Evaluating the effectiveness of lifestyle education for individuals at increased risk of type 2 diabetes and cardiovascular diseases (Halt2Diabetes): protocol for a repeated measures study

Abstract

Background

Halt2Diabetes is a prevention initiative in Flanders, Belgium, targeting the rising prevalence of type 2 diabetes (T2D) and cardiovascular diseases (CVDs). The program uses a two-step approach to identify high-risk individuals and guide them towards healthier lifestyles, reducing T2D and CVD risk factors. Building on evidence that lifestyle interventions can prevent both conditions by addressing modifiable risk factors, this research evaluates the effectiveness of Halt2Diabetes as a scalable real-world intervention and its potential as a blueprint for other contexts.

Methods

The Halt2Diabetes program identifies high-risk individuals using the Finnish Diabetes Risk Score (FINDRISC) through an online risk assessment tool. Eligible participants, referred by general practitioners, participate in six group-based lifestyle guidance sessions led by trained dietitians, focusing on healthy eating, physical activity, and behavior modification. This longitudinal study employs repeated measurements at five time points over 18 months. Primary outcomes include changes in body weight, waist circumference, and hemoglobin A1c (HbA1c) levels. Secondary outcomes encompass blood pressure, cholesterol levels, physical activity, and dietary habits. Data collection involves self-reported questionnaires, biochemical measures, and anthropometric assessments. Analysis will examine longitudinal changes in health and behavioral outcomes among participants.

Discussion

This study will provide evidence for the real-world effectiveness of Halt2Diabetes in promoting sustained lifestyle changes and reducing T2D and CVD risk factors. Despite challenges including participation barriers and data integration needs, the program’s scalable, digitalized risk assessment model shows promise for broader application. The findings will enhance understanding of effective T2D and CVD prevention strategies, offering insights for policymakers and healthcare providers to strengthen public health initiatives.

Trial registration

This study is registered with the ISRCTN registry, identifier: ISRCTN62149214 (https://www.isrctn.com/ISRCTN62149214). The date of registration is March 13, 2025.

Background

Type 2 diabetes (T2D) is an increasingly significant chronic condition in Europe and worldwide. It is closely associated with cardiovascular diseases (CVDs), as both are primarily linked to increasing rates of obesity, sedentary lifestyles, and an aging population [1–3]. According to the International Diabetes Federation (IDF), the global prevalence of diabetes, with T2D accounting for over 90% of cases, was estimated to be 1 in 9 adults in 2024 and is expected to rise to 1 in 8 adults by 2050 [4]. Moreover, symptoms of T2D often go unnoticed due to its latent progression, resulting in nearly half of the global population with T2D remaining undiagnosed until complications arise [5, 6]. Late diagnosis poses a significant burden on health and healthcare expenditures, highlighting the need to identify and target individuals at risk of T2D and CVDs [7–9].

Early identification of high-risk individuals enables the implementation of preventive measures to delay or prevent T2D and its associated complications [10]. Therefore, risk identification through screening programs is a crucial component in establishing appropriate interventions [11]. Prevention of T2D in high-risk populations typically focuses on correcting unhealthy lifestyle patterns and improving modifiable risk factors. Previous intervention studies provide strong evidence that lifestyle changes can successfully prevent the onset of T2D [12–15] and improve its risk factors [16–18]. Most of the studied interventions consist of healthcare professionals providing individual or group-based lifestyle education (e.g., dietary modification and physical activity) to high-risk individuals [19]. Although high-intensity lifestyle interventions types (such as frequent and highly personalized counseling) demonstrate better effectiveness for preventing T2D in controlled study environments, more sustainable and scalable low- to moderate-intensity interventions (like less frequent and more general lifestyle education) show promise in reducing T2D risk in real-world interventions [20, 21].

In Belgium, approximately 7% of the total population is diagnosed with diabetes, while an estimated 37% of individuals with diabetes remain undiagnosed, bringing the assumed prevalence to 10% [22]. Additionally, available estimates indicate that between 25 and 30% of adults over 45 in the region of Flanders (the Dutch-speaking part of Belgium) are at increased risk of developing T2D and associated CVDs [23]. However, implementing targeted prevention as a permanent practice in the healthcare system and the broader society remains challenging [24]. In the absence of a nationally coordinated program, the Halt2Diabetes initiative, coordinated by the patient organization Diabetes Liga, aims to prevent T2D and CVDs among high-risk individuals in Flanders. Supported by the Flemish Government, Halt2Diabetes is gradually being implemented across various regions in Flanders since 2019. It has been recognized as a scalable and promising prevention initiative by the EU’s Healthier Together initiative [25].

