Recruiting patients into trials in general practice: insights from the ENERGISED trial

Recruiting patients into trials in general practice: insights from the ENERGISED trial | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Recruiting patients into trials in general practice: insights from the ENERGISED trial Norbert Kral, Tomas Vetrovsky, Marketa Pfeiferova, Bohumil Seifert, and 17 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7775668/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 03 Mar, 2026 Read the published version in BMC Primary Care → Version 1 posted 10 You are reading this latest preprint version Abstract Background Recruiting patients into randomised controlled trials in general practice is challenging and carries a substantial risk of bias. The ENERGISED trial evaluated an mHealth physical activity intervention in patients with prediabetes or type 2 diabetes recruited through general practice. To minimise bias, the trial employed a systematic recruitment strategy in which general practitioners assessed the eligibility of patients from random stratified samples of their registers and sought consent from all those deemed eligible. This study aimed to analyse the recruitment process of the ENERGISED trial and identify sources of potential bias arising from general practitioners' eligibility assessments (selection bias) and patient consent (self-selection bias). Methods Patients with prediabetes or type 2 diabetes were randomly sampled from the registers of 28 Czech general practices using sex- and diagnosis-stratified lists. Eligibility was systematically assessed during routine visits, with general practitioners documenting reasons for ineligibility. All eligible patients were invited to participate, and reasons for non-consent were recorded. Logistic mixed-effects models were used to examine the influence of patient characteristics (age, sex, diagnosis) and general practitioner characteristics on eligibility and consent. Results Of 1,376 sampled patients, 1,138 (83%) were assessed, 792 (70% of assessed) were eligible, 348 (44% of eligible) consented and 343 were randomised. Older age was associated with lower odds of eligibility (OR 0.955, 95% CI 0.942–0.968; p < 0.001) and lower odds of consent among eligible patients (OR 0.972, 95% CI 0.958–0.986; p < 0.001). Ineligibility was most often due to digital barriers. Practices with older registered populations showed stronger age-related bias. Female practitioners and practices with more diabetes/prediabetes patients achieved significantly higher eligibility rates. Conclusions Systematic recruitment through general practice can reduce selection and self-selection bias, yet digital exclusion, particularly in older adults, persists. Future trials must proactively address digital literacy and age-related barriers to ensure representative participation in primary care research. Clinical trial recruitment Primary care Type 2 diabetes Physical activity Selection bias Digital exclusion Figures Figure 1 Figure 2 Background Recruiting patients into randomised controlled trials conducted in general practice remains a persistent challenge. Despite the increasing prevalence of chronic metabolic disorders such as type 2 diabetes and prediabetes, and their routine management in primary care, clinical trials often struggle to enrol representative patient populations. Recruitment may be hindered by a combination of practice-level workload constraints, trial-specific eligibility criteria, and patient-level barriers, including digital literacy, health beliefs, and functional limitations. These factors can introduce selection and self-selection bias, threatening both the internal and external validity of trial findings [1,2,3]. The ENERGISED trial evaluated a pragmatic behavioural intervention to increase physical activity among adults with prediabetes or uncomplicated type 2 diabetes, recruited through general practice. The intervention incorporated self-monitoring with a wrist-worn Fitbit tracker, phone counselling, and a digital support platform using just-in-time adaptive text messaging (JITAI), integrated into routine primary care across multiple sites in the Czech Republic. Comparable trials have shown that digital interventions for people with type 2 diabetes and prediabetes require intensive and flexible recruitment strategies to meet targets and maintain equity [4,5,6]. For example, previous studies of physical activity interventions in primary care identified a selection bias when general practitioners (GPs) preferentially pick those patients whom they believed to be able to use (e.g., highly educated patients) and to benefit from the intervention (e.g., patients with overweight and obesity) [7, 8]. To tackle this, the ENERGISED trial implemented a systematic recruitment procedure that limited GPs' discretion by using randomised, stratified patient lists generated from electronic health records. This approach aimed to minimise selection bias and ensure a more representative sample across age, sex, and diagnosis groups. Self-selection bias may occur when patients decline participation due to personal reasons—such as time constraints, discomfort with digital tools, or low perceived benefit—potentially limiting the representativeness of the sample [9]. To evaluate how this recruitment strategy worked in practice and to examine potential sources of selection and self-selection bias, we analysed recruitment data from the ENERGISED trial. Specifically, this study quantified recruitment flow from initial patient identification to randomisation, explored the influence of patient and GP characteristics on eligibility and consent, and investigated patterns of exclusion. By disentangling the drivers of recruitment attrition in a real-world setting, our findings provide practical and methodological insights to inform the design of future primary care trials—particularly those evaluating behavioural interventions involving digital tools [10, 11]. Objectives The aim of this study was to examine the recruitment process within the ENERGISED randomised controlled trial. The specific objectives were to: Describe the recruitment flow, including patient characteristics at each stage and reasons for ineligibility (potential selection bias) and non-consent (potential self-selection bias); Investigate selection and self-selection bias at the GP and patient level, respectively; and Investigate the relationship between GP characteristics, eligibility rates, and potential selection bias. Methods This study was conducted within the ENERGISED randomised controlled trial (ClinicalTrials.gov identifier NCT05351359), a 12-month pragmatic, multicentre trial evaluating an mHealth-enhanced physical activity intervention for adults with prediabetes or uncomplicated type 2 diabetes recruited through Czech general practices. The protocol was approved by the Ethics Committee of the General University Hospital in Prague (reference number: 49/20). Full details of the trial design, procedures, and statistical analysis plan have been published elsewhere [12,13]. Between April 2022 and April 2024, 343 patients were recruited through 28 participating GP practices. The present paper focuses exclusively on the recruitment process, including patient flow, eligibility assessments, and consent. Recruitment of general practices General practices were recruited through national GP conferences, a professional journal, direct e-mail invitations, and personal contacts, as described previously [12]. The original protocol envisaged participation of 21 practices, but this was later expanded to 28 to accelerate recruitment and ensure broader representation. The participating practices included 15 from urban areas and 13 from rural towns with fewer than 30,000 inhabitants, covering 9 of the 14 administrative regions of the Czech Republic. Participating GPs received remote training on study procedures and data entry and were compensated for their time (approximately €100 per patient completing the study). Patient eligibility Eligibility criteria were identical to those specified in the published ENERGISED protocol [12]. Patients were eligible if they met all of the following: (1) diagnosis of prediabetes or type 2 diabetes according to Czech guidelines for GPs (fasting plasma glucose 5.6–6.9 mmol/l or 2-h plasma glucose of 7.8–11.0 mmol/l after ingestion of 75 g of oral glucose load for the diagnosis of prediabetes fasting plasma glucose ≥ 7.0 mmol/l or 2-h plasma glucose ≥ 11.1 mmol/l after ingestion of 75 g of the oral glucose load for the diagnosis of type 2 diabetes); (2) age ≥ 18 years; (3) followed for prediabetes or diabetes by a participating GP (in the Czech Republic, GPs typically manage patients with uncomplicated type 2 diabetes with glycated haemoglobin (HbA1c) ≤ 53 mmol/mol who are not treated with insulin); (4) regular users of a mobile phone (not necessarily a smartphone), able and willing to answer calls and read text messages as part of the study; (5) able and willing to wear and use a wrist-worn Fitbit activity tracker for the study duration; and (6) provided written informed consent. As the Fitbit required a smartphone for initialisation, patients without their own smartphones were advised to ask a relative or friend to perform the setup. Exclusion criteria were: (1) unable to walk independently for any reason; (2) pregnant; (3) having a household member already recruited for thw trial; (4) living in a residential or nursing care home where the imposed regime could interfere with the intervention; or (5) having co-morbid conditions that would seriously affect adherence, including active malignancy; recent (< 3 months) myocardial infarction, coronary artery bypass graft or cerebrovascular accident; renal disease requiring dialysis; neurological condition (e.g., Parkinson disease); cognitive impairment, or significant hearing or visual impairment; hip or knee joint replacement within three months; or major surgery planned within the next 12 months. Because all study materials and procedures were in Czech, patients were excluded if they lacked sufficient Czech language proficiency to participate effectively. Recruitment process Before recruitment began, each GP generated a registry-based list of all patients with prediabetes or type 2 diabetes in their practice. From this list, an initial stratified random batch of 24 patients (sex 1:1, condition prediabetes:diabetes 1:2) was provided. GPs assessed eligibility of all the patients from these batches during their routine health check-ups conducted every 3 to 6 months, documented reasons for ineligibility, invited all eligible patients to participate and recorded reasons for non-consent. When a batch was exhausted, new random batches of 12 patients were issued as needed until the respective practice depleted its original list, reached 32 recruited patients, or the overall trial target was met (planned 340; actually 343 recruited). Patients who consented underwent baseline procedures, including seven days of wrist-worn accelerometry, followed by a second baseline visit where participants received a Fitbit activity tracker. After the second baseline visit, patients were randomly allocated in a 1:1 ratio to either the intervention or the active control arm. Recruitment stages Importantly, the study design required GPs to assess all patients from randomly selected lists, thereby minimising discretionary inclusion and reducing selection bias at the assessment stage. Recruitment losses were systematically tracked and categorised as either GP-determined ineligibility or patient-declared non-consent, allowing detailed insight into the factors contributing to recruitment attrition in a primary care trial. For the purpose of this study, we use the following terms to describe the successive recruitment stages: (1) total population, (2) sampled, (3) assessed, (4) eligible, (5) consented, and (6) randomised patients. The total population includes all patients with prediabetes or type 2 diabetes registered at participating GP practices. A list of these patients was generated by the GPs using computerised medical records before recruitment began. Sampled patients are those randomly selected from the total population, stratified by sex (female: male in a 1:1 ratio) and condition (prediabetes: diabetes in a 1:2 ratio). Each practice initially received a batch of 24 sampled patients. When this batch was exhausted, new random selections of 12 patients were generated and provided as needed. Assessed patients are those among the sampled patients who were assessed for eligibility by their GP. As recruitment was conducted opportunistically during routine health check-ups, patients who did not attend their scheduled visits during the recruitment period could not be assessed. Eligible patients are those assessed and confirmed to meet all eligibility criteria. Consented patients are eligible individuals who provided written informed consent to participate in the trial. Randomised patients are those who consented and were subsequently randomised. As the protocol involved an interval of at least one week between consent and randomisation, not all consenting patients were randomised. This study focuses exclusively on the recruitment process up to the point of randomisation. However, as both GPs' and patients' motivations may have been shaped by their expectations regarding the potential benefits and burdens of the trial procedures, a brief overview of the intervention and control conditions, as well as the trial visits and outcomes, is provided here. Intervention and control conditions Both trial arms involved a wrist-worn activity tracker and brief physical activity advice delivered by GPs during baseline visits, with additional components