Cases of pulmonary tuberculosis among the out-patients attending general health institutions in an Indian city*

Bull. Org. mond. Sant) 1970, 43, 35-40Bull. Wid Hlth Org. Cases of Pulmonary Tuberculosis Among the Out-patients Attending General Health Institutions in an Indian City * G. D. GOTHI,1 D. SAVIt,' G. V. J. BAILY3 & RUPERT SAMUEL' A study was undertaken in Bangalore City, India, to find out whether people with chest symptoms, including tuberculosis patients, attend general health institutions or report directly to tuberculosis clinics. The attendance for one day at 19 general dispensaries was investigated. Ofthe total of2506 eligible persons who were questioned about the presence of chest symptoms, 1170 admitted having symptoms suggestive of pulmonary tuberculosis and 20 cases and 31 suspected cases ofpulmonary tuberculosis were diagnosed. Thefindings indicate that tuberculosis patients do not by-pass the city health institutions. On the other hand, if these institutions take an active part in the tuberculosis programme and undertake diagnostic functions, or refer persons with chest symptoms to a central clinic, they can contribute substantially to case-finding. Urban areas in India generally have well-equipped tuberculosis clinics, manned by specialists, to deal with the problems of diagnosis and treatment of tuberculosis among the urban populations. It is presumed that city-dwellers who are aware of the existence of specialized tuberculosis institutions report directly to such institutions for treatment, thus by-passing the city general health institutions. It is further supposed that, when general health institutions are integrated into a city tuberculosis programme, their contribution to case-finding and treatment is generally very meagre. Information on the proportion of chest symptomatics5 among out- patients of general health institutions in cities assumes, therefore, a considerable importance. In the city of Bangalore, 3 specialized tuberculosis institutions (1 tuberculosis control centre and 2 sanatoria) serve the city and the surrounding rural area. There are, in addition, 4 large general hospitals, * From the National Tuberculosis Institute, Bangalore-3, India. Tuberculosis Specialist, National Tuberculosis Institute. 'Formerly Senior WHO Officer, National Tuberculosis Institute (1965-68). ' Medical Officer, National Tuberculosis Institute. 'Statistical Assistant, National Tuberculosis Institute. 'Persons with cough, pain in the chest, haemoptysis and chronic fever. 8 municipal health centres, 19 general dispensaries and many private medical practitioners working in- dependently. Between 1961 and 1966, a compre- hensive tuberculosis control programme was intro- duced in stages to serve an estimated population (in 1966) of 1.4 million persons. As part of this pro- gramme, the services rendered by the tuberculosis clinic to the general health institutions include the organization of visits of a mobile mass-miniature- radiography unit to the general hospitals and muni- cipal health centres, the free issue of antituberculosis drugs to all general dispensaries and the provision of diagnostic information to private practitioners about patients referred by them. Institutions without diagnostic facilities are expected to refer persons with chest symptoms to the tuberculosis control centre for diagnosis. The functioning and effective- ness of this system have been reported elsewhere (Andersen & Banerji, 1963; Nagpaul et al., 1970). In order to understand the situation better, it was considered important to investigate what proportion of persons with chest symptoms attends general dispensaries in Bangalore and, of this proportion, how many persons were previously examined and reported as " cases " and " suspected cases " (WHO Expert Committee on Tuberculosis, 1964) of tuber- culosis, or in whom a diagnosis of tuberculosis was made for the first time. 2541 - 35 36 G. D. GOTHI AND OTHERS OBJECTIVES The present study was undertaken to investigate the proportion of persons with chest symptoms among the out-patients attending general dispen- saries of Bangalore City, and the proportion among those with symptoms who present