Effect of a digital health behaviour change support system on cardiovascular disease risk in a randomized weight loss trial – Nature
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Advertisement
npj Digital Medicine volume 9, Article number: 560 (2026)
2108
Metrics details
Obesity is a risk factor for cardiovascular diseases (CVD). We evaluated a 52-week digital Health Behaviour Change Support System (HBCSS) with a one-year follow-up to treat overweight and Class I obesity. A total of 532 participants (BMI 27–35 kg/m2) were randomized into three groups: CBT-based group counselling and self-help guidance (SHG) delivered face-to-face, and usual care. These groups were further divided into HBCSS and non-HBCSS groups. The 10-year CVD risks were assessed using the FINRISK calculator. Baseline median overall CVD risks were similar between groups. In two-group analyses comparing HBCSS and non-HBCSS groups, after 12 months, CVD risks decreased in both groups. However, after 24 months, only the HBCSS group maintained significant reductions in overall risk (−0.40%, p < 0.001). Among participants with obesity, the HBCSS group demonstrated a sustained decrease in overall and coronary artery disease risk, while there were no CVD risk reductions in the non-HBCSS group. However, the differences between HBCSS and non-HBCSS groups were not significant in any analyses conducted, and there were no significant differences in six group analyses between intervention groups. The HBCSS group showed significant reductions in weight, BMI, and waist circumference (at 12 and 24 months) and in LDL cholesterol (at 12 months), compared with controls (p < 0.05). The HBCSS results in a decrease in CVD risk factors, which is reflected by a sustained reduction in calculated CVD risks, especially among participants with obesity.
Obesity increases the risk of many diseases, such as type 2 diabetes, cardiovascular diseases, cancer, and death1,2,3,4. In Europe, cardiovascular diseases (CVD) account for 45% of all deaths5,6. The most important factors explaining cardiovascular morbidity are hypertension, smoking, hyperlipidaemia, obesity, and diabetes, explaining more than half of the morbidity7. In Finland, around half of people over the age of 30 already have high blood pressure, and nearly one-third are living with obesity8.
The risk of cardiovascular diseases can be assessed by combining individual risk factors, such as blood pressure, cholesterol levels, and diabetes, and lifestyle factors, such as smoking. The overall risk may be clinically significant, even if the individual risk factors are only slightly elevated. In Finnish healthcare, the risk of disease in previously healthy patients is estimated using the FINRISK calculator, based on Finnish population cohorts9. Other widely used risk calculators include SCORE2 and LIFE-CVD210,11.
The main purpose of obesity treatment is to prevent further weight gain, prevent and treat obesity-related diseases, and improve work and functional capacity and quality of life12. Lifestyle intervention can be combined with other forms of treatment (e.g., medication or surgery) if lifestyle intervention alone does not achieve sufficient results12,13. Pharmacological treatment is currently one of the most effective conservative treatments; however, there are problems associated with weight regain after discontinuing medication14.
Digital interventions have a favourable effect on CVD risk factors, but there is less research on the primary prevention of CVD events15,16,17. The Health Behaviour Change Support System (HBCSS) is a web-based lifestyle change application that utilizes persuasive systems design (PSD) and cognitive behavioural therapy (CBT) methods to treat patients with overweight and obesity. We previously conducted a randomized clinical trial demonstrating the effectiveness of HBCSS in weight loss and reduction of metabolic syndrome, as well as its long-term legacy effect on reducing the need for blood pressure medications18,19,20. The primary objective of this study was to assess the impact of the HBCSS on cardiovascular disease risk using a CVD risk calculator.
Out of 532 participants at baseline, 108 dropped out between baseline and 12 months, and an additional 52 dropped out between 12 months and 24 months. Dropout rates did not differ between groups based on BMI category (obesity or overweight) (p > 0.05). The results are presented in Supplementary Fig. 2. One participant was excluded from the study due to an erroneous laboratory result to prevent bias from a significant random measurement error. Therefore, a total of 531 participants were analysed.