differing between groups. At the second baseline visit, all participants received a Fitbit Inspire 2 activity tracker and were told they could keep it after completing the study. They were encouraged to monitor their daily step count and gradually increase it by at least 3,000 steps above baseline through intentionally brisk walking. They were also advised to interrupt prolonged sitting every 30 minutes. Patients allocated to the intervention group received an additional mHealth intervention based on JITAI principles and delivered via the HealthReact platform. This platform utilised real-time data from the activity tracker to trigger automated text messages supporting behavioural change, including just-in-time prompts to increase walking pace (triggered after five consecutive minutes of walking) and interrupt prolonged sitting (sent after 30 minutes of inactivity). During the first six months (lead-in phase), these mHealth elements were supported by monthly phone counselling sessions. During the following six months (maintenance phase), the intervention was fully automated. Patients in the active control group received the same Fitbit tracker, physical activity prescription, and brief advice from their GP, but no mHealth or counselling components. Further details of the intervention content and delivery are available in the published protocol [12] and intervention development paper [14]. Trial visits and outcomes Trial procedures for both groups included five GP visits: two at baseline, followed by visits at 3, 6, and 12 months. In the Czech primary care setting, patients with type 2 diabetes typically attend their GPs every three months and those with prediabetes every six months. Therefore, participation required only one additional visit for patients with diabetes and two for those with prediabetes. The primary outcome was average daily step count measured by wrist-worn accelerometry over seven consecutive days. Secondary outcomes included additional accelerometry-derived metrics; anthropometric and functional measures (body mass index, waist circumference, blood pressure, and 30-second sit-to-stand test); blood tests; and patient-reported outcomes. Assessments were conducted at baseline and at 3, 6, and 12 months. Further details are available in the published protocol [12] and statistical analysis plan [13]. Statistical analysi s Descriptive statistics were used to summarise the characteristics of participating GPs and patients at each stage of the recruitment process. Continuous variables are reported as medians with interquartile ranges (IQR) or means with standard deviations (SD), and categorical variables as counts and percentages. To examine potential selection and self-selection bias, logistic mixed-effects models were used, with GP included as a random intercept to account for clustering at the practice level. Four binary outcomes were modelled: (1) consented vs. not consented among all sampled patients; (2) assessed vs. not assessed among sampled patients; (3) eligible vs. not eligible among those assessed; and (4) consented vs. not consented among those eligible. Each model included patient-level predictors: sex, age (calculated at the start of recruitment) and clinical condition (diabetes or prediabetes). In cases where a significant patient-level predictor of eligibility or consent was identified, we further explored its relationship with specific reported reasons for ineligibility or non-consent. For each reason, a separate logistic regression model was fitted, with the presence or absence of that reason as the binary outcome and the significant patient-level predictor as the independent variable. These models were restricted to the relevant subsample (ineligible or non-consenting patients). Odds ratios (ORs) with 95% confidence intervals (CI) and p -values were reported. To examine whether general practitioners’ characteristics are associated with eligibility rates and potential selection bias, a logistic mixed-effects model was constructed among assessed patients, with eligibility (yes/no) as the outcome and GP included as a random effect. GP-level predictors included sex, years since graduation, practice type (solitary vs. associated), practice location (urban vs. non-urban), number of registered patients, and the proportions of registered patients with diabetes or prediabetes, aged over 65 years, and the proportion of those who had undergone a preventive examination in the past two years. Model selection followed a stepwise approach. Starting from a full model including all patient-level and GP-level variables, non-significant fixed effects were removed sequentially, beginning with the predictor with the highest p -value. After obtaining a reduced model, two-way interaction terms between selected patient-level and GP-level variables were added to explore potential moderating effects. All trial data were collected using REDCap electronic data capture tools. Statistical analyses were conducted in R (version 4.4.2) using the lme4 package. Results Characteristics of participating GPs The trial involved 28 GPs from 28 practices, of whom 24 were women. The median number of years since graduation was 17 (IQR: 12 to 24). Thirteen GPs were based in cities with over 100,000 inhabitants, while 15 practised in smaller towns or rural areas. Twenty-one worked in solitary practices, and 7 in group or associated practices. The median number of registered patients per GP was 1,818 (IQR: 1,584 to 2,048). The mean proportion of patients registered for regular follow-up with diabetes or prediabetes was 4.5% (SD: 3.9). The mean proportion of patients aged over 65 years was 24.6% (SD: 6.9). In 2021, the mean proportion of patients who underwent a preventive health check was 25.7% (SD: 10.8). As preventive health checks are scheduled biennially, some patients may have attended in 2020 or subsequently in 2022, and the overall coverage is therefore higher than reflected in the 2021 data. Recruitment flow and patient characteristics Among all 3,579 patients with diabetes or prediabetes registered with the 28 participating GPs, a stratified random sample of 1,376 patients was selected. Of these, 1,138 (82.7%) were assessed for eligibility by their GPs, and 792 (69.6% of those assessed) were deemed eligible. Among the eligible patients, 348 (43.9%) consented to participate, and 343 (43.3%) were ultimately randomised, with 5 patients not showing up for the second baseline visit. The recruitment flow is summarised in Fig. 1 . Table 1 presents the characteristics of patients at each stage of the recruitment process: the total population, sampled patients, those assessed for eligibility, those deemed eligible, and those who consented. Figure 2 shows the number of patients assessed, found eligible, and consented per individual GP. Table 1 Characteristics of patients at each stage of the recruitment process: total population, sampled, assessed, eligible, and consenting patients Stage Sex Condition Count (n, %) Age (mean ± SD) Total Population Male Diabetes 950 (26.5%) 65.5 ± 11.4 Prediabetes 978 (27.3%) 60.0 ± 13.2 Female Diabetes 803 (22.4%) 70.1 ± 11.7 Prediabetes 848 (23.7%) 64.3 ± 13.6 Sampled Male Diabetes 478 (34.7%) 64.1 ± 11.0 Prediabetes 237 (17.2%) 59.2 ± 12.3 Female Diabetes 442 (32.1%) 68.7 ± 11.2 Prediabetes 219 (15.9%) 64.4 ± 13.6 Assessed Male Diabetes 393 (34.5%) 64.2 ± 11.0 Prediabetes 190 (16.7%) 59.1 ± 12.6 Female Diabetes 379 (33.3%) 68.5 ± 11.4 Prediabetes 176 (15.5%) 65.0 ± 13.8 Eligible Male Diabetes 272 (34.3%) 63.0 ± 10.4 Prediabetes 145 (18.3%) 57.0 ± 12.1 Female Diabetes 256 (32.3%) 66.7 ± 11.2 Prediabetes 119 (15.0%) 63.0 ± 12.7 Consented Male Diabetes 127 (36.5%) 62.7 ± 10.7 Prediabetes 62 (17.8%) 54.6 ± 12.1 Female Diabetes 109 (31.3%) 64.4 ± 10.7 Prediabetes 50 (14.4%) 60.6 ± 11.7 Selection and self-selection bias at the patient level To assess potential selection and self-selection bias, we first compared patients who consented to participate with those in the randomly sampled population. Increasing age was associated with lower odds of progressing through the stages of assessment, eligibility, and consent (OR = 0.965 per year, 95% CI: 0.954–0.976, p < 0.001). In contrast, there were no significant differences in sex or clinical condition between consenting and sampled patients. We then examined each stage of the recruitment process separately. No significant differences were found between assessed and sampled patients, suggesting no bias in the assessment stage. However, both the eligibility stage (selection bias) and the consent stage (self-selection bias) showed the same age-related pattern: older patients were less likely to be deemed eligible (OR = 0.955 per year, 95% CI: 0.942–0.968, p < 0.001) and less likely to consent (OR = 0.972 per year, 95% CI: 0.958–0.986, p < 0.001). Again, no significant differences were observed for sex or clinical condition at these stages. Reasons for ineligibility and non-consent The reasons for ineligibility are summarised in Table 2 . The most common reasons for ineligibility were an anticipated inability to use the wrist-worn Fitbit activity tracker throughout the study period (n = 126) and not being a regular user of a mobile phone or being unable to read text messages (n = 101). The age-related bias observed in the overall eligibility analysis was consistent across most specific reasons for ineligibility, with the only common reason not related to increasing age being registration with a diabetologist (Table 2 ). Table 2 Reported reasons for ineligibility and their association with patient age Reason for non-eligibility Count OR 95% CI p-value Unable to use Fitbit 126 1.11 1.08–1.14 0.0000 Not using a phone 101 1.12 1.08–1.16 0.0000 Registered with a diabetologist 60 1.00 0.97–1.02 0.7436 Comorbidities 52 1.05 1.02–1.09 0.0006 Not registered with the practitioner 50 1.05 1.02–1.08 0.0018 Unable to walk 40 1.14 1.09–1.19 0.0000 Institutionalised 16 1.06 1.00–1.12 0.0510 On insulin 13 1.01 0.95–1.08 0.6890 Not having pre/diabetes 12 0.97 0.93–1.01 0.1655 Household member already enrolled 10 1.06 0.99–1.14 0.0878 Not speaking Czech 9 0.97 0.91–1.04 0.3995 The reasons for non-consent are summarised in Table 3 . Among patients who were eligible but did not consent, the most frequently cited reasons were a lack of interest in a physical activity intervention (n = 172) and insufficient time (n = 158). Notably, the odds of non-consent due to lack of interest increased with age (OR = 1.02, 95% CI: 1.002–1.041, p = 0.03), whereas the odds of non-consent due to lack of time decreased with age (OR = 0.971, 95% CI: 0.954–0.99, p = 0.002). Table 3 Reported reasons for non-consent and their association with patient age Reason for non-consent Count OR 95% CI p-value Lack of interest 172 1.02 1.00–1.04 0.0303 Lack of time 158 0.97 0.95–0.99 0.0023 Not willing to use Fitbit 83 1.00 0.98–1.03 0.8404 Other 77 1.03 1.00–1.05 0.0324 Not willing to receive calls 51 1.00 0.97–1.03 0.8968 Privacy reasons 48 1.00 0.97–1.03 0.9205 Not willing to get messages 46 1.02 0.99–1.06 0.1517 Already wearing a tracker 24 0.96 0.93–0.99 0.0175 Influence of GP characteristics on eligibility rates and selection bias We next examined whether GP-level characteristics were associated with eligibility rates and whether they moderated potential selection bias. Based on the final model, patients assessed by female GPs had higher odds of being deemed eligible (OR = 5.38, 95% CI: 1.97–14.67, p = 0.001). Similarly, GPs with a higher proportion of registered patients with diabetes or prediabetes had greater eligibility rates (OR = 1.15 per percentage point, 95% CI: 1.04–1.28, p = 0.008). Moreover, GPs with a larger proportion of registered patients aged over 65 years exhibited a stronger age-related bias during the eligibility stage, as indicated by a significant interaction between patient age (per year) and the proportion of older registered patients (per percentage point) (interaction OR = 0.996, 95% CI: 0.995–0.998, p < 0.001). This interaction means that the negative effect of age on eligibility is stronger in GP practices with more older patients. In contrast, in practices with fewer older patients, the difference in eligibility between younger and older patients is smaller. No other GP-level characteristics were significantly associated with eligibility rates or with the strength of the age-related bias. Discussion Summary of main findings Increasing age was strongly associated with both ineligibility (selection bias) and non-consent (self-selection bias). Among patients assessed as ineligible, 36% were unable to use a Fitbit device and 29% did not use a mobile phone, two inclusion criteria that disproportionately affected older individuals. Other exclusion reasons included comorbidities, limited mobility, and institutionalisation. These findings underscore how apparently neutral criteria, especially those involving technology, can systematically exclude older or more vulnerable patients and limit trial generalisability [1,15,16]. In terms of self-selection, older patients most often declined participation due to lack of interest, while younger patients more frequently cited lack of time. This divergence reflects previously reported generational differences in motivation and perceived relevance of digital or behavioural health interventions [2,3]. Interestingly, no significant differences were observed by sex or by clinical condition (diabetes versus prediabetes), suggesting that these factors had little influence on patient engagement in this context. At the practice level, patients assessed by female GPs and by those in practices with a higher proportion of diabetes or prediabetes patients had significantly higher eligibility rates—findings aligned with existing evidence linking