Drawing valid inferences from case studies.

Although the case study is usually accorded an important heuristic role, its limitations as a basis for drawing valid conclusions about the efficacy of treatment are widely recognized. The major problem with a case study is the ambiguity related to the precise influences that are responsible for change. Threats to internal validity, normally ruled out in experimentation, make ambiguous the basis for therapeutic change. However, case studies may vary along several dimensions that dictate the extent to which specific threats to internal validity are addressed. The present article discusses the possibility of conceptualizing and conducting cases in such a way as to maximize the likelihood that internally valid conclusions can be drawn. Although case studies can never replace experimentation, the scientific yield from case reports might be improved in clinical practice where methodological alternatives are unavailable.