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.

    39/128 (page 37)
    Phase 3: Rejection of Individual Patients Who Qualified Phase 4: Patient Willingness or Refusal to Participate Chapter 2 Methods for Assessing Randomized Studies’ Generalizability to the Patient Population (Task 1) Remington (1989, p. I-67) has charged that “patient exclusion criteria represent a vast wasteland of clinical trial design.” He notes that, although investigators have been “relatively precise” in defining target groups and patient selection criteria, this has unfortunately not been the case for rejections of patients who had qualified under the initial criteria: “ .. Many [investigators] ... have been very crude in defining patient exclusion criteria ... in general ... [using only] a brief statement ... or such global phrases as ‘serious intercurrent disease.” Similarly, Liberati, Himel, and T.C. Chalmers (1986) report that fully two-thirds of the breast cancer treatment trials they assessed did not even mention having kept a log of patients rejected. Exclusion of individual patients can have important cumulative effects on the representativeness of the entire patient pool. To cite one example, in reviewing the British Medical Research Council Trial, Remington (1989, p. I-67) points out that, of the 46,000 patients identified as eligible for this trial on the basis of measured blood pressure, more than half were excluded. Apparently, the excluded patients were those with worse-than-average prognoses since “ultimate mortality rates [were] much lower than expected in general populations with similar elevations of blood pressure.” The Chalmers rating scale includes an item on each randomized study’s exclusions of patients; specifically, exclusions are judged on whether or not “a log [had been kept] of patients who had been seen by the investigators, but rejected before randomization as ineligible with listed reasons” (Liberati, Himel, and T.C. Chalmers, 1986, p. 945). Judgments of whether exclusions were adequately described—‘Yes,” “Partial,” “No,” or “Unknown”—are associated with a prescribed number of points (see T.C. Chalmers et al., 1981, p. 46). In certain illness or treatment areas, the reluctance of patients to participate in a randomized study (or of their physicians to refer them to the study) can seriously affect representativeness. Edlund, Craig, and Richardson (1985) call for primary investigators to provide more information on patient willingness to participate, noting that of 84 treatment studies reviewed, none had reported the number of refusers and nonrefusers.