GP characteristics to recruitment patterns [17]. Furthermore, the age-related decline in eligibility was most pronounced in practices with a higher proportion of older registered patients. This suggests that selection bias may be compounded in settings where both patient and practice characteristics contribute to disengagement or perceived unsuitability [11, 18]. Taken together, these results show that even with structured recruitment protocols and digital tools, trials in general practice remain susceptible to systematic exclusions—particularly of older adults—unless such barriers are explicitly anticipated and addressed during trial design. Comparison with existing literature While the association between older age and reduced participation in digital health interventions is well established, our study demonstrates this effect quantitatively and clearly within a trial recruitment process. Age-related barriers, particularly the inability to use a Fitbit device or mobile phone, were among the leading reasons for ineligibility. This strengthens previous observations regarding the digital divide and trial participation among older populations [1,15,16]. In addition to digital or physical limitations, older adults may decline participation due to a combination of psychological and perceptual factors. These include fatigue, fear of being overwhelmed by procedures, mistrust or confusion about research aims, and concerns about loss of autonomy or perceived burden [19, 20]. Prior studies have shown that older individuals often underestimate the relevance or benefit of participation or may feel that younger populations are more appropriate candidates for health interventions [21,22,23]. A key contribution of this study is the clear separation of GP-determined ineligibility from patient-declared non-consent. Eligibility was influenced not only by patient characteristics, but also by GP and practice-level factors. Patients assessed by female GPs and those in practices with a higher proportion of patients with diabetes or prediabetes tended to have higher eligibility rates. These findings should be interpreted with caution, given the small number of male GPs. These findings are consistent with research suggesting that gender-based communication styles and contextual familiarity with the condition may influence how clinicians interpret eligibility criteria and engage with patients [17]. Among patients who were eligible but declined participation, age-related differences in motivation were apparent. Older patients were more likely to report a lack of interest, while younger individuals more frequently cited time constraints, highlighting distinct generational patterns in perceived relevance and capacity to engage in behavioural research [2, 3]. Furthermore, the interaction between patient age and the age distribution of the GP’s practice revealed that age-related bias was amplified in practices with an older registered population. This finding suggests that recruitment bias may be systematically reinforced at the practice level, a nuance rarely captured in previous studies. From a general practice perspective, this may reflect the practical workload constraints in surgeries with a high proportion of older, often chronically ill patients. These practices may experience greater time pressure and complexity in consultations, reducing capacity to engage in extended recruitment conversations [11, 18]. As a result, GPs might, consciously or unconsciously, be more likely to exclude older patients, especially those perceived as less likely to benefit or more difficult to engage. Strengths and limitations A major strength of this study is its comprehensive and pragmatic recruitment framework. We applied a stratified random sampling approach to minimise clinician-driven selection and used mixed-effects modelling to distinguish patient-level from GP-level influences on recruitment outcomes. In addition, we collected structured, standardised data on reasons for both ineligibility and non-consent, allowing for a nuanced understanding of the specific mechanisms underlying recruitment attrition. Moreover, our conversion rate from invitation to randomisation notably surpasses earlier primary care walking trials such as PACE-Lift [24] and PACE-UP [25]. PACE-UP, which relied on postal invitations, achieved a recruitment rate of about 10%, while reported positive response rates for similar interventions, including PACE-Lift, ranged from 6% to 35%. In contrast, our face-to-face recruitment approach resulted in substantially higher participant engagement and uptake, underscoring both the feasibility and advantage of personalised, in-person strategies in general practice settings. However, several limitations should be acknowledged. First, although reasons for non-consent were self-reported, some technically framed reasons for ineligibility, such as anticipated inability or unwillingness to use a Fitbit device, may have reflected subjective preferences or assumptions rather than true objective barriers. This distinction is important, as it raises the possibility that some patients may have been excluded based on perceived rather than actual ineligibility. Second, although GP characteristics were included in the models, the GP sample was imbalanced, with only four male participants. This limits both the generalisability of findings regarding GP gender and the ability to examine interactions involving GP sex. Third, key sociodemographic and cognitive factors, such as educational attainment, digital literacy, or socioeconomic status, were not captured, which may confound or mediate some observed associations. Implications for research and practice These findings have practical implications for trial design in general practice. The high conversion rate from eligibility to randomisation suggests good initial acceptability of the study among patients, including those offered a digital intervention. However, the pronounced digital exclusion of older adults, combined with motivational differences, poses a significant threat to the external validity of research findings. The growing prevalence of diabetes and prediabetes, conditions frequently described as part of a global metabolic epidemic, further highlights the urgency of developing effective and scalable prevention strategies. General practitioners, who serve as the primary contact point for patients with or at risk of chronic conditions, are ideally positioned to support such interventions. Our findings offer insight into real-world recruitment dynamics that are likely to be relevant not only in the Czech context but also in other healthcare systems with ageing populations and increasing metabolic burden. Lastly, while our findings are likely to be representative of Czech primary care, differences in healthcare structure, digital access, and patient–GP relationships in other countries may limit broader applicability. Nevertheless, the challenges we identified, particularly those related to age, digital engagement, and recruitment dynamics, are likely relevant to many health systems. The systematic integration of GPs in the recruitment process provides a pragmatic and potentially replicable model for trial implementation internationally. Importantly, general practice in the Czech Republic differs from that in the UK and other countries. Most Czech practices are single-handed, with one GP responsible for all patient care, rather than team-based or group practices. Furthermore, patients with type 2 diabetes and prediabetes are typically managed directly by GPs, with minimal involvement of practice nurses or other allied professionals. These structural differences may influence recruitment feasibility and clinician–patient interactions, and should be considered when interpreting the transferability of our findings to other healthcare systems. Conclusions This study demonstrates that a systematic, GP-integrated recruitment strategy can meaningfully reduce selection bias and self-selection bias, improve standardisation across sites and achieve a high recruitment rate. However, even with this design, GP-level selection bias and patient-level self-selection bias, particularly those associated with age, persisted. Digital limitations were a major driver of ineligibility, while motivational misalignments, especially among older adults, contributed to lower consent. The ability to distinguish GP versus patient-level contributions enabled a more nuanced understanding of recruitment barriers in primary care trials. Although conducted in a single national context, the challenges identified, such as digital exclusion, age-related barriers, and motivational mismatches, are broadly applicable across primary care systems facing similar epidemiological and technological transitions. As the burden of type 2 diabetes and prediabetes continues to rise globally, the need for inclusive and scalable trial designs becomes increasingly urgent. Greater attention to recruitment equity and digital accessibility is essential to ensure that clinical research in general practice reflects the diversity of real-world patient populations. Abbreviations CI confidence interval GP general practitioner IQR interquartile range JITAI just-in-time adaptive intervention OR odds ratio SD standard deviation Declarations Ethics approval and consent to participate The study protocol has been approved by the Ethics Committee of the General University Hospital, Prague (No. 49/20), and the study was conducted in compliance with the principles of the Declaration of Helsinki. Informed consent to participate in the study was obtained from participants. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Funding This work was supported by the Czech Health Research Council, Ministry of Health of the Czech Republic (grant number NU21–09–00007). The funding source had no role in the conceptualisation, design, data collection, analysis, decision to publish, or preparation of the manuscript. Author Contribution Conceptualisation: NK, BS, MP, TV, TH, TY, MU, IM, DVD, SE, JP, MS, KJ, CW, PD. Data curation: NK, VC, KM, JK, TV, PD. Formal analysis: VC, TV, NK. Funding acquisition: TV, RC, BS. Investigation: MP, NK, KM, JK. Methodology: BS, NK, MP, VC, JK, AR, SE, DVD, IM, TH, JD, TY, TV, MU. Project administration: MS, BS, RC, KM, TV. Resources: TV, JN, JK, MP, NK, JD, KM, RC, BS, JD. Software: JK, RC. Supervision: BS, TH, TY, AR, TV, MU, JP, SE, DVD. Writing - original draft: NK, TH, TV. Writing - review & editing: all authors. Acknowledgements Not applicable. Data Availability The datasets generated during the current study are available from the corresponding author upon reasonable request. References McMurdo MET, Roberts H, Parker S Improving recruitment of older people to research through good practice. Age Ageing., Ford JG, Howerton MW, Lai GY et al. Barriers to recruiting underrepresented populations to cancer clinical trials: a systematic review. Cancer. 2008;112(2):228–242. Gul RB, Ali PA. Clinical trials: the challenge of recruitment and retention of participants. J Clin Nurs. 2010;19(1–2):227–33. Mira-Martínez S, Zamanillo-Campos R, Malih N, et al. Describing the initial results of a pragmatic, cluster randomized clinical trial to examine the impact of a multifaceted digital intervention for the prevention of type 2 diabetes mellitus in the primary care setting: intervention design, recruitment strategy and participants' baseline characteristics of the PREDIABETEXT trial. Front Endocrinol (Lausanne). 2025;16:1524336. Cotie P, Willms A, Liu S. Implementation of Behavior Change Theories and Techniques for Physical Activity Just-in-Time Adaptive Interventions: A Scoping Review. Int J Environ Res Public Health. 2025;22(7):1133. Brueton VC, Tierney JF, Stenning S, Strategies to improve retention in randomised trials. Cochrane Database Syst Rev., Plaete L, Huys J N. Process evaluation of an eHealth intervention implemented into general practice: general practitioners’and patients’views. Int J Environ Res Public Health. 2018; 15:1475. 8., Vetrovsky T, Cupka J, Dudek M, A pedometer-based walking intervention with and without email counseling in general practice: a pilot randomized controlled trial. BMC Public Health. 2018;18(1):635. 9., Toscos T, Drouin M, Pater J, Selection biases in technology-based intervention research: patients' technology use relates to both demographic and health-related inequities. J Am Med Inform Assoc. 2019;26(8–9):835–839. 10.Millar MM, Taft T, Weir CR et al. Clinical trial recruitment in primary care: exploratory factor analysis of a questionnaire to measure barriers and facilitators to primary care providers' involvement. BMC Prim Care. 2022;23(1):311. Man MS, Chaplin K, Mann C, et al. Improving the management of multimorbidity in general practice: protocol of a cluster randomised controlled trial (The 3D Study). BMJ Open. 2016;6(4):e011261. Vetrovsky T, Kral N, Pfeiferova M, et al. mHealth intervention delivered in general practice to increase physical activity and reduce sedentary behaviour of patients with prediabetes and type 2 diabetes (ENERGISED): rationale and study protocol for a pragmatic randomised controlled trial. BMC Public Health. 2023;23(1):613. Vetrovsky T, Kral N, Pfeiferova M, et al. mHealth intervention delivered in general practice to increase physical activity and reduce sedentary behaviour of patients with prediabetes and type 2 diabetes (ENERGISED): statistical analysis plan. Trials. 2025;26(1):166. Novak J, Jurkova K, Lojkaskova A, et al. Participatory development of an mHealth intervention delivered in general practice to increase physical activity and reduce sedentary behaviour of patients with prediabetes and type 2 diabetes (ENERGISED). BMC Public Health. 