CASE-FINDING BY MICROSCOPY 1 2. / 3. D. R. Nagpaul, D. M. Sav1c, K. P. Rao, 4 and G. V. J. Baily 5

Similar documents
WORLD HEALTH ORGANIZATION ORGANISATION MONDIALE DE LA SANTÉ ENGLISH ONLY. Working Paper No. 32

TECHNICAL AND OPERATIONAL APPRAISAL OF TUBERCULOSIS CASE-FINDING METHODS

Prevalence of infection among unvaccinated children for tuberculosis surveillance

A.S. BAGGA, M.S. KRISHNA MURTHY, K.R. RANGASWAMY AND M.S. KRISHNA MURTHY (From Tuberculosis Prevention Trial, Bangalore). surveys.

NOTES ON TUBERCULOSIS CONTROL IN NICARAGUA

IS CHEST X-RAY SCREENING FOR PULMONARY TUBERCULOSIS BY MASS RADIOGRAPHY- A COST-EFFECTIVE TOOL IN A MILITARY POPULATION?

Protocol of National TB Prevalence Survey in Cambodia

TWICE WEEKLY STREPTOMYCIN AND ISONIAZID REGIMEN WITHOUT SUPERVISION. J.L. BHATIA AND BALDEV RAJ (From Medical College, Amritsar)

M.A. SEETHA, G.E. RUPERT SAMUEL and V.B. NAIDU (From National TB Institute, Bangalore)

Background & objectives

POLICY FOR THE PREVENTION AND CONTROL OF TUBERCULOSIS

Managing the Revised National Tuberculosis Control Programme in Your Area. A Training Course. Modules 1 4

Tuberculosis prevalence survey in Kashmir valley

EFFECT OF STORAGE FOR THREE MONTHS AT DIFFERENT TEMPERATURES ON THE SENSITIVITY TO STREPTOMYCIN AND ISONIAZID OF CULTURES OF TUBERCLE BACILLI.

Key Words: Pulmonary Tuberculosis; Smear and Culture Findings; India METHODS

IMPACT OF IMPROVED TREATMENT SUCCESS ON THE PREVALENCE OF TB IN A RURAL COMMUNITY BASED ON ACTIVE SURVEILLANCE

Comparative Value of Sputum Smear Examination and Culture Examination in Assessing the Progress of Tuberculous Patients Receiving Chemotherapy *

MDR, XDR and Untreatable Tuberculosis and Laboratory Perspectives. Martie van der Walt TUBERCULOSIS EPIDEMIOLOGY & INTERVENTION RESEARCH UNIT

Pulmonary tuberculosis among contacts of patients with tuberculosis in an urban Indian population

TRENDS IN THE PREVALENCE OF PULMONARY TUBERCULOSIS OVER A PERIOD OF SEVEN AND HALF YEARS IN A RURAL COMMUNITY IN SOUTH INDIA WITH

Observations on the Protective Effect of

THE SOUTH AFRICAN SOCIETY OF OCCUPATIONAL MEDICINE

HEALTH ORGANISATIONS. National Health Programme

CONTROL OF TUBERCULOSIS. Note submitted by the Government of the United States of.america. Introduction

Guideline for the Surveillance of Pandemic Influenza (From Phase 4 Onwards)

MEASURE TO IMPROVE DETECTON OF SMEAR POSITIVE CASES UNDER RNTCP: COMPARISION OF COUGH 2WEEKS VS 3 WEEKS

(a) Infection control program. The facility must establish an infection control program under which it--

TB Prevalence Survey. -Census, Interview and X ray- Ikushi Onozaki MD, MPH STB, WHO Geneva

QUALITY CONTROL IN ISOLATION AND IDENTIFICATION OF MYCOBACTERIA FROM CLINICAL SPECIMENS

MedicalBiostatistics.com

Principle of Tuberculosis Control. CHIANG Chen-Yuan MD, MPH, DrPhilos

Appendix C. Recommendations for Counting Reported Tuberculosis Cases (Revised July 1997)

Epidemiology NTI Bulletin 2003, 39 / 1&2, 15-23

SOCIO-CULTURAL CONTEXT OF TB TREATMENT: A CASE STUDY OF SOUTHERN GUJARAT

TUBERCULOSIS IN NORTH ARCOT DISTRICT OF TAMIL NADU - A SAMPLE SURVEY

Clinical Practice Guideline

Achieving Polio Eradication in India. Emergency Preparedness and Response Plan 2011

A TUBERCULOSIS PREVALENCE SURVEY BASED ON SYMPTOMS QUESTIONING AND SPUTUM EXAMINATION

Impact of BCG vaccination on tuberculin surveys to estimate the annual risk of tuberculosis infection in south India