objective signs of pulmonary tuberculosis. METHODS AND MATERIAL All the 19 general dispensaries providing out- patient treatment were included in the study; each was studied for 1 working day only. The 19 working days were spread over 3 consecutive months during 1966. The possibility of seasonal variations in the out-patient attendance at these dispensaries was not taken into consideration. A team from the National Tuberculosis Institute (NTI), comprising a medical officer, a public health nurse, 3 interviewers (health visitors, untrained in the sociological interview technique), an X-ray tech- nician and a laboratory attendant, was detailed to carry out this study. So that the population of the area would not be especially attracted to a dispensary by the unusual diagnostic activity on the day of examination, the date and time of the visit of the NTI team was not revealed in advance. On the study day, the medical officer of the dispensary con- cerned learned about the purpose of the study and its methodology only 15 minutes before starting his usual out-patient work. All patients aged 10 years and over who attended the dispensary in person for treatment of any ailment on the study day, whether for the first time or for a repeat visit, were considered eligible for the study. After each eligible patient had been seen by the dis- pensary medical officer, one interviewer recorded the patient's registration particulars while the other took down a detailed history, first by " non-suggestive " questioning in order to elicit the presenting symp- toms and then by " suggestive" questioning to find out whether any specific chest symptoms were pre- sent. After the interview, a spot specimen of sputum was collected from each person, if he could produce it, and an X-ray examination of the chest was made. Those among the otherwise eligible patients who admitted during the interview that they had been attracted mainly by the presence of the mobile X-ray unit stationed outside the dispensary were later excluded from the final analysis (although they were examined). The investigations required for the study were carried out in such a way as to disturb the routine of the dispensary as little as possible. The collected sputum samples were examined by direct microscopy, culture and sensitivity tests at the NTI. The chest photofluorograms were examined independently by two readers at the NTI and differences were submitted to a referee for his deci- sion. The findings ofthe photofluorograms were clas- sified according to the system used in the District Tuberculosis Programme in India 1 and grouped into the following categories: (1) normal, (2) non- tuberculous, (3) probably or definitely tuberculous, (4) likely to be inactive or having indefinite shadows, to be kept under observation. Each case and suspect case diagnosed in this way was then checked against the tuberculosis case index maintained at the tuber- culosis centre to ascertain whether it was a new case not previously registered or one already known. FINDINGS Study population The following tabulation shows the population examined. Not eligible Number Out-patients aged 0-9 years and persons who attended to collect medicines for other patients 2 464 Eligible (1) Persons who attended for chest X-ray only; excluded from further analysis 567 (2) Out-patients aged 10 years or more who attended the dispensary for the relief of symptoms 2 506 Total registered out-patients in 19 dispensaries 5 537 The actual out-patient attendance at the 19 dispen- saries for 1 day each was 5537 persons, including 567 persons who admitted that they had been at- tracted by the mobile X-ray unit. Excluding those who attended for X-ray only on those days, the average daily attendance at each city dispensary in 1966 was 262, compared with 246 attendances per day at each dispensary in the year 1965. Though the study days did not cover all seasons, it may be observed that the average attendance on study days was close to the average daily attendance for the previous year. Out of the 5537 persons who attended, 2464 were not eligible for the study, either because they were below the age of 10 years or because they had