2024;24(1):927. Choi NG, DiNitto DM. The digital divide among low-income homebound older adults: Internet use patterns and attitudes. J Med Internet Res. 2013;2(5):e93. van Deursen AJ, Helsper EJ. The third-level digital divide: who benefits most from being online? Commun Inf Technol., Roter DL, Hall JA, Aoki Y. Physician gender effects in medical communication: a meta-analytic review. JAMA. 2002;288(6):756–764. 18., Salisbury C, Johnson L, Purdy S, Epidemiology and impact of multimorbidity in primary care: a retrospective cohort study. Br J Gen Pract. 2011;61(582):e12–e21. 19., Strayer TE, Hollingsworth EK, Shah AS, Why do older adults decline participation in research? Results from two deprescribing clinical trials. Trials. 2023;24(1):456. 20.Kabacińska, Sharma K, Kaye N, Halpern J, Karlawish SD, Berlin JH, Snipes JA, King SA DW,Disparate inclusion of older adults in clinical trials: priorities and opportunities for policy and practice change. Am J Public Health. 2010;100 Suppl 1(Suppl 1):S105-12. 23.Sugarman J, McCrory DC, Hubal RC. Getting meaningful informed consent from older adults: a structured literature review of empirical research. J Am Geriatr Soc. 1998;46(4):517 – 24. 24.Harris T, Kerry SM, Victor CR et al. A primary care nurse-delivered walking intervention in older adults: PACE (Pedometer Accelerometer Consultation Evaluation)-Lift cluster randomised controlled trial. PLoS Med. 2015;12(2):e1001783. 25.Harris T, Kerry S, Victor C,. A pedometer-based walking intervention in 45- to 75-year-olds, with and without practice nurse support: the PACE-UP three-arm cluster RCT. Health Technol Assess. 2018;22(37):1-274. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 03 Mar, 2026 Read the published version in BMC Primary Care → Version 1 posted Editorial decision: Revision requested 10 Dec, 2025 Reviews received at journal 05 Dec, 2025 Reviewers agreed at journal 05 Dec, 2025 Reviews received at journal 24 Nov, 2025 Reviewers agreed at journal 04 Nov, 2025 Reviewers invited by journal 15 Oct, 2025 Editor invited by journal 14 Oct, 2025 Editor assigned by journal 13 Oct, 2025 Submission checks completed at journal 11 Oct, 2025 First submitted to journal 11 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7775668","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":534909197,"identity":"8c79d516-bcab-4e49-a27d-f9d7a944345a","order_by":0,"name":"Norbert Kral","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Norbert","middleName":"","lastName":"Kral","suffix":""},{"id":534909198,"identity":"4f3c3201-572c-41f1-83b5-2b97380f616a","order_by":1,"name":"Tomas Vetrovsky","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIiWNgGAWjYPCDAwxyYOoBHjU8qIwDDMZgKoEULYkNIBqfFnv27sSPPxgOy9lLHz4m+ePM4fT5YYcfAm2xk9NtwGELz9nN0jwMh415+NLSpHluHM7deDvNAKgl2djsAA4tErkbpBkY0hJ7eHiMjRk+ALXMTgBpOZC4DbeWzT9/MKTVg7QY/vhwON1wdvoHQlq2SfAw2CTw8PAYPgA6LEFeOoeALWfObrPmMbAx7DnDlviY50y64QbpnIIDCQa4/cLe3rv55o8KCXn2HuYDB38cs5aXn52++cOHCjs5XFogwADOamYwOIAqQhDUMcg3EK96FIyCUTAKRgYAACASXoCikMM7AAAAAElFTkSuQmCC","orcid":"","institution":"Charles University","correspondingAuthor":true,"prefix":"","firstName":"Tomas","middleName":"","lastName":"Vetrovsky","suffix":""},{"id":534909199,"identity":"63643a0a-e0d0-46b8-9ac0-2feca63f740a","order_by":2,"name":"Marketa Pfeiferova","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Marketa","middleName":"","lastName":"Pfeiferova","suffix":""},{"id":534909200,"identity":"6b092106-c720-4967-9821-d696bacfd7d0","order_by":3,"name":"Bohumil Seifert","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Bohumil","middleName":"","lastName":"Seifert","suffix":""},{"id":534909201,"identity":"f7269cf2-a935-47b0-bb4c-f2ca58f13155","order_by":4,"name":"Vaclav Capek","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Vaclav","middleName":"","lastName":"Capek","suffix":""},{"id":534909202,"identity":"1f908558-94ac-47a8-8652-9489c37e61d1","order_by":5,"name":"Katerina Jurkova","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Katerina","middleName":"","lastName":"Jurkova","suffix":""},{"id":534909203,"identity":"46855449-fdbc-4ea4-9a28-cd48f29b1d40","order_by":6,"name":"Michal Steffl","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Michal","middleName":"","lastName":"Steffl","suffix":""},{"id":534909204,"identity":"53134b27-e5cf-4341-908f-32c2a8d6e14e","order_by":7,"name":"Richard Cimler","email":"","orcid":"","institution":"University of Hradec Kralove","correspondingAuthor":false,"prefix":"","firstName":"Richard","middleName":"","lastName":"Cimler","suffix":""},{"id":534909207,"identity":"11e82121-f53d-438a-9a6c-a77c265f877d","order_by":8,"name":"Jitka Kuhnova","email":"","orcid":"","institution":"University of Hradec Kralove","correspondingAuthor":false,"prefix":"","firstName":"Jitka","middleName":"","lastName":"Kuhnova","suffix":""},{"id":534909209,"identity":"e0c6e6d3-c5f8-4451-b8b1-5fd900791534","order_by":9,"name":"Michael Ussher","email":"","orcid":"","institution":"City St George’s University of London","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Ussher","suffix":""},{"id":534909210,"identity":"52954392-534e-45c9-a69f-aae480e908ba","order_by":10,"name":"Charlotte Wahlich","email":"","orcid":"","institution":"City St George’s University of London","correspondingAuthor":false,"prefix":"","firstName":"Charlotte","middleName":"","lastName":"Wahlich","suffix":""},{"id":534909211,"identity":"5ce52686-ade1-492b-9a75-aaff62efc59e","order_by":11,"name":"Katerina Malisova","email":"","orcid":"","institution":"Palacky University Olomouc","correspondingAuthor":false,"prefix":"","firstName":"Katerina","middleName":"","lastName":"Malisova","suffix":""},{"id":534909212,"identity":"79298f89-781e-4cf4-81ca-3a6666d20b6b","order_by":12,"name":"Jana Pelclova","email":"","orcid":"","institution":"Palacky University Olomouc","correspondingAuthor":false,"prefix":"","firstName":"Jana","middleName":"","lastName":"Pelclova","suffix":""},{"id":534909213,"identity":"add6b6d5-228b-48c9-a0da-704318d5cc98","order_by":13,"name":"Jan Dygryn","email":"","orcid":"","institution":"Palacky University Olomouc","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"","lastName":"Dygryn","suffix":""},{"id":534909214,"identity":"e36479b2-8414-4ab3-a60e-9842ba51cc10","order_by":14,"name":"Steriani Elavsky","email":"","orcid":"","institution":"University of Ostrava","correspondingAuthor":false,"prefix":"","firstName":"Steriani","middleName":"","lastName":"Elavsky","suffix":""},{"id":534909215,"identity":"816b36ab-7d06-431a-bcbd-8c5adf224117","order_by":15,"name":"Iris Maes","email":"","orcid":"","institution":"Ghent University","correspondingAuthor":false,"prefix":"","firstName":"Iris","middleName":"","lastName":"Maes","suffix":""},{"id":534909216,"identity":"69e6ae15-f326-4543-90a5-60f3559891fb","order_by":16,"name":"Delfien Dyck","email":"","orcid":"","institution":"Ghent University","correspondingAuthor":false,"prefix":"","firstName":"Delfien","middleName":"","lastName":"Dyck","suffix":""},{"id":534909217,"identity":"33273875-0992-46ab-8420-d5ffaf0234db","order_by":17,"name":"Alex Rowlands","email":"","orcid":"","institution":"University of Leicester","correspondingAuthor":false,"prefix":"","firstName":"Alex","middleName":"","lastName":"Rowlands","suffix":""},{"id":534909218,"identity":"70510aa8-9949-4f4f-8b55-d6c642789bd2","order_by":18,"name":"Tom Yates","email":"","orcid":"","institution":"University of Leicester","correspondingAuthor":false,"prefix":"","firstName":"Tom","middleName":"","lastName":"Yates","suffix":""},{"id":534909220,"identity":"a2e5c8aa-1fca-4bff-983c-4db599e0de35","order_by":19,"name":"Petr Dvorak","email":"","orcid":"","institution":"Institute of Health Information and Statistics of the Czech Republic","correspondingAuthor":false,"prefix":"","firstName":"Petr","middleName":"","lastName":"Dvorak","suffix":""},{"id":534909221,"identity":"6ac9b661-eec7-45b6-9792-652bdcef6442","order_by":20,"name":"Tess Harris","email":"","orcid":"","institution":"City St George’s University of London","correspondingAuthor":false,"prefix":"","firstName":"Tess","middleName":"","lastName":"Harris","suffix":""}],"badges":[],"createdAt":"2025-10-03 18:23:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7775668/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7775668/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12875-026-03218-4","type":"published","date":"2026-03-03T15:58:06+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":94681601,"identity":"d25d734f-b0f8-4a2c-a59c-d71dab41a271","added_by":"auto","created_at":"2025-10-29 14:52:27","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":59484,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscriptclean.docx","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/3a29c068019a81baccb2435c.docx"},{"id":94681596,"identity":"6b02a0c3-0fc0-4dda-976e-25a0462dc179","added_by":"auto","created_at":"2025-10-29 14:52:27","extension":"json","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19698,"visible":true,"origin":"","legend":"","description":"","filename":"8f12e972f0b8405c8bf8ee6cad2245f6.json","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/23ad9014eab47a49f4a4bfbf.json"},{"id":94727991,"identity":"7ad317b3-0f59-4fc8-af34-b2c7fc6ee8ea","added_by":"auto","created_at":"2025-10-30 07:02:42","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102515,"visible":true,"origin":"","legend":"","description":"","filename":"8f12e972f0b8405c8bf8ee6cad2245f61enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/a99a032e812f15f7d934ef81.xml"},{"id":94681603,"identity":"785cecbb-4093-4b93-a5b6-40dd6c5f216a","added_by":"auto","created_at":"2025-10-29 14:52:27","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16419,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/45f59eda345cf3ffd0348344.pdf"},{"id":94681602,"identity":"a95378a9-23a5-4a65-8e4c-6ed20944b6fa","added_by":"auto","created_at":"2025-10-29 14:52:27","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20633,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/51016dcffe030f3d57194ba2.pdf"},{"id":94728556,"identity":"2d537ad2-2d2f-486c-bf01-8d8d471009b0","added_by":"auto","created_at":"2025-10-30 07:04:01","extension":"xml","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":98162,"visible":true,"origin":"","legend":"","description":"","filename":"8f12e972f0b8405c8bf8ee6cad2245f61structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/ff26d0619da684f1e95c8c79.xml"},{"id":94681606,"identity":"4ca1e8c2-2e02-4f7a-91df-b8c712294df7","added_by":"auto","created_at":"2025-10-29 14:52:27","extension":"html","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":112149,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/113dc42e2893177e56ec7a34.html"},{"id":94681590,"identity":"ab405b4e-7436-49da-a347-1ce55c89ec2f","added_by":"auto","created_at":"2025-10-29 14:52:27","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":19195,"visible":true,"origin":"","legend":"\u003cp\u003eRecruitment flowchart with the number of individuals and their mean age at each stage\u003c/p\u003e","description":"","filename":"Figure1181.png","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/9cdd1cef42cb3f53e326fcea.png"},{"id":94728483,"identity":"b718d76b-0380-4bc7-bab2-2a5ee733344f","added_by":"auto","created_at":"2025-10-30 07:03:54","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":101163,"visible":true,"origin":"","legend":"\u003cp\u003eNumber of patients assessed, eligible, and consenting per individual general practitioner\u003c/p\u003e\n\u003cp\u003eNote: Horizontal lines indicate the mean number of patients assessed, eligible, and consented.\u003c/p\u003e","description":"","filename":"Figure1182.png","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/48c533822caefc5976832546.png"},{"id":104251910,"identity":"cb2e8d8d-b4ea-459f-a909-42bfa86b9f24","added_by":"auto","created_at":"2026-03-09 16:16:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1245691,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7775668/v1/84ec7daf-a37c-4ee7-8111-1d3f1d4c67cd.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Recruiting patients into trials in general practice: insights from the ENERGISED trial","fulltext":[{"header":"Background","content":"\u003cp\u003eRecruiting patients into randomised controlled trials conducted in general practice remains a persistent challenge. Despite the increasing prevalence of chronic metabolic disorders such as type 2 diabetes and prediabetes, and their routine management in primary care, clinical trials often struggle to enrol representative patient populations. Recruitment may be hindered by a combination of practice-level workload constraints, trial-specific eligibility criteria, and patient-level barriers, including digital literacy, health beliefs, and functional limitations. These factors can introduce selection and self-selection bias, threatening both the internal and external validity of trial findings [1,2,3].\u003c/p\u003e\u003cp\u003eThe ENERGISED trial evaluated a pragmatic behavioural intervention to increase physical activity among adults with prediabetes or uncomplicated type 2 diabetes, recruited through general practice. The intervention incorporated self-monitoring with a wrist-worn Fitbit tracker, phone counselling, and a digital support platform using just-in-time adaptive text messaging (JITAI), integrated into routine primary care across multiple sites in the Czech Republic. Comparable trials have shown that digital interventions for people with type 2 diabetes and prediabetes require intensive and flexible recruitment strategies to meet targets and maintain equity [4,5,6]. For example, previous studies of physical activity interventions in primary care identified a selection bias when general practitioners (GPs) preferentially pick those patients whom they believed to be able to use (e.g., highly educated patients) and to benefit from the intervention (e.g., patients with overweight and obesity) [7, 8]. To tackle this, the ENERGISED trial implemented a systematic recruitment procedure that limited GPs' discretion by using randomised, stratified patient lists generated from electronic health records. This approach aimed to minimise selection bias and ensure a more representative sample across age, sex, and diagnosis groups. Self-selection bias may occur when patients decline participation due to personal reasons—such as time constraints, discomfort with digital tools, or low perceived benefit—potentially limiting the representativeness of the sample [9].