INTENSIFIED TB CASE FINDING

Sputum grading as a predictor of treatment outcome of new sputum smear positive tuberculosis patients in Khammam Tuberculosis Unit

Reasons for vaccine acceptance: parents and girls perspectives

TB & HIV. GOVERNMENT OF INDIA NATIONAL TUBERCULOSIS INSTITUTE No.8, Bellary Road, Bangalore INDIA

Evaluation of Malaria Surveillance System in Hooghly district of West Bengal - India

Ethiopia. Targeted Tuberculosis Case Finding Interventions in Six Mining Shafts in Remote Districts of Oromia Region in Ethiopia PROJECT CONTEXT

Background paper number 7

MEMORANDUM. Re: Guidance for follow-up of newly-arrived Individual with a Class B1 Tuberculosis Extrapulmonary Tuberculosis

Performance of RNTCP NTI Bulletin 2005,41/3&4,

Appendix B. Recommendations for Counting Reported Tuberculosis Cases (Revised July 1997)

MEMORANDUM. Re: Guidance for follow-up of newly-arrived individual with Class B1 Tuberculosis Pulmonary Tuberculosis, no treatment

Active community surveillance of the impact of different tuberculosis control measures, Tiruvallur, South India,

Guidance on Matching Funds: Tuberculosis Finding the Missing People with TB

EQUIVALENCE OF ACID ALONE OR ACID-ALCOHOL AS DECOLOURIZING AGENT IN ZIEHL - NEELSEN METHOD

National TB Prevalence survey in Myanmar. By Dr. Thandar Lwin Programme Manager National TB Programme Department of Health

Analysis of TB prevalence surveys

Th e experience of the Cattaraugus County Bureau

ANNEXURE A: EXPLANATORY NOTES ON THE DMR 164 REPORTING ON HIV AND TB FORM

CORRELATES OF DELAYED INITIATION OF TREATMENT AFTER CONFIRMED DIAGNOSIS UNDER RNTCP: A CROSS SECTIONAL STUDY IN AHMEDABAD MUNICIPAL CORPORATION, INDIA

TB IN EMERGENCIES. Disease Control in Humanitarian Emergencies (DCE)

TUBERCULOSIS AND ITS CONTROL IN VIETNAM. Report submitted by the Government of Vietnam

Targeted Testing and the Diagnosis of. Latent Tuberculosis. Infection and Tuberculosis Disease

NATIONAL HEALTH MISSION OF INDIA. Dr. Rajesh Kumar, MD PGIMER School of Public Health Chandigarh (India)

IJEMR December Vol 7 Issue 12 - Online - ISSN Print - ISSN

Why do people in India avoid government doctors? Debasis Barik 1, Sonalde Desai 2 and Reeve Vanneman 3

Tuberculosis Procedure ICPr016. Table of Contents

PREVALENCE OF DRUG RESISTANCE IN PATIENTS WITH PULMONARY TUBERCULOSIS PRESENTING FOR THE FIRST TIME WITH SYMPTOMS AT CHEST CLINICS IN INDIA*.

11/3/2009 SECOND EDITION Madhukar Pai McGill University. ISTC Training Modules Introduction

TOG The Way Forward

Title: Meta-analysis of Individual patient Data (IPD) of Patients with INH (mono or poly-drug) Resistant Tuberculosis.

Version No. 7 Date: July Please send comments or suggestions on this glossary to

IMMUNIZATION COVERAGE OF INFANTS RURAL-URBAN DIFFERENCE IN KERALA

Appendix B. Recommendations for Counting Reported Tuberculosis Cases (Revised July 1997)

Active case finding. Care Pathways: Seek Find Follow

International Standard for Athlete Evaluation. July 2015

CMH Working Paper Series

INFECTION PREVENTION AND CONTROL POLICY AND PROCEDURES Sussex Partnership NHS Foundation Trust (The Trust)

Table 9. Policy for Tuberculosis Surveillance and Screening

Tuberculosis - clinical forms. Dr. A.Torossian,, M.D., Ph. D. Department of Respiratory Diseases

TB/HIV prevention. Implementation science and TB prevention

BCG. 120 Kul Bhushan: ASSESSMENT OF BCG VACCINATION IN INDIA: THIRD REPORT Indian J Med Res 1960, 48,

TB Clinical Guidelines: Revision Highlights March 2014

Open Access Article pissn eissn

Action Plan of China Malaria Elimination ( ) Page 2

4:ll.J1 ~I 41 ~.J,II,,1'1 'IJ".:J ~.J,II...:.CJI WORLD HEALTH ORGANIZATION ORGANISA'TION MONDIALE DE LA SANT~

"GUARDING AGAINST TUBERCULOSIS IN INSTITUTIONAL FACILITIES"

