Carroll E. Palmer

On the methodology of investigations of etiologic factors in chronic diseases

Iv1 UCH of our knowledge concerning etiologic factors in chronic diseases has been derived from uncontrolled, or poorly controlled, observations. Experimentation on human beings is difficult if not impossible in investigations of causes of diseases and conditions of long duration, especially those requiring strict randomization. Occasionally a set of circumstances occurs which enables the alert investigator to utilize observations in such a manner as to approximate and simulate planned experimentation. The classic and oft-quoted observations on cholera by Snow represent an example of an occurrence of this type. Such an advantageous combination of circumstances, however, is indeed rare. The search for meaningful associations, in the main, will have to be guided by observations of phenomena as they occur, with little, if any, interference on the part of the observer. Progress in the identification of causative factors in the important chronic diseases will depend to a great extent on our ability to utilize such observations and on the development of more rigorous methods and rules of analysis which will increase the likelihood of correct interpretation and will minimize the effect of extraneous factors inherent in uncontrolled observations. The multiple approach, i.e., the investigation of the effect of a specific factor on a disease through a number of different independent types of observations, if each independently indicates the same relationship, reduces the chance that results are due to extraneous factors and increases the significance of the findings, However, the observations must be indeed different in type and not merely different manifestations of the same set of phenomena all subject to the same limi.. tations. Moreover, each observation must be accurate and reliable, and the relationship implied in each of the different types of observation must be valid in itself.

Height, Weight, Tuberculous Infection, and Tuberculous Disease

Investigation of body build and its relation to prior tuberculous infection and to subsequent tuberculous disease was carried out among 823,199 recruits who entered the navy from 1958 through 1967. Results confirm earlier studies which showed that body build is associated with tuberculosis. Morbidity Increased with height and decreased with weight. In addition, physique is not associated with the likelihood of becoming infected with Mycobacterium tuberculosis. The very substantial association between body build and tuberculosis is concentrated on the susceptibility of the individual to develop recognizable disease after he has been infected. Morbidity is over three times greater for underweight than for overweight men, while the frequency of infection in the two groups is the same. It is suggested that body build and susceptibility to developing tuberculous disease are interrelated familial or stock traits.

Experimental and epidemiologic basis for the interpretation of tuberculin sensitivity

Summary Epidemiologic evidence that nontuberculous infections cause many of the positive tuberculin reactions observed in human beings is corroborated by results of experimental studies of guinea pigs reported in the present paper. Some of the animals were infected with virulent tubercle bacilli, some with the nonphotochromogenic mycobacterium of the “Battey” type, and some were not infected. Intracutaneous tests with standard mammalian tuberculin (PPD-S) and the PPD antigen prepared from the Battey organism (PPD-B) showed that although cross-reactions occur with both types of infection, the homologous antigen usually produced a larger reaction than the heterologous antigen. Frequency histograms of the sizes of reactions to PPD-S for theoretical populations containing various proportions of tuberculous infected, Battey infected, and uninfected guinea pigs, revealed patterns of sensitivity similar to those observed in human populations. Parallel studies of patients in tuberculosis hospitals from whom only typical tubercle bacilli or Battey-type organisms had been recovered gave results like those found in the guinea pigs. Studies of healthy U.S. Navy recruits from different parts of the country, and of general population groups in this country and abroad, indicate that cross-reactions to tuberculin are highly prevalent in some geographic areas and much less prevalent in others; moreover, the frequency of cross-reactions varies with the prevalence of sensitivity to PPD-B. Whether the organism chiefly responsible for these cross-reactions is the Battey organism, or some other related organism (or organisms), is still undetermined. It is clear, however, that when the frequency of tuberculin cross-reactions is high and of specific tuberculous reactions is low, few of the tuberculous infected can be identified by the size of the tuberculin reaction alone. Simultaneous testing with PPD-S and other PPD antigens can then be expected to help determine the source of the tuberculin sensitivity, just as simultaneous testing with histoplasmin and coccidioidin helps to distinguish between the two fungus infections.

Experience with BCG vaccine in the control of tuberculosis among North American Indians.