The Halt2Diabetes program takes a two-step approach to identifying individuals at increased risk for T2D and associated CVDs, and guiding them towards healthier lifestyles. High-risk individuals are detected through an online risk assessment tool, which utilizes the Finnish Diabetes Risk Score (FINDRISC) [26], and subsequently invited to a general practitioner (GP) consultation. Although this screening tool is freely accessible for everyone, its use in the local healthcare system is not systematic and relies heavily on awareness campaigns. Following GP examination, eligible high-risk individuals are invited to participate in group-based lifestyle intervention sessions led by professional dietitians. The sessions focus on key modifiable risk factors, emphasizing healthy eating, physical activity, and other lifestyle changes. Over 100 dietitians have been trained to facilitate these sessions. To date, nearly 250 session series have been conducted, involving approximately 1,500 participants.

As part of the JACARDI project (Joint Action on Cardiovascular Diseases and Diabetes) [27], this research seeks to evaluate the effectiveness of Halt2Diabetes. Importantly, this real-world intervention study will offer valuable insights into the program’s blueprint potential for other regions and contexts. This paper outlines the protocol for identifying and enrolling high-risk individuals in the study, along with the observational research design featuring pre- and repeated post-intervention measurements. The study envisages the following specific objectives.

1. To assess the impact of the lifestyle intervention as part of the Halt2Diabetes program on behavioral changes and health outcomes.

2. To explore the varying effects of the Halt2Diabetes lifestyle intervention across different participant subgroups, considering factors such as sociodemographic background, health literacy, and diabetes knowledge.

3. To examine the determinants of behavioral changes and health outcomes in the Halt2Diabetes lifestyle intervention.

4. To evaluate the extent of the Halt2Diabetes program’s reach by analyzing data from risk assessment tools and intervention participation records.

Methods

Study design

This study will employ a repeated measures research design to assess individual health and behavioral changes in participants of the Halt2Diabetes lifestyle intervention over an 18-month observation period, with participants examined on five occasions from the start of their participation. The primary and secondary outcomes cover changes in both health and behavioral outcomes during this period, as outlined in Table 1. To gain insights into how the lifestyle intervention affects health and behavioral outcomes over time, this study relies on the theoretical framework of the Behavior Change Wheel (BCW) [28], which considers three components of behavioral change: capability, opportunity, and motivation that are influenced by different intervention types. Following BCW theory, lifestyle interventions target specific aspects of capability and motivation to drive changes in healthy lifestyle behaviors [29]. Figure 1 presents a comprehensive overview of the entire study flow as outlined in this methods section.

Table 1.

Primary and secondary outcomes of the study

Primary outcomes	Body weight
	Waist circumference
	Hemoglobin A1c (HbA1c)
	Healthy dietary score (HDI)
Secondary outcomes	Body mass index (BMI)
	Blood pressure (systolic and diastolic)
	Cholesterol (blood lipids)
	Total physical activity
	Tobacco use
	Diabetes knowledge
	Health literacy

Fig. 1.

Study flowchart

Study population

The Halt2Diabetes program is implemented at the level of primary care zones in Flanders, which are 60 organizational entities encompassing primary care practices across multiple geographically clustered municipalities [30]. The program is gradually expanding to ultimately cover the entire Flemish territory. This study will be conducted in the 31 primary care zones currently participating across Flanders, encompassing an estimated 1,500,000 individuals within the 45–80 age group targeted by Halt2Diabetes [31]. Recent data from an online risk assessment tool indicate that 31.9% of users older than 45 are at high risk of developing T2D or CVDs (FINDRISC score of 12 or higher) [32], suggesting that approximately 480,000 individuals are eligible to participate in the intervention study area.

