Management of suicide-related behaviors in a hospital is challenging. This article (1) describes integration of an electronic suicide risk notification system to improve assessment of psychiatric inpatients, (2) details the manner in which these alerts complement standard of care, and (3) provides support of using aggregate data to inform administrative decision-making. Complementing routine clinical care and under the supervision of an assessment coordinator, adult inpatients at a specialty psychiatric hospital complete a computerized battery of outcome assessments throughout the course of their hospitalization. A critical-item response notification system for suicide-related behaviors was implemented within the larger, assessment architecture, sending an alert e-mail to unit staff if patients endorsed suicidal ideation on any 1 of 6 critical items. Analysis of aggregate data over a 19-month period reveal a linear trend of increasing rate of suicide alerts from October 2012 to April 2013 (Phase A) with a stabilization at the heightened level from July 2013 to April 2014 (Phase B), R = 0.697, P = .007. Findings suggest that more nuanced training in the management of suicide-related behavior may be necessary and that traditional approaches to staffing may need to accommodate patient acuity. The communication innovation of this system is in line with the Joint Commission's emphasis on designing and implementing patient-centered systems that enhance quality of care, including improved safety.
The Menninger Clinic, Houston, Texas (Drs Madan, Mahoney, Allen, Ellis, Hardesty, Oldham, and Fowler); Menninger Department of Psychiatry and Behavioral Sciences, Baylor College of Medicine, Houston, Texas (Drs Madan, Allen, Ellis, Hardesty, Oldham, and Fowler).
Correspondence: Alok Madan, PhD, MPH, The Menninger Clinic, 12301 South Main St, Houston, TX 77035 (firstname.lastname@example.org).
This research was partially supported by the Menninger Foundation and McNair Medical Institute. Dr Madan is a McNair Scholar. The study follows the guidelines on good publication practices. The study sponsors were not involved in any aspect of the research activities and did not approve the specific protocol or manuscript. Thus, the authors were independent from study sponsors in the context of the research.
None of the other authors have any other relevant conflicts of interest to declare.