Table 1 presents the baseline characteristics of the participants. At baseline, the mean (SD) age was 45.9 (10.0) years; 50.7% of participants were male, and 49.3% were female. Among all participants, the median 10-year overall cardiovascular risk was 2.51% (IQR 1.10 to 4.40), coronary artery disease risk was 1.43% (0.55 to 2.86), and stroke risk was 0.97% (0.47 to 1.58). Participants with obesity (BMI 30–35 kg/m2) had FINRISK values 1.03, 1.04, and 1.14 times higher at baseline than participants with overweight (BMI 27–29.99 kg/m2), respectively. There were no statistically significant differences between HBCSS and non-HBCSS groups at baseline.
The changes (mean (95% CI)) in weight and cardiovascular risk factors after 12 and 24 months are presented in Table 2. After 12 months, the HBCSS group lost more weight (−2.30 kg (−2.91 to −1.68), p < 0.001) compared with the non-HBCSS group (−0.76 kg (−1.39 to −0.13), p = 0.019), with a significant difference (p < 0.001) between groups. Similar patterns were observed for BMI (−0.76 vs. −0.27 kg/m², p < 0.001 between groups) and waist circumference (−2.42 vs. −0.68 cm, p < 0.001 between groups). Although both groups showed significant reductions in systolic and diastolic blood pressure, the differences between the groups were not statistically significant (p = 0.676 and p = 0.717, respectively). LDL cholesterol level was significantly reduced in the HBCSS group (−0.17 mmol/l (−0.26 to −0.09), p < 0.001), compared with the non-HBCSS group (−0.01 mmol/l (−0.10 to 0.08), p = 0.824), with p = 0.011 between groups. There was no difference between the groups for other cardiovascular risk factors (Table 2).
At 24 months, many improvements observed at 12 months were sustained, especially in the HBCSS group. In the HBCSS group, weight, BMI, and waist circumference did not change significantly between 12 and 24 months (p = 0.705, p = 0.715 and p = 0.294), with measured values at 24 months −1.74 kg, −0.64 kg/m2 and −2.92 cm, respectively; all p < 0.001 compared with baseline (with p = 0.012, p = 0.005, and p = 0.005, between groups, respectively). Systolic and diastolic blood pressure remained significantly reduced in both groups. The reduction in LDL cholesterol was not sustained at 24 months in the HBCSS group.
Among all participants, both groups demonstrated significant reductions (in percentage points) in overall 10-year risk, coronary artery disease risk, and stroke risk at 12 months (Fig. 1, Supplementary Table 1). At 24 months, only the HBCSS group maintained significant reductions in overall 10-year risk (1.74-fold), coronary artery disease risk (1.71-fold), and stroke risk (1.57-fold) compared with the non-HBCSS group, which showed no significant reductions. However, the differences between groups were not statistically significant.
Panel A represents participants with a BMI of 27–35 kg/m2; panel B represents participants with a BMI of 30–35 kg/m2. *p < 0.05, in comparison with baseline. Risk changes are presented as percentage points. Blue bars, non-HBCSS groups; green bars, HBCSS groups. Lighter colours represent the change at 12 months and darker colours at 24 months.
Among participants with obesity (BMI 30–35), at 12 months, the reductions in overall 10-year risk, coronary artery disease risk, and stroke risk were significant only in the HBCSS group (Fig. 1, Supplementary Tables 2). However, the differences between the groups did not reach statistical significance (overall CVD risk p = 0.073, CAD risk p = 0.067, and stroke risk p = 0.248). At 24 months, the reduction in the HBCSS group remained significant in the overall and coronary artery disease risk, with no statistically significant differences between the groups.
The changes were also analysed among the six groups of the original study design, and the results were heterogeneous. There were no significant differences between groups or compared with the control arm for the overall 10-year CVD-risk, CAD risk, or stroke risk (Supplementary Table 3). Also, the results of the sensitivity analyses were similar. Those results are not shown. In addition, no statistically significant difference was found between the groups (p > 0.05) in analyses examining the usage data (Supplementary Fig. 3).
