To analyse the consequences of broadening DSM-IV criteria for generalized anxiety disorder (GAD) on the utilization of health care resources and corresponding costs.

Multicentre, prospective and observational study conducted in outpatient psychiatric clinics selected at random and weighted by geographical density of population. Patients with GAD according to DSM-IV criteria and subjects with anxiety symptoms fulfilling broader criteria were compared. Broadening criteria was considered 1-month of excessive or non-excessive worry and only 2 associated symptoms listed on DSM-IV for GAD. Socio-demographic data, medical history and health care resources and corresponding costs were recorded during a 6-month period.

A total of 3,549 patients were systematically recruited; 12.8% excluded because not fulfilling inclusion criteria, 1,815 patients in DSM-IV criteria group (DG) and 1,264 in broad criteria group (BG). Both groups were similar on their sociodemographic characteristics at baseline. Type of treatments prescribed at psychiatric clinics during the study were similar in frequency; anti-depressives (77.0% in DG vs. 75.3% in BG, ns), benzodiazepines (71.5% vs. 67.2% respectively, ns), and anti-convulsants (72.1% vs. 67.0% respectively, ns). Health care resources utilization were statistically reduced to a similar extent in both groups as a consequences of treatments yielding to a cost-of-illness in the 6-month period of 1,196 (1,158) and 1,112 (874), respectively; p=0.304.

In a large sample of subjects, broadening of GAD criteria could lead to earlier diagnosis that would not be associated necessarily to an increase in health care resources utilization or costs to the National Health System.