Objective: To estimate the direct and indirect costs of osteoarthritis (OA) in an active population, and to identify factors significantly influencing these expenditures.
Methods: A cohort of 3,440 subjects employed by the Liège City Council was followed prospectively for 6 months. Subjects were asked to report monthly OA related health resource utilization (contacts with health professionals, medical examinations, drug consumption, etc.) and absence from work. Health related quality of life (HRQOL) was evaluated at baseline using the Medical Outcomes Study Short-form 36 (SF-36). Logistic regression analysis identified factors associated with the probability that the individual incurred costs, and multiple regression identified factors influencing the magnitude of these costs.
Results: A total of 1,811 subjects filled in at least one questionnaire (response rate 52%). The mean duration of followup was 3.46 months. Self-reported prevalence of OA was 34.1%. The mean total direct costs were 44.5 euros per OA patient-month. Contacts with health professionals, medical examinations, drugs, and hospital stays accounted for 23.7 euros, 8.7 euros, 6.7 euros, and 4.9 euros, respectively, per OA patient-month. The average number of sick-leave days was 0.8 per OA patient-month. From a payer's perspective, this loss of productivity represented a mean cost of 64.5 euros per OA patient-month. We also recorded 0.02 mean days off work per active subject-month due to informal care by relatives, yielding a mean cost of 1.8 euro per active subject-month for the employer. Poorer scores for most of the dimensions of the SF-36 at baseline were significantly associated with greater likelihood of incurring direct and indirect costs and with higher costs among subjects who reported costs. If we consider the overall cohort of active subjects, the burden of OA related to the direct and indirect costs was 15.2 euros and 23.8 euros, respectively, per active subject-month.
Conclusion: Direct and indirect costs attributable to OA are substantial, with productivity related costs being predominant. Poorer HRQOL was a major determinant of these expenditures.