Which version of the General Health Questionnaire should be used in community studies?

Literature pertinent to the use of the various versions of the General Health Questionnaire (GHQ) is briefly surveyed. The literature indicates that the 60-item version has the highest reliability and validity coefficients, the lowest misclassification rate, and the highest sensitivity and specificity. Using data from a previously published community survey in Perth using the 60-item GHQ, comparisons are made between use of the 60-, 30-, 20- and 12-item versions, including community prevalence rates, simultaneous identification of cases, disagreement rates and patterns of community rates with age. The 60-item version gives the lowest prevalence rates. It is argued that the 60-item GHQ is probably the best as it gives the lowest misclassification rates, the smallest standard error of estimated prevalence rates and allows measurement of certain subscales not contained within the shorter versions.