Systematische Übersicht und Bewertung digitaler Interventionen zur Diabetesprävention und –versorgung

Background: The prevalence of type 2 diabetes mellitus (T2D) and the number of patients with comorbidities like hypertension, dyslipidemia and cardiovascular diseases are increasing worldwide. Evidence-based medicine uses the best available evidence from systematic research to make decisions about the care of individual patients. The systematisation and appraisal of evidence are done in care guidelines, which in turn aim to guide the application of effective diabetes prevention and care interventions in different age groups and settings. Patients with T2D need continuous and individualised care. They are therefore seen as the ideal target group for the use of digital health interventions like telemedicine. However, heterogeneous patient populations, telemedicine phenotypes and settings hamper the evaluation of digital health interventions. Comparing study results to provide evidence-based recommendations is further complicated by the diversity of applied study designs. Therefore, there is a need for a systematic review of the current state of research while considering the described variability. In line with this aim five research studies were conducted. Objective: The overall objective of this thesis was, to identify current needs of patients with diabetes (publication 1), to systematically analyse the effectiveness of different diabetes prevention and care interventions (publications 2+3) and to evaluate digital diabetes prevention and care interventions (publications 4+5). Material und Method: To analyse current needs of patients with diabetes, three substudies were conducted in the beginning (publication 1). They consisted of a standardised survey of experts to analyse existing chronic care programs, an expert workshop to identify patients’ needs and an online survey to prioritise the categorised needs dimensions seen from the perspective of patients and health care providers. Two literature overviews were performed to analyse the best available evidence in diabetes prevention and care. An umbrella review analysed the available evidence to identify effective interventions of blood sugar regulation on cardiovascular risk (publication 2). Study quality was assessed using OQAQ (Overview Quality Assessment Questionnaire). Afterwards, a literature overview aimed to identify effective measures of population-based prevention and communication strategies to provide recommendations for policy makers on how to prevent diabetes in different age groups and settings (publication 3). In a next step, digital diabetes prevention and care interventions were summarised. To evaluate digital health interventions with more than one active function, a study protocol was developed. It describes the evaluation of a hypothetical gamification-based smartphone application for weight loss in overweight and obese adolescents (publication 4). As a last step, an umbrella review (publication 5) systematically analysed the effectiveness of telemedicine interventions in diabetes, dyslipidaemia and hypertension. Potentially relevant records had to analyse the effectiveness of telemedicine on clinical outcomes under real-life conditions in patients with one of the defined target diseases using either a systematic review or meta-analysis based on RCTs. Results of meta-analyses and their subgroup analyses were used to identify effective components or other characteristics (e.g. intensity or frequency of feedback). Overall certainty of outcomes was assessed using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) tool. Results and Implications: The standardised survey on current care models indicated that a lack of national guidelines, cost-ineffective and non-individualised health care as well as long waiting periods were criticised. Education of patients, communication within the team and with the patient, prevention and health promotion as well as the accessibility of services were significantly more important to patients when compared to health care providers. The identified differences in priorities support the early assessment of these preferences. The umbrella review on the potential of blood sugar regulation for the reduction of cardiovascular risk identified 44 records which were of good quality (OQAQ-median = 17). The results suggest that pharmacological and non-pharmacological interventions have the potential to improve cardiovascular outcomes. When deciding for a certain intervention as well as its intensity, baseline blood pressure and cardiovascular risks of the patient should be considered. Guidelines on cardiovascular prevention should take into account pathophysiological mechanisms as well as individual lifestyle interventions. While effective measures for individual level prevention including physical activity and diet programs were found, available evidence for population-based intervention was scarce and insufficient. The literature overview conducted afterwards identified evidence-based interventions for population-based prevention, including taxation of unhealthy products and specific prevention strategies in certain settings (e.g. kindergartens, schools). These strategies may contribute to the development of policies and governmental regulations for the prevention of diabetes in different age groups and settings. To evaluate a digital mobile health intervention consisting of more than one component, a study protocol for a single-centre, two-arm, triple-blinded, randomised controlled trial following the CONSORT recommendations was developed. The intervention consists of a smartphone application that provides both tracking and gamification elements for lifestyle change. The control group uses an identically designed application, which solely features the tracking of health information. It appears favourable to use RCTs for proof of concept assessments, to evaluate the effectiveness of an app or specific components in controlled settings. The fifth publication of this thesis shows that telemedicine may lead to significant and clinically relevant reductions of HbA1c (≤ -0,5 %) in patients with T2D. The identified reduction rates are comparable to those of non-pharmacological and even some pharmacological interventions. Extracted subgroup analyses showed that certain population and intervention characteristics seem to be associated with improved clinical benefits. This applies to interventions with a rather short duration (< 6 months) and those with frequent or intensive feedback. Age (< 55 years), time since diagnosis (< 8,5 years) and baseline HbA1c (> 8,0 %) were identified as population characteristics favourable for clinically relevant improvements of HbA1c. In addition to the characteristics, future updates of guidelines should carefully consider the low levels of certainty as indicated by the low GRADE results. The present thesis provides a systematic overview of effective measures in diabetes prevention and care. Overall, there is a potential for the early and structured assessment of patients’ preferences. The systematisation and appraisal of the best available evidence on the effectiveness of telemedicine in patients with diabetes and associated comorbidities revealed areas for the update of present guidelines. There is a need for methodologically robust studies on the effectiveness of telemedicine in specific populations and in consideration of combined digital health components. The results and identified research needs have the potential to motivate future studies.

Level thesis:degree_level

thesis.doctoral

Grantor dc:publisher

Technische Universität Dresden

Year

2020

Author dc:creator

• Timpel, Patrick

Contributors dc:contributor

• Schwarz, Peter
• Schmitt, Jochen

• Diabetes
• Telemedizin
• Evidenz
• Digitalisierung
• Versorgungsforschung
• telemedicine
• evidence
• digitalization
• healthcare research