Social Awareness NTI Bulletin 2005,41/1&2, 11-17

Interview National health database crucial to providing preventive care to vulnerable sections. S. Rajendran Feb 17, 2017

CHAPTER 2. Literature review

CHAPTER 14 ORAL HEALTH AND ORAL CARE IN ADULTS

Biology and Medicine

Let s Talk TB A Series on Tuberculosis, A Disease That Affects Over 2 Million Indians Every Year

Recently two of the Senior Lady Gynecologists at a well known Govt. Hospital in Bangalore were diagnosed with advanced stage Breast Cancer.

PREVALENCE OF HIV INFECTION AND RISK FACTORS OF TUBERCULIN INFECTION AMONG HOUSEHOLD CONTACTS IN AN HIV EPIDEMIC AREA: CHIANG RAI PROVINCE, THAILAND

Blindness Prevalence Rates in Egypt

TB Program Management San Antonio, Texas November 5-7, 2008

Clinical aspects of tuberculosis with directly observed treatment in Mehsana district India

HEALTH SYSTEM STRENGTHENING UNDER THE NATIONAL RURAL HEALTH MISSION (NRHM) IN INDIA

International Standards for Tuberculosis Care Barbara J. Seaworth, MD August 22, 2007

Xpert MTB/RIF use for TB diagnosis in TB suspects with no significant risk of drug resistance or HIV infection. Results of Group Work

Kenya Perspectives. Post-2015 Development Agenda. Tuberculosis

Transcription:

.,, ' ) 'WORLD HEALTH ORGANIZATION ORGANISATION DE LA SANTE MONDIALE WHO/TB/Techn.Information/68.63 ENGLISH ONLY CASE-FINDING BY MICROSCOPY 1 by 2. / 3 D. R. Nagpaul, D. M. Sav1c, K. P. Rao, 4 and G. V. J. Baily 5 INTRODUCTION The two main tuberculosis control measures nowadays are BCG vaccination and chemotherapy. BCG vaccination has been accepted in many countries as the basis of tuberculosis control programmes, and will have to be continued in these countries until the disease has been eliminated as a serious public health problem. Case-finding cannot reduce the tuberculosis problem in a community, but must form part of a comprehensive control programme employing all the available means. What, then, is the point of discussing case-finding - and particularly case-finding by microscopy - separately from the tuberculosis control programmes as a whole? Microscopy is believed to be the most practicable and effective case-finding method for most developing countries. The present study was therefore undertaken to evaluate this method from the technical and operational points of view. control are limited, the highest priority should be This policy is dictated by epidemiological, socio Where resources for tube~culosis given to treating infectious cases. logical and economic considerations. Epidemiological considerations The WHO Expert Committee on Tuberculosj.s, in its eighth report, defined a case as follows: "from the epidemiological point of view, a 'case' of pulmonary tuberculosis means a person suffering from bacteriologically confirmed disease All other possible 1 This paper, which was read at the Nineteenth International Tuberculosis Conference, Amsterdam, 3-7 October 1967, will be published in the proceedings of that Conference. It is issued in the WHO/TB/Techn.Information series by kind permission of the International Union against Tuberculosis. 2 Director, National Tuberculosis Institute, Bangalore, India. 3 Senior WHO Officer, National Tuberculosis Institute, Bangalore, India. 4 Senior Bacteriologist, National Tuberculosis Institute, Bangalore, India. 5 Medical Officer, National Tuberculosis Institute, Bangalore, India. The issue of this document does not constitute formal publication. lt should not be reviewed, abstracted or quoted without the agreement of the World Health Organization. Authors alone are responsible for views expressed in signed articles. Ce document ne constitue pas une publication. 11 ne doit faire l'objet d'aucun compte rendu ou resume ni d'aucune citation sans l'autorisation de!'organisation Mondiale de la Sante. Les opinions exprimees dans les articles signes n'engagent que leurs auteurs.

