Prevalence of suicidal behaviours in two Australian general population surveys: methodological considerations when comparing across studies

Abstract

PURPOSE: To investigate whether methodological differences between two Australian general population surveys have the capacity to affect the apparent prevalence rates of suicidal ideation and suicide attempts.

METHODS: 609 Wave 1 of the Personality and Total Health (PATH) Through Life Project participants, and 83 participants derived from the 1997 National Survey of Mental Health and Wellbeing (NSMHWB) met the criteria for inclusion (suicidal ideation/suicide attempt). Analysis involved Chi-square and binary logistic regression.

RESULTS: Twelve-month prevalence rates for suicidal ideation and suicide attempt were 8.2%, (95% CI = 7.6-8.8) and 0.8% (95% CI = 0.6-1.0) for PATH (N = 7,485), and contrast with 2.9% (95% CI = 2.6-3.2) and 0.3% (95% CI = 0.2-0.5) for NSMHWB (N = 10,641) samples, respectively. While notable discrepancies are apparent between the prevalence statistics, both sets of statistics are within the bounds of other Australian and international studies. Parallel rate disparities for suicidal ideation are found across age-by-gender groups. Aside from differences in the basic prevalence rates, surveys have analogous age-by-gender profiles for suicidal ideation.

CONCLUSIONS: While it is possible that samples are representative of the populations from which they are derived, 12-month prevalence rate discrepancies between PATH and NSMHWB surveys are likely to originate from demographic and survey methodology differences. Where investigations employ different methodologies, especially in relation to modes of survey administration and the assessment items utilised, a cautious approach should be taken when comparing findings.