Reliability of alcohol use indices. The Lifetime Drinking History and the MAST.

TLDR

Alcohol use indices are widely employed for diagnosis, treatment decisions, and outcome evaluation, but their reliability can be influenced by assessment context, detoxification status, and the specific alcohol‑related behaviors measured. The study examines factors that influence test‑retest reliability of the Michigan Alcoholism Screening Test and the Lifetime Drinking History, a structured interview that quantifies lifetime alcohol consumption patterns. The authors assess test‑retest consistency of these instruments to determine the stability of reported drinking behavior across assessment occasions.

Abstract

are often used as a means of collecting information on drinking behavior for the diagnosis of alcohol misuse, for making treatment decisions and for evaluating treatment outcome. However, various elements can influence the reliability and validity of these data, including the assessment context, detoxication status and the kinds of alcohol-related behaviors measured (1). In the present study, factors which influence test-retest reliability are examined for the widely used Michigan Alcoholism Screening Test (MAST) (2) and for the Lifetime Drinking History (LDH), 2 a structured interview that quantifies lifetime patterns of alcohol consumption. Reliability may be defined as the consistency of an individual's reporting of drinking behavior both within a single assessment (internal consistency) as well as between two assessment occasions (test-retest). Reliability addresses only the reprodueibility or stability of indices related to alcohol use, and is a necessary but not sufficient condition for the validity of a measure. Since the extent of measurement error sets a mathematical upper bound on any