page 2 sufferers from tuberculosis, i.e., those in whom the disease has not been confirmed bacteriologically, would be classified as 'suspect cases' and would remain so classified unless or until the presence of tubercle bacilli or some other etiology was established" (World Health Organization, 1964). In other words, ~and suspect cases are two distinctly different epidemiological entities. The implications of this concept are being studied in a longitudinal survey by the National Tuberculosis Institute (NTI) in Bangalore (unpublished data). The survey was started in May 1961 in three "taluks" (sub-divisions) of the rural part of Bangalore District, Mysore State. Out of the total of 734 villages, 133, with a total population of 77 000, were randomly selected. Children less than five years old were excluded from the survey. The methods used were: tuberculin testing, radiography, bacteriological examination (microscopy of sputum smears, and sputum culture), testing of mycobacterial sensitivity to drugs, and testing for the identification of mycobacteria. The same study population was examined again 18 months later. There was no organized control programme in the survey area that could have interfered with the natural trend of tuberculosis incidence in that community. The bateriological status after 12 months, i11 various categories of the study population from 119 villages, is shown in Table 1: TABLE 1. BACTERIOLOGICAL STATUS 12 MONTHS AFTER THE FIRST EXAMINATION Initial diagnosis Number in group Culture-positive 12 months later % A. Radiologically normal (according 28 961 0.08 to two readers) (28 376) B. Radiologically abnormal (unconfirmed by culture) (a) Non-tuberculous or inactive 3 308 0.44 (according to both readers) (2 878) (b) Active or probably active 623 0.58 (according to one reader) (459) (c) Active or probably active 496 4.84 (confirmed by umpire reader) (358) c. Radiologically abnormal (confirmed bacteriologically) (a) Culture-positive only 98 35.19 * (72) (b) Smear-positive, confirmed by 81 42.96 * culture ( 45) * Difference not significant. followed up. Figures in brackets represent persons Roelsgaard et al. (1964), on the basis of data from WHO-assisted tuberculosis prevalence. surveys in Africa, have found that the prevalence of infection among children and teenagers living in household contact with cases of tuberculosis was higher than in those who lived in

WHO/TB/Techn. l.jformation/68. 63 page 3 households where there were no cases. The prevalence of infection in households with a suspect case was between these two extremes, but closer to that in the non-tuberculous than in the tuberculous households. In comparing the infectiousness of cases, Raj Narain et al. (1966) have shown that the proportions of infected children aged 0 to 14 years in households with cases, suspect cases and no cases were 41%, 20% and 12% respectively. This material indicates a striking difference between the infectiousness of cases and that of suspect cases. However, other studies provide strong evidence that the potential risk of acquiring infection from cases confirmed on culture only is considerably lower than from cases with tubercle b~cilli detectable by direct microscopy. Shaw & Wynn-Williams (1954) have demonstrated that, among children 0 to 14 years of age who lived in contact with smear-positive, culture-positive and suspect cases, and among those who were not in contact with such cases, the prevalence of infection was 65.2%, 26.8%, 17.6% and 22.1% respectively. Grzybowski & Alien (1964) studied the prevalence of active tuberculosis among 1116 close contacts of persons suffering from pulmonary tuberculosis. The prevalence was 6.5% when the sputum of the index case was positive by direct microscopy, 1.3% when the bacilli from the index case were found only on culture, and 1.1% when the culture was negative. These different rates of prevalence may indeed reflect differences in the infectivity of the index cases if one assumes an equal duration of exposure. These examples illustrate the difference in epidemiological significance between cases and suspect cases. They show also that the infectiousness of smear-positive cases was higher than that of cases found positive only by culture. Economic considerations Since, in any country, the resources allocated to tuberculosis control are limited, they should be used in the most rational way. Unfortunately, few countries appear to have tried to derive the maximum benefit out of the resources available to them for tuberculosis control. For epidemiological reasons, it is clear that the finding of cases detectable by direct microscopy deserves top priority in a control programme. A similar conclusion is reached when economic implications are taken into consideration. Rough estimates have been made at NTI of the cost of all operations (including the use of transport for supervision) involved in finding one sputum-positive case by different methods, in actually existing situations, in groups with varying prevalence of tuberculosis. Table 2 gives a comparison of the different case-finding methods according to cost: TABLE 2. COST (IN US$) OF DIAGNOSING ONE CASE BY DIFFERENT METHODS Examination Approximate cost of one examination Approximate cost of diagnosing one case Direct microscopy 0.21 3.4 Culture examination 0.49 12.1 70-mm x-ray film (Static unit) 0.26 3.5 70-mm X-ray film (Mobile unit) 0.50 73.0