We have previously demonstrated that the HBCSS intervention effectively reduces weight for up to two years, supports long-term weight management, and reduces the need for antihypertensive medications for at least four years after the intervention18,19. In addition, the HBCSS effect in reducing the prevalence of metabolic syndrome has been previously described20. Here, we demonstrate that both HBCSS and non-HBCSS groups demonstrated decreased CVD risks in a two-group analysis after 12 months. After 24 months, only the HBCSS group maintained significant reductions in CVD risks. However, the differences between the HBCSS and non-HBCSS groups were not statistically significant.
When examining individual risk factors, the HBCSS intervention significantly reduced weight, waist circumference, and LDL cholesterol compared with the control group. However, as the overall risk of cardiovascular disease is a combination of individual risk factors, it is essential to look beyond the individual risk factors to the whole21. Additionally, when using a risk calculator, patients can more easily understand the estimated CVD risk.
FINRISK reflects the 10-year risk of developing and dying from specific CVD events. A risk level of <5% is considered low, and >10% is considered high9. The results obtained with the FINRISK calculator closely correlate with the actual prevalence, as the original FINRISK study found that in a low-risk group (< 5%), the disease prevalence was 1.7% among men and 1.5% among women, respectively9. The study population consisted primarily of middle-aged, apparently healthy individuals without known atherosclerotic disease. Therefore, the absolute risk percentages obtained at baseline and at 12 and 24 months are relatively low, compared with the Finnish population average22. However, even in such a healthy study population, the risk changes are significant, particularly among participants with obesity. This suggests that the higher the patient’s baseline risk, the greater the absolute benefit of the HBCSS intervention. From an economic perspective, preventing a single myocardial infarction saves an average of more than €20,000 in Finland during the first year alone23.
A recently published, extensive cohort study demonstrates that weight loss through midlife lifestyle changes is beneficial for long-term survival and mortality24. Obesity has been shown to significantly increase the risk of coronary heart disease and stroke, as well as other cardiometabolic and cardiovascular diseases, including atrial fibrillation25,26. HBCSS aims to guide individuals in making long-term lifestyle changes by altering thoughts, feelings, and behaviours, leading to weight loss. The HBCSS also aims to raise awareness about the impact of one’s lifestyle, feelings, and environment on their weight, health, and well-being. In addition to weight management, changes in eating behaviour can also lead to healthier eating habits. FINRISK assesses CVD risk based on individual-defined risk factors. Still, the actual risk of cardiovascular disease may also be influenced by well-being-related factors and biomarkers that cannot be assessed with FINRISK but can be influenced by HBCSS27,28.
Risk calculators are most applicable to the population on which they are based. As the trial was performed in Finland, it was reasonable to use the FINRISK calculator. It is also recognized by the Finnish National Current Care Guidelines. One study compared the previous version of the FINRISK calculator with other CVD risk calculators, including SCORE and Framingham22. The FINRISK and the Framingham model account for both mortality and morbidity, whereas the original SCORE model focused solely on mortality. The current SCORE2 provides a composite of cardiovascular mortality, non-fatal myocardial infarction, and non-fatal stroke. The Framingham morbidity data also include milder coronary cases, whereas the FINRISK model considers only major coronary events as outcomes. Therefore, although the results obtained with the different calculators are mostly similar, the various definitions of the end events yield slightly different risk scores22.
The original randomized trial design, along with the relatively large number of participants and the measurements controlled by healthcare professionals, supports the reliability of the results in this study. Confounding factors were minimized; however, the lack of more detailed family history and cardiovascular risk data may affect the baseline risk level. It should be acknowledged that there is an overlap between weight loss and the variables that constitute the CVD risk scores. Therefore, changes in the FINRISK score should be interpreted as a composite measure of the metabolic benefits of weight loss. As the inclusion criteria for this trial were BMI 27 to 35, the results reflect changes in the risk levels associated with overweight and mild, class I obesity.
In addition to offering a scalable solution for weight loss, our findings suggest that the HBCSS may lower the risk of cardiovascular diseases, especially in patients living with class I obesity. Further research is warranted to demonstrate the effectiveness of the HBCSS on the occurrence of actual CVD events.
