The 130th Annual Meeting of APHA

Abstract #41547

David Dausey, MPhil, Department of Health Policy, Yale University, 17 Pardee Place, New Haven, CT 06515, (203) 389-0795, David.Dausey@Yale.edu, Rani Desai, PhD, Psychiatry, Yale University, 950 Campbell Avenue, West Haven, CT 06516, and Robert Rosenheck, MD, Psychiatry and Public Health, Yale University, Building 8, 950 Campbell Ave, West Haven, CT 06516.

This study examined whether three groups of variables were associated with suicide among psychiatric inpatients in the year following discharge: (1) sociodemographic factors, (2) individual quality of care measures, and (3) institutional quality of care measures. We utilized administrative data on 121,752 psychiatric inpatients from across the US discharged from an inpatient psychiatric program in the VA Health Care System between January 1, 1994 and December 31, 1998. VA administrative data were merged with data from the National Death Index to determine mortality and causes of death. Proportional hazards models were fit to test the study objectives while controlling for covariables. There were 481 total suicides for an overall suicide rate of 40.59 per 10,000 person years. After controlling for significant sociodemographic variables (gender, age, race, level of service connected disability, and diagnosis), quality of care measures were significantly associated with suicide risk. Specifically, lengths of stay shorter than 15 days, poorer continuity of care, and lack of outpatient care within 30 days after discharge were significantly associated with increased suicide rates. Institutional qualities of care measures were not associated with suicide risk. Quality of care is significantly associated with suicide risk. The implications are three-fold: (1) suicide rates may be a useful quality indicator for large systems of care; (2) modifiable factors such as length of stay may be associated with an individuals risk of suicide, and (3) there may be little effect of facility level quality of care above and beyond the effect of individual quality of care.

Learning Objectives: At the end of the session, the participant will be able to

• List 5 risk factors for sucide among the mentally ill
• Articulate situations in which suicide rates might be used as quality of care measures
• Develop a protocol to investigate quailty of care with suicide as an outcome in large systems of care

Keywords: Suicide, Health Care Delivery

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: Veteran's Affairs Health Care System
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.