page 4 The cost of direct microscopy has been estimated on the basis of unpublished NTI data connected with the assessment of district tuberculosis programmes. In these programmes, direct microscopy is done for a few persons at a time, by general paramedical staff at peripheral health institution~ under the supervision of a district tuberculosis centre. Approximately 6.2 per cent. (see Table 6) of persons reporting symptoms suggestive of pulmonary tuberculosis would be smear-positive by direct microscopy undertaken at peripheral health institutions at a cost per case of $ 3.4. Facilities for sputum culture exist only at state tuberculosis control centres. One of the aims of providing culture facilities is to find cases that would be missed by direct microscopy. Persons reporting symptoms are first screened by mass miniature radiography and those with X-ray shadows are offered sputum examination by microscopy and culture. On the basis of estimates given in an unpublished NTI document on state tuberculosis control centre laboratories, the cost of finding a case by culture alone works out at $ 12.1. However, the cost of diagnosing a case for the purpose of treatment should include not only the expense of the actual examination, but also that of retrieving the patient for treatment. Patients found by direct microscopy are put on treatment immediately, whereas those who are culture-positive will need to be retrieved when the result of culture becomes available. Thus, operationally, the cost of culture for finding additional cases would be higher still. Each district tuberculosis centre in India is provided with a static mass miniature radiography unit, These centres provide X-ray examination for microscopy-negative persons referred to them by peripheral health institutions, as well as for those who attend spontaneously. Generally speaking, 30 per cent. of such persons appear to be tuberculous according to X-ray examination, but only 25 per cent. of these are confirmed as 'cases' by direct microscopy. Again, according to unpublished NTI data connected with the assessment of district tuberculosis programmes, it would cost $ 3.5 to diagnose one microscopypositive case through a static X-ray unit at a district tuberculosis centre. In contrast, the cost of finding one case by mobile mass miniature radiography is very high because the operational costs are quite out of proportion to the yield of cases. However, the casefinding situations described are not strictly comparable as regards cost, The full significance of these costs would be revealed only if they were expressed as a single component of the full chain of operations under programme conditions: diagnosis - treatment - treatment results. An operational model of this is at present being worked out at NTI. Sociological considerations From the planning point of view, it is essential to classify the awareness of chest symptoms among persons with tuberculosis as (a) mere consciousness of one or more symptoms; (b) awareness of one or more symptoms, causing worry; or (c) awareness of symptoms, leading to action, A person may suffer direct economic loss through an illness such as tuberculosis; may be deprived of certain rigpts; or, in extreme cases, have to change his profession (e.g., teacher). If someone knows the significance of certain symptoms and is worried about them, he may seek medical advice soon after their onset. Indeed, in countries where mass casefinding is carried out regularly and frequently, it happens that tuberculosis is often diagnosed between case-finding rounds in persons who have reported symptoms. Toman (1966) reports that, in Czechoslovakia during the period 1963-1964, l 118 068 per~ons over 15 years of age were examined, and 1387 cases of tuberculosis were detected as a result. A striking fact is that 42% of these cases were found among the 6% of persons examined because of chest symptoms.

page 5 EVALUATION OF DIRECT MICROSCOPY UNDER PROGRAMME CONDITIONS In a district tuberculosis programme in India, case-finding consists in (a) direct microscopy of spot sputum specimens collected from persons who have reported symptoms suggesting pulmonary tuberculosis, and who have presented themselves at rural general health institutions that are equipped with microscopes; and (b) referring abacillary persons with chest symptoms to the district tuberculosis centre for a 70-mm X-ray. The evaluation of direct microscopy was carried out by NTI in two ways; l. Technical (the reliability of the method when performed at a peripheral health institution); and 2. Operational (the workload and corresponding case yield); In addition, the case-finding potentialities of the method were estimated. Technical evaluation In the district tuberculosis programme in India, most peripheral health institutions carry out direct microscopy of sputum specimens. Paramedical personnel working in these institutions are trained in the techniques of direct microscopy by a qualified laboratory technician attached to the district tuberculosis centre. The staff of the centre is responsible also for regular supervision of the direct microscopy done in the peripheral institutions. 1 Technical evaluation of the d1rect microscopy performed at these institutions was undertaken by the NTI in Bangalore District during the period September 1963 - May 1965. The district, excluding the metropolitan area, extends over 3046 square miles and, at the time of the 1961 census, had a population of about 1.3 millions in its 13 towns and 2477 villages. Nine peripheral institutions were selected randomly, and sputum slides that had been prepared and read by the paramedical personnel in the field were brought to the laboratory of the National Tuberculosis Institute for re-examination. The same sputum specimens were also brought to the NTI laboratory, where duplicate smears as well as cultures were made. The material was analysed in respect of over- and under-reading of smears. In order to estimate the extent of over-reading at peripheral institutions, specimens originally found to be culture-negative at the NTI were "read" at the nine selected peripheral institutions and then again at the National Tuberculosis Institute (Table 3). l Rao, K. P. et al. (1967) Assessment of sputum examination in a district tuberculosis programme (to be published).

