Effectiveness of Wearable Device-based Intervention on Glycemic Control in Patients with Type 2 Diabetes: A System Review and Meta-Analysis

Personalized Health Prediction AI Models Using Transfer Learning and Strategic Overfitting on Wearable Device Data

The increasing availability of wearable device data provides an opportunity for developing personalized models for health monitoring and condition prediction. Unlike conventional approaches that rely on pooled data from diverse individuals, our study explores the strategy of intentionally overfitting models to personal data and subsequently applying a transfer learning technique to refine performance for each user. We predicted Next-Day Condition (NDC) and Next-Day Emotion (NDC) while considering diverse features such as physical activity, sleep patterns, environmental context, and self-reported measures. Initial experiments showed that models trained at the sample level performed better on evaluation data but failed to generalize effectively during external validation. In contrast, our personalized learning approach, initiated with a pre-trained model, significantly enhanced accuracy within ten days of incremental user-specific training. Although generalization across the entire cohort diminished after individual tailoring, extended individualized training increased the overall predictive accuracy for each participant's personal data. The interpretation of feature importance using Shapley's additive explanations revealed substantial variability in the features influencing predictions across individuals, emphasizing the need for tailored health models. These findings highlight the potential of combining intentional overfitting and transfer learning in constructing high-performance user-specific predictive models from wearable data. Future research should expand the number of participants, extend the training period, and refine these methods to bolster personalized digital health solutions.

Multidisciplinary lifestyle treatment for type 2 diabetes in 12 European countries: protocol for a quasi-experimental study

BACKGROUND

The incidence and prevalence of type 2 diabetes (T2DM) are expected to continue rising. T2DM causes life-threatening, disabling and costly complications, and significantly reduces quality of life and life expectancy. The burden of T2DM can be reduced using comprehensive lifestyle modifications. The aim of this study is to evaluate the applicability and cost-effectiveness of a multicomponent, multidisciplinary lifestyle program in 22 European regions and to generate guidelines for transfer to European health care systems.

METHODS

A quasi-experimental study (without a control group) will be conducted to evaluate the CARE4DIABETES program, which is based on the Reverse Diabetes 2Now best practice. The program will involve more than 120 healthcare professionals and 860 people with T2DM from 12 European countries - Belgium, Bulgaria, Finland, Hungary, Italy, Greece, Malta, Poland, Portugal, Slovakia, Slovenia and Spain. Patients will be enrolled based on clinical criteria and motivation for change. The program will have two phases, an intensive phase (6 months) with face-to-face and online training to achieve behavioral change, and an online aftercare phase (6 months) to consolidate changes. The program will be evaluated for impact, sustainability and cost-effectiveness using a combination of validated questionnaires at baseline, six months and one year after the start of the intervention.

CLINICAL TRIAL NUMBER

Trial registration number: ISRCTN62063346.

Wearable Devices for Supporting Chronic Disease Self-Management: Scoping Review

BACKGROUND

People with chronic diseases can benefit from wearable devices in managing their health and encouraging healthy lifestyle habits. Wearables such as activity trackers or blood glucose monitoring devices can lead to positive health impacts, including improved physical activity adherence or better management of type 2 diabetes. Few literature reviews have focused on the intersection of various chronic diseases, the wearable devices used, and the outcomes evaluated in intervention studies, particularly in the context of primary health care.

OBJECTIVE

This study aims to identify and describe (1) the chronic diseases represented in intervention studies, (2) the types or combinations of wearables used, and (3) the health or health care outcomes assessed and measured.

METHODS

We conducted a scoping review following the Joanna Briggs Institute guidelines, searching the MEDLINE and Web of Science databases for studies published between 2012 and 2022. Pairs of reviewers independently screened titles and abstracts, applied the selection criteria, and performed full-text screening. We included interventions using wearables that automatically collected and transmitted data to adult populations with at least one chronic disease. We excluded studies with participants with only a predisposition to develop a chronic disease, hospitalized patients, patients with acute diseases, patients with active cancer, and cancer survivors. We included randomized controlled trials and cohort, pretest-posttest, observational, mixed methods, and qualitative studies.