\u003c/p\u003e\u003cp\u003eTo evaluate how this recruitment strategy worked in practice and to examine potential sources of selection and self-selection bias, we analysed recruitment data from the ENERGISED trial. Specifically, this study quantified recruitment flow from initial patient identification to randomisation, explored the influence of patient and GP characteristics on eligibility and consent, and investigated patterns of exclusion. By disentangling the drivers of recruitment attrition in a real-world setting, our findings provide practical and methodological insights to inform the design of future primary care trials—particularly those evaluating behavioural interventions involving digital tools [10, 11].\u003c/p\u003e\n\u003ch3\u003eObjectives\u003c/h3\u003e\n\u003cp\u003eThe aim of this study was to examine the recruitment process within the ENERGISED randomised controlled trial. The specific objectives were to:\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eDescribe the recruitment flow, including patient characteristics at each stage and reasons for ineligibility (potential selection bias) and non-consent (potential self-selection bias);\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eInvestigate selection and self-selection bias at the GP and patient level, respectively; and\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eInvestigate the relationship between GP characteristics, eligibility rates, and potential selection bias.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e"},{"header":"Methods","content":"\u003cp\u003eThis study was conducted within the ENERGISED randomised controlled trial (ClinicalTrials.gov identifier NCT05351359), a 12-month pragmatic, multicentre trial evaluating an mHealth-enhanced physical activity intervention for adults with prediabetes or uncomplicated type 2 diabetes recruited through Czech general practices. The protocol was approved by the Ethics Committee of the General University Hospital in Prague (reference number: 49/20). Full details of the trial design, procedures, and statistical analysis plan have been published elsewhere [12,13]. Between April 2022 and April 2024, 343 patients were recruited through 28 participating GP practices. The present paper focuses exclusively on the recruitment process, including patient flow, eligibility assessments, and consent.\u003c/p\u003e\u003ch3\u003eRecruitment of general practices\u003c/h3\u003e\u003cp\u003eGeneral practices were recruited through national GP conferences, a professional journal, direct e-mail invitations, and personal contacts, as described previously [12]. The original protocol envisaged participation of 21 practices, but this was later expanded to 28 to accelerate recruitment and ensure broader representation. The participating practices included 15 from urban areas and 13 from rural towns with fewer than 30,000 inhabitants, covering 9 of the 14 administrative regions of the Czech Republic. Participating GPs received remote training on study procedures and data entry and were compensated for their time (approximately €100 per patient completing the study).\u003c/p\u003e\u003ch3\u003ePatient eligibility\u003c/h3\u003e\u003cp\u003eEligibility criteria were identical to those specified in the published ENERGISED protocol [12]. Patients were eligible if they met all of the following: (1) diagnosis of prediabetes or type 2 diabetes according to Czech guidelines for GPs (fasting plasma glucose 5.6–6.9 mmol/l or 2-h plasma glucose of 7.8–11.0 mmol/l after ingestion of 75 g of oral glucose load for the diagnosis of prediabetes fasting plasma glucose ≥ 7.0 mmol/l or 2-h plasma glucose ≥ 11.1 mmol/l after ingestion of 75 g of the oral glucose load for the diagnosis of type 2 diabetes); (2) age ≥ 18 years; (3) followed for prediabetes or diabetes by a participating GP (in the Czech Republic, GPs typically manage patients with uncomplicated type 2 diabetes with glycated haemoglobin (HbA1c) ≤ 53 mmol/mol who are not treated with insulin); (4) regular users of a mobile phone (not necessarily a smartphone), able and willing to answer calls and read text messages as part of the study; (5) able and willing to wear and use a wrist-worn Fitbit activity tracker for the study duration; and (6) provided written informed consent. As the Fitbit required a smartphone for initialisation, patients without their own smartphones were advised to ask a relative or friend to perform the setup.\u003c/p\u003e\u003cp\u003eExclusion criteria were: (1) unable to walk independently for any reason; (2) pregnant; (3) having a household member already recruited for thw trial; (4) living in a residential or nursing care home where the imposed regime could interfere with the intervention; or (5) having co-morbid conditions that would seriously affect adherence, including active malignancy; recent (\u0026lt; 3 months) myocardial infarction, coronary artery bypass graft or cerebrovascular accident; renal disease requiring dialysis; neurological condition (e.g., Parkinson disease); cognitive impairment, or significant hearing or visual impairment; hip or knee joint replacement within three months; or major surgery planned within the next 12 months. Because all study materials and procedures were in Czech, patients were excluded if they lacked sufficient Czech language proficiency to participate effectively.\u003c/p\u003e\u003ch3\u003eRecruitment process\u003c/h3\u003e\u003cp\u003eBefore recruitment began, each GP generated a registry-based list of all patients with prediabetes or type 2 diabetes in their practice. From this list, an initial stratified random batch of 24 patients (sex 1:1, condition prediabetes:diabetes 1:2) was provided. GPs assessed eligibility of all the patients from these batches during their routine health check-ups conducted every 3 to 6 months, documented reasons for ineligibility, invited all eligible patients to participate and recorded reasons for non-consent. When a batch was exhausted, new random batches of 12 patients were issued as needed until the respective practice depleted its original list, reached 32 recruited patients, or the overall trial target was met (planned 340; actually 343 recruited).\u003c/p\u003e\u003cp\u003ePatients who consented underwent baseline procedures, including seven days of wrist-worn accelerometry, followed by a second baseline visit where participants received a Fitbit activity tracker. After the second baseline visit, patients were randomly allocated in a 1:1 ratio to either the intervention or the active control arm.\u003c/p\u003e\u003ch3\u003eRecruitment stages\u003c/h3\u003e\u003cp\u003eImportantly, the study design required GPs to assess all patients from randomly selected lists, thereby minimising discretionary inclusion and reducing selection bias at the assessment stage. Recruitment losses were systematically tracked and categorised as either GP-determined ineligibility or patient-declared non-consent, allowing detailed insight into the factors contributing to recruitment attrition in a primary care trial.\u003c/p\u003e\u003cp\u003eFor the purpose of this study, we use the following terms to describe the successive recruitment stages: (1) total population, (2) sampled, (3) assessed, (4) eligible, (5) consented, and (6) randomised patients.\u003c/p\u003e\u003cp\u003eThe \u003cem\u003etotal population\u003c/em\u003e includes all patients with prediabetes or type 2 diabetes registered at participating GP practices. A list of these patients was generated by the GPs using computerised medical records before recruitment began.\u003c/p\u003e\u003cp\u003e\u003cem\u003eSampled\u003c/em\u003e patients are those randomly selected from the total population, stratified by sex (female: male in a 1:1 ratio) and condition (prediabetes: diabetes in a 1:2 ratio). Each practice initially received a batch of 24 sampled patients. When this batch was exhausted, new random selections of 12 patients were generated and provided as needed.\u003c/p\u003e\u003cp\u003e\u003cem\u003eAssessed\u003c/em\u003e patients are those among the sampled patients who were assessed for eligibility by their GP. As recruitment was conducted opportunistically during routine health check-ups, patients who did not attend their scheduled visits during the recruitment period could not be assessed.\u003c/p\u003e\u003cp\u003e\u003cem\u003eEligible\u003c/em\u003e patients are those assessed and confirmed to meet all eligibility criteria.\u003c/p\u003e\u003cp\u003e\u003cem\u003eConsented\u003c/em\u003e patients are eligible individuals who provided written informed consent to participate in the trial.\u003c/p\u003e\u003cp\u003e\u003cem\u003eRandomised\u003c/em\u003e patients are those who consented and were subsequently randomised. As the protocol involved an interval of at least one week between consent and randomisation, not all consenting patients were randomised.\u003c/p\u003e\u003cp\u003eThis study focuses exclusively on the recruitment process up to the point of randomisation. However, as both GPs' and patients' motivations may have been shaped by their expectations regarding the potential benefits and burdens of the trial procedures, a brief overview of the intervention and control conditions, as well as the trial visits and outcomes, is provided here.\u003c/p\u003e\u003ch2\u003eIntervention and control conditions\u003c/h2\u003e\u003cp\u003eBoth trial arms involved a wrist-worn activity tracker and brief physical activity advice delivered by GPs during baseline visits, with additional components differing between groups. At the second baseline visit, all participants received a Fitbit Inspire 2 activity tracker and were told they could keep it after completing the study. They were encouraged to monitor their daily step count and gradually increase it by at least 3,000 steps above baseline through intentionally brisk walking. They were also advised to interrupt prolonged sitting every 30 minutes.\u003c/p\u003e\u003cp\u003ePatients allocated to the intervention group received an additional mHealth intervention based on JITAI principles and delivered via the HealthReact platform. This platform utilised real-time data from the activity tracker to trigger automated text messages supporting behavioural change, including just-in-time prompts to increase walking pace (triggered after five consecutive minutes of walking) and interrupt prolonged sitting (sent after 30 minutes of inactivity). During the first six months (lead-in phase), these mHealth elements were supported by monthly phone counselling sessions. During the following six months (maintenance phase), the intervention was fully automated.\u003c/p\u003e\u003cp\u003ePatients in the active control group received the same Fitbit tracker, physical activity prescription, and brief advice from their GP, but no mHealth or counselling components.\u003c/p\u003e\u003cp\u003eFurther details of the intervention content and delivery are available in the published protocol [12] and intervention development paper [14].\u003c/p\u003e\u003ch3\u003eTrial visits and outcomes\u003c/h3\u003e\u003cp\u003eTrial procedures for both groups included five GP visits: two at baseline, followed by visits at 3, 6, and 12 months. In the Czech primary care setting, patients with type 2 diabetes typically attend their GPs every three months and those with prediabetes every six months. Therefore, participation required only one additional visit for patients with diabetes and two for those with prediabetes.\u003c/p\u003e\u003cp\u003eThe primary outcome was average daily step count measured by wrist-worn accelerometry over seven consecutive days. Secondary outcomes included additional accelerometry-derived metrics; anthropometric and functional measures (body mass index, waist circumference, blood pressure, and 30-second sit-to-stand test); blood tests; and patient-reported outcomes. Assessments were conducted at baseline and at 3, 6, and 12 months. Further details are available in the published protocol [12] and statistical analysis plan [13].\u003c/p\u003e\u003ch2\u003e\u003cb\u003eStatistical analysi\u003c/b\u003es\u003c/h2\u003e\u003cp\u003eDescriptive statistics were used to summarise the characteristics of participating GPs and patients at each stage of the recruitment process. Continuous variables are reported as medians with interquartile ranges (IQR) or means with standard deviations (SD), and categorical variables as counts and percentages.