Predictors of Suicide in First-Episode Affective and Nonaffective Psychotic Inpatients: Five-Year Follow-Up of Patients From a Catchment Area in Vitoria, Spain

Ana González-Pinto, M.D., Ph.D.; Ana Aldama, Ph.D.; Cristina González, Ps.; Fernando Mosquera, M.D.; Marta Arrasate, M.D.; and Eduard Vieta, M.D., Ph.D.

Objective: To determine the baseline demographic and clinical characteristics associated with suicide attempts over 5 years following a first admission for psychosis and to assess the relationship between suicidal behavior during 5-year follow-up and clinical factors.

Method: All inpatients consecutively admitted between February 1997 and January 1999 with a first psychotic episode from a specific catchment area in Spain were included at baseline; they were followed up yearly over a 5-year period with an extensive protocol that included the Structured Clinical Interview for DSM-IV, the Positive and Negative Syndrome Scale, and the 21-item Hamilton Rating Scale for Depression. The primary outcome measure was suicide attempts. Comorbidity with alcohol and drug abuse was recorded, as were all suicidal acts made by patients over the follow-up period. Relationships between suicidal outcome and baseline clinical features were examined. Logistic regression modeling was used to test the significance and independence of associations of relevant factors to suicidal status.

Results: Of 83 first-episode psychotic patients, 14.5% displayed suicidal behavior within the 5 years following the first admission, and 2.4% died by suicide. Suicide rate was 0.48%/year and attempt rate was 2.89%/year, with a 1.5-fold (95% CI = 1.07 to 2.22) greater risk for each depressive symptom at index episode and with an 8-fold (95% CI = 1.45 to 44.40) higher risk among patients with baseline stimulant abuse (cocaine and amphetamine).

Conclusions: Patients with a first-episode psychotic disorder seemed to be a high-risk population for suicidal behavior. Depressive symptoms during the index psychotic episode and comorbidity with stimulant abuse at baseline were relevant predictive factors for suicidal behavior during the first years of first affective and nonaffective psychotic episodes.

(J Clin Psychiatry 2007;68:242-247)

Received Nov. 25, 2005; accepted June 26, 2006. From the Stanley Institute International Mood-Disorders Research Center, Vitoria (Drs. González-Pinto, Aldama, Mosquera, and Arrasate and Ms. González); and the Bipolar Disorders Program, Stanley Research Center, Hospital Clinic, University of Barcelona, IDIBAPS (Institut d'Investigacions Biomèdiques August Pi i Sunyer), Barcelona (Dr. Vieta), Spain.

Preparation of this article was supported by Health Research Funds from the Spanish Government (FIS: PI021297; PI05/2761, G03/032; CM04/00101; CM05/00127), European Regional Development Funds, and the Health Department of the Basque Government (2004/11016). The research psychiatric departments in Santiago Apóstol Hospital and Barcelona University Hospital Clinic are supported by the Stanley Research Foundation (03-RC-003).

Drs. Gonzáles-Pinto, Aldama, Mosquera, and Arrasate and Ms. González report no other financial relationships relevant to the subject of this article. Dr. Vieta has received grant/research support from Bial and grant/research support and honoraria from AstraZeneca, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, Janssen-Cilag, Novartis, Pfizer, Sanofi, and Servier; and serves on the speakers or advisory boards for Almirall, AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Janssen-Cilag, Lundbeck, Novartis, Organon, Pfizer, Sanofi, Servier, and UCB.

Corresponding author and reprints: Ana González-Pinto, M.D., Ph.D., Director, Stanley International Mood Disorders Research Center, Hospital Santiago Apóstol, University of the Basque Country, Olaguibel 29, Vitoria 01004, Spain (e-mail: agonzalez@hsan.osakidetza.net).