The Prevent Metabolic Syndrome (PrevMetSyn) was designed as a randomized intervention trial that included a population-based sample of participants with overweight or class I obesity (BMI 27–35 kg/m2). The procedures in this trial followed the guidelines of the Declaration of Helsinki29. The study design is registered under ClinicalTrials.gov with registration ID NCT01959763. The Ethics Committee of the North Ostrobothnia Hospital District, Oulu, Finland, has approved the PrevMetSyn study with decision 29/2012 (March 26, 2012) and amendment statements dated November 23, 2012, February 18, 2013, and May 7, 2018.
The research data is stored at Oulu University Hospital, and access is limited to members of the research group. Individual identities are pseudonymized, and data are processed using research ID numbers. The participants gave written consent and had the option to withdraw from the study at any time. The principles of Good Clinical Practice were adhered to. The storage and use of software and personal data comply with the General Data Protection Regulation (GDPR).
Participants for this trial were recruited by sending an invitation letter to 11,400 people aged 20–60 years living in Oulu, Finland, according to the Population Registry. A total of 1065 people responded to the invitation and were assessed for eligibility. After exclusion, 532 volunteers with a BMI between 27 and 35 kg/m2, internet access, and no treatment for obesity or contraindications to weight loss were included. Exclusion criteria included uncontrolled health factors, such as abnormal laboratory values (thyroid, kidney, and liver function tests), or clinically significant illness that contraindicated weight loss or physical activity18.
The 532 enrolled participants were assigned to two groups that received different intensities of face-to-face group counselling or to a control arm. Each of these three groups was then further randomized into web-based HBCSS users and non-users (Supplementary Fig. 1). This paper focuses on the HBCSS intervention, and the groups are categorized into two groups: those with and those without HBCSS (Fig. 2).
Cardiovascular disease risks were assessed at baseline, 12 months, and 24 months. HBCSS, health behaviour change support system.
Face-to-face group counselling included cognitive-behavioural therapy (CBT) based counselling and self-help guidance (SHG). CBT-based group counselling included eight 90-min group sessions with two clinical nutritionists per group, aimed at helping participants recognize thoughts, feelings, and behaviours related to eating and weight, and to change them to support weight management and weight loss. The groups had eight to nine participants. Mindfulness exercises, psychoeducation, and individual, pair, and group exercises were used during the sessions, along with home tasks. CBT methods included techniques to identify and cope with dysfunctional thoughts, to recognize high-risk situations related to uncontrolled or unplanned eating, and problem-solving skills. Self-monitoring and planning were utilized to follow changes. SHG group counselling included two 90-min sessions based on the Elvira counselling model, utilizing the constructive learning theory and stage of change model30,31. SHG group sessions were led by a registered nurse and aimed to improve self-help skills to change lifestyle and eating habits. The control group received information about weight management and metabolic syndrome, which is the usual care provided in primary healthcare.
The HBCSS was a web-based information system designed for this trial, based on PSD combined with CBT methods. The HBCSS was a stand-alone 52-week programme designed to guide users toward a healthier lifestyle, improved eating behaviours, and weight loss. Self-monitoring was an important part of the system and included tools to track lifestyle progress, such as weight, meal rhythm, and physical activity. The HBCSS provided weekly information on health, weight loss, and exercises to recognize, change, and cope with eating, weight, and lifestyle-related issues. The software features of the PSD were designed to support users in performing their primary tasks, enhance computer-human interaction, and increase perceived credibility and social influence32,33. Participants interacted with the web-based information system weekly throughout the 52-week intervention. The content of the HBCSS was based on current knowledge of eating psychology, nutrition, and obesity treatment.
The original trial design included baseline and 12- and 24-month visits. Later, an additional 60-month visit was added19. Height, weight, and waist circumference were measured, and blood samples were collected by a nurse at the research centre of Oulu University Hospital, Finland. The weight change analysis, as outlined in the primary protocol, along with information on the treatment arms, sample size calculations, and randomization, has been presented in previous publications18,19,20.