\u003c/p\u003e\u003cp\u003eTo examine potential selection and self-selection bias, logistic mixed-effects models were used, with GP included as a random intercept to account for clustering at the practice level. Four binary outcomes were modelled: (1) consented vs. not consented among all sampled patients; (2) assessed vs. not assessed among sampled patients; (3) eligible vs. not eligible among those assessed; and (4) consented vs. not consented among those eligible. Each model included patient-level predictors: sex, age (calculated at the start of recruitment) and clinical condition (diabetes or prediabetes). In cases where a significant patient-level predictor of eligibility or consent was identified, we further explored its relationship with specific reported reasons for ineligibility or non-consent. For each reason, a separate logistic regression model was fitted, with the presence or absence of that reason as the binary outcome and the significant patient-level predictor as the independent variable. These models were restricted to the relevant subsample (ineligible or non-consenting patients). Odds ratios (ORs) with 95% confidence intervals (CI) and \u003cem\u003ep\u003c/em\u003e-values were reported.\u003c/p\u003e\u003cp\u003eTo examine whether general practitioners’ characteristics are associated with eligibility rates and potential selection bias, a logistic mixed-effects model was constructed among assessed patients, with eligibility (yes/no) as the outcome and GP included as a random effect. GP-level predictors included sex, years since graduation, practice type (solitary vs. associated), practice location (urban vs. non-urban), number of registered patients, and the proportions of registered patients with diabetes or prediabetes, aged over 65 years, and the proportion of those who had undergone a preventive examination in the past two years. Model selection followed a stepwise approach. Starting from a full model including all patient-level and GP-level variables, non-significant fixed effects were removed sequentially, beginning with the predictor with the highest \u003cem\u003ep\u003c/em\u003e-value. After obtaining a reduced model, two-way interaction terms between selected patient-level and GP-level variables were added to explore potential moderating effects.\u003c/p\u003e\u003cp\u003eAll trial data were collected using REDCap electronic data capture tools. Statistical analyses were conducted in R (version 4.4.2) using the lme4 package.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eCharacteristics of participating GPs\u003c/h2\u003e\u003cp\u003eThe trial involved 28 GPs from 28 practices, of whom 24 were women. The median number of years since graduation was 17 (IQR: 12 to 24). Thirteen GPs were based in cities with over 100,000 inhabitants, while 15 practised in smaller towns or rural areas. Twenty-one worked in solitary practices, and 7 in group or associated practices.\u003c/p\u003e\u003cp\u003eThe median number of registered patients per GP was 1,818 (IQR: 1,584 to 2,048). The mean proportion of patients registered for regular follow-up with diabetes or prediabetes was 4.5% (SD: 3.9). The mean proportion of patients aged over 65 years was 24.6% (SD: 6.9). In 2021, the mean proportion of patients who underwent a preventive health check was 25.7% (SD: 10.8). As preventive health checks are scheduled biennially, some patients may have attended in 2020 or subsequently in 2022, and the overall coverage is therefore higher than reflected in the 2021 data.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eRecruitment flow and patient characteristics\u003c/h2\u003e\u003cp\u003eAmong all 3,579 patients with diabetes or prediabetes registered with the 28 participating GPs, a stratified random sample of 1,376 patients was selected. Of these, 1,138 (82.7%) were assessed for eligibility by their GPs, and 792 (69.6% of those assessed) were deemed eligible. Among the eligible patients, 348 (43.9%) consented to participate, and 343 (43.3%) were ultimately randomised, with 5 patients not showing up for the second baseline visit. The recruitment flow is summarised in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the characteristics of patients at each stage of the recruitment process: the total population, sampled patients, those assessed for eligibility, those deemed eligible, and those who consented. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the number of patients assessed, found eligible, and consented per individual GP.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCharacteristics of patients at each stage of the recruitment process: total population, sampled, assessed, eligible, and consenting patients\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStage\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSex\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCondition\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eCount (n, %)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAge (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e\u003cb\u003eTotal Population\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e950 (26.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e65.5\u0026thinsp;\u0026plusmn;\u0026thinsp;11.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e978 (27.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e60.0\u0026thinsp;\u0026plusmn;\u0026thinsp;13.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e803 (22.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e70.1\u0026thinsp;\u0026plusmn;\u0026thinsp;11.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e848 (23.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e64.3\u0026thinsp;\u0026plusmn;\u0026thinsp;13.6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e\u003cb\u003eSampled\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e478 (34.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e64.1\u0026thinsp;\u0026plusmn;\u0026thinsp;11.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e237 (17.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e59.2\u0026thinsp;\u0026plusmn;\u0026thinsp;12.3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e442 (32.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e68.7\u0026thinsp;\u0026plusmn;\u0026thinsp;11.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e219 (15.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e64.4\u0026thinsp;\u0026plusmn;\u0026thinsp;13.6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e\u003cb\u003eAssessed\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e393 (34.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e64.2\u0026thinsp;\u0026plusmn;\u0026thinsp;11.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e190 (16.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e59.1\u0026thinsp;\u0026plusmn;\u0026thinsp;12.6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e379 (33.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e68.5\u0026thinsp;\u0026plusmn;\u0026thinsp;11.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e176 (15.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e65.0\u0026thinsp;\u0026plusmn;\u0026thinsp;13.8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e\u003cb\u003eEligible\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e272 (34.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e63.0\u0026thinsp;\u0026plusmn;\u0026thinsp;10.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e145 (18.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e57.0\u0026thinsp;\u0026plusmn;\u0026thinsp;12.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e256 (32.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e66.7\u0026thinsp;\u0026plusmn;\u0026thinsp;11.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e119 (15.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e63.0\u0026thinsp;\u0026plusmn;\u0026thinsp;12.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e\u003cb\u003eConsented\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e127 (36.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e62.7\u0026thinsp;\u0026plusmn;\u0026thinsp;10.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e62 (17.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e54.6\u0026thinsp;\u0026plusmn;\u0026thinsp;12.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDiabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e109 (31.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e64.4\u0026thinsp;\u0026plusmn;\u0026thinsp;10.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePrediabetes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e50 (14.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e60.6\u0026thinsp;\u0026plusmn;\u0026thinsp;11.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eSelection and self-selection bias at the patient level\u003c/h2\u003e\u003cp\u003eTo assess potential selection and self-selection bias, we first compared patients who consented to participate with those in the randomly sampled population. Increasing age was associated with lower odds of progressing through the stages of assessment, eligibility, and consent (OR\u0026thinsp;=\u0026thinsp;0.965 per year, 95% CI: 0.954\u0026ndash;0.976, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In contrast, there were no significant differences in sex or clinical condition between consenting and sampled patients.\u003c/p\u003e\u003cp\u003eWe then examined each stage of the recruitment process separately. No significant differences were found between assessed and sampled patients, suggesting no bias in the assessment stage. However, both the eligibility stage (selection bias) and the consent stage (self-selection bias) showed the same age-related pattern: older patients were less likely to be deemed eligible (OR\u0026thinsp;=\u0026thinsp;0.955 per year, 95% CI: 0.942\u0026ndash;0.968, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and less likely to consent (OR\u0026thinsp;=\u0026thinsp;0.972 per year, 95% CI: 0.958\u0026ndash;0.986, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Again, no significant differences were observed for sex or clinical condition at these stages.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eReasons for ineligibility and non-consent\u003c/h2\u003e\u003cp\u003eThe reasons for ineligibility are summarised in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The most common reasons for ineligibility were an anticipated inability to use the wrist-worn Fitbit activity tracker throughout the study period (n\u0026thinsp;=\u0026thinsp;126) and not being a regular user of a mobile phone or being unable to read text messages (n\u0026thinsp;=\u0026thinsp;101). The age-related bias observed in the overall eligibility analysis was consistent across most specific reasons for ineligibility, with the only common reason not related to increasing age being registration with a diabetologist (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eReported reasons for ineligibility and their association with patient age\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReason for non-eligibility\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCount\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eOR\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eUnable to use Fitbit\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e126\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.08\u0026ndash;1.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNot using a phone\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e101\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.08\u0026ndash;1.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRegistered with a diabetologist\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.97\u0026ndash;1.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.7436\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eComorbidities\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.02\u0026ndash;1.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0006\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNot registered with the practitioner\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.02\u0026ndash;1.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0018\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eUnable to walk\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.09\u0026ndash;1.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eInstitutionalised\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.00\u0026ndash;1.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0510\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eOn insulin\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.95\u0026ndash;1.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.6890\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNot having pre/diabetes\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.93\u0026ndash;1.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.1655\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eHousehold member already enrolled\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.99\u0026ndash;1.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0878\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNot speaking Czech\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.91\u0026ndash;1.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.3995\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe reasons for non-consent are summarised in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Among patients who were eligible but did not consent, the most frequently cited reasons were a lack of interest in a physical activity intervention (n\u0026thinsp;=\u0026thinsp;172) and insufficient time (n\u0026thinsp;=\u0026thinsp;158). Notably, the odds of non-consent due to lack of interest increased with age (OR\u0026thinsp;=\u0026thinsp;1.02, 95% CI: 1.002\u0026ndash;1.041, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03), whereas the odds of non-consent due to lack of time decreased with age (OR\u0026thinsp;=\u0026thinsp;0.971, 95% CI: 0.954\u0026ndash;0.99, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eReported reasons for non-consent and their association with patient age\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReason for non-consent\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCount\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eOR\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eLack of interest\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e172\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.00\u0026ndash;1.