INT J TUBERC LUNG DIS 2(9):S9–S15 © 1998 IUATLD
Expanding the WHO tuberculosis control strategy: rethinking the role of active case-finding C. J. L. Murray, J. A. Salomon Harvard Center for Population and Development Studies, Cambridge, Massachusetts, USA
EACH YEAR, there are more than 6.5 million new cases of tuberculosis and more than 2 million deaths from tuberculosis worldwide.1 Tuberculosis is the seventh most important cause of global premature mortality and disability,2 and is projected to remain one of the ten leading causes of disease burden even in the year 2020.3 Because of a powerful interaction between tuberculosis and the human immunodeficiency virus (HIV), the tuberculosis incidence rate is rising in sub-Saharan Africa and may rise in Asia. Increasing drug resistance, which could lead to worse treatment outcomes, has also been reported.4 Using directly observed therapy, short-course (DOTS), cure rates of 80% to 90% have been achieved for passively-diagnosed cases of smear-positive pulmonary tuberculosis.5 Analyses of national programmes in Malawi, Tanzania, Mozambique and in ten provinces of China have shown that this strategy is both effective and cost-effective.6–8 Based on the successes of these programmes, the World Health Organization (WHO) adopted DOTS as its strategy for global tuberculosis control. Since the WHO championed the DOTS strategy, uptake by national programmes has been slow; the WHO reports that only 11% of new smear-positive pulmonary tuberculosis cases are enrolled in DOTS programmes worldwide.9 Elsewhere, we have developed regional models of the epidemiology of tuberculosis to examine the effects of alternative global control strategies.10 Using these models, we have shown that under a range of plausible assumptions about DOTS uptake, there will be 199 to 241 million new cases and 67 to 87 million deaths from tuberculosis over the period 1998 to 2030. Based on the enormous remaining burden that is anticipated even with continued expansion of DOTS, we have evaluated two sets of extensions to the WHO strategy: 1) aggressive applications of existing technologies, such as active case-finding, mass preventive therapy, and preventive therapy targeted to HIV-positive individuals; and 2) development and application of new tools, such as improved diagnos-
tics, ultra-short-course treatment regimens and various forms of vaccines. In particular, our regional models suggest that active case-finding, as an extension to the WHO DOTS strategy, could reduce tuberculosis mortality by one-quarter to one-third over the next several decades. This finding accords well with the fact that tuberculosis programmes that have achieved extremely high treatment completion rates through DOTS—such as the World Bank project areas in China—may have case-detection rates of well below 50%.7 Clearly, raising case-detection in such areas will have a major impact. Despite the long-standing dogma that there is no role for active case-finding, this paper more carefully examines the potential benefits of a range of strategies based on active detection. We estimate, in each region, the maximum costs at which active casefinding strategies would still be cost-effective, and discuss the managerial implications of these extensions to global control strategies.
METHODS Figure 1 depicts the basic structure of the epidemiological model used in this study. Complete details of the model and its application to various regions are discussed elsewhere.10 For this paper, it is critical to note two particular features of the model: 1) it allows each clinical form of disease to have a separate case detection rate; and 2) the average delay from the onset of symptoms to diagnosis may be adjusted independently from the proportion of cases diagnosed. To capture the profound interaction between tuberculosis and HIV, the model actually consists of two submodels—one for the HIV sero-negative population, and one for the HIV sero-positive population. Based on patterns of tuberculosis epidemiology, the world has been divided into five regions, adapted from the World Bank regions used in the Global Burden of Disease Study (GBD)8: 1) EME—Established Market Economies; 2) FSE—Formerly Socialist Econ-
Correspondence to: Dr Christopher J L Murray, Harvard Center for Population and Development Studies, 9 Bow Street, Cambridge, MA 02138, USA. Tel: (1617) 496 3230. Fax: (1617) 496 3227.