The efficacy of BCG in lowering the incidence and mortality of tuberculosis in America Indians was tested on 3007 persons on 7 reservations in Arizona Wyoming North and South Dakota and Alaska in a double-blind design. Participants were selected on the basis of negative tuberculin reaction and chest x-rays but presence of a Koch reaction could not be determined in controls. The test groups received 0.1-0.15 ml BCG intracutaneously and the controls saline; all were followed for 6 years with annual tuberculin tests and chest x-rays. The groups did not differ in age amount of exposure to tuberculosis or follow-up. Mortality in the control group was higher than in the vaccinated group as judged by number of deaths 60 vs. 43 and by death rate per 1000 person-years 7.2 vs. 3.8. 28 deaths were attributed to tuberculosis in the controls compared to 4 among the BCG groups. Severity of infection was greater in controls including numbers of extrapulmonary tuberculosis advanced pulmonary lesions minimal lesions and enlarged hilar glands. Incidence of new cases of tuberculosis decreased with time in the treated group while that in controls remained constant. There was some variation in effectiveness of the different lots of vaccine.

Nationwide histoplasmin sensitivity and histoplasmal infection

RESULTS of skin testing with histoplasmin and the diagnosis of clinical cases of histoplasmosis have made it clear that where people live is of prime significance in the risk of beicoming infected with the fungus Histopla8ma capsulatum. The purpose of this paper is twofold: to depict State-to-State variations in indigenous sources of sensitivity to histoplasmin, as measured by the prevalence of reactions among young men who have lived all their lives in one State, and to distinguish, in some regions, between the reactions caused by infection with Histoplasima and cross-reactions attributable to infection with other agents. In dealing with geographic units as large as States, the presence of small and perhaps isolated areas of both high and low prevalence will be obscured. State figures, on the other hand, can provide a broad picture of geographic variations across the 3,000 miles from the Atlantic to the Pacific coasts. Several earlier reports from this office have described nationwide variations in the prevalence of histoplasmin sensitivity and variations in the sizes of reactions observed in different geographic areas (1-4). These reports were based on material collected in several studies of young adults during 1945-52. More recently, it has been possible to obtain material in a single study, in which the same histoplasmin, techniques, and procedures were used for giving skin tests, at the same time and place, to large numbers of young men (Navy recruits) from all parts of the country. Results of this newer study confirm and extend the conclusions drawn from the previous material, which was collected, on the average, about 10 years earlier.

IDENTIFICATION OF THE TUBERCULOUS-INFECTED BY SKIN TESTS*

An ideal skin test for identifying persons with tuberculous infection would elicit a reaction in all of those infected, and only those infected with M . tuberculosis. The present tuberculin test fulfills the first requirement very satisfactorily. In appropriate dosage, it gives a reaction in almost every individual who has had a tuberculous infection. Moreover, there is no indication that i t gives reactions in persons who are not infected at all. It fails to fulfill the second requirement, however, in that it may give cross reactions in persons infected with other (nontuberculous) mycobacteria. The first reactions recognized as cross reactions were those elicited only by large doses (100 or 250 Tuberculin Units ( T U ) ) of tuberculin, and the discontinuance of such doses eliminated a large proportion of the cross reactions. It soon became apparent, however, that not all reactions, even to the 5 TU dose, were specific, that is, caused by tuberculous infection. Some reactions, particularly the small ones, could be interpreted in no other way than as cross reactions. In an attempt to reduce the number of persons with cross reactions in the group regarded as having a tuberculous infection, the dividing point used to separate “reactors” from “nonreactor~” was raised; from 5 mm, to 6 mm, then to 8 mm, and recently to 10 mm. It was recognized, however, that this procedure did not effect a perfect separation of the tuberculous-infected from the nontuberculous. As the dividing point was pushed up, an increasing percentage of the persons having only cross reactions to tuberculin was eliminated from the “reactor” group, but so also was an increasing percentage of the tuberculous-infected. Some laboratory results illustrate the difficulty of using just the tuberculin test to distinguish individuals with tuberculoiis infection from those infected with other mycobacteria. FIGURE 1 shows frequency distributions of reaction sizes to Purified Protein Derivative (PPD-St ) tuberculin in three groups of experimentally infected guinea pigs. In the upper section are reactions in animals infected with virulent tubercle bacilli. Virtually all react to tuberculin, with rather large reactions distributed in a form approximating a normal frequency curve. In the middle section are reactions in animals infected with a Battey type of “atypical” mycobacteria. The cross reactions in these animals are less frequent and smaller than the specific reactions in the tuberculous animals. However, some of the reactions in each group fall within the same size range. There is no single point on the size scale above which all reactions are specific, and below which all are cross reactions. In the bottom section is a mixed group of animals, half of them infected with tubeicle bacilli, and half with Battey bacilli. The shape of the distribution in this group, showing an excess of smaller reactions compared with the tuberculous animals, would of itself indicate that not all the animals are tuberculous-infected. But with the tuberculin test alone, it is not possible to efficiently separate the tuberculous from the Battey infected animals.

