SOME ASPECTS OF A TUBERCULOSIS PREVALENCE SURVEY IN A SOUTH INDIAN DISTRICT.

As a preliminary to a number of pilot projects intended to assess the efficiency of various tuberculosis control measures, a sample prevalence survey was undertaken in the Tumkur District of Mysore State, South India. Such surveys would be greatly simplified if correlations could be established between the infection rates, as determinated by a positive tuberculin reaction, and the prevalence rates for radiologically active and bacteriologically confirmed disease. The paper discusses various indices of infection and radiological disease. Prevalence rates for these indices are presented and their degree of correlation with each other and with the prevalence of bacteriologically confirmed disease is determined. The value of the various indices is discussed in the light of these results. Although the correlations do not justify a clear-cut selection of the best possible indices, they provide a useful basis for further work. The survey also revealed certain interesting variations in prevalence that suggest the desirability of stratification in sampling when making surveys in future.