We assessed the overall risk of cardiovascular diseases at baseline and at 12 and 24 months after the intervention period using the Finnish CVD risk calculator, FINRISK. The FINRISK is primarily designed to help clinicians assess an apparently healthy patient’s overall 10-year cardiovascular risk, which can be challenging9. It is actively used in Finland and has been utilized nationwide since 2007. The FINRISK calculator is based on extensive data from Finnish population surveys over the decades, including risk factor and morbidity data. It calculates a patient’s 10-year risk (%) for coronary artery disease, stroke, and an overall risk, which is the risk of having either of the two.
The FINRISK considers the patient’s sex, age, total cholesterol (mmol/l), high-density lipoprotein (HDL) cholesterol (mmol/l), smoking status, diabetes status, systolic blood pressure (mmHg), and family history of myocardial infarction and stroke. In our data, information on family history was not available. There is no generally accepted coefficient to describe the mean history of cardiovascular risk at the mean age of our study population. As this study examines intra-individual changes and family history is not a changing risk factor in this context, the coefficient was set to 0 in the formula.
The FINRISK calculator is designed for risk calculations for individuals aged 30 to 74. Although most participants fell within this age range, a few were under 30. They were included in the analysis because the change in risk is the study’s primary analysis. Additionally, age was fixed at the baseline when analysing changes from baseline to 12 and 24 months to minimize confounding variables that could affect the CVD risk scores.
The primary outcome was the absolute change (in percentage points) in CVD risk between baseline and 12 and 24 months. Changes in risks between HBCSS and non-HBCSS groups were analysed among all participants using the intention-to-treat (ITT) principle.
A linear mixed-effects model (LMM) was used to assess changes in FINRISK values from baseline to 12 and 24 months. LMM was also used to analyse changes in cardiovascular risk factors over the follow-up period. Group and time were used as fixed effects34. The model also included a group and time interaction term to evaluate differences in mean change over time between groups, adjusting for age and sex. Missing data were assumed to be missing at random. As the LMM allows analysis with unbalanced datasets, no imputation was needed for the primary analyses. The analyses were also conducted using an LMM with imputed values based on the return-to-baseline (RTB) method, and with completers only, as sensitivity analyses. Changes in FINRISK values from baseline to 12 and 24 months across six groups of the original study design were also analysed as a sensitivity analysis.
Differences in baseline characteristics between groups were analysed using Paired Samples t tests and Chi-square tests. Baseline CVD risk differences between groups were analysed using the Mann-Whitney U test.
IBM SPSS Statistics version 26.0 (IBM Corp., Armonk, NY, USA) was used for the analyses. p-values less than 0.05 were considered statistically significant.
The data supporting this study’s findings are available from the corresponding author upon request. The data are not publicly available due to privacy or ethical restrictions.
Malone, J. I. & Hansen, B. C. Does obesity cause type 2 diabetes mellitus (T2DM)? Or is it the opposite? Pediatr. Diabetes 20, 5–9 (2019).
Article PubMed Google Scholar
Ng, M. et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet Lond. Engl. 384, 766–781 (2014).
Article Google Scholar
Renehan, A. G., Tyson, M., Egger, M., Heller, R. F. & Zwahlen, M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet Lond. Engl. 371, 569–578 (2008).
Article Google Scholar
Truesdale, K. P. et al. Changes in risk factors for cardiovascular disease by baseline weight status in young adults who maintain or gain weight over 15 years: the CARDIA study. Int. J. Obes. 30, 1397–1407 (2006). 2005.
Article CAS Google Scholar
Townsend, N. et al. Cardiovascular disease in Europe: epidemiological update 2016. Eur. Heart J. 37, 3232–3245 (2016).
Article PubMed Google Scholar
Di Cesare, M. et al. The Heart of the World. Glob. Heart 19, 11 (2024).
Article PubMed PubMed Central Google Scholar
Global Cardiovascular Risk Consortium et al. Global Effect of Modifiable Risk Factors on Cardiovascular Disease and Mortality. N. Engl. J. Med. 389, 1273–1285 (2023).
Article Google Scholar
Koponen, P., Borodulin, K., Lundqvist, A., Sääksjärvi, K. & Koskinen, S. Terveys, toimintakyky ja hyvinvointi Suomessa – FinTerveys 2017 -tutkimus. (2018).
