Repeated Diagnoses of Lifetime Alcohol Use Disorders in a Prospective Study: Insights into the Extent and Nature of the Reliability and Validity Problem
Article first published online: 22 AUG 2013
Copyright © 2013 by the Research Society on Alcoholism
Alcoholism: Clinical and Experimental Research
How to Cite
Haeny, A. M., Littlefield, A. K. and Sher, K. J. (2013), Repeated Diagnoses of Lifetime Alcohol Use Disorders in a Prospective Study: Insights into the Extent and Nature of the Reliability and Validity Problem. Alcoholism: Clinical and Experimental Research. doi: 10.1111/acer.12237
- Article first published online: 22 AUG 2013
- Manuscript Accepted: 2 JUL 2013
- Manuscript Received: 19 FEB 2013
- NIAAA. Grant Numbers: F31 AA019596, T32 AA13526, R01 AA13987, R37 AA07231, KO5 AA017242
- Lifetime Diagnosis;
- Lifetime Prevalence;
- Alcohol Use Disorder;
- Negative Prevalence
Prior research indicates that assessments of lifetime alcohol use disorders (AUDs) show low sensitivity and are unreliable when assessed by a single, retrospective interview. This study sought to replicate and extend previous research by calculating the lifetime prevalence rate of AUDs using both single retrospective assessments of lifetime diagnosis and repeated assessments of both lifetime and past-year diagnoses over a 16-year period within the same high-risk sample. In addition, this study examined factors that contributed to the consistency in reporting lifetime AUDs over time.
Using prospective data, the reliability and validity of lifetime estimates of alcohol dependence and AUD were examined in several ways. Data were drawn from a cohort of young adults at high and low risk for alcoholism, originally ascertained as first-time college freshmen (N = 489 at baseline) at a large, public university and assessed over 16 years.
Compared with using a single, lifetime retrospective assessment of DSM-III disorders assessed at approximately age 34, lifetime estimates derived from using multiple, prospective assessments of both past-year and lifetime AUD were substantially higher (25% single lifetime vs. 41% cumulative past-year vs. 46% cumulative lifetime). This pattern of findings was also found when conducting these comparisons at the symptom level. Further, these results suggest that some factors (e.g., symptoms endorsed, prior consistency in reporting of a lifetime AUD, and family history status) are associated with the consistency in reporting lifetime AUDs over time.
Based on these findings, lifetime diagnoses using a single measurement occasion should be interpreted with considerable caution given they appear to produce potentially large prevalence underestimates. These results provide further insight into the extent and nature of the reliability and validity problem with lifetime AUDs.