A study was undertaken with the following objectives: (1) To estimate the proportion of population that would be available for resurvey after 5 years. (2) To ascertain five years later the fate of persons with X-ray pathology. (3) To compare the prevalence of tuberculosis in the villages at an interval of 5 years. Population of 15 of the 31 villages from the Madanapalle zone, was selected for this study. About 9,500 persons were registered and 7,200 were X-rayed at the initial survey. Five years later the same population was re-examined and nearly 70% were available for X-ray examination. Sputa were collected from persons with abnormal X-ray shadows interpreted as such by either of the two readers. Two spot samples were collected within an interval of 1-3 days and were examined by direct smear and by culture.

Analysis of the data shadow showed that: (1) There was no significant difference in the prevalence rates i.e., 3.6 and 4.6 per thousand respectively at two points of time. (2) During the interval, 30% of active cases had died and 20% were still active at the end of 5 years. (3) There was almost complete turn over of the bacillary cases during the 5 years interval.

It is a common observation that in epidemiological surveys all those eligible for various examinations (tuberculin, X-ray and sputum examinations) do not attend them. If the 'non-attenders' differ from the 'attenders' the true situation of the problem may not be known. In this paper, the prevalence of infection, bacillary cases and suspect cases at II survey for both attenders and non-attenders of the I survey from longitudinal study conducted by National Tuberculosis Institute, Bangalore, are compared.

It was observed that in spite of repeated attempts, nearly 1/5th of the population did not attend examinations. The non- response group during I survey was examined at the subsequent survey and both response and non- response groups at the preceding survey were compared. It was found that in respect of prevalence of infection and bacillary disease, the two groups did not differ, but the mortality and emigration was higher among the non- response group. Higher mortality among non-attenders may be due to the fact that the group contained more sick people. The higher emigration among non-attenders due to small error even to the extent of 0.5% at the stage of census taking by registering a non-resident as permanent resident of the village would highly boost the rate of emigration among non- attenders. The difference in the indices of crude mortality and emigration rates becomes narrower and narrower as coverages for examinations increase. The analysis indicates that every attempt should be made to obtain as high a coverage as possible in order to obtain valid estimates of epidemiological indices in a population survey.