Risk assessment and recruitment

As indicated in Fig. 1, the main entry point to the Halt2Diabetes program is an online risk assessment tool (Gezondheidskompas, https://www.gezondheidskompas.be/) developed in Dutch by the Flemish Government. The tool is freely accessible, optimized for different devices, and has been completed over 150,000 times since its launch in 2019. It includes questions to calculate a FINDRISC score (age, BMI, waist circumference, daily physical activity, daily consumption of vegetables, fruits, and berries, use of blood pressure medication, history of elevated blood glucose, and family history of diabetes) [26]. Additionally, it gathers information on gender, smoking habits, postal code, family history of CVDs and personal history of diabetes and CVDs. Users are also asked how they learned about the tool. All information is collected anonymously.

Upon completing the questionnaire, all users receive a personalized lifestyle report. Those aged 45 or older with a FINDRISC score of 12 or higher are advised to review their risk assessment with a GP in their primary care zone (cf. Fig. 1). A unique code allows GPs to access the risk assessment results, or users may forward the results directly via the online tool. Individuals identified as high risk can only participate in the intervention after being referred by a GP to the Halt2Diabetes lifestyle guidance sessions. Alternatively, the risk assessment can be completed by a GP or another healthcare professional with the patient, either online or using a paper-and-pencil version, before referral to the intervention. GPs may also refer individuals to the Halt2Diabetes lifestyle guidance sessions based on other parameters, such as a history of gestational diabetes or stress hyperglycemia.

For this intervention study, participants will be recruited at the beginning of their first Halt2Diabetes lifestyle guidance session, following the GP consultation and upon provision of informed consent. Study information will be provided to participants prior to recruitment. We aim to reach the targeted high-risk population through established social media and local communication strategies to raise awareness about the Halt2Diabetes program and enhance study participation.

Inclusion and exclusion criteria

To participate in the study, individuals must meet the following criteria: (1) be at high risk of developing T2D (FINDRISC score ≥ 12 or as assessed by a GP), (2) have received a GP referral to lifestyle guidance sessions, (3) reside in or near a participating primary care zone included in the study, and (4) be willing and able to provide written informed consent. Participants will be excluded from the study if they (1) have a previous diagnosis of diabetes or CVD(s) (excluding hypertension), (2) have physical or psychological disabilities that prevent participation in a lifestyle guidance program, or (3) have insufficient knowledge of Dutch to follow the lifestyle guidance sessions.

The Halt2Diabetes lifestyle intervention

The intervention consists of six face-to-face group-based lifestyle guidance sessions (cf. Table 2), each lasting approximately 90 min, led by a trained dietitian. The free-of-charge series includes five main sessions and one follow-up session for each participant. Groups range from 5 to 12 participants, and sessions are decentralized, with each participating primary care zone organizing its sessions independently. The sessions aim to promote healthier eating habits, increase physical activity, and encourage healthy behavior modifications, tailored to individuals with high-risk profiles. The Halt2Diabetes lifestyle intervention is modeled after previously tested programs in studies across various countries [16–18, 33, 34].

Table 2.

Summarized content of the lifestyle guidance sessions

Session	Content
1	The first session introduces participants to the conversational code used during the sessions, establishing a supportive group environment. Discussions cover the risks of T2D, an overview of the disease, and participants’ current eating habits and motivations for healthy eating. Baseline assessments are conducted to inform future progress evaluations.
2	In the second session, participants review their past efforts to eat healthier and set personalized goals and action plans, guided by the nutritional pyramid. Emphasis is placed on realistic goal-setting and practical strategies for long-term dietary changes.
3	The third session focuses on evaluating the progress of participants’ action plans, addressing challenges such as relapse, and introducing the physical activity triangle. Additional topics like reading food labels, understanding nutri-scores, or revising action plans are covered based on group needs.
4	During the fourth session, participants continue to evaluate their progress and refine their action plans. Practical applications, such as shopping exercises, and discussions on topics selected by the group are included.
5	The fifth session involves a final evaluation of participants’ action plans and focuses on long-term sustainability. Advanced activities like cooking workshops or further exploration of group-selected topics may be included.
6	The program concludes with a follow-up session conducted by the dietitian to monitor ongoing progress, address emerging challenges, and provide continued support, ensuring the sustainability of healthy eating habits developed during the program.
Optional components	Practical skills: focused activities include meal and menu planning, creating shopping lists, and practicing healthy shopping habits.
	Behavioral tools: SWOT analysis to identify strengths, weaknesses, opportunities, and threats related to dietary habits, and strategies for modifying the environment to facilitate healthy choices.
	Interactive cooking workshops: balanced meal preparation through hands-on cooking exercises, including mindful eating practices and understanding different types of hunger.
	Customized group discussions: topics based on group preferences and needs, such as relapse prevention, physical activity promotion, or advanced meal planning.

