Abstract
The prevalence of mood disorders has increased globally. People with mood disorders have been found to use more health services than the general population, although a mood disorder diagnosis does not necessarily entail utilization of health services. This heterogeneity in health services utilization could make it difficult for governments to plan resources to meet the needs of people with mood disorders. A patient-level linked database from residents of Quebec, Canada was used to model care trajectories of people who self-reported having been diagnosed with a mood disorder. The data from the Canadian Community Health Survey were linked to health administrative data for a 21-year period. We used latent class analysis and latent profile analysis to group people into categories. Four care trajectories were identified using the latent class analysis: 1) people who only used services of a general practitioner; 2) people having seen a psychiatrist or having at least one ED visit or hospitalization; 3) people consulting other types of specialists; 4) null utilization. The latent profile analysis on medical services yielded four profiles, with average numbers of services of 41, 33, 7, and 1, while that on hospitalization yielded two profiles, with 20 % of the population having had at least one hospitalization and the remainder none. By classifying people into service utilization groups, these methods enable determining needs for a given population and can support resource allocation for health care decision makers.