Diabetes, cardiovascular events, and lifestyle : from epidemiology to clinic

Abstract

Today, the health care system plays an important role in the prevention of diseases, not least through support of a healthy lifestyle, including being physically active. Given the development of digital tools together with the importance of healthy lifestyle habits in prevention of most diseases, mHealth with self-management interventions has the potential to play a greater role than ever before. The growing implementation of digital solutions in health care shows a need for flexible and remote support as a complement to regular visits to the health care provider.

Physical activity (PA), a modifiable lifestyle factor, has been shown to be associated with a decreased risk of cardiovascular disease (CVD) in previous epidemiological studies. Some have suggested that PA affects the risk of CVD in women and men differently, and different types of PA may have different effects. Type 2 diabetes (T2D) and CVD share several characteristics including risk factors. Following an aging population and increases in risk factors ascribed to lifestyle, the prevalence of T2D is expected to arise in the coming decades. Currently, there is a rapid development in lifestyle promoting smartphone apps and devices for self-management targeted to persons with T2D. However, few previous app interventions in persons with T2D have targeted objectively measured daily PA alone. Moreover, an elevated blood pressure (BP) is the leading risk factor for CVD, and a common comorbidity in persons with T2D. Hence, a normotensive BP in persons with T2D is of great importance, who are also likely to use digital devices for self-measurement.

This thesis aims to study the association between total physical activity (TPA) and leisure time physical activity (LPA), and risk of stroke and myocardial infarction (MI) in women and men (study I), to study if a smartphone app intervention promoting PA by daily steps can improve time spent in moderate-to-vigorous intensity physical activity (MVPA), Health Related Quality of Life (HRQoL), and several other clinical variables in persons with T2D (study II-III), and to validate two automatic BP monitors with a Bluetooth function against manual BP monitoring in persons with T2D (study IV).

Study I was a prospective cohort study based on the Swedish National March Cohort (SNMC), following 31,580 individuals from 1997 to 2016. TPA and LPA were self-reported in the baseline questionnaire and outcomes of incident cases of MI and stroke were derived from national registers. A 22% lower risk of MI was seen in women in the highest tertile of TPA compared to the lowest tertile (HR: 0.78; 95% CI: 0.63–0.97). Among men, being in the highest tertile of LPA was associated with a lower risk of MI compared to belonging to the lowest tertile (HR: 0.78; 95% CI: 0.62–0.98), and a lower risk of stroke was seen among men in the highest tertile compared to the lowest (HR: 0.78; 95% CI: 0.61–0.99).

Study II-III was based on data from the DiaCert-study, a two-armed randomized controlled trial including 181 women and men with T2D. The intervention group was given access to the step promoting smartphone app DiaCert at baseline and 3 months onwards in addition to standard care, while the control group received standard care only. Outcomes of objectively measured MVPA (min/day) using accelerometers (primary outcome), BP, body mass index (BMI), waist circumference, HbA1c, blood lipids, and HRQoL (secondary outcomes) were assessed at baseline and at follow-ups at 3 and 6 months. The intervention had a positive effect and showed improvements in three HRQoL health concepts; role limitations caused by physical health problems, (−16.9; 95% CI −28.5 to −5.4), role limitations caused by emotional problems (−13.9; 95% CI −25.8 to −2.1), and emotional well-being (−5.7; 95% CI −10.4 to −1.0), in the intervention group compared to the control group after 3 months of intervention. No effect was observed on neither the primary outcome MVPA after 3 months, nor any of the clinical variables measured at 3 and 6 months.

In Study IV, two automatic BP monitors were validated against manual BP monitoring using baseline data from the DiaCert-study. The mean difference between the automatic monitor Beurer BM 85 and the manual BP monitor was 11.1 (SD 11.2) mmHg for systolic BP and 8.0 (SD 8.1) mmHg for diastolic BP. For the automatic monitor Andersson Lifesense BDR 2.0, the corresponding mean difference was 3.2 (SD 10.8) mmHg for systolic BP and 4.2 (SD 7.2) mmHg for diastolic BP.

In conclusion, our results highlight potential differences regarding LPA and TPA on the risk of MI and stroke between the sexes. The step promoting app intervention DiaCert showed an effect on health concepts reflecting both physical and emotional HRQoL, but no effect was found for MVPA or any of the cardiometabolic markers measured. Hence, future research is needed to conclude what type of mHealth solution that would not only improve HRQoL, but also be effective in supporting PA and improve cardiometabolic factors in this patient group. The validation of the two automatic BP monitors showed that one of the monitors differed in measurements within what could be clinically acceptable (Andersson Lifesense BDR 2.0), while the other did not (Beurer BM 85). Taken together, our results show that evaluation of the efficacy of mHealth PA interventions and validation of automatic BP monitors for home management are of importance to ensure the quality of the care, before implementing those in the health care system.