attended to collect medicine on behalf of another 1 District Tuberculosis Officer's Manual in a District Tuberculosis Programme, SM/l, p. 25. A limited number of copies of this mimeographed document are available on request to the National Tuberculosis Institute, Bangalore-3, India. PULMONARY TUBERCULOSIS AMONG OUT-PATIENTS ATTENDING CLINICS IN BANGALORE person. A further 567 persons admitted that they had been attracted by the X-ray unit. After exclu- sion of these groups, 2506 persons were considered for the final analysis. Coverage While 100% of the eligible population was inter- viewed satisfactorily, only 96% was X-rayed and only 51% was able to produce sputum specimens, in spite of their best efforts (Table 1). TABLE I SPUTUM AND X-RAY EXAMINATION COVERAGE OF STUDY POPULATION BY AGE AND SEX Sputum X-ray Age and sex Total sample examinationeligible examined made patients No. % No. % 10-24 years 1 217 455 37.4 1174 96.5 25-44 years 744 475 63.8 711 95.6 >45 years 545 355 65.1 519 95.2 Total 2 506 1 285 51.3 2 404 95.9 Male 1412 741 52.5 1 352 95.8 Female 1 094 544 49.7 1 052 96.2 Proportion of out-patients with symptoms Of the 2506 out-patients, 1170 (47%) reported that they had come primarily for the relief of chest symptoms. Of the rest who attended mainly to seek relief for other ailments, 198 (8y%) persons on non- suggestive questioning admitted having 1 or more chest symptoms, and 222 (8 Y.) did so on direct sug- gestive questioning (leading questions). In other words, the proportion of persons with chest symp- toms who spontaneously sought relief of these symptoms was only 47% of the study patients. However, this proportion rose to 63% when careful questioning was resorted to. The possibility of an overestimate is discussed below. Cases of tuberculosis among out-patients The radiological and bacteriological status of the patients, distributed by age and sex, is shown in Table 2. Only 49 persons (2%) among the 2506 eligible persons were considered to have X-ray evidence of active or probably active pulmonary tuberculosis. Of these, 18 were confirmed bacterio- logically, leaving the rest as suspect cases (sputum x w U) a z W 0 z z 0 U) z C] I 0 U) Lu z uJ 0. Z U) a.0 Lu I_ I- CD z z Lu U) Lu 0 0 m tu 0 W) I- -0 % e= Xmc. >,a 0.0 WA_ IL o LE " E Co 0 O0 Z1% '0&_ Q AeX0 Ez c 0 D >, :a .0 .0 0. DL Oe*W 0 0.-C 0 U) x 0 10 CE Z U) m E s 0.X C U) 0 E = _.. c U)C 00 U)C= Q a 00 coc004 r-In ar- c00 I_- 4 II of I o r- Oco v- 004 -- S' _- D 0 a Xa E 0 W E Q= 11S X E a C 0:_c 0.a U) x e) U0. 0 0 6 z a 0 00 040 I00 o04 S f aain VI'VI c0 0404_ 0o0 04 04 0m 0-4'- CO 0oco O00 __ C,'* sa i 09 co%qb ° E I0 04_ W.t CDC5 to 0 r- -04 0404 - 1 -I II so 0i 0404 In U)If'-' ~~ _CN9c-ACIA, a0 0 ° A\ 0 a 0 0 0 .(a LL U. U. 1D 1@ iD c ' a 0 I- U. 37 e EE CD 0 co e 5 0 c 0 E E a a e a co1 E 0 0 0 0. 0 z aa e S a C) S 0 e E e e C0 0 0 e a ae0 E :5 Q 0. U) a I C4 I C G. D. GOTHI AND OTHERS TABLE 3 PROGRAMME STATUS AND SPUTUM STATUS OF THE 51 CASES DIAGNOSED IN THE STUDY Cases Programme Suspect cases (sputum-positive on smear and culture) negative) Isoniazid- lsonlazid- Culture- sensitive resistant negative Known to the programme 7 2 4 - 13 New cases 24 10 2 2 38 Total 31 12 6 ] 2 51 samples of 10 patients were not examined). In 2 cases, the sputa were found to be positive although the chest skiagrams were read as " normal " by both readers, independently and jointly. Another 51 pa- tients were placed in group (4) whereas 96 were clas- sified as presenting some, presumably non-tuber- culous, pathology. When these results were matched against the case index maintained by the tuberculosis control centre, it appeared that, out of 20 cases and 31 suspected cases of pulmonary tuberculosis thus diagnosed at the general dispensaries, 13 (6 cases and 7 suspected cases) were already known, whereas the remaining 38 were new tuberculosis patients (Table 3). The prevalence rate of sputum-positive cases represents 0.8% of the dispensary-going population above the age of 10 years. Drug sensitivity status (isoniazid) Of the 20 cases discovered, only 18 yielded a positive culture and 6 of them belonged to the group already on the case index of the tuberculosis control