Inconsistent Reporting Affects the Link Between Substance Use and Suicide Attempt in Adolescents.

Although past research has documented substantial rates of inconsistent reporters on substance use and suicidality, most studies within these fields overlook this issue. This study (a) estimated the prevalence of inconsistent reporters on substance use and suicide attempt and (b) assessed the extent to which including or excluding these respondents affects the link between these variables. The study involved a representative sample of 5,592 secondary-school attendees ( = 15.9; 52.6% female) and included measures of alcohol use, inebriation experience, cannabis use, cigarette use, and suicide attempt. Prevalence estimates were computed based on (1) raw data, (2) a domain-specific exclusion of inconsistent reporters, and (3) the removal of all inconsistent reporters. Logistic regressions were conducted alternately with raw data and without inconsistent reporters to assess how inconsistent reporting bears on the link between substance use and suicide attempt. About 26% of the participants provided inconsistent information on substance use and about 15%, on suicide attempt. In total, 36% of the participants were identified as inconsistent reporters. Excluding inconsistent reporters affected the prevalence of substance use and suicide attempt in general and the prevalence of abstainers in particular. Their removal also affected the link between substance use and suicide attempt: inebriation experience and cannabis use were no longer significantly associated with suicide attempt once inconsistent reporters were excluded. Inconsistent reporting should be more systematically considered in research on substance use and suicidality. Researchers and policy makers should interpret findings from studies neglecting this key issue with caution.