Trials) 2010 statement: extension to randomized pilot and feasibility trials, which provides a framework for transparent reporting of small-scale randomized studies with exploratory outcomes [33] (Checklist 1).

People accessing CEP services across various sites in South Australia were recruited for this study. Recruitment sites included the Queen Elizabeth Hospital, Hampstead Rehabilitation Centre, GP Plus Healthcare Centre Elizabeth, the University of South Australia Exercise Physiology Clinic, iNform Health and Fitness Solutions, and Fleurieu Exercise Physiology. A target sample size of 40 (20 per arm) was deemed sufficient based on results from previous studies using similar interventions [21,34]. Eligibility criteria included being a newly referred or current CEP service client and proficiency in reading and writing English. People without access to an internet-enabled device, or with insufficient IT skills or cognitive function to operate a web-based questionnaire, were not eligible to participate.

Convenience sampling with active recruitment was used to recruit participants [35]. The study was promoted to potential participants through face-to-face interaction with researchers and CEP staff at participating sites, as well as via email. Individuals who expressed interest were contacted by the research team and screened for eligibility. Those who met the eligibility criteria were provided with a Participant Information Sheet and asked to provide informed written consent. They were then asked to complete a brief questionnaire, either in hard copy or online via REDCap (Vanderbilt University), capturing demographic (age and sex) information and CEP service client status (current or newly referred).

The FESS questionnaire was developed by members of the research team (MM and NH), both CEPs, and was informed by over 30 years of combined clinical experience as well as existing literature on lifestyle-related health behaviors [1-4] and behavior change principles [3,17-24]. Items were generated to reflect key modifiable lifestyle domains addressed in CEP practice, including exercise as a key domain, alongside related areas, such as diet, stress management, and sleep. These additional domains, while not a key focus, are often considered in practice because they influence clients’ ability to engage in physical activity and achieve overall health outcomes.

The questionnaire was designed as a pragmatic clinical tool to support behavioral reflection and tailored feedback within the intervention, rather than as a psychometrically validated, domain-specific instrument. The questionnaire comprises 23 items. Items 1 to 8 of the questionnaire address food-related behaviors, items 9 to 12 focus on exercise, items 13 to 18 focus on stress, and items 19 to 23 focus on sleep. The response options are predominantly categorical (22 items), with only 1 item using a numerical response format. Each response option is assigned a numerical score; these are then summed to produce a total score out of 100, with higher scores reflecting healthier behaviors. As respondents progress through the questionnaire, they are presented with health advice and tips specific to each item and its associated health domain. For example, within the exercise domain, respondents are provided with guidance such as: “The amount, type, and intensity of your exercise, along with your sedentary behaviors, are strong predictors of health. A combination of strength and aerobic exercise, together with breaking up prolonged sitting, are key behaviors that support health.” Within the stress domain, advice includes messages such as: “It’s not only the amount of stress you are under but how you react to it which influences your health. Developing behaviors to improve resilience have been shown to influence longevity. They can also impact our motivation to exercise as well as influence our food choices.”

The intervention was conducted between August and December 2022. Participants were randomly allocated to 2 groups using block randomization to ensure balanced distribution by age, sex, and whether they were a newly referred or current CEP service client. One group was required to complete the FESS questionnaire 3 times per week (3-day group), while the other group completed the FESS questionnaire daily (Everyday group). Both groups completed the questionnaire for a period of 12 weeks. Participants were prompted to complete the FESS questionnaire via automated email reminders administered through Keap, a customer relationship management platform. Participants in the 3-day group received the email 3 times per week—twice on weekdays and once on a weekend—while participants in the everyday group received an email each day of the week. Participants were trained by CEP staff at participating sites on how to access and complete the questionnaire. Upon completing the questionnaire, participants received their score out of 100. These results could then be discussed with CEP staff as part of the participant’s ongoing support. No additional or study-specific training was provided to CEPs involved in the study regarding diet, sleep, or stress management beyond their standard professional training and scope of practice. CEPs delivered usual care consistent with standard CEP services, primarily focused on exercise assessment, prescription, and delivery. No additional engagement, counseling, or behavioral support related to diet, sleep, or stress was provided outside of routine practice, and any discussions were limited to general advice consistent with relevant Australian guidelines and within CEPs’ scope of practice.