centre (Table 3). Of these 6 known cases, only 2 (33 %) were sensitive to isoniazid, and 4 (67%.) were resistant. Among the remaining 12 newly detected culture-positive cases, 10 (83%) were sensitive to isoniazid and 2 (17%) were resistant. Of the 2 newly diagnosed resistant cases, 1 had a history of treatment from a private source; in the other, although the patient totally denied having received antituberculosis treatment, the very extensive radio- logical disease pointed to earlier treatment. DISCUSSION A 1-day sample may not be considered to be representative of the average population that makes use of dispensaries over a 1-year period. It is also possible that the time spent by health visitors in taking a detailed history, and the presence of a mobile X-ray unit in the precincts of the dispensary, may have influenced the proportion of persons with symptoms among the out-patients who attended on examination days. Despite the above limitations, the study has produced some striking information. The first finding is that even in a sophisticated city population in South India, many persons with chest symptoms first contact general dispensaries for relief. The percentage of such symptomatics has been reported to be as high as 47% (spontaneous) or 16% (elicited). Baily et al. (1968) found that 6y%-7% of the new out-patients aged 10 years and over attending rural dispensaries had chest symptoms that had lasted a week or more. The significance of the very high rate of persons with chest symptoms found in this study among those attending the 19 dispensaries in Bangalore City is subject to some speculation. In view of these conflicting findings, a trained social investigator was sent again a year later to each of the 19 dispensaries for 1 day in order to obtain another estimate of the proportion of persons with chest symptoms among general out-patients in dispensaries, discounting those attracted by a mobile X-ray unit. The social investigator did not interview the patients directly but stationed himself near the dispensary medical officer and noted the information given by patients. A much lower estimate (22%) of the proportion of persons with chest symptoms among patients attending urban dispensaries was obtained. This discrepancy, how- ever, does not alter the finding that many persons with chest symptoms attend general dispensaries as frequently in the city as in rural areas. This is sup- 38 PULMONARY TUBERCULOSIS AMONG OUT-PATIENTS ATEENDING CLINICS IN BANGALORE 39 ported by another study (Nagpaul et al., 1970), in which it was observed that many patients prior to visiting the city centre attend general dispensaries. Despite the high attendances at general dispensa- ries, an analysis of the records of the Lady Willing- don Tuberculosis Demonstration and Training Cen- tre revealed that only 683 patients referred from such institutions had attended the specialized tuberculosis services for further examination dur- ing a period of 3 months in the year 1965. This suggests that, for unknown reasons, either general dispensaries and hospitals do not refer many persons with symptoms to the specialized services for diag- nosis, or if they do, then the referred patients fail to report there. There appears to be, therefore, some justification for equipping general dispensaries in the city with microscopes, as is also recommended for rural dispensaries. General dispensaries in the city could then play the same role in a co-ordinated system as rural dispensaries now play in the District Tuberculosis Programme (Piot, 1962; Nagpaul, 1967). Another significant finding of the study is the high case-yield from out-patient attendances at general dispensaries. Diagnostic work for 1 day at all the 19 dispensaries led to the discovery of 14 new cases of tuberculosis (Table 3) and 24 radiologically sus- pected cases among patients aged 10 years or more. If simple microscopy were introduced in city dispen- saries, it could be expected that 10 new cases a day would be diagnosed by this simple means. Although the work-load in respect of sputum examinations would be high, 4 more cases could be added by sputum cultures. In this way, the dispensaries would