The content of the lifestyle guidance sessions focuses on educating participants about T2D, dietary habits, and lifestyle choices. To facilitate effective behavioral change, the sessions are grounded in the Self-Determination Theory, which emphasizes the importance of autonomous action, self-efficacy, and social relatedness for sustained behavioral change [35]. The intervention therefore entails coaching strategies that focus on individual goal setting, self-monitoring, and action and coping planning.

During the sessions, instructors use techniques such as nudging and engaging communication to encourage behavioral changes. The first five sessions gradually transition from knowledge sharing to practical applications through goal setting and action planning. Throughout the sessions, there is an emphasis on self-monitoring progress and refining action plans. The series concludes with a follow-up session three months later to evaluate sustained progress and provide ongoing support. Optional components may include customized group discussions, practical skills training (e.g., meal planning, healthy shopping habits, or food context management), and interactive cooking workshops, depending on the specific needs of participants.

Dietitians delivering the Halt2Diabetes lifestyle guidance sessions receive comprehensive training focused on hands-on program implementation and motivational techniques. A detailed manual outlines mandatory and adaptable activities. Ongoing support for dietitians includes quarterly review meetings for updates, experience sharing, addressing challenges, and exchanging practical tips. Feedback from these meetings is used to refine training materials and ensure continuous improvement.

Participant timeline

Figure 1 depicts the study participant timeline. The intervention, consisting of six lifestyle guidance sessions, will take place during the first six months of the study. Throughout this intervention phase, participants will be observed at three measurement points: t₀ at baseline (during the first session), t₁ after three months (during the fifth session), and t₂ after six months (during the sixth session). Following the intervention, participants will be assessed at two follow-up points: t₃ after 12 months, and t₄ after 18 months. Each measurement involves repeated data collection, such as biochemical measures. At t₀, single-time surveyed information, including sociodemographic characteristics at baseline, will be collected.

We anticipate that post-intervention participant retention (i.e., at t₃ and t₄) will be challenging as participants will no longer attend lifestyle guidance sessions. To address this, participants will receive a financial incentive upon completing the t₂ questionnaire. Continued participation offers the benefit of regular self-testing of key anthropometric parameters (e.g., weight and waist circumference). To increase response rates during the follow-up phase, questionnaires will be distributed using a mixed-mode delivery method. Initially, respondents will receive a paper version of the questionnaire, aiming to reduce selective non-response, particularly among subgroups with lower digital skills and limited internet access. In case of non-response, participants will receive a reminder letter and email including a personalized hyperlink to the web-based survey. Up to three reminders via email or phone contact may be used. In-person interviews can be considered as a last resort.

Study measures and data collection

Data will be collected at five time points for each participant using validated instruments. The data are organized into six modules: contextual data, behavioral determinants, behavioral outcome measures, anthropometric outcome measures, biochemical outcome measures, and risk assessment data. Table 3 outlines these categories, their corresponding indicators, concepts, and observation time points. Most data will be collected repeatedly to enable longitudinal comparison, with the exception of contextual and risk assessment data. This longitudinal approach allows for assessment of changes in behavioral and health outcomes (cf. Table 1) over time within the same individuals. Data collection will involve both anthropometric and biochemical measurements as well as individual questionnaires administered in paper-and-pencil and web-based formats.

Table 3.

