Behavioral headache research: methodologic considerations and research design alternatives.

Behavioral headache treatments have garnered solid empirical support in recent years, but there is substantial opportunity to strengthen the next generation of studies with improved methods and consistency across studies. Recently, Guidelines for Trials of Behavioral Treatments for Recurrent Headache were published to facilitate the production of high-quality research. The present article compliments the guidelines with a discussion of methodologic and research design considerations. Since there is no research design that is applicable in every situation, selecting an appropriate research design is fundamental to producing meaningful results. Investigators in behavioral headache and other areas of research consider the developmental phase of the research, the principle objectives of the project, and the sources of error or alternative interpretations in selecting a design. Phases of clinical trials typically include pilot studies, efficacy studies, and effectiveness studies. These trials may be categorized as primarily pragmatic or explanatory. The most appropriate research designs for these different phases and different objectives vary on such characteristics as sample size and assignment to condition, types of control conditions, periods or frequency of measurement, and the dimensions along which comparisons are made. A research design also must fit within constraints on available resources. There are a large number of potential research designs that can be used and considering these characteristics allows selection of appropriate research designs.