Sampling in medical research

The spectrum of coronary heart disease in a community of 30,000. A clinicopathologic study

The high rate of necropsy in deaths among permanent residents of Rochester, Minnesota, provided a unique opportunity to study the prevalence of coronary heart disease and the frequency and mode of death resulting from this disease. In this community of approximately 30,000 population, necropsy was done in 73 per cent of all resident deaths during the years 1947 through 1952. Included in this group were 691 necropsies that represented 67 per cent of the 1,026 deaths of persons 20 years of age or older.

Coronary heart disease caused death in 221 patients (23 per cent of all necropsies and 32 per cent of all necropsies on adults). These 221 coronary deaths in adults represented 41 per cent of the men and 22 per cent of the women. The coronary deaths were attributed to acute coronary failure (sudden death) in 94 patients (43 per cent), acute myocardial infarction in 87 patients (39 per cent), congestive heart failure in 32 patients (14 per cent) and thromboembolism in eight patients (4 per cent). The 87 patients dying during acute myocardial infarction died of congestive heart failure (30 per cent), myocardial rupture (24 per cent), acute coronary failure (23 per cent), and thromboembolism (14 per cent), with the remaining 9 per cent dying of a combination of acute myocardial infarction and additional serious systemic disease.

The greatest number of deaths from coronary heart disease occurred during the seventh decade of life in men and the eighth decade in women. At least one coronary artery exhibited from 25 to 100 per cent obstruction from atherosclerosis in 513 hearts (74 per cent of the adult necropsies).

These necropsies, representing two thirds of all deaths in adults in this community, disclosed that significant coronary-artery disease was present in three of four adults and was the cause of death in four of 10 men and two of 10 women.

Generalizability of clinical studies conducted at tertiary care medical centers: a population-based analysis

The Marshfield Epidemiologic Study Area (MESA), a geographically defined population registry at one of the participating sites in SUPPORT (a multicenter study of the care of seriously ill hospitalized patients) permitted assessment of generalizability in that study. On the basis of age- and sex-specific rates of enrollment of SUPPORT patients in MESA, we estimate that about 400,000 patients per year would fulfill SUPPORT eligibility criteria in the United States. However, an estimated 925,000 patients, particularly the elderly and those with impairments in their activities of daily living (ADLs), have SUPPORT-like illnesses annually, but do not receive the aggressive care required for study enrollment. The absence of patients not interested in aggressive care in tertiary care-based studies is compounded by the overrepresentation of patients referred from distant areas to the tertiary care center. Such patients tended to be older and to have different diseases than patients in MESA. Care should be taken in generalizing results from clinical and epidemiologic studies conducted at tertiary care centers.

Token swap test of significance for serial medical data bases

Established tests of statistical significance are based upon the concept that observed data are drawn randomly from a larger, perhaps infinite source population. The significance value, p, is the probability that the observations are drawn from a source population satisfying the null hypothesis; if p is small enough (less than 5 percent, 1 percent, etc.), then the null hypothesis is rejected. Serial medical data bases, such as a hospital clinic intake or autopsy case accessions, often do not have an identifiable source population from which they are randomly drawn. In an effort to make a reasonable interpretation of these less-than-ideal data, this report introduces a "token swap" test of significance, in which the usual paradigm of repeated drawing from a source population is replaced by a paradigm or misclassification within the observed data themselves. The token swap test consists of rearranging the data into a balanced distribution, and determining the disparity between the observed and the balanced distribution of data. In a two-by-two contingency table, patients are represented as "tokens" distributed into four "cells." Significance is determined by the proportion of "token swaps" that are able to transform the balanced table into the observed table. The token swap test was applied to three series of autopsy observations, and gave results roughly comparable to the corresponding (two-tail) chi-square and one-tail Fisher exact tests. The token swap test of significance may be a useful alternative to classic statistical tests when the limiting assumptions of a retrospective, serial medical data base are present.

Enrollment bias and variation in clinical manifestations: a review of consecutive cases of familial paroxysmal polyserositis

The effect of systematic enrollment bias was studied in a series of 158 patients with familial paroxysmal polyserositis (F.P.P.) from a special clinic in a university hospital. The characteristics of 79 patients seen for the first time at this special clinic in its first 16 months of operation, were compared with the characteristics of 79 patients presented during the last 6 yr of its operation. As a group the patients studied first at the clinic had a more severe form of the disease. This group had a significantly larger number of patients with multiple clinical manifestations, with proteinuria and amyloidosis. In addition, there were more males and more patients with positive family history of the disease in this earlier group. It is probable that much of the temporal variation in these clinical characteristics between patients studied in the earlier and later periods can be explained by case selection originating from differentials in referral patterns. Thus, a larger proportion of the earlier patients were referred following an episode of hospitalization. Differences in referral pattern, in case selection and in the sources of data, can account for the significant variation in the frequency of the various clinical manifestations in the different published F.P.P. case series from several countries. The effect of enrollment bias on the results of clinical studies and the importance of a population base for such studies is emphasized.

Pulmonary infection and rheumatoid arthritis

Five of eleven patients with bronchiectasis and/or cystic fibrosis developed a polyarthritis with positive tests for rheumatoid factor. Possible mechanisms of this complication are discussed.

Planning for clinical research

The need to initiate research projects in physiotherapy is being recognized throughout the world, while a growing interest is being shown by physiotherapists in undertaking studies appropriate to the areas of medicine. Many writers have urged the profession to accept its responsibility for contributing to the advancement of knowledge and it is to be hoped that an increasing number of physiotherapists will respond to this challenge.