Some of the parameters relating to duration of infected status and incidence of disease have been measured by analysing the data collected from the five year study. Between 1961-68, 119 villages in Bangalore district with total average population of about 62,000 were surveyed at intervals of 1, 3 and 5 years from the first survey. All persons were tuberculin tested with 1 TU RT 23 and those aged 5 years or more were X-rayed. Sputum of those persons showing any X-ray abnormality were collected and examined for AFB. Persons with X-ray abnormality but bacteriologically negative or with normal X-ray in all the preceding surveys, and who became culture positive with X-ray abnormality in the current survey were termed as "New cases". New cases who had shown 10 mm or more reaction to 1 TU RT 23 at I Survey were considered infected previously. New cases, tuberculin negative at I survey but who showed an increase of 16 mm or more between two consecutive surveys were considered infected midway between the two surveys.

Of the 42 new cases diagnosed from among the newly infected during 5 years, 81% came from those infected within one year. Incidence rate of cases among those who were infected within one year was about 5 times more than those infected earlier than one year. Incidence of cases steadily decreased with the increase in the duration of infection. Further, it was found that one fourth of all newly diagnosed cases came from the newly infected persons. However, the size of the pool of previously infected persons in a community being much larger, at least 72% of the new cases came from the reservoir of previously infected persons. The incidence of disease among the newly infected was almost the same in the three age groups i.e., 5-14, 15-34 and 35 years or more. But, the ratio of the incidence rates for the newly infected and the previously infected decreased from 13 for the age group 5-14 to 3 for the age group 35 years and above. In other words, the incidence of disease among the newly infected in the age group 5-14 was thirteen times more than for the previously infected in the same age-group whereas in the age-group 35 years and above, the incidence among newly infected was only thrice that among the previously infected.
Out of the 160 new cases diagnosed during the three repeat surveys, 21 per cent cases came from among those who were infected on the average for one year or less. This is almost in conformity with the hypothesis that one-fourth of all new active cases come from new infections less than a year old.

A study was carried out mainly to find out the prevalence and incidence of sputum negative active pulmonary tuberculosis (suspect cases) among 35,876 persons aged 5 years and above in rural areas of Bangalore district during 1968-72. Two surveys (I & II) at an interval of 3 months, succeeded by a follow up examination of the X-ray abnormals of the earlier surveys, were conducted in the same villages. Examinations at each survey consisted of tuberculin test, X-ray and sputum examinations. X-rays were interpreted individually at the time of each survey by single picture interpretation method and subsequently by Joint Parallel Reading (JPR) method to arrive to a diagnosis. In the JPR method X-ray readings and their comparison was done by a panel of three X-ray readers with full knowledge of age, sex, result of sputum examination and tuberculin test of each person with chest abnormality at any of the three surveys.

On a single picture interpretation the overall prevalence rate of suspect disease was found to be 5.4 per thousand at I survey and 4.59 per thousand at II survey. There was no significant difference in the overall age and sex specific prevalence rates of suspect disease between I & II surveys. Incidence of suspect disease at the end of 3 months was 2.24 per thousand. By JPR method the prevalence rates of suspect disease was 3.2 per thousand at I survey and 3.6 per thousand at II survey. The prevalence rates by single picture method were overestimated to the extent of 38% at I survey and 19% at II survey when compared with those found by JPR method. At I survey prevalence rates on JPR method was significantly lower than by single picture method. This was not so at II survey. Similarly, incidence rate of 0.2 per thousand of suspect disease on JPR was about 1/10th of that found by single picture method.

The incidence of bacteriologically positive cases in 6 months from among suspect cases on JPR was found to be 28%. Majority (76%) of non-tuberculous or inactive tuberculous shadows continued to remain as such after 6 months and about a quarter (23%) became normal. Incidence of bacteriologically positive cases from this group was minimal. Of 19,640 persons with normal X-rays 134 (0.7%) developed new shadows in 3 months; 103 (0.5%) cleared after 2-12 weeks (fleeting shadows). Mis-interpretation of the latter as active tuberculous may falsely boost the estimates of suspect disease to the extent of about 5%.