be capable of finding a substantial proportion of the tuberculosis cases still unknown to the clinic. This is estimated below. It was encouraging to learn that 83% of new cases among persons contacting the city dispensaries were sensitive to isoniazid compared with 74 %Y of the new cases among those attending the specialized tuberculosis clinic in Bangalore City (Rao, 1967). This difference is, however, not statistically sig- nificant. The third important finding is that, out of the tuberculosis cases discovered at all general dis- pensaries, 30% were already known to the city tuberculosis programme (out of the 20 cases, 6 were already known). This finding is interpreted as fol- lows: in spite of the existence and influence of specialized tuberculosis institutions in a city, patients in whom tuberculosis has already been diagnosed under the programme do continue to visit the general dispensaries. This may be either because they fail to derive any special benefit from attending the specialized centre or because they prefer to take treatment from general health institutions nearer home. Thus, the general health services in a city emerge as institutions exerting considerable in- fluence and attraction on the public and on known cases. It is tempting to estimate what case-yield could be expected from dispensaries during any one year, and what contribution they would make to the total number of cases diagnosed in the city during the same period in an integrated case-finding pro- gramme. If all the dispensaries functioned through- out the year at the same intensity as on the study days, that is, on about 312 working days, they would have the potentiality of contributing about 3000 direct-smear-positive cases to the programme. In view of the influence probably exerted by the mobile X-ray unit used by the special investigation team, and possibly by seasonal variations, it may be more realistic to place the expected case-yield at 1500 in a year, to which the yearly yield of 940 cases at the tuberculosis clinic (unpublished records of Lady Willingdon Tuberculosis Demonstration and Training Centre) should be added. With an estimated prevalence of about 5000 cases in the city 1 and an estimated annual incidence of about 1670 cases (one-third of the prevalence) 2 such a potential case- yield is nearly half of the prevalence and substan- tially higher than the annual incidence. The impor- tance of general dispensaries in the tuberculosis pro- gramme is evident because the case-yield of the city centre alone is far below the estimated incidence. Considering also that out-patients attending general dispensaries are mostly city residents, it appears justi- fiable that all general dispensaries should participate in tuberculosis case-finding and treatment within the city. Whereas the findings of this study indicate the 1 The population of Bangalore City was estimated to be 1.4 million in 1966, based on an annual rate of population increase since the 1961 census of 2.01 %. Of this population, 1.2 million persons will be in the 5 years or more age-group(86% in Bangalore district census of 1961). Applying the bacillary prevalence rate of 4.33 per thousand persons, that is, the average prevalence rate of all cities sampled in the National Sample Survey in 1957 (Indian Council of Medical Research, 1959, p. 53) to the estimated population, the expected prevalence of bacillary cases in Bangalore City is 5037. In another survey, it was found that 97°% of such cases were in the 10 years or more age-group. Thus the number of cases in the 10 years or more age-group will be 4866, or approximately 5000. ' Preliminary analysis of longitudinal survey data from the National Tuberculosis Institute, Bangalore. 