Vartiainen, E. et al. Sepelvaltimotaudin ja aivohalvauksen riskin arviointi FINRISKI 2.0-laskurilla. Suom Lääk L 75, 2778–2782 (2020).
Google Scholar
Conroy, R. M. et al. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur. Heart J. 24, 987–1003 (2003).
Article PubMed CAS Google Scholar
Hageman, S. H. J. et al. Prediction of individual lifetime cardiovascular risk and potential treatment benefit: development and recalibration of the LIFE-CVD2 model to four European risk regions. Eur. J. Prev. Cardiol. 31, 1690–1699 (2024).
Article PubMed PubMed Central Google Scholar
The Finnish Medical Society Duodecim. Obesity in children, adolescents and adults. Current Care Guidelines. (2025).
Grunvald, E. et al. AGA Clinical Practice Guideline on Pharmacological Interventions for Adults With Obesity. Gastroenterology 163, 1198–1225 (2022).
Article PubMed CAS Google Scholar
West, S. et al. Weight regain after cessation of medication for weight management: systematic review and meta-analysis. BMJ 392, e085304 (2026).
Article PubMed PubMed Central Google Scholar
Widmer, R. J. et al. Digital health interventions for the prevention of cardiovascular disease: a systematic review and meta-analysis. Mayo Clin. Proc. 90, 469–480 (2015).
Article PubMed PubMed Central Google Scholar
Xu, Y. et al. The efficacy of mobile health in alleviating risk factors related to the occurrence and development of coronary heart disease: A systematic review and meta-analysis. Clin. Cardiol. 44, 609–619 (2021).
Article PubMed PubMed Central Google Scholar
Sequi-Dominguez, I. et al. Effectiveness of Mobile Health Interventions Promoting Physical Activity and Lifestyle Interventions to Reduce Cardiovascular Risk Among Individuals With Metabolic Syndrome: Systematic Review and Meta-Analysis. J. Med. Internet Res. 22, e17790 (2020).
Article PubMed PubMed Central Google Scholar
Teeriniemi, A.-M. et al. A randomized clinical trial of the effectiveness of a Web-based health behaviour change support system and group lifestyle counselling on body weight loss in overweight and obese subjects: 2-year outcomes. J. Intern. Med. 284, 534–545 (2018).
Article PubMed Google Scholar
Turkkila, E. et al. Five-year follow-up of a randomized weight loss trial on a digital health behaviour change support system. Int. J. Obes. 49, 949–953 (2025).
Article Google Scholar
Seo, Y.-G. et al. Lifestyle counselling by persuasive information and communications technology reduces prevalence of metabolic syndrome in a dose–response manner: a randomized clinical trial (PrevMetSyn). Ann. Med. 52, 321–330 (2020).
Article PubMed PubMed Central Google Scholar
Jaspers, N. E. M. et al. Prediction of individualized lifetime benefit from cholesterol lowering, blood pressure lowering, antithrombotic therapy, and smoking cessation in apparently healthy people. Eur. Heart J. 41, 1190–1199 (2020).
Article PubMed PubMed Central CAS Google Scholar
Vartiainen, E., Laatikainen, T., Peltonen, M. & Salomaa, V. Verisuonisairauksien riskiarvio riippuu menetelmästä: vertailussa FINRISKI, SCORE ja Framingham. Suom Lääk L 65, 2079–2082 (2010).
Google Scholar
Iversen, T. & Häkkinen, U. Comparative treatment costs for patients with acute myocardial infarction between Finland and Norway. Nord. J. Health Econ. 6, 58–79 (2019).
Article Google Scholar
Strandberg, T. E. et al. Weight Loss in Midlife, Chronic Disease Incidence, and All-Cause Mortality During Extended Follow-Up. JAMA Netw. Open 8, e2511825 (2025).
Article PubMed PubMed Central Google Scholar
Orozco-Ruiz, X. & Kostev, K. Sex and Body Mass Index differences in the incidence of vascular and cardiometabolic diseases: A retrospective cohort study including 233 730 patients in Germany. Diabetes Obes. Metab. dom.70104 https://doi.org/10.1111/dom.70104 (2025).
