Mobile Health Applications and Physical Activity Adherence among Urban Pakistani Adults: A Multi-City Descriptive Study
Main Article Content
Abstract
Background: Physical inactivity is a major public health challenge in Pakistan, where fewer than 15% of adults meet WHO recommendations for weekly physical activity. Mobile health (mHealth) applications have emerged as scalable, low-cost tools for health promotion; however, their effectiveness within Pakistan's distinct sociocultural and infrastructural context remains largely unexplored.
Objective: To characterize the prevalence and patterns of mHealth physical activity (PA) application use, evaluate user experience and application quality, assess adherence trajectories, and identify barriers and determinants of adherence among Pakistani adults.
Methods: A descriptive cross-sectional design with an embedded qualitative component was employed. A purposively stratified sample of 384 adults (mean age: 31.7 years; 50% female) was recruited across Lahore, Karachi, and Islamabad. Instruments included the International Physical Activity Questionnaire–Short Form (IPAQ-SF), Mobile Application Rating Scale (MARS), Technology Acceptance Model constructs, and a validated barrier checklist. Descriptive statistics and binary logistic regression were performed using SPSS v25.0; qualitative data from 30 semi-structured interviews were analyzed thematically.
Results: Sixty-two percent of participants (61.7%; 95% CI: 56.7–66.5%) reported current or past use of PA applications. App users demonstrated significantly higher mean physical activity (2,214 vs. 1,287 MET-min/week; p<0.001). Applications received moderate quality ratings (MARS total: 3.42/5.0), with Functionality scoring highest (3.71) and Engagement lowest (3.18). High adherence (≥4 days/week) was maintained by 48.5% of users. Significant predictors of adherence included perceived usefulness (aOR=2.31), app personalization (aOR=2.14), social features (aOR=1.98), and exercise self-efficacy (aOR=1.87), while female gender was a negative predictor (aOR=0.54; p=0.001). Primary barriers included inconsistent internet connectivity (67.4%), English-language interfaces (54.2%), affordability concerns (42.1%), and cultural incongruence (38.7%).