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0303\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eLack of time\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e158\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.95\u0026ndash;0.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0023\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNot willing to use Fitbit\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.98\u0026ndash;1.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.8404\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eOther\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.00\u0026ndash;1.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0324\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNot willing to receive calls\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.97\u0026ndash;1.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.8968\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePrivacy reasons\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.97\u0026ndash;1.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.9205\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNot willing to get messages\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.99\u0026ndash;1.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.1517\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAlready wearing a tracker\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.93\u0026ndash;0.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.0175\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eInfluence of GP characteristics on eligibility rates and selection bias\u003c/h2\u003e\u003cp\u003eWe next examined whether GP-level characteristics were associated with eligibility rates and whether they moderated potential selection bias. Based on the final model, patients assessed by female GPs had higher odds of being deemed eligible (OR\u0026thinsp;=\u0026thinsp;5.38, 95% CI: 1.97\u0026ndash;14.67, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001). Similarly, GPs with a higher proportion of registered patients with diabetes or prediabetes had greater eligibility rates (OR\u0026thinsp;=\u0026thinsp;1.15 per percentage point, 95% CI: 1.04\u0026ndash;1.28, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.008).\u003c/p\u003e\u003cp\u003eMoreover, GPs with a larger proportion of registered patients aged over 65 years exhibited a stronger age-related bias during the eligibility stage, as indicated by a significant interaction between patient age (per year) and the proportion of older registered patients (per percentage point) (interaction OR\u0026thinsp;=\u0026thinsp;0.996, 95% CI: 0.995\u0026ndash;0.998, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). This interaction means that the negative effect of age on eligibility is stronger in GP practices with more older patients. In contrast, in practices with fewer older patients, the difference in eligibility between younger and older patients is smaller.\u003c/p\u003e\u003cp\u003eNo other GP-level characteristics were significantly associated with eligibility rates or with the strength of the age-related bias.\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003eSummary of main findings\u003c/h2\u003e\u003cp\u003eIncreasing age was strongly associated with both ineligibility (selection bias) and non-consent (self-selection bias). Among patients assessed as ineligible, 36% were unable to use a Fitbit device and 29% did not use a mobile phone, two inclusion criteria that disproportionately affected older individuals. Other exclusion reasons included comorbidities, limited mobility, and institutionalisation. These findings underscore how apparently neutral criteria, especially those involving technology, can systematically exclude older or more vulnerable patients and limit trial generalisability [1,15,16].\u003c/p\u003e\u003cp\u003eIn terms of self-selection, older patients most often declined participation due to lack of interest, while younger patients more frequently cited lack of time. This divergence reflects previously reported generational differences in motivation and perceived relevance of digital or behavioural health interventions [2,3]. Interestingly, no significant differences were observed by sex or by clinical condition (diabetes versus prediabetes), suggesting that these factors had little influence on patient engagement in this context.\u003c/p\u003e\u003cp\u003eAt the practice level, patients assessed by female GPs and by those in practices with a higher proportion of diabetes or prediabetes patients had significantly higher eligibility rates\u0026mdash;findings aligned with existing evidence linking GP characteristics to recruitment patterns [17]. Furthermore, the age-related decline in eligibility was most pronounced in practices with a higher proportion of older registered patients. This suggests that selection bias may be compounded in settings where both patient and practice characteristics contribute to disengagement or perceived unsuitability [11, 18].\u003c/p\u003e\u003cp\u003eTaken together, these results show that even with structured recruitment protocols and digital tools, trials in general practice remain susceptible to systematic exclusions\u0026mdash;particularly of older adults\u0026mdash;unless such barriers are explicitly anticipated and addressed during trial design.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eComparison with existing literature\u003c/h2\u003e\u003cp\u003eWhile the association between older age and reduced participation in digital health interventions is well established, our study demonstrates this effect quantitatively and clearly within a trial recruitment process. Age-related barriers, particularly the inability to use a Fitbit device or mobile phone, were among the leading reasons for ineligibility. This strengthens previous observations regarding the digital divide and trial participation among older populations [1,15,16].\u003c/p\u003e\u003cp\u003eIn addition to digital or physical limitations, older adults may decline participation due to a combination of psychological and perceptual factors. These include fatigue, fear of being overwhelmed by procedures, mistrust or confusion about research aims, and concerns about loss of autonomy or perceived burden [19, 20]. Prior studies have shown that older individuals often underestimate the relevance or benefit of participation or may feel that younger populations are more appropriate candidates for health interventions [21,22,23].\u003c/p\u003e\u003cp\u003eA key contribution of this study is the clear separation of GP-determined ineligibility from patient-declared non-consent. Eligibility was influenced not only by patient characteristics, but also by GP and practice-level factors. Patients assessed by female GPs and those in practices with a higher proportion of patients with diabetes or prediabetes tended to have higher eligibility rates. These findings should be interpreted with caution, given the small number of male GPs. These findings are consistent with research suggesting that gender-based communication styles and contextual familiarity with the condition may influence how clinicians interpret eligibility criteria and engage with patients [17].\u003c/p\u003e\u003cp\u003eAmong patients who were eligible but declined participation, age-related differences in motivation were apparent. Older patients were more likely to report a lack of interest, while younger individuals more frequently cited time constraints, highlighting distinct generational patterns in perceived relevance and capacity to engage in behavioural research [2, 3].\u003c/p\u003e\u003cp\u003eFurthermore, the interaction between patient age and the age distribution of the GP\u0026rsquo;s practice revealed that age-related bias was amplified in practices with an older registered population. This finding suggests that recruitment bias may be systematically reinforced at the practice level, a nuance rarely captured in previous studies. From a general practice perspective, this may reflect the practical workload constraints in surgeries with a high proportion of older, often chronically ill patients. These practices may experience greater time pressure and complexity in consultations, reducing capacity to engage in extended recruitment conversations [11, 18]. As a result, GPs might, consciously or unconsciously, be more likely to exclude older patients, especially those perceived as less likely to benefit or more difficult to engage.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003eStrengths and limitations\u003c/h2\u003e\u003cp\u003eA major strength of this study is its comprehensive and pragmatic recruitment framework. We applied a stratified random sampling approach to minimise clinician-driven selection and used mixed-effects modelling to distinguish patient-level from GP-level influences on recruitment outcomes. In addition, we collected structured, standardised data on reasons for both ineligibility and non-consent, allowing for a nuanced understanding of the specific mechanisms underlying recruitment attrition.\u003c/p\u003e\u003cp\u003eMoreover, our conversion rate from invitation to randomisation notably surpasses earlier primary care walking trials such as PACE-Lift [24] and PACE-UP [25]. PACE-UP, which relied on postal invitations, achieved a recruitment rate of about 10%, while reported positive response rates for similar interventions, including PACE-Lift, ranged from 6% to 35%. In contrast, our face-to-face recruitment approach resulted in substantially higher participant engagement and uptake, underscoring both the feasibility and advantage of personalised, in-person strategies in general practice settings.\u003c/p\u003e\u003cp\u003eHowever, several limitations should be acknowledged. First, although reasons for non-consent were self-reported, some technically framed reasons for ineligibility, such as anticipated inability or unwillingness to use a Fitbit device, may have reflected subjective preferences or assumptions rather than true objective barriers. This distinction is important, as it raises the possibility that some patients may have been excluded based on perceived rather than actual ineligibility. Second, although GP characteristics were included in the models, the GP sample was imbalanced, with only four male participants. This limits both the generalisability of findings regarding GP gender and the ability to examine interactions involving GP sex. Third, key sociodemographic and cognitive factors, such as educational attainment, digital literacy, or socioeconomic status, were not captured, which may confound or mediate some observed associations.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003eImplications for research and practice\u003c/h2\u003e\u003cp\u003eThese findings have practical implications for trial design in general practice. The high conversion rate from eligibility to randomisation suggests good initial acceptability of the study among patients, including those offered a digital intervention. However, the pronounced digital exclusion of older adults, combined with motivational differences, poses a significant threat to the external validity of research findings.\u003c/p\u003e\u003cp\u003eThe growing prevalence of diabetes and prediabetes, conditions frequently described as part of a global metabolic epidemic, further highlights the urgency of developing effective and scalable prevention strategies. General practitioners, who serve as the primary contact point for patients with or at risk of chronic conditions, are ideally positioned to support such interventions. Our findings offer insight into real-world recruitment dynamics that are likely to be relevant not only in the Czech context but also in other healthcare systems with ageing populations and increasing metabolic burden.\u003c/p\u003e\u003cp\u003eLastly, while our findings are likely to be representative of Czech primary care, differences in healthcare structure, digital access, and patient\u0026ndash;GP relationships in other countries may limit broader applicability. Nevertheless, the challenges we identified, particularly those related to age, digital engagement, and recruitment dynamics, are likely relevant to many health systems. The systematic integration of GPs in the recruitment process provides a pragmatic and potentially replicable model for trial implementation internationally.\u003c/p\u003e\u003cp\u003eImportantly, general practice in the Czech Republic differs from that in the UK and other countries. Most Czech practices are single-handed, with one GP responsible for all patient care, rather than team-based or group practices. Furthermore, patients with type 2 diabetes and prediabetes are typically managed directly by GPs, with minimal involvement of practice nurses or other allied professionals. These structural differences may influence recruitment feasibility and clinician\u0026ndash;patient interactions, and should be considered when interpreting the transferability of our findings to other healthcare systems.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study demonstrates that a systematic, GP-integrated recruitment strategy can meaningfully reduce selection bias and self-selection bias, improve standardisation across sites and achieve a high recruitment rate. However, even with this design, GP-level selection bias and patient-level self-selection bias, particularly those associated with age, persisted. Digital limitations were a major driver of ineligibility, while motivational misalignments, especially among older adults, contributed to lower consent. The ability to distinguish GP versus patient-level contributions enabled a more nuanced understanding of recruitment barriers in primary care trials.