Citation

Booth CL. Am. J. Prev. Med. 2014; 47(Suppl): S106-S114.

Affiliation

Suicide Prevention Branch, Division of Prevention, Traumatic Stress, and Special Programs, Substance Abuse and Mental Health Services Administration, Rockville, Maryland. Electronic address: chelsea.booth@samhsa.hhs.gov.

Copyright

(Copyright © 2014, Elsevier Publishing)

DOI

10.1016/j.amepre.2014.05.021

PMID

25145727

Abstract

BACKGROUND: The Research Prioritization Task Force of the National Action Alliance for Suicide Prevention conducted a stakeholder survey including 716 respondents from 49 U.S. states and 18 foreign countries.

PURPOSE: To conduct a qualitative analysis on responses from individuals representing four main stakeholder groups: attempt and loss survivors, researchers, providers, and policy/administrators. This article focuses on a qualitative analysis of the early-round, open-ended responses collected in a modified online Delphi process, and, as an illustration of the research method, focuses on analysis of respondents' views of the role of life and emotional skills in suicide prevention.

METHODS: Content analysis was performed using both inductive and deductive code and category development and systematic qualitative methods. After the inductive coding was completed, the same data set was re-coded using the 12 Aspirational Goals (AGs) identified by the Delphi process.

RESULTS: Codes and thematic categories produced from the inductive coding process were, in some cases, very similar or identical to the 12 AGs (i.e., those dealing with risk and protective factors, provider training, preventing reattempts, and stigma). Other codes highlighted areas that were not identified as important in the Delphi process (e.g., cultural/social factors of suicide, substance use).

CONCLUSIONS: Qualitative and mixed-methods research are essential to the future of suicide prevention work. By design, qualitative research is explorative and appropriate for complex, culturally embedded social issues such as suicide. Such research can be used to generate hypotheses for testing and, as in this analysis, illuminate areas that would be missed in an approach that imposed predetermined categories on data.

Language: en