Modules of data collection, indicators or concepts per module and time of observation

Data module	Indicator or concept	t−1	t0	t1	t2	t3	t4
		Pre-study	Month 1	Month 3	Month 6	Month 12	Month 18
Contextual data	Sociodemographic information		✓
	Individual history on diabetes and CVD		✓	✓	✓	✓	✓
	Family history on diabetes and CVD		✓			✓
	Pathway to lifestyle guidance sessions		✓
	Participation in other lifestyle programs		✓	✓	✓	✓	✓
Behavioral determinants	Diabetes knowledge		✓		✓		✓
	Health literacy		✓		✓		✓
	Self-efficacy		✓		✓		✓
	Coping skills		✓		✓		✓
	Attitudes and outcome expectations		✓		✓		✓
	Risk perception		✓		✓		✓
	Goals and intentions		✓
Behavioral outcome measures	Healthy dietary score		✓	✓	✓	✓	✓
	Total physical activity		✓	✓	✓	✓	✓
	Tobacco use		✓	✓	✓	✓	✓
Anthropometric outcome measures	Body weight and height		✓	✓	✓	✓	✓
	Waist circumference		✓	✓	✓	✓	✓
	Blood pressure (systolic and diastolic)		✓	✓	✓
Biochemical outcome measures	Hemoglobin A1c (HbA1c)		✓	✓	✓	✓	✓
	Cholesterol (blood lipids)		✓	✓	✓	✓	✓
Risk assessment data	Sociodemographic information	✓
	Body mass index (BMI)	✓
	Finnish Diabetes Risk Score (FINDRISC)	✓
	Systematic Coronary Risk Evaluation (SCORE)	✓

Contextual data

Contextual information will include sociodemographic characteristics collected at baseline (t₀): gender, age, education, migration status, partnership status, employment, subjective income, family size, and language use. These data will facilitate assessment of participant characteristics, examination of selective attrition, and analysis of variation in intervention effects. Participants’ personal history of T2D and CVD will be tracked throughout the study (t₀-t₄) to monitor new diagnoses, while family histories will be collected at baseline and follow-up (t₀ and t₃) to compute FINDRISC scores. Additionally, data on pathways to the lifestyle guidance sessions will be gathered at baseline (t₀), and participants’ attendance at other lifestyle interventions will be monitored throughout the study period (t₀-t₄).

Behavioral determinants

Following BCW theory, the Halt2Diabetes lifestyle intervention primarily targets motivation and capability as key components to achieve behavioral and health outcome changes. The study will measure behavioral determinants of two subcomponents—psychological competencies (capability) and reflective drivers (motivation)—at t₀, t₂, and t₄, corresponding to areas addressed by the lifestyle guidance sessions [29]. Each behavioral determinant will be assessed using validated and tailored instruments.

Psychological competencies will be measured through the following determinants: diabetes knowledge (22-item Diabetes Knowledge Questionnaire Revised) [36], health literacy (6-item European Health Literacy Survey Questionnaire) [37], and coping skills (4-item Brief Resilient Coping Scale) [38]. Reflective drivers will be captured through: self-efficacy (28-item Self-Rated Abilities for Health Practices Scale) [39], diabetes-tailored attitudes and outcome expectations (7-item Intrinsic Motivation Inventory subscale adapted from the Center for Self-Determination Theory) [40–42], diabetes risk perception (8-item Risk Perception Survey for Developing Diabetes) [43], and goals and intentions (self-categorized and open-ended questions). It should be noted that diabetes knowledge and health literacy can serve as both determinants and outcome measures in this study.

Behavioral outcome measures

Repeated collection of behavioral outcomes will represent a core element of the study (t₀-t₄). Dietary behavior will be assessed using the 18-item Finnish Food Intake Questionnaire (D2D-FIQ), which covers a comprehensive range of food and drink consumption [44]. This questionnaire generates a validated composite diet quality score, the Healthy Diet Index (HDI), to measure adherence to health-promoting dietary patterns [45]. Physical activity and tobacco use data will be collected using the WHO STEPS instrument [46]. The physical activity component will be based on the 16-item International Physical Activity Questionnaire (IPAQ) [47], which enables examination of the extent to which participants meet WHO global recommendations on physical activity. Tobacco use assessment will be limited to two questions focusing on current use patterns.

Anthropometric outcome measures and blood pressure

Anthropometric outcomes will be self-reported by study participants at all time points (t₀-t₄), focusing on key indicators: height, weight, and waist circumference. Participants will receive repeated instruction on correct measurement procedures during group sessions to ensure data quality. Body mass index (BMI) is calculated as weight in kilograms divided by the square of height in meters.