Guh, D. P. et al. The incidence of co-morbidities related to obesity and overweight: A systematic review and meta-analysis. BMC Public Health 9, 88 (2009).
Article PubMed PubMed Central Google Scholar
Sin, N. L. The Protective Role of Positive Well-Being in Cardiovascular Disease: Review of Current Evidence, Mechanisms, and Clinical Implications. Curr. Cardiol. Rep. 18, 106 (2016).
Article PubMed PubMed Central Google Scholar
Hilvo, M. et al. Ceramide stearic to palmitic acid ratio predicts incident diabetes. Diabetologia 61, 1424–1434 (2018).
Article PubMed CAS Google Scholar
World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. JAMA 310, 2191 (2013).
Laitinen, J. et al. Feasibility of videoconferencing in lifestyle group counselling. Int. J. Circumpolar Health 69, 500–511 (2010).
Article PubMed Google Scholar
Nevanperä, N., Keränen, A.-M., Ukkola, O. & Laitinen, J. Effects of Group Counseling Transmitted Through Videoconferencing on Changes in Eating Behaviors. J. Nutr. Educ. Behav. 47, 555–559.e1 (2015).
Article PubMed Google Scholar
Oinas-Kukkonen, H. & Harjumaa, M. Persuasive Systems Design: Key Issues, Process Model, and System Features. Commun. Assoc. Inf. Syst. 24, (2009).
Oinas-Kukkonen, H. A foundation for the study of behavior change support systems. Pers. Ubiquitous Comput. 17, 1223–1235 (2013).
Article Google Scholar
Hox, J., Moerbeek, M. & Van De Schoot, R. Multilevel Analysis. https://doi.org/10.4324/9780203852279 (Routledge, 2010).
Download references
We would like to thank the research staff and the participants for their dedication and contributions to this study. The trial was supported by Suorsa Healthcare Foundation and the Finnish Foundation for Cardiovascular Research. The funders had no role in writing and publishing the paper. Open Access funding is provided by the University of Oulu (including Oulu University Hospital). This research is connected to the DigiHealth, Fibrobesity, and strategic profiling projects at the University of Oulu, Finland. The project is supported by the Academy of Finland (project numbers Profi5: 326291 and Profi6: 336449).
Research Unit of Biomedicine and Internal Medicine, University of Oulu, Oulu, Finland
Eero Turkkila, Heta Merikallio, Markku J. Savolainen, Laura Heikkilä, Tuire Salonurmi & Janne Hukkanen
Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland
Eero Turkkila, Heta Merikallio, Markku J. Savolainen, Laura Heikkilä, Tuire Salonurmi, Anna-Maria Teeriniemi & Janne Hukkanen
Oulu Advanced Research on Service and Information Systems, University of Oulu, Oulu, Finland
Harri Oinas-Kukkonen
Research Unit of Health Sciences and Technology, University of Oulu, Oulu, Finland
Tuire Salonurmi
Psychiatry, University of Oulu and Oulu University Hospital, Oulu, Finland
Terhi Jokelainen
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
M.J.S., H.M., J.H., A-.M.T., T.S., T.J., and H.O.K. participated in the study design, and L.H., T.S., T.J., and A-.M.T. contributed to data collection. E.T. conducted the statistical analyses and wrote the first draft. All the other authors participated in preparing the manuscript. All authors read and approved the final manuscript.
Correspondence to Eero Turkkila.
H.O.-K. and M.J.S. (founders) are shareholders of Onnikka Health Oy that owns the intellectual property rights of the HBCSS described in this paper. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and permissions
Turkkila, E., Merikallio, H., Savolainen, M.J. et al. Effect of a digital health behaviour change support system on cardiovascular disease risk in a randomized weight loss trial. npj Digit. Med. 9, 560 (2026). https://doi.org/10.1038/s41746-026-02747-7
Download citation
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41746-026-02747-7
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
Collection
Advertisement
npj Digital Medicine (npj Digit. Med.)
ISSN 2398-6352 (online)
© 2026 Springer Nature Limited
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