\u003c/p\u003e\u003cp\u003eAlthough conducted in a single national context, the challenges identified, such as digital exclusion, age-related barriers, and motivational mismatches, are broadly applicable across primary care systems facing similar epidemiological and technological transitions. As the burden of type 2 diabetes and prediabetes continues to rise globally, the need for inclusive and scalable trial designs becomes increasingly urgent. Greater attention to recruitment equity and digital accessibility is essential to ensure that clinical research in general practice reflects the diversity of real-world patient populations.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCI\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003econfidence interval\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGP\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003egeneral practitioner\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIQR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003einterquartile range\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eJITAI\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003ejust-in-time adaptive intervention\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eOR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eodds ratio\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSD\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003estandard deviation\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study protocol has been approved by the Ethics Committee of the General University Hospital, Prague (No. 49/20), and the study was conducted in compliance with the principles of the Declaration of Helsinki. Informed consent to participate in the study was obtained from participants.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis work was supported by the Czech Health Research Council, Ministry of Health of the Czech Republic (grant number NU21\u0026ndash;09\u0026ndash;00007). The funding source had no role in the conceptualisation, design, data collection, analysis, decision to publish, or preparation of the manuscript.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eConceptualisation: NK, BS, MP, TV, TH, TY, MU, IM, DVD, SE, JP, MS, KJ, CW, PD. Data curation: NK, VC, KM, JK, TV, PD. Formal analysis: VC, TV, NK. Funding acquisition: TV, RC, BS. Investigation: MP, NK, KM, JK. Methodology: BS, NK, MP, VC, JK, AR, SE, DVD, IM, TH, JD, TY, TV, MU. Project administration: MS, BS, RC, KM, TV. Resources: TV, JN, JK, MP, NK, JD, KM, RC, BS, JD. Software: JK, RC. Supervision: BS, TH, TY, AR, TV, MU, JP, SE, DVD. Writing - original draft: NK, TH, TV. Writing - review \u0026amp; editing: all authors.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe datasets generated during the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eMcMurdo MET, Roberts H, Parker S Improving recruitment of older people to research through good practice. Age Ageing., Ford JG, Howerton MW, Lai GY et al. Barriers to recruiting underrepresented populations to cancer clinical trials: a systematic review. Cancer. 2008;112(2):228\u0026ndash;242.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGul RB, Ali PA. Clinical trials: the challenge of recruitment and retention of participants. J Clin Nurs. 2010;19(1\u0026ndash;2):227\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMira-Mart\u0026iacute;nez S, Zamanillo-Campos R, Malih N, et al. Describing the initial results of a pragmatic, cluster randomized clinical trial to examine the impact of a multifaceted digital intervention for the prevention of type 2 diabetes mellitus in the primary care setting: intervention design, recruitment strategy and participants' baseline characteristics of the PREDIABETEXT trial. Front Endocrinol (Lausanne). 2025;16:1524336.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCotie P, Willms A, Liu S. Implementation of Behavior Change Theories and Techniques for Physical Activity Just-in-Time Adaptive Interventions: A Scoping Review. Int J Environ Res Public Health. 2025;22(7):1133.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrueton VC, Tierney JF, Stenning S, Strategies to improve retention in randomised trials. Cochrane Database Syst Rev., Plaete L, Huys J N. Process evaluation of an eHealth intervention implemented into general practice: general practitioners\u0026rsquo;and patients\u0026rsquo;views. Int J Environ Res Public Health. 2018; 15:1475. 8., Vetrovsky T, Cupka J, Dudek M, A pedometer-based walking intervention with and without email counseling in general practice: a pilot randomized controlled trial. BMC Public Health. 2018;18(1):635. 9., Toscos T, Drouin M, Pater J, Selection biases in technology-based intervention research: patients' technology use relates to both demographic and health-related inequities. J Am Med Inform Assoc. 2019;26(8\u0026ndash;9):835\u0026ndash;839. 10.Millar MM, Taft T, Weir CR et al. Clinical trial recruitment in primary care: exploratory factor analysis of a questionnaire to measure barriers and facilitators to primary care providers' involvement. BMC Prim Care. 2022;23(1):311.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMan MS, Chaplin K, Mann C, et al. Improving the management of multimorbidity in general practice: protocol of a cluster randomised controlled trial (The 3D Study). BMJ Open. 2016;6(4):e011261.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVetrovsky T, Kral N, Pfeiferova M, et al. mHealth intervention delivered in general practice to increase physical activity and reduce sedentary behaviour of patients with prediabetes and type 2 diabetes (ENERGISED): rationale and study protocol for a pragmatic randomised controlled trial. BMC Public Health. 2023;23(1):613.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVetrovsky T, Kral N, Pfeiferova M, et al. mHealth intervention delivered in general practice to increase physical activity and reduce sedentary behaviour of patients with prediabetes and type 2 diabetes (ENERGISED): statistical analysis plan. Trials. 2025;26(1):166.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNovak J, Jurkova K, Lojkaskova A, et al. Participatory development of an mHealth intervention delivered in general practice to increase physical activity and reduce sedentary behaviour of patients with prediabetes and type 2 diabetes (ENERGISED). BMC Public Health. 2024;24(1):927.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoi NG, DiNitto DM. The digital divide among low-income homebound older adults: Internet use patterns and attitudes. J Med Internet Res. 2013;2(5):e93.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Deursen AJ, Helsper EJ. The third-level digital divide: who benefits most from being online? Commun Inf Technol., Roter DL, Hall JA, Aoki Y. Physician gender effects in medical communication: a meta-analytic review. JAMA. 2002;288(6):756\u0026ndash;764. 18., Salisbury C, Johnson L, Purdy S, Epidemiology and impact of multimorbidity in primary care: a retrospective cohort study. Br J Gen Pract. 2011;61(582):e12\u0026ndash;e21. 19., Strayer TE, Hollingsworth EK, Shah AS, Why do older adults decline participation in research? Results from two deprescribing clinical trials. Trials. 2023;24(1):456. 20.Kabacińska, Sharma K, Kaye N, Halpern J, Karlawish SD, Berlin JH, Snipes JA, King SA DW,Disparate inclusion of older adults in clinical trials: priorities and opportunities for policy and practice change. Am J Public Health. 2010;100 Suppl 1(Suppl 1):S105-12. 23.Sugarman J, McCrory DC, Hubal RC. Getting meaningful informed consent from older adults: a structured literature review of empirical research. J Am Geriatr Soc. 1998;46(4):517\u0026thinsp;\u0026ndash;\u0026thinsp;24. 24.Harris T, Kerry SM, Victor CR et al. A primary care nurse-delivered walking intervention in older adults: PACE (Pedometer Accelerometer Consultation Evaluation)-Lift cluster randomised controlled trial. \u003cem\u003ePLoS Med.\u003c/em\u003e 2015;12(2):e1001783. 25.Harris T, Kerry S, Victor C,. A pedometer-based walking intervention in 45- to 75-year-olds, with and without practice nurse support: the PACE-UP three-arm cluster RCT. Health Technol Assess. 2018;22(37):1-274.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-primary-care","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"famp","sideBox":"Learn more about [BMC Primary Care](https://bmcprimcare.biomedcentral.com/)","snPcode":"","submissionUrl":"https://author-welcome.nature.com/12875","title":"BMC Primary Care","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Clinical trial recruitment, Primary care, Type 2 diabetes, Physical activity, Selection bias, Digital exclusion","lastPublishedDoi":"10.21203/rs.3.rs-7775668/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7775668/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eRecruiting patients into randomised controlled trials in general practice is challenging and carries a substantial risk of bias. The ENERGISED trial evaluated an mHealth physical activity intervention in patients with prediabetes or type 2 diabetes recruited through general practice. To minimise bias, the trial employed a systematic recruitment strategy in which general practitioners assessed the eligibility of patients from random stratified samples of their registers and sought consent from all those deemed eligible. This study aimed to analyse the recruitment process of the ENERGISED trial and identify sources of potential bias arising from general practitioners' eligibility assessments (selection bias) and patient consent (self-selection bias).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003ePatients with prediabetes or type 2 diabetes were randomly sampled from the registers of 28 Czech general practices using sex- and diagnosis-stratified lists. Eligibility was systematically assessed during routine visits, with general practitioners documenting reasons for ineligibility. All eligible patients were invited to participate, and reasons for non-consent were recorded. Logistic mixed-effects models were used to examine the influence of patient characteristics (age, sex, diagnosis) and general practitioner characteristics on eligibility and consent.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003e Of 1,376 sampled patients, 1,138 (83%) were assessed, 792 (70% of assessed) were eligible, 348 (44% of eligible) consented and 343 were randomised. Older age was associated with lower odds of eligibility (OR 0.955, 95% CI 0.942\u0026ndash;0.968; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and lower odds of consent among eligible patients (OR 0.972, 95% CI 0.958\u0026ndash;0.986; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Ineligibility was most often due to digital barriers. Practices with older registered populations showed stronger age-related bias. Female practitioners and practices with more diabetes/prediabetes patients achieved significantly higher eligibility rates.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eSystematic recruitment through general practice can reduce selection and self-selection bias, yet digital exclusion, particularly in older adults, persists. Future trials must proactively address digital literacy and age-related barriers to ensure representative participation in primary care research.\u003c/p\u003e","manuscriptTitle":"Recruiting patients into trials in general practice: insights from the ENERGISED trial","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-29 14:52:22","doi":"10.21203/rs.3.rs-7775668/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-10T16:28:32+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-05T11:10:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"305742611061726503426056982360151572993","date":"2025-12-05T10:48:03+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-24T07:49:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"126470361847482438482214990343505763892","date":"2025-11-04T07:55:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-15T15:56:05+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-14T08:54:12+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-13T08:48:30+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-11T08:18:12+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Primary Care","date":"2025-10-11T08:14:16+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-primary-care","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"famp","sideBox":"Learn more about [BMC Primary Care](https://bmcprimcare.biomedcentral.com/)","snPcode":"","submissionUrl":"https://author-welcome.nature.com/12875","title":"BMC Primary Care","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ad706e90-2f43-4d97-8249-7c9dec72a08a","owner":[],"postedDate":"October 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-09T16:12:19+00:00","versionOfRecord":{"articleIdentity":"rs-7775668","link":"https://doi.org/10.1186/s12875-026-03218-4","journal":{"identity":"bmc-primary-care","isVorOnly":false,"title":"BMC Primary Care"},"publishedOn":"2026-03-03 15:58:06","publishedOnDateReadable":"March 3rd, 2026"},"versionCreatedAt":"2025-10-29 14:52:22","video":"","vorDoi":"10.1186/s12875-026-03218-4","vorDoiUrl":"https://doi.org/10.1186/s12875-026-03218-4","workflowStages":[]},"version":"v1","identity":"rs-7775668","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7775668","identity":"rs-7775668","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}