Systolic and diastolic blood pressure measurements will be conducted by trained fieldwork nurses using electronic sphygmomanometers in a subset of highly engaged Halt2Diabetes regions. Resting blood pressure will be measured twice from the right arm in a sitting position. Due to logistical constraints, blood pressure measurements will be limited to three time points: t₀, t₁, and t₂.

Biochemical outcome measures

Biochemical outcomes will be measured in the same subset of highly engaged Halt2Diabetes regions. Trained fieldwork nurses will collect venous blood samples during lifestyle guidance sessions at t₀-t₂ and will be responsible for specimen collection and transport to a standardized, quality-controlled clinical laboratory for analysis. These measurements will assess participants’ glycated hemoglobin levels (HbA1c) and cholesterol profile (high- and low-density lipoprotein, and triglycerides).

At follow-up time points t₃ and t₄, Tasso devices (Tasso, Inc., https://www.tassoinc.com/) will be used to collect biochemical parameters (HbA1c and blood lipids) through a minimally invasive, self-administered method. This device functions as a blood lancet that collects liquid blood samples in standard collection tubes, enabling participants to perform the procedure at home. A small-scale validation study will be conducted prior to implementation to confirm the accuracy and feasibility of this collection method.

Risk assessment data

In addition to primary study data, aggregate-level pre-study data (t₋₁) will be retrieved from the online risk assessment tool to enable calculation of participation rates in lifestyle guidance sessions among high-risk individuals. By comparing characteristics of study participants with those of risk assessment tool users, we will identify which high-risk groups are under- or overrepresented in the intervention. The following aggregate information from risk assessment tool users will be available for comparison: age, gender, FINDRISC score, Systematic Coronary Risk Evaluation (SCORE), and BMI.

Sample size calculation

To align with the repeated measures study design, GLIMMPSE (https://glimmpse.samplesizeshop.org/) sample size calculation software is used1 [48]. The GLIMMPSE calculation procedure accounts for expected changes in key outcomes within individuals over time, while considering the correlation of repeated measurements in the same participant [49, 50]. To obtain relevant parameters, such as repeated measures means and their standard deviations, our sample size calculations rely on results from comparable studies on healthy lifestyle interventions in high-risk populations: the Greater Green Triangle (GGT) Diabetes Prevention Project (Australia) [16], the Stop Diabetes study (StopDia, Finland) [17], and SLIM iMplementation Experience Region Noord- and Oost-Gelderland (SLIMMER, The Netherlands) [18]. Correlation matrices are defined based on relevant literature to accommodate correlations between repeated measurements. Sample size calculations are conducted with a significance level of 0.05 (𝛼) and a desired power of 0.80 (1−𝛽), using the Hotelling Lawley Trace test due to its equivalence to a mixed model test. The calculation assumes a null hypothesis that the differences between all pairs of repeated measures means are the same.

For self-reported outcomes of interest, we estimate a minimum of 162 completed cases (i.e., individuals participating at all five measurement points), based on the largest sample size for primary outcomes (weight). The sample size range for other outcomes is between 49 and 1372. To mitigate up to 50% attrition during the 18-month observation period3, the baseline number of participants will be increased to 324. Because of logistical and budgetary constraints, data collection on blood pressure and blood samples (HbA1c and blood lipids) will be conducted only in a subsample. We estimate a minimum of 87 completed cases for triglycerides, with subsample sizes ranging between 19 and 76 for other blood measurement outcomes. Accounting for 60% attrition, this will result in a baseline of 218 participants. If there is a high dropout rate during the initial intervention stage, additional recruitment may be considered.

Given that we will test for multiple response outcomes, the risk of a type 1 error (i.e., finding a result with p < 0.05 (𝛼) by chance) increases. Therefore, the analysts will adopt a conservative approach in interpreting the results.

Statistical analysis

Analysis will focus on longitudinal changes in health and behavioral outcomes. Descriptive analysis will be performed, with unadjusted explorations of change over time using time plots and repeated measures ANOVA. Linear mixed models (LMMs), adjusted for confounding factors, will be used to analyze longitudinal changes within individuals. These models nest repeated measurements within individuals to account for clustering, providing insights into average changes during and after the intervention. This approach will offer a comprehensive understanding of temporal trends, considering general patterns and individual variations. Individual covariates will be used to explain variation between participants. Models can be expanded to test consistency across subgroups (e.g., by gender or education), though sample sizes are not designed for specific subgroup analysis.