In a five year epidemiological survey conducted by National Tuberculosis Institute (NTI) from 1961 to 1968, the population was mainly classified into three epidemiological groups (i) with no radiological abnormalities seen in the lungs (Group N) (ii) having X-ray shadows of non-tuberculous etiology or tuberculosis etiology but judged as inactive (Group M) and (iii) with shadows of tuberculosis etiology judged possibly or definitely active but negative on culture (Group S). The objective of this paper is to compare the characteristics of cases coming from the above three groups (N, M and S) in respect to bacillary disease status (a) at the time of diagnosis and (b) after a lapse of time (Fate). Out of the total 172 new cases diagnosed during three follow ups, 70 were diagnosed between I & II surveys, 40 between II and III and 62 between III-IV surveys. In the two 18 months follow up periods, 45 of the total new cases had come from Group N, 31 cases from Group M and 34 cases from Group S, corresponding figures for 24 months follow up (III & IV surveys) were 26, 26 and 10 respectively.

In the 18 months follow up it was observed that proportion of new cases positive on culture in the three groups were not significantly different. Comparison of fate of cases coming from three groups were similar in terms of cure, death and culture positivity. The findings point out clearly that not only development of disease but also the fate of cases is independent of pre diagnosis status of the new cases.

From all the 3 groups, disease developed more rapidly in some cases than in others. This reveals that tuberculosis cases are not an uniform entity from the point of view of development of the disease and cure.

The incidence rates of sputum positive pulmonary tuberculosis (cases) from the five year follow ups of a rural population done by National Tuberculosis Institute were reported on the basis of studies at intervals of one and a half to two years. Information on fate of cases was also likewise reported. These parameters appear to be imprecise since incidence and fate of cases at shorter intervals were not taken into account. Thus, the information on incidence of pulmonary tuberculosis in India is meager as compared to that on prevalence of disease. Therefore, a study mainly to find out the incidence and fate of cases at shorter intervals of 3-6 months was undertaken in 87 randomly selected villages of Nelamangala sub-division, Bangalore district which was one of the 3 sub-divisions where repeated epidemiological surveys had been conducted between 1961-68. The sample of villages in the present investigation was other than that included in the earlier report. Organized Case-finding, anti-tuberculosis treatment and BCG vaccination neither existed nor could be provided in the area till the completion of the study. The present study was conducted between 1968-1972.

This study conducted among 30,576 persons has shown that incidence of cases over a period of three months was 0.99 per thousand and was not much different from the annual rate of 1.03 per thousand reported on the basis of repeated surveys at longer intervals. That the three months rates were not a quarter of the annual rates meant that the procedure of calculating incidence rates on the basis of surveys done at varying intervals after adjusting for the interval had to be used with great caution. The study of fate of cases showed that cases converted or reverted even at shorter intervals and this appeared to be going on continually in the community. However, incidence of cases and cure and death from among the existing as well as the fresh cases kept on balancing each other so that the prevalence rates of cases studied at shorter or at longer intervals did not show variations.

Mortality from tuberculosis is an important epidemiological parameter for defining the problem of tuberculosis in any country. But due to lack of systematic recording and reporting system, precise information on cause of death is not available in our country. An attempt has been made to estimate the case fatality of tuberculosis cases as well as mortality of cases diagnosed in a longitudinal study conducted from 1961-68 in Bangalore district. Crude mortality of cases is defined as the ratio of total deaths observed among cases to the total number of cases observed, while case fatality is defined as the ratio of deaths that have occurred due to tuberculosis to the total number of cases investigated.

The overall observed annual crude mortality was 14.8%, while among culture positive smear positive (C+S+) it was 21%. An upward trend was seen with the increase in the age. The overall annual crude mortality among culture positive smear negative (C+S-) cases was 9.5% which is significantly lower than that among C+S+ cases. The death rates among old and new cases at the end of 18 months were 16.7% and 13.7% respectively. No statistical difference was found in the crude mortality either among old and new cases or in relation to the interval of diagnosis. Case fatality due to tuberculosis was computed by calculating the deaths among non-tuberculosis population of the same area and during same period and eliminated from the total deaths observed among tuberculosis cases. The case fatality of tuberculosis was found to be 13.3%. It was further observed that out of the total 38 deaths among cases, 89% were due to tuberculosis and 11% were due to non-tuberculosis causes.