Participants in both groups were required to complete an assessment version of the FESS questionnaire, which was administered via SurveyGizmo (Alchemer, Inc) at 3 time points: prior to commencing the intervention (baseline) and at 6 and 12 weeks after the intervention began. The assessment FESS questionnaire was identical to the intervention version, except that it excluded health advice and tips. At each assessment point, participants were asked to complete the assessment questionnaire 3 times within a week.

In addition, participants in the everyday group were invited to wear a GENEActiv triaxial accelerometer (ActivInsights) for 7 consecutive days; participation in this component of the study was voluntary. To minimize the likelihood of user error, participants were provided with at least a 1-week acclimatization period with FESS prior to being issued the accelerometers. A total of 6 GENEActiv accelerometers were used in the study and deployed to participants across 3 rollout rounds. Accelerometers were calibrated to a sampling frequency of 100 Hz for the 7-day data collection period [36]. Devices were programmed to commence recording on the morning of delivery to each participant. Following delivery, participants were encouraged to wear the accelerometer for at least 5 consecutive days, including a weekend day, to optimally capture physical activity. Accelerometer data were then used to calculate total physical activity expressed as counts per minute (CPM).

The study protocol was approved by the Central Adelaide Local Health Network Human Research Ethics Committee (ID: 16356) and the University of South Australia’s Human Research Ethics Committee (ID: 204757), and all participants provided written informed consent prior to participation.

Forty-one participants (23 female) with a median age of 65 (IQR 58‐70) years were assessed for eligibility; all were enrolled in the study, provided consent, and completed the FESS assessment at baseline, with none excluded. Table 1 shows the demographic characteristics of participants by group. Of these, 21 were randomized to the 3-day group and 20 to the everyday group. Not all participants completed the assessment FESS questionnaire at each assessment time point; participant flow is shown in Figure 1.

This study examined changes in health behaviors and compliance across all domains of the FESS questionnaire and assessed the validity of the exercise domain specifically, as an adjunct to services provided by CEPs. The preliminary findings suggest that the FESS questionnaire may be a useful tool for supporting positive changes in health behaviors across several health behavior domains. Differences between groups suggest that the daily use of the questionnaire may encourage stronger behavior change; however, compliance was higher for participants who completed it 3 times per week. Accelerometer data further supported the utility of the tool, with low rates of overreporting exercise and underreporting SB. Taken together, these outcomes highlight the potential of the FESS questionnaire as a low-cost and convenient digital platform that warrants further investigation for the promotion of healthy lifestyle behaviors.

The positive behavior changes observed in the study align with existing evidence that digital interventions can support healthier lifestyle choices [21-24]. In particular, the significant improvements observed in food, stress, and sleep-related behaviors highlight that the digital tool may have supported positive change in these domains, even when delivered alongside indirectly related service (ie, exercise physiology services). Given that most digital health promotion strategies have been found to focus on only one domain of health, typically physical activity, nutrition, or sexual health [41,42], these findings are particularly noteworthy. They demonstrate that a digital behavior change tool addressing multiple domains of health, including those less commonly targeted such as sleep and stress, may be able to be successfully integrated as an adjunct to exercise-focused services, rather than being limited to exercise alone. The reasons underlying these observed outcomes may be due to the design of the intervention itself. The FESS questionnaire’s capacity to prompt reflection and offer practical tips, combined with the opportunity to review results with a CEP, reflects features consistent with established aspects of successful behavior change interventions [32] and may have reinforced positive behaviors and reassured participants that they were meeting healthy behavior guidelines. This reassurance may have alleviated stress, which in turn, could have contributed to improvements in sleep.

Statistically significant changes were observed in a limited number of items (6): serves of vegetables, liters of water, level of stress, effect of stress on health, exposure to sunlight, and hours of sleep. This may reflect the already positive levels of healthy behaviors reported at baseline: most participants consumed a balanced diet, drank alcohol in moderation, did not smoke, and maintained adequate sleep and sleep hygiene practices [43-45]. It is notable, however, that despite the FESS questionnaire being delivered in conjunction with services provided by CEPs, exercise was the one domain where there were no significant improvements. There may be several explanations for this result. First, similar to the other FESS questionnaire domains, the level of exercise undertaken by participants reported at baseline was already high. While direct comparisons between these exercise data and physical activity levels in the general population are complicated by methodological differences, the levels of exercise reported by participants suggest that they would fall within the upper quartile of exercise engagement compared to the general Australian adult population [46]. Second, participants in this study were either a newly referred or existing CEP service client, and therefore, were likely to be on a prescribed exercise program. Taken together, these findings suggest a ceiling effect for healthy behaviors that may have limited the potential for further measurable improvements in these participants.

