Hazard and determinants of dropout and rehospitalization in patients with obesity after residential rehabilitation

Abstract

Purpose: To identify clinical and sociodemographic factors that predict follow-up discontinuation and rehospitalisation after multidisciplinary residential rehabilitation for severe obesity, thereby defining high-risk patient profiles and guiding tailored retention strategies. Methods: We retrospectively followed 1,851 adults with obesity discharged from a multidisciplinary residential programme between 2015 and 2018 (median BMI 42 kg m⁻²). Dropout, defined as more than twelve months without contact, was studied with discrete-time survival models; time to rehospitalisation was analysed with Cox regression. Results: Within twelve months 1,513 patients (87%) discontinued follow-up. Each five-year increase in age lowered drop-out risk (HR 0.97, 95% CI 0.94–0.99, p = 0.004); diabetes had a similar protective effect (HR 0.89, 0.79–1.00, p = 0.0455). Rehospitalisation occurred in 591 patients (32%). Risk increased with age (5-years increment; HR = 1.05, 95% CI 1.01–1.09, p = 0.0191), baseline BMI (HR = 1.04, 95% CI 1.03–1.05, p < 0.0001), diabetes (HR = 1.22, 95% CI 1.02–1.30, p = 0.0306) and eating disorders (HR = 1.48, 95% CI 1.07–2.05, p = 0.0193). Discussion: Maintaining the benefits of residential rehabilitation is important. In our cohort, 87% of patients dropped out of follow-up within one year and 32% were readmitted. Two distinct profiles emerged: younger and non-diabetic subjects were prone to dropout, while patients with higher BMI, diabetes, or eating disorders were at higher risk of rehospitalization. Early identification of these groups may suggest flexible, technology-assisted follow-up for working-age patients and integrated metabolic-psychiatric care for complex cases, safeguarding outcomes and optimizing resources.