Predictive analytics models have generated both excitement and debate over their use as a tool for suicide prevention. The authors in a study published in Psychiatry Services wanted to get the patient perspective. If a physician tells someone he or she is a suicide risk, how would they perceive that information? To find out, the authors surveyed high-risk veterans hospitalized in a psychiatric inpatient unit.

The researchers recruited patients from a large Veteran’s Affairs  facility at high risk for being identified by the predictive program Recovery Engagement and Coordination for Health — Veterans Enhanced Treatment (REACH VET). Out of 220 patients approached, 102 surveys were returned.

The survey included demographic information and 3 vignettes that described conversations between clinicians and patients about a suicide prevention program. A 5-point, 5-item scale followed each vignette.

All 3 vignettes were rated neutral to very caring by 80% of respondents. Script C was rated as neutral to very caring by 95% (N=88) of respondents. The overall difference in vignette ratings was significant (x2=11.17, df=2, P =.004). Post hoc comparisons were not significant.


Continue Reading

All 3 vignettes were rated neutral to very helpful by 90% of respondents, neutral to very informative by at least 88% of respondents, and neutral to very encouraging by at least 78% of respondents.

Most participants (78%) said hearing one of these scripts would not make them feel hopeless or helpless.

Advice for clinicians included tips such as “be genuine and compassionate” and “really care.” A few had concerns about using a predictive analytics program: Why did it flag me? Will my treatment change?

The fact that the results may not generalize to all REACH VET patients or the general population was one of the limitations of this study.

“Although these results support the feasibility of using predictive analytics for suicide prevention, this preliminary study represents only a first step; more research is needed to examine a variety of stakeholder opinions utilizing improved methods,” the authors concluded.

Reference

Reger MA, Ammerman BA, Carter SP, et al. Patient Feedback on the Use of Predictive Analytics for Suicide Prevention. Psychiatr Serv. 2020 Nov 3:appips202000092. doi: 10.1176/appi.ps.202000092