page 6 TABLE 3, OVER-READING OF CULTURE-NEGATIVE SPECIMENS BY PERIPHERAL HEALTH INSTITUTIONS PHI Culture- % read as % re-read as Code negative at smear-positive smear-positive No. NTI at PHI's at NTI l 306 1.6 1.3 2 233 3.4 0.4 3 159 4.4 4.4 4 156 1.3 1.3 5 lob ll.l 1.9 6 lll 2.7 0.9 7 loo l.o l.o 8 84 o.o 1.2 9 196 o.o o.o Total l 453 2.6 1.3 Out of the 1453 specimens originally culture-negative, 2.6% were read as microscopypositive by the peripheral institutions. On the other hand, 1.3% of the same negative specimens were also read as positive when re-examined at the NTI. More detailed analysis shows that this difference in respect of over-reading was mainly attributable to one of the peripheral institutions: No. 5. Had this institution performed with the same efficiency as the other eight, over-reading by the peripheral institutions as a whole could have been reduced to 1.9%, when the difference between the estimated over-reading at the peripheral institutions and at the NTI would cease to be of any importance. In order to estimate the extent of under-reading, culture-positive specimens from the NTI were examined by direct microscopy at the peripheral health institutions and again at the NTI (Table 4), TABLE 4. UNDER-READING OF CULTURE-POSITIVE SPECIMENS BY PERIPHERAL HEALTH INSTITUTIONS PHI Culture- % read as % re-read as Code positive at smear-negative smear-negative No. NTI at PHI's at NTI l 101 26.7 25.7 2 21 33.3 38.1 3 23 30.4 21.7 4 22 86.4 40.9 5 15 40.0 40.0 6 16 31.2 25.0 7 15 46.7 33.3 8 lo 80.0 30.0 9 5 20.0 20.0 Total 228 38.2 29.4

page 7 It may be seen that, of 228 culture-positive specimens, 38.2% were read as microscopynegative by the peripheral institutions and 29.4% were re-read as negative by the NTI. This difference was attributable mainly to two centres (numbers 4 and 8). Under-reading by the other seven rural institutions was nearly as low as at the NTI, that is, about 30%. It is clear from these experiments that, with proper supervision and corrective training from the experienced laboratory technicians of district tuberculosis centres, these peripheral institutions could become highly proficient at direct microscopy. Operational evaluation An operational field study was carried out from February 1966 to December 1966 in Tumkur District, Mysore State, South India, in order to estimate the diagnostic workload and case yield by direct microscopy performed at peripheral health institutions. 1 Tumkur District, in size and density of population, corresponds to an average Indian district (4058 square miles in area; 2392 villages; 11 towns; population 1 522 000). In addition to the general hospital and the district tuberculosis centre in the chief town of the district, there are 94 peripheral health institutions, run by the Government. These institutions are of different types, depending on their functions. In the district in question, three different types are recognized: primary health centres, health units, and rural dispensaries. The primary health centres and the health units perform both curative and preventive functions, whereas rural dispensaries have only curative functions. The preventive functions carried out by primary health centres include smallpox immunization, sanitation, family welfare, maternity and child health - all systematically organized so that each primary health centre covers an area with a population of about 75 000 distributed among some loo villages. The preventive functions performed by a health unit are limited to a population of about 20 000. In addition, all peripheral health institutions have general out-patient departments. The Tumkur District tuberculosis programme was organized two years before this study began. It is the district tuberculosis centre which bears the main responsibility for organizing and supervising the programme. The centre also runs a tuberculosis clinic serving mainly the population of the town and some nearby villages. Finally, it provides radiography for patients who are referred to it from peripheral institutions. This, then, is the only specialized tuberculosis centre in the district. Of the 94 peripheral institutions, 55 participate in the district tuberculosis programme: 38 of these are "microscopy centres", i.e., they are equipped to carry out direct microscopy of sputum specimens; the other 17, known as "referring centres", are not provided with microscopes, but collect sputa on the spot and send the smears to the nearest microscopy centre for examination. The 55 participating institutions are entrusted with the domiciliary treatment of tuberculosis patients. A stratified random sample consisting of 15 institutions - eight microscopy centres and seven referring centres - was selected for this study. The study period for all the selected institutions was one month or 26 consecutive working days. During the study period, the total out-patient attendance in all age-groups was 14 881. Because of the very low expectation of cases in the age-group 0 to 9 years, this group was excluded from the study, leaving a total study population of 10 792 (Table 5). 1 Baily, G. V. J. et al. (1967) Potential case-yield by direct microscopy (to be published).