The data from a longitudinal survey conducted in Bangalore district from 1961-1968 by National Tuberculosis Institute was analysed to find out the prevalence, incidence and fate of suspect cases. In brief, the survey was conducted in 119 randomly selected villages in three taluks of Bangalore district and repeated within the next five years. At each survey, eligible population was subjected to tuberculin, X-ray & sputum smear and culture examinations.

The overall prevalence rate of suspect cases among persons aged five years and more was 1.06% at I survey, 0.68%, 0.49% and 0.43% at II, III and IV survey respectively. In males, the prevalence rate was 1.19% at I survey & 0.62% at IV survey corresponding figures for females were 0.94% and 0.24% respectively. A decline of prevalence of suspect cases from 1.06% at I survey to 0.43% at IV survey was observed. The overall incidence of suspect cases was 0.16% between I & II surveys, 0.10% between II & III, and 0.06% between III & IV surveys. The overall as well as age specific annual incidence rates between III & IV surveys were significantly less than that between I & II surveys. At all the three intervals the incidence increased with the age. Incidence of suspect cases in males was more than that in females. Change in disease status over a period of time is termed as "fate". The disease status was classified as (i) cure (ii) continued to be suspect case (iii) converted into bacillary cases and (iv) dead. The percentage of cure (51.9%, 53.2% and 50.3%) and conversion into bacillary cases(7.2%,5.8% and 5.4%) were almost the same at all the three intervals. But the percentage of those who remained suspect cases reduced from 33.5% at the end of 18 months to 17.5% at the end of 60 months. On the other hand, the death rate increased from 7.4% at the end of 18 months to 26.8% at the end of 60 months. The decreasing trend of continuing to be suspect cases at the rate of 10% between two observations, appears to be corresponding to the increasing trend in the death rate as seen from the observations made at the three intervals.

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.

It has been reported that a substantial proportion of the new cases arise from the previously infected population. Hence, it appears that exogenous reinfection and/or endogenous reactivation play a major role in the development of post-primary disease. Though the risk of disease associated with exogenous reinfection and endogenous reactivation has not been computed in Indian conditions, the data collected during a longitudinal study by National TB Institute, Bangalore was analysed to estimate the above mentioned risk rates.

The risk of disease associated with exogenous reinfection was 6.55% per year compared with 0.21% yearly due to endogenous reactivation. To test the validity of the computed risk rates these were applied to the interval between the 3rd and 4th surveys. It was then estimated that 64 new cases should have been diagnosed in that survey interval as against 57 cases actually diagnosed. It was also estimated that 1.9% of the total population would be having recent infection, 1.3% would be previously infected with recent reinfection and 32.7% with previous infection but no recent infection leaving 64.1% who are not infected at all (uninfected). Among the new cases diagnosed, 28% would have progressive primary disease, 41% cases arise due to exogenous reinfection and 31% due to endogenous reactivation. In other words, the 1.9% population with recent infection contributes 28% of the total new cases, the 1.3% reinfected population contributes 41% and the 32.7% previously infected population contributes the remaining 31% of the total new cases.

Screening of the population by Mass Miniature Radiography (MMR) followed by sputum examination by culture of the X-ray abnormals is the customary method for arriving at the prevalence rate of cases in the community. It is not possible to use this methodology by states to carry out prevalence surveys in these areas, even if they desire to evaluate the effect of anti tuberculosis measures implemented by them. Therefore, simpler means of screening population through chest symptom for sputum examination has been studied by National Tuberculosis Institute (NTI). The objectives of the present investigation were to find out the prevalence of bacillary cases by screening the population through identification of chest symptomatics by Social Investigators (Sls) or General Health Workers (GHWs) compared to that by MMR. In a peri urban area 10 kms away and around Bangalore city all the villages were listed and of the 60 villages were selected on the basis of a sample random sample. Of them, 30 were covered by Sls of NTI and the other 30 by GHWs of the state government. The methodology adopted was that (1) After census taking and registration of the entire population aged 15 years and above, Sls questioned the persons house to house for presence of cardinal chest symptoms of any duration. All chest symptomatics were subjected to MMR and sputum examination. (2) Similar methodology was adopted by GHWs in the other 30 villages allotted to them. (3) Without knowing the symptom status of all the registered persons, aged 15 years and more belonging to all the 60 villages, were subjected to MMR and from among those having X-ray abnormalities, to sputum examination.

