1. A review of the few studies so far made on the high mortality from tuberculosis among elderly men, and a consideration of the available statistics, indicate that urbanization is one of the principal factors responsible.2. In the present inquiry, which was confined to London, 445 newly diagnosed cases of tuberculosis in men over 40, together with the same number of paired controls, were studied by means of a questionnaire and of personal interview.3. It was found that the tuberculous patients differed significantly from the controls in the following characteristics; Scots, Irish, Welsh, or foreign nationality; single, widower or divorced; resident in common lodging houses or hostels; inadequate or special diet; history of gastrectomy; a winter cough; shortness of breath; insufficient sleep; and heavy drinking and smoking. On the other hand, overtime or night work, diabetes, rheumatoid arthritis, asthma, and mental illness were distributed fairly evenly in the two groups.4. Unfortunately no group of elderly women exists in this country living under the same sort of conditions as the elderly men studied here, so that it was impossible to determine how far the various factors considered were responsible for the high rate of late adult male tuberculosis. A study, however, of a population of established civil servants living under ordinary conditions revealed little difference between the observed rates of tuberculosis and those expected on the basis of national notification figures for men and women in the older age groups.5. It appears that a low standard of personal hygiene, associated especially with heavy smoking and drinking and residence in loading houses, predispose to the development of tuberculosis in the elderly male. Part of the evil effect of living in common lodging houses in particular may be due to the increased risk of exposure to tuberculous infection that it entails.6. It is tentatively concluded that the casual workers of an urbanized community are one of the principal reservoirs of tuberculous infection in large towns, and since there is no numerically comparable female population, this group, and its immediate male contacts, account in large measure for the difference between the male and female tuberculosis rates in the older age groups.This work was initiated during the tenure of a Prophit Scholarship of the Royal College of Physicians, and completed with the aid of a grant from the Medical Research Council.I am indebted to members of the Prophit Committee of the Royal College of Physicians for their support and encouragement, and most particularly to Dr G. S. Wilson, Director of the Public Health Laboratory Service, under whose guidance the work was carried out. Figures relating to the incidence of tuberculosis in the Civil Service are published by kind permission of Dr W. E. Chiesman, Treasury Medical Adviser, and Dr M. C. W. Long, Dr J. W. Parks, and Dr H. Stannus Stannus, whose departmental records were used to compute the figures.I am greatly indebted to the consultants and staff of the seventeen chest clinics who co-operated in the investigation, for their interest and help in tracing patients, and to the medical superintendents of numerous sanatoria and chest hospitals, and to the surgeons who permitted me to interview patients under their care as controls.I should like also to acknowledge the assistance received from the medical officers of health of the metropolitan boroughs who kept me informed of notifications from lodging houses in their areas, and supplied information about the accommodation.
Two sources of bias explain errors in facial age estimation