A follow-up study of tuberculosis in former student nurses: I. Methods of locating and obtaining information from a study population of 25,752

Abstract With present-day interest rapidly turning toward the chronic diseases and disabilities, investigations calling for observations on large numbers of persons for periods of time measured in decades rather than years or months, will undoubtedly become an increasingly prominent feature of public health research. And for the kinds of investigations that require wide geographic coverage or involve study populations which become widely dispersed before the follow-up is completed, the practicability of being able to trace individuals successfully becomes a question of immediate concern. The present paper bears directly on that question, and indicates that persons “lost” from observation, even in a large study population, can be reduced to a negligible number. The present paper, the eighth in a series on the study of tuberculosis in student nurses, describes the methods used to locate and obtain information in the follow-up of a widely scattered population of 25,752 former student nurses, after an interval ranging from four to ten years since they had last been observed in an earlier study. All but nine were located, and only 44 did not wish to answer the questions put to them. Experience in the present study brings out a number of problems and mistakes which, in retrospect, point toward ways of making a follow-up study more efficient. One of the most critical in anticipating a follow-up is to obtain from each study participant the names and addresses of several persons, preferably adult male relatives or friends, through whom the participant may later be traced. Second, it is probably sufficient in many studies of this kind to take a fair-sized random sample of the final 5, 10, or 20 per cent of the total study population that are difficult to locate or do not readily respond. And finally, the present kind of follow-up cannot be attempted without a well-trained field staff, experienced in methods of tracing persons.

Skin sensitivity of healthy young Adults to the fungus antigen, haplosporangin.

The fungus isolated and named Haplosporangium parvum by EMIVIONS & ASHBURN in 1942 (8), and recently renamed Emmonsia parva (3,9), has been found in the lungs of wild rodents and small mammals in the northern and western parts of the United States and Canada by numerous investigators (1, 4, 9, 11, 12), in Italy by CIrERRI (3), in England by MCDIARMID & AUSTWlCK (20) and TEVlS (23), and in such scattered areas as Africa, Ecuador, Finland, France, Japan, Korea, Norway, Sweden, and Yugoslavia by JELLISON and associates (9, 13-18). Whether or not the animals can be considered infected is debatable because the organism seemingly does not multiply in the natural host--i t simply becomes enlarged. As far as is known, no case of haplomycosis has yet been reported in a human being. The question of whether man, too, may unknowingly share the role of host to this curious invader thus remains unanswered. Several years ago, we found what appeared to be histoplasmin cross reactions in a fairly large percentage of the young men, Navy recruits, coming from the eastern part of Texas and the adjacent states (5, 22). Coccidioidomycosis could not be incriminated as a significant source of the cross sensitivity because very few recruits from that area react to coccidioidin. And results of subsequent skin test studies in Navy recruits with a blastomyces antigen speak against blastomycosis as the causative agent (7). Our interest in this problem was one of the reasons for including haplosporangin in our studies of skin sensitivity to antigens prepared from a number of different pathogenic and possibly pathogenic fungi. Results of tests with haplosporangin in several thousand Navy recruits coming from all parts of the United States, and in several smaller groups of persons in this country and in Egypt, are

A follow-up study of tuberculosis in student nurses. II. Mortality from all causes in the first decade after entering training.

Abstract In a study population of nearly 26,000 young women who were placed under observation as student nurses during the years 1943–1949 and participated in a follow-up study in 1952–1953, there were a total of ninety-six deaths from all causes—a rate of 52 as compared with an expected rate of 100 per 100,000 person-years of observation. That the over-all observed rate was about one-half the expected rate can be ascribed largely to the very low rates in the study population during the early years of the observation period when most of the women were in training and the first few years after they left school. Toward the end of the follow-up period, however, the observed death rates were about the same as those in a comparable group in the general population. With regard to specific causes of death, the main findings in the present study were the extremely low death rates from tuberculosis and the prominence of violent causes of death, particularly those from accidents and suicides.

An American View of BCG Vaccination

The United States is one of the few countries in the world in which BCG has not been used extensively for the control of tuberculosis. A brief review of recent history may help those in other countries to understand the factors which initially kept the United States from adopting BCG and the additional factors which make it even less likely now.