RESULTS

After the removal of 1987 duplicates, we screened 4540 titles and abstracts. Of the remaining 304 articles after exclusions, we excluded 215 (70.7%) full texts and included 89 (29.3%). Of these 89 texts, 10 (11%) were related to the same interventions as those in the included studies, resulting in 79 studies being included. We structured the results according to chronic disease clusters: (1) diabetes, (2) heart failure, (3) other cardiovascular conditions, (4) hypertension, (5) multimorbidity and other combinations of chronic conditions, (6) chronic obstructive pulmonary disease, (7) chronic pain, (8) musculoskeletal conditions, and (9) asthma. Diabetes was the most frequent health condition (18/79, 23% of the studies), and wearable activity trackers were the most used (42/79, 53% of the studies). In the 79 included studies, 74 clinical, 73 behavioral, 36 patient technology experience, 28 health care system, and 25 holistic or biopsychosocial outcomes were reported.

CONCLUSIONS

This scoping review provides an overview of the wearable devices used in chronic disease self-management intervention studies, revealing disparities in both the range of chronic diseases studied and the variety of wearable devices used. These findings offer researchers valuable insights to further explore health care outcomes, validate the impact of concomitant device use, and expand their use to other chronic diseases.

TRIAL REGISTRATION

Open Science Framework Registries (OSF) s4wfm; https://osf.io/s4wfm.

A knowledge-based decision support system to support family doctors in personalizing type-2 diabetes mellitus medical nutrition therapy

BACKGROUND

Type-2 Diabetes Mellitus (T2D) is a growing concern worldwide, and family doctors are called to help diabetic patients manage this chronic disease, also with Medical Nutrition Therapy (MNT). However, MNT for Diabetes is usually standardized, while it would be much more effective if tailored to the patient. There is a gap in patient-tailored MNT which, if addressed, could support family doctors in delivering effective recommendations. In this context, decision support systems (DSSs) are valuable tools for physicians to support MNT for T2D patients - as long as DSSs are transparent to humans in their decision-making process. Indeed, the lack of transparency in data-driven DSS might hinder their adoption in clinical practice, thus leaving family physicians to adopt general nutrition guidelines provided by the national healthcare systems.

METHOD

This work presents a prototypical ontology-based clinical Decision Support System (OnT2D- DSS) aimed at assisting general practice doctors in managing T2D patients, specifically in creating a tailored dietary plan, leveraging clinical expert knowledge. OnT2D-DSS exploits clinical expert knowledge formalized as a domain ontology to identify a patient's phenotype and potential comorbidities, providing personalized MNT recommendations for macro- and micro-nutrient intake. The system can be accessed via a prototypical interface.

RESULTS

Two preliminary experiments are conducted to assess both the quality and correctness of the inferences provided by the system and the usability and acceptance of the OnT2D-DSS (conducted with nutrition experts and family doctors, respectively).

CONCLUSIONS

Overall, the system is deemed accurate by the nutrition experts and valuable by the family doctors, with minor suggestions for future improvements collected during the experiments.

Wearable monitoring device based on an internet management platform improves metabolic parameters in type 2 diabetes patients: a prospective pilot study

OBJECTIVES

This pilot study aimed to prospectively investigate the effects of a wearable monitoring device, based on an Internet management platform, on the comprehensive management of type 2 diabetes mellitus (T2DM) patients.

METHODS

A total of 120 hospitalized patients with T2DM were enrolled and randomly divided into the control group and the intervention group. Patients in the control group only received conventional diabetes treatments, while patients in the intervention group were provided with a wearable monitoring device in addition to conventional diabetes treatments. Moreover, the wearable device could connect to an Internet platform for diabetes management and upload self-monitoring data. All patients were followed for 3 months. The changes in parameters representing glucose metabolism, blood lipids, renal function, and patient satisfaction were compared between the two groups. All results were analyzed on an intention-to-treat basis.

