Category: Suicide and Self-Injury
Suicide is the 10th leading cause of death in the U.S. (CDC, 2013). In order to ultimately predict and prevent this devastating outcome, it is critical that both researchers and clinicians be able to effectively ask about and receive accurate reports of suicidal ideation (SI) from those affected. Prior research suggests that fewer than half of people with SI disclose their thoughts spontaneously and/or after probing (Eskin et al., 2015). Additionally, research identifies a number of incentives (e.g., desire to be understood/supported, trust in confidentiality) and impediments (e.g., perceived stigma, concerns about unwanted treatment such as hospitalization or medication) to accurate reporting of SI (Hom et al., 2017).
To further explore how people report SI, we examined the consistency in participant response across multiple modes of probing. Modes of probing included: phone interview, in-person interview, non-confidential self-report, and confidential self-report. Participants included 200 veterans recruited from the local Boston community. Participant consistency in reporting lifetime SI and SI recency were examined among those recruited as “suicide ideators” (i.e., reported active SI in the past 30 days via phone interview) and “depressed controls” (i.e., reported no active SI in their entire lifetime and screened positive for depression via phone interview). Most were male (89.5%) and White (72.5%), with a mean age of 43 (range 19 to 81 years). SI items came primarily from the Self-Injurious Thoughts and Behaviors Interview (Nock et al., 2007).
Data show a number of inconsistencies in reporting lifetime presence and recency of SI across both suicide ideators and depressed controls. For instance, among depressed controls, who reported no lifetime SI via phone interview, 12.5% reported lifetime SI via in-person interview and 24.1% reported lifetime SI via confidential self-report. To examine one potential explanation for these discrepancies, we looked at participant self-reports of understanding and responding to questions. Overall, participants reported carefully reading and accurately responding to research questions with ease (79.2%), or carefully reading and accurately responding though finding it hard “to give one number that best describes me” (19.5%). Only a small percentage reported answering inaccurately (1.3%). Despite general participant efforts to understand questions and answer accurately, inconsistencies in reporting SI exist. Additional analyses examine unique factors associated with inconsistent responders, including demographic factors (e.g., age, gender, race/ethnicity), suicidal severity (i.e., Beck Scale for Suicide Ideation) and mental health functioning (i.e., using the PHQ-9 and Psychache Scale).
Results suggest that despite general efforts to understand questions and to answer accurately, some people respond inconsistently when asked about SI across multiple modes of probing. Analyses describe inconsistencies, explore characteristics of the inconsistent responders, and make recommendations for how researchers and clinicians can most effectively ask about and receive accurate information about SI.
Charlene Deming– Clinical Psychology PhD Candidate, Harvard University, Cambridge
Franchesca Ramirez– Clinical Research Coordinator, Harvard University, Cambridge, Massachusetts
Julia Harris– Graduate Student, University of Utah, Utah
Andrew Huckins-Noss– Harvard University
Matthew Nock– Professor, Harvard University, Cambridge, Massachusetts
Clinical Research Coordinator