page 8 TABLE 5. MICROSCOPY-POSITIVE CASES AMONG NEW OUT-PATIENTS WITH CHEST SYMPTOMS Number microscopy- Number of % with chest positive among patients symptoms those with symptoms Males 7 204 6.9 36 Females 3 588 6.4 9 Both sexes 10 792 6.7 45 Statistical analysis reveals that the proportion of patients with chest symptoms increases with age. Out of the 10 792 newout-patients, 724 (6,7%) had chest symptoms of varying duration. A sputum specimen was collected from each patient with symptoms, and all the patients were examined. Forty-five new cases were diagnosed. Of the 724 persons with chest symptoms, 352 (24 cases) presented themselves at microscopy centres and 192 (21 cases) at referring centres. In order to detect all 45 cases, 724 sputum specimens had to be examined in 208 working days at eight microscopy centres, that is, just under four smear examinations per day per centre. Yet by concentrating only on the patients who spontaneously reported a cough with a duration of 14 days or longer, the workload could have been reduced to two smear examinations per centre per day, and only two of the 45 cases would have been missed (Table 6). TABLE 6. OPTIMIZATION OF CASE-FINDING BY DIRECT MICROSCOPY IN PERSONS WITH CHEST SYMPTOMS Persons with chest symptoms Number of persons (examinations) % Number of cases smear-positive % I of less than 14 days' 241 33.2 l 2.2 Persons I duration spontaneously reporting a I of more than cough I 14 days' 381 52.6 43 95.5 I duration Persons with a cough (as eliciteq by an interviewer) 101 13.9 l 2.2 Persons with a pain in the chest only l 0.3 - - Total 724 100.0 45 100.0

WHO/TB/Techn.Information/68,63 page 9 All 724 patients presenting symptoms were referred also to the district tuberculosis centre for radiological examination by a static X-ray unit (70-mm camera) (Table 7). TABLE 7. PATIENTS' RESPONSE WHEN REFERRED FOR RADIOLOGICAL EXAMINATION Total Number Number smearpositive Smear-negative Number % Patients with chest symptoms 724 45 679 loo Willing to go 490 39 451 66 Reported at district tuberculosis centre 76 11 65 10 Returned to peripheral institution to learn result of radiological examination 55 10 45 7 Of the patients with chest symptoms, 490 (66%) were willing to contact the district tuberculosis centre, but only 76 (10%) actually went there and only 55 (7%) eventually returned to the referring institution to learn the result of their radiological examination. Among the persons who visited the district tuberculosis centre, there were only 10 cases and 5 suspect cases. Case-finding potentialities An attempt was made to estimate how many of the cases that would be sputum-positive on direct microscopy could be detected if, in the rural part of Tumkur district, all 55 rural institutions worked in the same way. The rural population is estimated to be 1 466 000. Of these, some 1 033 000 are 10 or more years of age, and it is believed that these include about 2890 cases that would be positive on direct microscopy. In addition, approximately 1340 cases would be positive on culture only. The 55 rural health institutions would diagnose annually about 60% of the cases that would be microscopy-positive and about 41% of all cases including those that would be culture-positive (Table 8). TABLE 8. EXPECTED ANNUAL CASE YIELD IN THE RURAL PART OF TUMKUR DISTRICT (POPULATION AGED 10 YEARS OR OVER: 1 033 000) Prevalence Number Estimated annual case yield a percentage of cases as Total smear-positive 2 890 60.2 Total cases including culture-positive 4 230 41.1

