Objective-To determine the prevalence of potentially inappropriate long term prescribing in general practice in the United Kingdom.Design-Review of 62 studies of the appropriateness of prescribing identified from seven electronic databases, from reference lists, and by hand searching ofjournals. A nominal group of 10 experts helped to define the appropriateness of prescribing.Setting-General practice in the United Kingdom.Main outcome measures-Prevalences of 19 indicators of inappropriate long term prescribing representing five dimensions: indication, choice of drug, drug administration, communication, and review.Results-Prevalences ofpotentially inappropriate prescribing varied by indicator and chronic condition, but drug dosages outside the therapeutic range consistently recorded the highest rates. The lowest rates were generally associated with indicators of the choice of the drug, except cost minimisation. Communication is studied less frequently than other dimensions of prescribing appropriateness.Conclusions-The evidence base to support allegations of widespread inappropriate prescribing in general practice is unsound. Although inappropriate prescribing has occurred, the scale of the problem is unknown because oflimitations associated with selection of a standard, publication bias, and uncertainty about the context of prescribing decisions. Opportunities for cost savings and effectiveness gains are thus unclear. Indicators applicable to individual patients could yield evidence ofprescribing appropriateness.