The intervention evaluation will also analyze behavioral determinants, shedding light on the pathway between lifestyle guidance and changes in health risk factors. Multilevel structural equation modeling (MSEM) will be used, combining principles of SEM and multilevel modeling to analyze hierarchical data. MSEM allows for the inclusion of latent variables (e.g., self-efficacy) and modeling complex variable paths.

Study participation response will be evaluated by assessing the number of individuals who enroll in lifestyle guidance sessions, agree to participate, meet inclusion criteria, and provide informed consent. Response patterns will be compared between online risk assessment tool use (t₋₁) and study participation (t₀). Attrition will be calculated by subtracting the number of participants at each follow-up (t₁-t₄) from the baseline (t₀). Logistic regression and survival analysis will be employed to explore selective attrition, aiding in interpreting the main results.

Most analyses will be conducted using Stata, with MPlus being used for MSEM analyses.

Data management

Only researchers will have access to collected study data, with all identifiable information removed. An identification key will be securely stored, accessible only to the fieldwork team in direct contact with participants. During the project, all datasets and data management procedures will be stored in an encrypted file. Each personal computer with data will be protected by updated antivirus and firewall software. The encrypted file will be backed up on a protected storage server provided by the University of Antwerp. This repository will also preserve all encrypted project datasets and files for 20 years after the research project concludes.

Ethical issues

This study has received approval from the Ethical Committee of the University Hospital Antwerp (BUN B3002025000014) by March 2025 in accordance with Good Clinical Practice (GCP) guidelines. The approval is reviewed annually based on researchers’ reports. The study adheres to the ethical principles outlined in the Declaration of Helsinki. Written informed consent will be required from all participants. Upon signing up for Halt2Diabetes, participants will receive a detailed information letter, providing sufficient time to review the study’s objectives, procedures, risks, and benefits before the first lifestyle guidance session. Participation in the study will be voluntary, and individuals can take part in the Halt2Diabetes lifestyle guidance program without enrolling in the intervention study. Participants can withdraw their consent at any time during or after the study, with the option to have their individual data removed. They will be allowed to continue their usual care throughout the study and will be referred to professional healthcare if study blood parameters indicate concerning results.

Dissemination of study results

The study results will be published in high-quality, peer-reviewed international scientific journals with open access. Findings will be disseminated at both international and national scientific meetings. Pre-final versions of publications and conference abstracts will be available through the University of Antwerp’s institutional repository in correspondence to its publication policy. As part of the JACARDI project, the main study conclusions will be reported and interpreted in terms of the scalability and sustainability of the Halt2Diabetes program. Key results will also be shared with program stakeholders and valorized through the communication channels of the patient organization Diabetes Liga.

Discussion

This study aims to evaluate the existing Halt2Diabetes program, a health prevention initiative that uses a two-step approach to identify individuals at increased risk of T2D and associated CVDs, and guide them towards healthier lifestyles. It builds on robust evidence that T2D can be prevented by addressing modifiable risk factors through lifestyle interventions. This study contributes to existing research by: (1) assessing the effectiveness of an established real-world lifestyle intervention, (2) exploring the intervention effects across different subpopulations, (3) uncovering the behavioral determinants of these effects, and (4) evaluating the intervention’s reach. Overall, this research will provide key insights into the functioning of Halt2Diabetes, a proven scalable and sustainable prevention initiative in a non-clinical context that engages citizens, policymakers, and other stakeholders [25]. Scientifically, this study promises to elucidate the pathways through which the intervention leads to behavioral change in a real-world setting.