page 10 No doubt, if such a large proportion of the total pre\alence were kept under control continuously, one might expect a substantial reduction of the tuberculosis problem in the area. CONCLUSIONS Epidemiological considerations demonstrate clearly the high significance of microscopypositive cases in the transmission of tuberculous infection to the susceptible population. There is strong evidence that patients from whom bacilli can be isolated only by culture represent a lower risk to the community. Suspect cases have generally a good prognosis, and very few of them are likely to become infectious. Such epidemiological ranking of various categories of tuberculosis cases and suspects forms a sound basis for planning and for the allocation of available resources. There is evidence that, in countries where tuberculosis is prevalent, it affects the rural population just as much as the people living in crowded cities. As the population in those countries is predominantly rural, case-finding efforts oriented only or mainly towards city dwellers cannot exert the desired effect on the problem. This makes it mandatory t6 provide antituberculosis services all over the country. Simple and reliable diagnostic facilities for tuberculosis, such as direct microscopy, must therefore be integrated into the existing general rural health network. If the prevalence of tuberculosis in a country is high, then the total pool of infected persons in the community (being a function of cases) must also be large, as has been demonstrated by prevalence surveys. Infected individuals from this pool may develop the clinically active disease at any time. This is a continuous process which renders permanent diagnostic facilities an absolute necessity. Mass mobile case-finding facilities alone would detect only few of these cases. Between mass case-finding rounds, the infectious pool would continue to produce cases uninterruptedly. On the other hand, a firmly established network of permanent facilities for direct microscopy, integrated into peripheral health institutions, would leave few cases undetected. For economic and epidemiological reasons, infectious cases deserve the first priority in a tuberculosis control programme, and it is by rendering these cases bacteriologically quiescent that the greatest benefit will be derived from the available resources. Culture is of advantage in cases that are less infectious. Suspects, that is, persons with chest X-ray shadows, are obviously the third priority, especially in the early stages of development of a community-oriented programme. radiography is prohibitively expensive. The cost of detecting on~ case by mass mobile From the sociological considerations, it may be concluded that tuberculosis causes considerable suffering. Among the patients in rural areas, it is those who are the most aware of their symptoms who seek help from health institutions. The sociological implications for the planner of tuberculosis control services are far-reaching. It is the "felt need", caused by suffering, that must be satisfied first. The constant pressure that thousands of patients with chest symptoms exert on peripheral health institutions justifies the strengthening of these institutions so as to satisfy the demands made on them. Not only would this course alleviate human suffering, but it would increase public confidence in the health services. As a result, many people who normally would only have worried about their symptoms would be prompted to action. Operational considerations have demonstrated that the total number of out-patients includes a substantial proportion of patients presenting themselves at peripheral health institutions on account of symptoms suggesting pulmonary tuberculosis. By careful selection, about 96% of the cases in persons spontaneously presenting themselves could be found. This yield could be achieved by examining, on an average, only two sputum specimens daily at each institution, and such a "workload" should be quite acceptable to the staff working at the rural level.

WHO/TB/Techn.Information/68,63 page 11 The great operational advantage of diagnosis by direct microscopy is that it enables domiciliary treatment to be started immediately. In this way the loss of patients that otherwise would result from a delay between diagnosis and treatment can be virtually eliminated, The operation of culture facilities is difficult. In a programme that is based on case-finding in persons who complain of symptoms, the additional yield by culture is not great. Culture examination is complicated and requires highly trained staff. Finally, there is considerable delay between the collection of sputum and the receipt of culture results by which time the suspects who have turned out to be culture-positive are dispersed all over the area covered by the programme. Such persons are difficult to trace and retrieve for treatment, In the present study, patients responded poorly to the opportunity that was given to them of being referred to a district tuberculosis centre for radiological examination. If this is a typical response, the role of X-ray examination in a community-oriented programme would seem to be severely limited, The radiological service for the tuberculosis programme in the district concerned serves mainly patients residing in the town and in a few nearby villages. It remains to be seen whether it would not be more profitable to equip some conveniently located centres with multi-purpose X-ray units that would serve many more purposes besides the detection of chest abnormalities. Last, but not least, case-finding is merely a step towards treatment which is the actual control measure. From the operational point of view, merging suspect cases with true cases of tuberculosis results in an artificial increase in the number of patients to be treated on a domiciliary basis by the already strained peripheral health institutions, On the basis of the technical considerations, the following conclusions may be drawn: The fear that- as in the misinterpretation of radiological lung abnormalities - direct microscopy of sputum specimens could result in diagnostic errors, especially when it is performed by paramedical staff who have undergone only short training, has not been substantiated. In the present study, the observed over- and under-reading of slides was attributable mainly to a few workers who failed to attain the required technical standard. Most of them were as proficient as the "standard" microscopist. In an integrated programme, such deficiencies could be eliminated easily through constant supervision by a qualified laboratory technician from a supervisory centre, Moreover, it is likely that, in developing countries, it will soon be possible to train technicians from mass campaigns as multi-purpose laboratory workers and assign them to peripheral health institutions, REFERENCES Andersen, S. & Banerji, D. (1963) Bull. Wld Hlth Org., 29, 685-700 Grzybowski, s. & Allen, E. A. (1964) Amer. Rev. resp. Dis., 90, 707-720 Raj Narain, Nair, S. s., Ramanatha Rao, G. & Chandrasekhar, P. (1966) Bull. Wld Hlth Org., 34, 639-654 Roelsgaard, E., Iversen, E. & Bl~cher, C. (1964) Bull. Wld Hlth Org., 30, 459-518 Shaw, I, B. & Wynn-Williams, N. (1954) Amer. Rev. Tuberc,, 69, 724-231 Toman, K. (1966) World Health Organization document EURO 305/5 World Health Organization, Expert Committee on Tuberculosis (1964) Wld Hlth Org, techn. Rep. Ser., 290