Abstract
Newly developed validated frailty assessment instrument (FAI) was used to assess the prevalence of frailty in Colombo district, Sri Lanka. Criteria to diagnose frailty were developed giving weights according to the magnitude of the eigenvalues of principal components in domains of FAI yielded from factor analysis. Cross sectional descriptive and analytical study was conducted among 1620 elders selected by multistage cluster sampling method. Prevalence of frailty in Colombo district was 14.9% (95%CI:13.2%- 16.6%). Prevalence rates of frailty were higher in males (17.7%, 95%CI:15.0% - 20.4%) and elders below the age of 75 years (11.3%, 95%CI: 9.4% - 13.2%) than females (12.4%, 95%CI: 10.2% - 14.6%) and elders aged 75 years or more (21.6%, 95%CI: 18.2% - 25.0%). Age, gender, marital status, education, past medical history, physical activities, nutrition, Body Mass Index (BMI), activities of daily living (ADL) and instrumental activities of daily living (IADL), memory, living index, environment index, economic status, depression, duration of non communicable diseases (NCDs), number of drugs using, social contacts, social support and consumption of medical facilities during the preceding six months are significantly associated with frailty status at 5% level. However, number of drugs using (OR–0.86, 95%CI:0.77–0.95), admissions to hospitals (OR–4.86, 95%CI:2.92–8.08), age (OR–1.06, 95%CI:1.0 –1.09), IADL (OR–1.13, 95%CI:1.09–1.18), unsatisfactory environment (OR – 3.46, 95%CI: 2.17–5.51) and economic dependency (OR–7.09, 95%CI:2.87–17.51) were significant correlates at 5% level in logistic regression models. Prevalence rates are relatively high and early interventions to prevent frailty are recommended.