The success of Halt2Diabetes depends on both the accurate identification of individuals at risk and their active participation in lifestyle interventions. Effective communication strategies are crucial for the reach and uptake of risk identification, both within and outside the professional healthcare context. However, one of the main challenges is that awareness campaigns often fail to effectively reach less empowered population groups, particularly those with ethnically diverse backgrounds and lower socioeconomic positions [51]. In addition, the current digitalized risk assessment tool may be less tailored to these groups. Community-based and outreaching screening initiatives could be promising in overcoming these barriers to risk identification [52, 53]. As a next step, strong participation of high-risk individuals in lifestyle interventions is essential for effective prevention. Currently, there is no connection between risk assessment data and information on participation in the Halt2Diabetes lifestyle guidance. This results in a lack of detailed information to monitor and evaluate intervention participation comprehensively. Therefore, there is a need for a more elaborate data infrastructure to develop strategies that increase participation among individuals at high risk, with this evaluative study being a first step in this direction. Additionally, integrating Halt2Diabetes as a targeted prevention initiative within the healthcare system is challenging [24], which limits the program’s effectiveness in encouraging participation.

The potential benefits of Halt2Diabetes strongly rely on its ability to ensure participants adhere to a healthy lifestyle. One of the challenges during the lifestyle guidance sessions is retaining participants, as dropout rates are considerable. Although Halt2Diabetes offers a free-of-charge lifestyle intervention to boost participation, this also lowers the threshold for opting out. To promote sustained lifestyle changes, programs like Halt2Diabetes focus on enhancing participants’ motivation and capability to adapt. However, while the intervention’s emphasis on autonomous goal-setting and self-monitoring supports short-term behavior modification, its long-term effectiveness remains uncertain [54].

Currently, the program lacks tools to monitor long-term participant trajectories, including the link with routine clinical and biochemical data to measure outcomes. Limited compatibility with the healthcare system also hinders coordinated long-term action plans for high-risk individuals involving other healthcare providers. Additionally, the lifestyle intervention adopts a one-size-fits-all approach, overlooking the specific needs of individuals or population groups [55]. Personalized approaches could enhance participant commitment and, consequently, the program’s effectiveness. Addressing these challenges would help to build on the current intervention, improving participant adherence and ultimately ensuring the sustainability of valuable intervention effects.

Despite its challenges and limitations, scalable initiatives such as Halt2Diabets are crucial to tackle the rising prevalence of preventable non-communicable diseases like T2D and CVDs. In a context of limited resources for prevention and staff shortages in the healthcare sector, it is vital that the Halt2Diabetes model is both effective and financially sustainable to serve as a blueprint for other prevention initiatives. Notably, it relies on a digitalized risk assessment approach that can be applied to a range of other health conditions (e.g., cancer, dementia, etc.). Through this study, we aim to provide evidence for the real-world effectiveness of Halt2Diabetes in modifying lifestyle behavior and reducing T2D risk factors.

Abbreviations

T2D

Type 2 Diabetes

CVD(s)

Cardiovascular Disease(s)

IDF

International Diabetes Federation

FINDRISC

Finnish Diabetes Risk Score

GP

General Practitioner

BCW

Behavior Change Wheel

BMI

Body Mass Index

HbA1c

Hemoglobin A1c

HDI

Healthy Diet Index

IPAQ

International Physical Activity Questionnaire

WHO

World Health Organization

STEPS

STEPwise approach to NCD risk factor surveillance

D2D-FIQ

Finnish Food Intake Questionnaire

MSEM

Multilevel Structural Equation Modeling

LMM

Linear Mixed Models

GCP

Good Clinical Practice

JACARDI

Joint Action on Cardiovascular Diseases and Diabetes

EU

European Union

HaDEA

European Health and Digital Executive Agency

Author contributions

This is a shared first author paper as J.V. and W.L. contributed equally to the study. W.L., J.V., F.M., and J.V.O. designed the study. E.W., K.D., L.D., M.M., R.V., J.Va., A.L., S.V., and I.E. assisted in designing the study. J.Va., A.L., S.V., J.V., F.M., and W.L. coordinate both the intervention implementation and study data collection. J.V., W.L., and F.M. contributed to writing the manuscript. All other authors provided feedback and approved the final manuscript version.

Funding

This study is part of JACARDI - Joint Action on Cardiovascular Disease and Diabetes, which has received funding from the EU4Health Programme 2021–2027 under Grant Agreement 101126953. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Health and Digital Executive Agency (HaDEA). Neither the European Union nor the granting authority can be held responsible for them.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The current study is approved by the Ethical Committee of the University Hospital Antwerp (BUN B3002025000014) and all methods were performed in accordance with the relevant guidelines and regulations (Declarations of Helsinki). Written informed consent will be required from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.