Another key finding of this study was the contrast between participant compliance and outcomes across the 2 groups. Consistent with previous research [47-49], compliance was lower among those asked to access the FESS questionnaire daily compared to those who accessed it 3 days per week, and it declined in both groups over time. Despite this, there was evidence that more frequent access to the questionnaire was associated with greater positive changes in health behaviors (although it should be noted that only 3 items on the questionnaire reached statistical significance). This aligns with findings from a systematic review and meta-analysis on attrition and adherence in smartphone-delivered interventions for mental health, which examined 10 trials that assessed whether more frequent intervention use was linked to better health outcomes and found mixed results [47]. Taken together, the findings of this study and previous research suggest that while more frequent engagement may have greater potential to promote behavior change, it may be difficult to sustain and result in only moderate additional benefits. Allowing flexibility in frequency of use in this and other digital platforms may therefore be important for optimizing long-term engagement and, hence, behavior change. However, further research is needed to determine the optimal frequency of exposure to digital behavior change platforms.

Finally, we found that the accelerometer data support the validity of the exercise domain of the FESS questionnaire. The relatively low rates of overreporting equal to, or greater than, MVE provide an important first step in validating the tool, given the well-documented inaccuracies of self-report questionnaires, particularly for physical activity [50-52]. In contrast, self-reported SB, consistent with previous research [53], appeared more error-prone, with a higher proportion of participants underreporting sedentary time. It is important to acknowledge, however, that the validation process had limitations: comparisons were limited to 2 items within the exercise domain; the validation involved a small number of participants (n=11); and the accelerometer-derived metrics and the questionnaire responses do not capture the same constructs. These factors may have contributed to discrepancies between the 2 measures.

Additionally, this study has several other limitations. Most notably, the absence of a control group limits the ability to determine whether the observed changes are attributable to the FESS questionnaire itself or simply to routine interactions with a CEP. Convenience sampling may have introduced selection bias [54]. Participants were recruited from CEP clinics and were likely to be more motivated and health conscious than the general population. Eligibility criteria included both newly referred and existing CEP clients, who may differ in motivation, readiness for behavior change, and engagement with services. However, comparative analyses between these 2 subgroups were not performed, and they were treated as a single sample in this study. As such, potential differences in responsiveness to the intervention between newly referred and usual clients cannot be determined. Furthermore, in addition to the potential reporting bias inherent in self-report questionnaires, as highlighted above, participants were encouraged to discuss their behaviors with their CEP, potentially introducing observer bias [55]. Importantly, a similar FESS questionnaire was used both as an intervention tool and as the outcome assessment measure. Repeat exposure to the instrument may have influenced participants’ responses over time, with observed changes potentially partially reflecting increased awareness rather than actual behavior change. However, because the questions have good face validity, being simple reports of easily recalled behaviors, it is unlikely that repeat exposure would change participant reporting. Finally, the intervention of 12 weeks, while adequate for measuring short-term behavior change, is not sufficient to provide insights into the longer-term maintenance of behaviors.

As a proof-of-concept study, these preliminary, hypothesis-generating findings indicate that it is worthwhile to address these limitations by incorporating a control group within a larger, adequately powered randomized controlled trial; recruiting more diverse samples through a multifaceted sampling strategy; conducting subgroup analyses to examine potential differences between newly referred and existing CEP clients; extending follow-up to assess long-term behavior maintenance; and validating the tool across all domains following best practice guidelines [56]. In addition, cost-effectiveness analyses to determine the scalability and economic value of integrating the FESS questionnaire into standard care practices may be warranted.

This study observed positive changes in multiple health behaviors during the implementation of the FESS web-based questionnaire alongside exercise-focused services. These preliminary, hypothesis-generating findings suggest that CEP services may benefit from embedding structured digital tools to complement traditional face-to-face interactions. By providing education, promoting reflection and accountability between sessions, and addressing a broader range of lifestyle factors, such tools may enhance the care that CEPs deliver. However, larger controlled trials are required to confirm these findings and determine the effectiveness of this approach.