40 G. D. GOTHI AND OTHERS potentialities of case-finding, the emphasis must be on operational factors such as the (1) participation of doctors in charge of general dispensaries in the proper selection of out-patients with chest symptoms for further examination; (2) determination of the place and method of investigation: (a) sputum examination by direct microscopy at the health institution itself, (b) preparation of sputum smears at the health institution and their examination at the main tuberculosis centre, (c) referring the patient directly to the centre for examination; (3) seeking patients' co-operation in the urban referral system; (4) investigation of the practicability of retrieving patients from whom sputum specimens were col- lected and found positive; (5) provision of additional equipment and personnel for the main centre or public health laboratory to permit the increased work-load to be supported. A study designed to obtain precise information on these points is strongly indicated. ACKNOWLEDGEMENTS The authors are grateful to Dr D. R. Nagpaul, Director, National Tuberculosis Institute, for his guidance and constructive suggestions; to Mr S. S. Nair, Seniop Statistical Officer, for his valuable suggestions; to Dr K. P. Rao, for undertaking the bacteriological work pertaining to the study; to Dr Puttabutta, District Health Officer, Bangalore District, for permitting the study to be undertaken at the city dispensaries; and to Dr Susai Mary, Superintendent, Lady Willingdon Tuberculosis Demonstration and Training Centre, who allowed the use of the city centre's records and took the newly diagnosed cases into her care for treatment. The authors are most appreciative also of the hard work put into this study by the field teams. RISUMt CAS DE TUBERCULOSE PULMONAIRE PARMI LES MALADES NON HOSPITALISES FREQUENTANT DES ETABLISSEMENTS DE SOINS GENERAUX DANS UNE VILLE DE L'INDE Une enquete a 6t6 men6e en 1966 a Bangalore afin d'evaluer la proportion des personnes presentant des symptomes pulmonaires ou atteintes de tuberculose pul- monaire parmi les patients non hospitalises traites au cours d'une journ6e dans 19 dispensaires generaux. Sur 2506 sujets interroges, 1170 (47%) ont indique comme motif principal de la consultation l'existence de symptomes pulmonaires et 49 (2%) ont ete reconnus a l'examen radiologique atteints de tuberculose pulmonaire evolutive ou probablement evolutive. Le diagnostic a ete confirm6 par l'examen microscopique direct dans 18 cas - auxquels sont venus s'ajouter 2 cas a image radio- logique normale mais a expectoration positive -, les 31 personnes restantes 6tant class6es comme . cas sus- pects *. Parmi ces 20 cas confirm6s et ces 31 cas suspects, 13 (6 cas confirm6s et 7 cas suspects) etaient d6ja inscrits dans les registres du Centre de lutte antituberculeuse. Au total, 38 cas 4 r6ellement nouveaux * ont et6 d6pistes; la proportion des malades excr6teurs de bacilles par rap- port A 1'ensemble des malades de plus de 10 ans 6tait de 0,8 %. Parmi les 20 sujets donnant des 6talements directs positifs, 18 ont fourni egalement des cultures positives. On comptait parmi eux 12 cas r6ellement nouveaux, dont 2 (17%) etaient porteurs de bacilies resistants i l'isoniazide. Bien que limit6e dans le temps, l'enquete montre que, meme dans un milieu urbain 6volu6 du sud de l'Inde, beaucoup de personnes pr6sentant des symptbmes pul- monaires s'adressent en premier lieu aux 6tablissements de soins gen6raux; ces derniers, de ce fait, contribuent efficacement au d6pistage de la tuberculose. L'activit6 d'un jour de 19 de ces formations m6dicales, a Bangalore, a conduit a d6couvrir 14 cas confirm6s et 24 cas suspects. Le role des dispensaires g6n6raux urbains dans le pro- gramme global de depistage et de traitement de la tuber- culose merite d'etre soulign6 et intensifi6. REFERENCES Baily, G. V. J., Savic, D., Gothi, G. D., Naidu, V. B. & Nair, S. S. (1967) Bull. Wld Hlth Org., 37, 875-892 Andersen, S. & Banerji, D. (1963) Bull. Wid Hlth Org., 29, 685-700 Indian Council of Medical Research (1959) Tuberculosis in India, a sample survey (1955-1958), New Delhi, p. 53 (Special Report Series, No. 34) Nagpaul, D. R. (1967) Indian J. Tuberc., 14, 186-198 Nagpaul, D. R., Vishwanath, M. K. & Dwarakanath, G. (1970) Bull. Wld Hlth Org., 43, 17-34 Piot, M. A. (1962) Indian J. Tuberc., 9, 151-156 Rao, K. P. (1967) A comparative study of the prevalence of drug resistance to major antituberculosis drugs. In: Proceedings ofthe 22nd Tuberculosis and Chest Diseases Workers' Conference, Hyderabad, p. 55 WHO Expert Committee on Tuberculosis (1964) Wld Hlth Org. techn. Rep. Ser., No. 290, p. 8