HIV risk behavior self-report reliability at different recall periods

AIDS Behav. 2010 Feb;14(1):152-61. doi: 10.1007/s10461-009-9575-5. Epub 2009 May 28.


Few studies have investigated the optimal length of recall period for self-report of sex and drug-use behaviors. This meta-analysis of 28 studies examined the test-retest reliability of three commonly used recall periods: 1, 3, and 6 months. All three recall periods demonstrated acceptable test-retest reliability, with the exception of recall of needle sharing behaviors and 6-months recall of some sex behaviors. For most sex behaviors, a recall period of 3 months was found to produce the most reliable data; however, 6 months was best for recalling number of sex partners. Overall, shorter periods were found to be more reliable for recall of drug-use behaviors, though the most reliable length of recall period varied for different types of drugs. Implications of the findings and future directions for research are discussed.

Publication types

  • Meta-Analysis
  • Research Support, N.I.H., Extramural

MeSH terms

  • Female
  • HIV Infections / epidemiology*
  • Health Behavior
  • Humans
  • Male
  • Memory, Short-Term*
  • Reproducibility of Results
  • Risk-Taking*
  • Substance-Related Disorders / epidemiology
  • Surveys and Questionnaires*
  • Time Factors