We examined the use of theory in studies identified in a systematic review of rigorous evaluations of clinical practice guideline dissemination and implementation strategies. The full methods and results of the systematic review are available elsewhere . Briefly, we searched Medline, EMBASE, Health Star, the Cochrane Controlled Trials Register, and SIGLE (System for Information on Grey Literature in Europe) using a highly sensitive search strategy developed for the Cochrane Effective Practice and Organisation of Care (EPOC) group between 1976 and 1998 . Searches were not restricted by language or publication type. We included cluster and individual randomized controlled trials, controlled clinical trials, controlled before and after studies, and interrupted time series that evaluated any guideline dissemination or implementation strategy targeting physicians and that reported an objective measure of provider behavior and/or patient outcome. Two reviewers independently screened the search results and assessed studies against the inclusion criteria. Disagreements were resolved by consensus. The final sample included 285 reports of 235 studies yielding 309 comparisons of guideline dissemination and implementation strategies.
For the purposes of the current study, we identified whether included studies had used a theory to inform the design of an intervention and/or the interpretation of the results. A study was considered to have used a theory if the authors stated that they had done so within the report of the study, preferably with a source reference and/or an explanation of how the theory was proposed to explain the phenomenon to which it had been applied. Where a study described a framework or approach that appeared to be theoretically based, but the authors had not explicitly stated that they had used a theory, a decision was made by two reviewers regarding whether or not the study should be classified as theory-based or not. PD read all papers to identify whether or not they used theory. In cases of uncertainty, papers were considered by a second reviewer (AW), and a consensus was reached about whether or not these studies should be classified as having used theory or not.
We classified all papers using a descriptive framework that considered the level of theory use and the stage at which theory was used (Appendix 1). Level of theory use reflects the intensiveness of use of theory within studies. Studies judged to have used theories were classified within the first two categories ('explicitly theory based'; 'some conceptual basis'). Studies using individual constructs from theories, e.g., knowledge, attitude, self-efficacy, that were not reported within a theoretical framework were classified as 'construct(s) (unrelated to theory)'. In cases where a study employed more than one theory, each instance of theory use was classified separately using the framework. Studies in the first three categories--explicitly theory-based; some conceptual basis; constructs (unrelated to theory)--were further classified according to the stage of the research process at which the theory (or construct) was used. For the purposes of the review, the stage of use categories were treated as being mutually exclusive, i.e., each instance of theory use was assigned to one category only (the first stage of the research process at which the theory has been integrated). This is not to say that studies employing a theory at one stage could not, or did not, use the same theory at any other stage.