Objective and subjective physical activity assessment in obese patients with eating disorders

Abstract Background Discrepancies between objective and subjective measures of physical activity have been reported in healthy populations but remain unexplored in adults with obesity and psychosomatic disorders. Exercise is key for reducing these disorders, yet physiological and psychosomatic factors may complicate valid activity assessments in this population. Here, we compare self-reporting with actigraph data in a clinical setting. Methods Over two years, 27 obese patients (21 female, 6 male; 43±14 yrs; 41.3±5.7 kg·m−2) with eating disorders (EDE-Q) were recruited from an Austrian Psychosomatic Center. Total physical activity was assessed using the metabolic equivalent (MET min/w) and sitting time (min/d). Objective data were collected using the SenseWear® wristband (SWA) during weeks 3-4 of patients’ clinical stay. SWA was worn for ≥18 h/d on ≥ 4 days, including one weekend day. Patients self-administered the International Physical Activity Questionnaire (IPAQ-SF) in alignment with the ∼7 days of SWA wear. Concurrent validity - SWA as a criterion - was assessed using regression analyses, and proportional bias was examined with regression and equality line similarity analyses. Results Comparison showed an overestimation of total activity (+45%; p<.001) and underestimation of sitting time (-28%; p<.001) with IPAQ-SF vs SWA. Within-patient correlations were poor for MET (r=.92, p<.001) and impractical for sitting time (r=.27, p=.018), with a proportional bias for both outcomes (p≤.001). Converting objective to subjective measures yielded typical errors of 232 (total activity) and 90 min (sitting time). Conclusions The observed disparities emphasize the need to integrate sensor-based assessment of physical activity into clinical practice or tailored public health initiatives for adult obesity and psychosomatic disorders. However, further research is needed to evaluate the long-term impact of incorporating accurate SWA data on intervention efficacy in this vulnerable group. Key messages • There was strong evidence for meaningful differences in objective vs subjective physical activity assessments in obese patients with eating disorders. • Integrating objective measures into treatment seems crucial to improve exercise-guided interventions and address the global obesity and psychosomatic disorders pandemic.