It was found that GHWs had identified the same proportion of the persons either having general symptoms or having chest symptoms from the general population, as Sls. Prevalence rates of culture positive as well as smear positive cases were similar by any of the three methods i.e., 0.18%, 0.23% & 0.25% respectively. Prevalence rates of smear positive cases obtained through symptom questioning, either by Sls or GHWs, were more or less similar to the estimates obtained by the more comprehensive screening method of MMR and/or symptom questioning. The culture positive prevalence rate following MMR screening was 0.25%, which was lower than the rates observed in other surveys. The paper discusses the possible hypothesis that could explain the observation. It also presents correction factors to compute rates comparable to the best estimate i.e., that obtained through comprehensive screening by MMR and/or symptom questioning, followed by sputum culture.

Mass miniature radiography (MMR) is the usual tool for population screening in tuberculosis case prevalence surveys. However, this facility is not available at most centres in India. An attempt was made to study the feasibility of carrying out sputum positive case prevalence survey in a population by introducing methodological variation in the screening, in order to select those eligible for sputum test without resorting to the customary use of MMR for the purpose. The study was carried out in Bangalore rural district during 1984-1986. The area was the same as for six earlier prevalence surveys conducted since 1961. The population aged up to 44 years was tuberculin tested. Persons with test induration size of = 10 mm were eligible for sputum examination, besides all those aged over 45 years were eligible. It was observed that 78.4% of the registered population (29400) in the age group 10 years and above were required to undergo sputum examination by the present method of screening leading to a very high work load of sputum examination necessitating deployment of additional sputum cultures. Thus, the purpose of pre selection for sputum examination was hardly fulfilled. Further, a high contamination rate was observed. The changed screening procedure in this survey made comparison with the earlier data difficult.

The overall prevalence rate of cases was 438/100,000 in persons aged 10 years and above, while smear positive prevalence rate was 68/100,000. The observed prevalence rate was similar to earlier surveys, while smear positive prevalence rate was much lower. In conclusion, the screening methodology was found to be operationally unfeasible, ineffective and counterproductive to complicate the survey procedure in the quest for simplicity.

An operational study to understand the influence of training of Peripheral Health Institution (PHI) Medical Officers (MOs) at District Tuberculosis Centre (DTC) in comparison to on the job training in their own PHIs in carrying out case-finding activity, was carried out in districts of Mysore, Mandya, Bellary and Hassan of Karnataka State in 1980 81. These districts are now being referred as I, II, III and IV respectively. From each district, 20 Microscopy Centres (MCs) were selected. All the selected MCs of the above four districts after stratified random allocation were divided into two groups, i.e., A & B. The MOs of Group A of each district were trained for 2 days in Case-finding activity at the respective DTCs by District Tuberculosis Officer and District Health Officer, while the MOs of Group B were given on the job training as per manual. In all, 108 MOs: 52 in Group A and 56 in Group B were under study. The performance of each PHI was monitored in terms of number of new Out patient Attendance, selection of chest symptomatics for sputum examination and number of smear positive cases detected, for a period of 12 months after the training.

At the end of one year it was observed that there was a boosting in case detection in districts I and III, no effect in district II and negative effect in district IV. The efficiency in districts I and III was higher by methodology A. It was enhanced from 7.6% pre-training efficiency to 16.7% after training and in district III, 18% to 65.8%. The enhancement with methodology B was from 5.5% to 8.1% in district I and from 19.1% to 43.2% in district III. The average increase by amalgamating all the four districts was from 8.5% to 17.8% with methodology A and from 9.7% to 12.3% with methodology B. There was a suggestion of better improvement through methodology A, which, however, did not attain statistical significance.

In the districts under study, Case-finding was at a very low ebb. Systematic training by either of the two methodologies, did improve the activity in I and II i.e., in two of the four districts. In districts II and IV other variables might also have been at work e.g., training variables of knowledge, skill and communication abilities of DTOs who were trainers could have influenced the outcome.