Cross design synthesis : a new strategy for medical effectiveness research : report to Congressional requesters / United States General Accounting Office.

  • United States. General Accounting Office.
Date:
[1992]
Catalogue details

Licence: Public Domain Mark

Credit: Cross design synthesis : a new strategy for medical effectiveness research : report to Congressional requesters / United States General Accounting Office. Source: Wellcome Collection.

    40/128 (page 38)
    Improving Methodological Assessments Chapter 2 Methods for Assessing Randomized Studies’ Generalizability to the Patient Population (Task 1) Depending on the recruitment mode, patients (or primary physicians) may find it easy to avoid participation in randomized studies. In such instances, investigators cannot count individual refusals, but potential problems can be indicated by difficulties in achieving recruitment targets, as has occurred in randomized studies comparing lumpectomy with mastectomy (Taylor et al., 1984), or by low rates of participation in targeted clinics or hospitals, as has occurred in trials evaluating the intracranial/extracranial bypass (Barnett et al., 1987). An example of the potential importance of patient refusals is Schooler’s (1980, p. 30) description of a trial evaluating a drug used to prevent relapse in schizophrenia: “The hypothesis to be tested was that guaranteeing receipt of medication would — significantly decrease the number of patients who would relapse, and would also delay relapse for those who did ultimately relapse. Contrary to prediction, there were no differences in relapse rate between the two groups....[P]atients whose medication taking was controlled relapsed as early and as often as those who had to take oral medication daily. “This study has been criticized on sampling grounds. It has been suggested that a significant treatment effect was not found because patients who refuse to enter a drug trial are the same patients who will be noncompliant with treatment, and that the restriction of the sample to those who consented to be studied also restricted the study to subjects who would take oral medication and therefore not show a treatment effect.” (Emphasis added.) Representativeness of patients participating in a randomized study can be threatened by each of the four phases of the patient selection process—the target group definition, recruitment mode, rejection of individual patients, and patient willingness or unwillingness to participate. Alternatively, success in each phase can ensure representativeness. One example of a trial that:comes close to ensuring representation in every phase would be the placebo trials that tested the Salk polio vaccine. These trials targeted all public schoolchildren in grades 1 through 3, recruited from all schools in communities at high risk of polio and apparently did not reject any of the children who appeared at the clinics. Further,