RESULTS

One hundred twenty subjects met all criteria and agreed to participate in this study. During the follow-up period, 5 and 4 subjects were lost to follow-up in the intervention and control groups, respectively. Compared with the control group, the blood glucose of the intervention group decreased significantly after 3 months (p < 0.05). Subgroup analysis found that females, those younger than 60 years, with baseline glycated hemoglobin A1c (HbA1c) levels of 8% or greater, and patients with good adherence showed significant improvements in HbA1c (p < 0.05). However, there was no significant difference in blood lipid and renal function. The intervention group showed a better adherence rate to blood glucose, comprehensive adherence rate, and diabetes treatment satisfaction (p < 0.05). One subject in the intervention group and two subjects in the control group reported mild hypoglycemia. No other adverse events such as infections and skin allergies occurred in the two groups.

CONCLUSION

The intervention of a wearable monitoring device based on an Internet management platform significantly improved blood glucose control in T2DM patients, as well as the overall adherence rate and patient satisfaction with treatment.

CLINICAL TRIAL REGISTRATION

NCT04973644.

Longitudinal Studies of Wearables in Patients with Diabetes: Key Issues and Solutions

Glucose monitoring is key to the management of diabetes mellitus to maintain optimal glucose control whilst avoiding hypoglycemia. Non-invasive continuous glucose monitoring techniques have evolved considerably to replace finger prick testing, but still require sensor insertion. Physiological variables, such as heart rate and pulse pressure, change with blood glucose, especially during hypoglycemia, and could be used to predict hypoglycemia. To validate this approach, clinical studies that contemporaneously acquire physiological and continuous glucose variables are required. In this work, we provide insights from a clinical study undertaken to study the relationship between physiological variables obtained from a number of wearables and glucose levels. The clinical study included three screening tests to assess neuropathy and acquired data using wearable devices from 60 participants for four days. We highlight the challenges and provide recommendations to mitigate issues that may impact the validity of data capture to enable a valid interpretation of the outcomes.

Effectiveness of Wearable Activity Monitors on Metabolic Outcomes in Patients With Type 2 Diabetes: A Systematic Review and Meta-Analysis

OBJECTIVE

Wearable activity monitors are promising tools for improving metabolic outcomes in patients with type 2 diabetes mellitus (T2DM); however, no uniform conclusive evidence is available. This study aimed to evaluate the effects of the intervention using wearable activity monitors on blood glucose, blood pressure, blood lipid, weight, waist circumference, and body mass index (BMI) in individuals with T2DM.

METHODS

Two independent reviewers searched 4 online databases (PubMed, Cochrane Library, Web of Science, and Embase) to identify relevant studies published from January 2000 to October 2022. The primary outcome indicator was hemoglobin A1c (HbA1c), and the secondary outcome indicators included physical activity (steps per day), fasting blood glucose, triglyceride, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, total cholesterol, systolic blood pressure, diastolic blood pressure, BMI, waist circumference, and weight.

RESULTS

A total of 25 studies were included. The HbA1c level (standardized mean difference [SMD], -0.14; 95% confidence interval [CI], -0.27 to -0.02; P = .02; I² = 48%), BMI (SMD, -0.16; 95% CI, -0.26 to -0.05; P = .002; I² = 0), waist circumference (SMD, -0.21; 95% CI, -0.34 to -0.09; P < .001; I² = 0), and steps/day (SMD, 0.55; 95% CI, 0.36-0.94; P < .001; I² = 77%) significantly improved.

CONCLUSION

Wearable activity monitor-based interventions could facilitate the improvement of the HbA1c level, BMI, and waist circumference and increase in physical activity in individuals with T2DM. Wearable technology appeared to be an effective tool for the self-management of T2DM; however, there is insufficient evidence about its long-term effect.