The degree of the risk of infection and disease in man from drug resistant strains of mycobacterium tuberculosis is not clear. An increase in the prevalence of primary resistance indicates the extent of such risk while an increase of secondary or acquired resistance could be considered as a problem of the individual patient and may reflect limitations of his treatment.

The present report describes the prevalence of strains with acquired or primary resistance or of sensitive strains found in 3 successive surveys in a sizable random sample of village in a south Indian district. Changes in the status of cases with such strains from one survey to another and their infectivity among household contacts are also described. The prevalence of tuberculosis infection among household contacts of cases with acquired resistance to isoniazid was significantly higher than those with primary resistance or with sensitive culture. This was probably due to the longer duration of sputum positivity of isoniazid resistant strains at the time of diagnosis. But infectivity as judged by the incidence of new infection among household contacts was generally less for cases with acquired or primary resistance than for cases with sensitive cultures, though the difference observed was not statistically significant. A large number of culture positive cases especially those with primary resistance had no radiological evidence of active pulmonary tuberculosis. The prevalence of primary resistance was high in certain categories of cases and the differences between cases with primary resistance and those with acquired resistance were many and large. It was suggested that this could be due to the primary resistant cultures being those of atypical mycobacteria, despite positivity in the niacin test. There was a significant increase in the number of cases with acquired resistance to isoniazid at the third survey owing to the irregular treatment and supply of INH alone after the second round. The prevalence of primary resistance at the three rounds was almost the same.