Research Briefs

Balas, E.A., Austin, S.M., Ewigman, B.G., and others (1995). "Methods of randomized controlled clinical trials in health services research." (AHCPR grant HS07268). Medical Care 33(7), pp. 687-699.

The authors analyzed the methodologies employed by health services researchers who conduct randomized controlled clinical trials to evaluate the cost, quality, and effectiveness of health care services. They created and validated a streamlined quality evaluation tool and used it to analyze 101 trials from the Columbia Registry of Controlled Clinical Trials. Of a possible score of 100, the trials received an average score of 54.8. Significant quality deficiencies showed up in sample size, description of case selection, data on possible adverse effects, analysis of secondary effect variables, and retrospective analysis. The researchers found their quality evaluation tool to be useful as a checklist that could prompt investigators to provide pertinent information when designing a trial and reporting its results.

Bombardier, C., Melfi, C.A., Paul, J., and others (1995). "Comparison of a generic and a disease-specific measure of pain and physical function after knee replacement surgery." (AHCPR grant HS06432). Medical Care 33(4), pp. AS131-AS144.

Both generic and disease-specific health status instruments are commonly used to assess patients' outcomes. According to this Total Knee Replacement Patient Outcomes Research Team (PORT) study, these instruments are complementary, and both are needed to fully assess patient outcomes. The PORT researchers used a sample of patients 67 to 99 years of age who had undergone knee replacement surgery 2 to 7 years earlier. They compared patients' scores on a generic health-related quality-of-life (HRQOL) SF-36 measure with their scores on the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index. The distribution of scores in pain, physical function, and overall score were consistently higher on the WOMAC than on the SF-36, indicating less disability from arthritis than from other conditions after knee surgery in this elderly population. The WOMAC index was better able to discriminate persons with more or less severe knee problems, whereas the SF-36 discriminated better among persons with varying levels of overall health status and coexisting health problems.

Glanz, M., Klawansky, S., Stason, W., and others (1995, June). "Biofeedback therapy in poststroke rehabilitation: A meta-analysis of the randomized controlled trials." (AHCPR poststroke guideline development contract 282-91-0085). Archives of Physical Medicine and Rehabilitation 76, pp. 508-515.

Approximately 150,000 individuals die each year as a result of strokes, and 60 percent of stroke survivors suffer some degree of long-term disability. Evidence for the efficacy of rehabilitative efforts for stroke patients has been lacking. These researchers pooled results of a meta-analysis of randomized controlled trials of biofeedback therapy to assess its efficacy in poststroke rehabilitation. Biofeedback was applied to a paretic (weak or paralyzed) limb of patients in the rehabilitative phase of their illness. Patients in both treatment and control groups received standard physical therapy. According to the researchers, results do not support the efficacy of biofeedback in restoring the range of motion of hemiparetic joints, that is, joints on one side of the body that are weak or paralyzed. The researchers caution, however, that their findings are based on relatively few studies with small individual sample sizes and that the studies themselves raise questions in terms of population homogeneity and the randomization process used.

Guadagnoli, E., Cleary, P.D., and McNeil, B.J. (1995). "The influence of socioeconomic status on change in health status after hospitalization." (AHCPR grant HS06341). Science and Medicine 40(10), pp. 1399-1406.

An association between socioeconomic status (SES) and health status has been well established. Poorer, less educated patients experience higher rates of serious illness, spend more time in bed because of illness, and report fewer health care visits than higher SES patients. A recent study found that hospitalization has little effect on gaps in health status that exist between high and low SES patients. The researchers used hospital records and a posthospitalization questionnaire to examine the influence of SES on the health status of 1,962 patients recently discharged from six university-affiliated teaching hospitals following stays for chest pain and four types of surgery (coronary artery bypass grafting, total hip replacement, gallbladder removal, and prostatectomy). They found that on admission to the hospital, lower SES patients were less able to function (activities of daily living, social activities, and work performance) compared with higher SES patients. Yet afterwards, both high and low SES patients with chest pain improved minimally compared with prehospital functioning, and their degree of improvement did not vary significantly by education or income level. Unlike patients with chest pain, most surgical patients improved in functioning following hospitalization, but differences in improvement between highest and lowest income groups were small to moderate.

Melfi, C., Holleman, E., Arthur, D., and Katz, B. (1995). "Selecting a patient characteristics index for the prediction of medical outcomes using administrative claims data." (AHCPR grant HS06432). Journal of Clinical Epidemiology 48(7), pp. 917-926.

Different indexes have been developed to provide a common classification scheme for patient characteristics—such as disease severity, comorbidities, resource needs, and stability—for use in administrative databases to study the outcomes of medical care. These authors examined the utility of four indexes to predict length of hospital stay and 30-day mortality for patients undergoing total knee replacement surgery between 1985 and 1989. They compared the Deyo-adapted Charlson Index, the Relative Intensity Score derived from Patient Management Categories (PMCs), the Patient Severity Level derived from PMCs, and the number of diagnoses (up to five) listed in the Medicare claims data. They found that all of the indexes were an improvement over the baseline models of length of stay and mortality. The Relative Intensity Score and Patient Severity Level indexes resulted in the greatest improvement in measures of model fit; the number of diagnoses (Medicare claims data) performed well and did not suffer from problems associated with miscoding on claims data.

Morishita, L., Boult, C., Ebbitt, B., and others (1995, June). "Concurrent validity of administering the geriatric depression scale and the physical functioning dimension of the SIP by telephone." Journal of the American Geriatric Society 43(6), pp. 680-683.

The Sickness Impact Profile (SIP) and the Geriatric Depression Scale (GDS), which measure health status and symptoms of depression, respectively, have been used extensively to assess older persons. Usually, these instruments are administered in person, but home interviews are expensive and sometimes frightening to older adults. Interviews away from home are complicated and time-consuming. As a result, there is growing interest in administering these instruments by telephone. This study, cosponsored by the Agency for Health Care Policy and Research and the National Institute of Aging, compared telephone and in-person interviews using these instruments in a sample of patients recruited from a university-based geriatrics clinic in 1994. When interviewed in person, more than one-third of the sample reported that their general health was "fair" or "poor." In person, the subjects' mean GDS score was 5.7 (not depressed), and their mean SIP score for physical functioning was 11.1 (moderately impaired). The concurrent validity of administering the GDS by telephone was 90 percent compared with 96 percent for the SIP. The mean interview time required for administering the GDS and the SIP was 13.4 minutes in person and 11.5 minutes by telephone. The researchers conclude that telephone administration of health status measures among community-dwelling older adults is more convenient and less expensive than in-person interviews, but it may be less useful in older persons with significant hearing, physical, or cognitive impairment.

Osuch, J.R., Anthony, M., Bassett, L.W., and others (1995, June). "A proposal for a national mammography database: Content, purpose, and value." American Journal of Radiology 164, pp. 1329-1334.

Marietta Anthony, Ph.D., a former AHCPR staff member now with the U.S. Food and Drug Administration, and colleagues propose a national mammography database. They suggest using a centralized, computerized method of data collection consisting of two parts: a national mammography audit and a system for monitoring and tracking patients. The audit would collect and analyze medical audit data of individual mammography practices at a national level as a critical step in improving the interpretive component of mammography. The monitoring and tracking component would be a centralized system that provides women and physicians with a recruitment and followup mechanism to optimize participation in mammography services. However, the researchers point out that unique scientific, legal, and fiscal concerns must be addressed before establishment of this database. For example, the onus would be placed on practices to provide the data, track patients, and report on outcomes over a 1-year followup period.

AHCPR Publication No. 96-0010
Current as of October 1995