Digital Health in Chronic Conditions: Promoting Physical Activity, Assessing Digital Health Literacy, and Profiling Readiness

Introduction
Chronic conditions are the leading causes of global morbidity and mortality and their prevalence continues to rise. A major common behavioral risk factor for chronic conditions such as depression, anxiety, heart disease, Chronic Obstructive Pulmonary Disease (COPD), osteoarthritis (OA), hypertension, and type 2 diabetes is physical inactivity. Being physically inactive increases morbidity and mortality risks, while regular physical activity can slow progression and reduce the severity of these conditions. Despite these benefits, physical inactivity remains widespread, highlighting the need for innovative health strategies to alleviate the growing burden on healthcare systems and economic resources. Digital health solutions, such as mobile applications (apps), websites, or wearables, have emerged. They offer promising avenues to address this challenge by enabling personalized remote care, empowering people to manage their health, and facilitating scalable health interventions. However, the effectiveness of digital health interventions in promoting physical activity for people with chronic condition management is not yet fully understood. It may depend on digital health literacy and digital readiness. Therefore, it is essential to investigate the effect of these digital technologies and assess digital health literacy and digital readiness to tailor digital health solutions for diverse populations, ensuring that these successfully improve physical activity and help reduce rather than increase health disparities. 

Aim
This thesis aims to assess aspects of digital health usage to leverage health outcomes for individuals with chronic conditions with an emphasis on physical activity. It focuses on the benefits and harms of digital health intervention that promotes physical activity among people with chronic conditions. It also examines the relationship between digital health literacy, physical activity levels, and the interaction of chronic conditions. Additionally, it investigates digital readiness profiles among people with hip or knee OA and identifies the factors differentiating various digital readiness levels.

Methods
Paper I was a systematic review and meta-analysis of randomized controlled trials (RCTs), searching MEDLINE, Embase, CINAHL, and the Cochrane Central Register until October 2022. Eligible RCTs evaluated digital health interventions targeting physical activity versus usual care in adults with conditions such as depression, anxiety, ischemic heart disease, heart failure, COPD, hip or knee OA, hypertension, or type 2 diabetes. Primary outcomes were objectively measured physical activity and physical function, while secondary outcomes included subjectively measured physical activity, physical function, anxiety, depression, health-related quality of life (HRQOL), and adverse events at two time points: the end of the interventions and the follow-up closest to 12 months. Data were pooled using random effects models, and the risk of bias and evidence certainty was assessed with the Cochrane Risk of Bias 2 tool (RoB 2) and the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach.

Paper II was a cross-sectional study based on data from the Health Survey in Region Zealand, Denmark. Digital health literacy was measured using three eHealth Literacy Questionnaire (eHLQ) subscales (measuring the ability to process health information, feeling safe and in control, and engagement motivation). Physical activity was assessed by the Nordic Physical Activity Questionnaire-short form, compliance with the World Health Organization (WHO) minimum recommendation for physical activity, and the Saltin-Grimby Physical Activity Level Scale. Associations between digital health literacy and physical activity levels were analyzed using logistic regression, controlling for demographic and health variables. Interactions with long-term health conditions (somatic and mental health) were also evaluated.

Paper III was a cross-sectional study with data from the Good Life with osteoArthritis in Denmark (GLA:D) register. Digital readiness was measured using the eHealth Readiness Scale (eHRS). Latent class analysis identified digital readiness profiles, and multinomial logistic regression examined sociodemographic and health characteristics associated with profile membership.

Results
The systematic review in Paper I included 130 RCTs and found that digital health interventions significantly increased objectively measured physical activity (Standard mean differences (SMDs): 0.29 (95% confidence interval (CI) 0.21 to 0.37) at intervention end; 0.17 (95% CI 0.04 to 0.31) at follow-up) and physical function (SMDs: 0.36 (95% CI 0.12 to 0.59) at intervention end; 0.29 (95% CI 0.01 to 0.57) at follow-up) compared to usual care. Similar effects were found for the secondary outcomes but these were not sustained at follow-up. The digital health interventions had higher non-serious adverse event risks (risk ratio (RR): 1.31 (95% CI 1.11 to 1.55) at the interventions end but not at follow-up (RR 1.35 95% CI 0.05 to 0.66) or for serious adverse events (RRs: 0.89 (95% CI 0.76 to 1.04; 0.76 95% Ci 0.56 to 2.54). High heterogeneity and poor study quality resulted in moderate to very low certainty of evidence. 

Paper II included 19,231 participants and found that higher digital health literacy across all three eHLQ scales was positively associated with engaging in >30 minutes weekly moderate-to-vigorous physical activity (Odds Ratios (ORs): 1.13–1.25, p≤0.012), be compliant to physical activity recommendations (ORs 1.08–1.33, p≤0.025), and being self-reported active (ORs 1.24–1.54, p<0.001), even after adjusting for covariates. No interactions were observed with long-term health conditions. Most had high digital health literacy scores, but people with more long-term health conditions and a desire for support to increase physical activity had the lowest digital health literacy scores.

Paper III assessed digital readiness in 3,555 participants and found that 53% were confident using internet technologies, 32% felt it improved efficiency, and 26% used devices to track their lifestyles. Latent class analysis identified low, intermediate, and high readiness profiles. Compared to the high profile, the low profile was associated with older age (OR 1.96, 95% CI 1.71-2.24), being female (OR 0.72, 95% CI 0.57-0.90), lower education (OR 0.62, 95% CI 0.45-0.88), living alone (OR 1.39, 95% CI 1.11-1.76), and more comorbidities (OR 1.10, 95% CI 1.04-1.17). While the intermediate profile largely mirrored this, being in the intermediate was less likely among obese individuals (0.70, 95% CI 0.54-0.91) and those with higher walking speed (0.70, 95% CI 0.50-0.98) compared to the high profile. Being sedentary was more likely in the high profile compared to both low and intermediate (OR: low: 1.63, 95% CI 1.32 to 2.02 and intermediate: 1.21 95%CI 1.02 to 1.42)

Conclusion
This thesis revealed that digital health interventions effectively enhance physical activity and physical function among individuals with various chronic conditions. Improvements in anxiety, depression, and HRQOL were, however, only seen at the end of the interventions. The risk of nonserious adverse events during these interventions highlights the need for careful management. Higher digital health literacy was associated with increased physical activity levels, emphasizing the importance of assessing and promoting digital health literacy when utilizing digital health solutions. Identifying distinct digital readiness profiles indicates that digital readiness should be considered to optimize the adoption of digital health solutions in clinical settings. It is essential to tailor interventions to diverse personal needs to ensure that digital health solutions contribute to reducing health disparities, particularly among older adults, those with lower education levels, and people managing multiple chronic conditions. Support or alternative options for those not ready to engage with digital health solutions may be needed to ensure equitable access and effective participation in health interventions. Future research should focus on improving reporting, ensuring sustainable effects, and developing strategies to screen, measure, and enhance digital health literacy and digital readiness.