S10
The International Journal of Tuberculosis and Lung Disease
omies of Europe; 3) Asia—composed of India, China and Other Asia and Islands; 4) SSA—Sub-Saharan Africa; and 5) LAC/MEC—a combination of Latin America and the Caribbean with the Middle Eastern Crescent. For each region, a set of time-dependent parameters has been chosen so that the resulting models match closely the levels and trends in incidence, prevalence, annual risk of infection (ARI) and mortality reported by the WHO11,12 and the GBD.13 For each region, we have selected only one model; clearly, other combinations of parameter values and shifts in these values over time might also provide adequate fits on known trends in tuberculosis. Uncertainly analysis for a similar model for the United States, however, indicates that evaluations of the impact of different interventions are largely unaffected by the choice of parameter values that fit past trends.14
porate not only the future uncertainty regarding DOTS uptake, but also uncertainty as to the trajectory of the HIV epidemic and the potential effects of high BCG coverage. Each of the alternative strategies described below has been evaluated incrementally on the baseline, best-case and worst-case DOTS scenarios. The results of this uncertainty analysis are available elsewhere.10 We have evaluated four extensions to the DOTS strategy based on active case-finding. For each scenario, we have estimated incidence and mortality in the HIV-negative and HIV-positive populations, from 1998 to 2050. 1. Active case-finding, symptomatic (ACF-Sx). The entire population would be screened for respiratory symptoms every 7 years (approximately 14% of the population every year). It is assumed that the capacity to implement this strategy would develop gradually over the next 10 years. The 3% to 4% reporting symptoms15 would be subject to sputum examination. Using this algorithm, we expect to detect twothirds of prevalent smear-positive cases.16 Ninety per cent of detected cases are assumed to start treatment, with the same cure rates as in the baseline DOTS assumptions. 2. Active case finding using mass miniature radiography (ACF-MMR). The entire population would be screened every 7 years using mass miniature radiography, followed by sputum examination for those with suspicious lesions. It is assumed that capacity to implement this strategy will be built over the next 10 years. This screening approach would detect 95% of the prevalent smear-positive cases,16 with 90% of detected cases starting treatment. Smear-negative X-ray suggestives meeting strict clinical criteria (2 to 3 weeks persistent cough and other symptoms consistent with tuberculosis) would also be treated empirically. We assume (conservatively) that one-third of true smearnegatives would be detected and 80% of these cases enrolled in treatment. 3. Single-cycle active case-finding (ACF-Sx-1 and ACF-MMR-1). For both the symptomatic and MMR screening algorithms, we have also examined the potential impact of just one complete cycle of screening. Because capacity will develop gradually, this initial cycle would require 12 years to achieve full coverage of the population.
Global control strategies The likely impact of pursuing only the WHO DOTS strategy was evaluated, with three different variants based on a range of assumptions about the extent to which DOTS is adopted, adequately funded and given sufficient managerial attention by governments. In all three scenarios, it is assumed that DOTS uptake will continue to increase over the next two decades and will level off by 2020. We have defined a baseline, best-case and worst-case DOTS scenario, which incor-
Dalys and maximum willingness-to-pay Deaths averted through active case-finding strategies have been used to estimate Disability-Adjusted Life Years (DALYs). The average age of death in HIVnegative and HIV-positive tuberculosis cases in each region were based on the results reported in the Global Burden of Disease Study.13 Following the methods described elsewhere,17 DALYs averted through the alternative strategies were estimated using the average age of death and the life expectancy at that age in
Figure 1 The basic epidemiological model of tuberculosis. The model includes nineteen states: uninfected (U); infected subject to fast or slow breakdown (IF and IS, respectively); superinfected subject to fast breakdown (SF); INH recipient subject to slow breakdown (HS); untreated cases (Ci,Uj ), where the index i takes on values 1, 2 and 3 (smear-positive pulmonary, smearnegative pulmonary and extra-pulmonary, respectively) and the index j takes on values 1 and 2 (fast and slow rates of diagnosis, respectively); treated cases (Ci,Tk ), where the index i takes on value 1, 2 and 3 (as above) and the index k takes on values 1 and 2 (good treatment and bad treatment, respectively); recovered cases subject to fast or slow relapse (RF and RS, respectively). For a detailed description of the equations and parameters which specify the model, see Reference 10.
Rethinking the role of active case-finding
each region. Total DALYs averted over the period 1998 to 2050 have been calculated in present value terms using a 3% discount rate. The World Bank has established the importance of comparing the cost-effectiveness of interventions across the health sector in formulating policy.8,18,19 Interventions that cost less than $100 per DALY averted are considered extremely cost-effective. In middle-income countries such as Mexico, society’s willingness-to-pay to avert a DALY will be much higher.20 Most societies should be willing to pay at least the annual per capita income to avert one DALY (i.e., to gain one year of healthy life); in many cases, societies will be willing to pay several multiples of income per capita for each DALY averted.21 In this analysis, we assume conservatively that the maximum willingness-to-pay per DALY is equal to the 1998 GNP per capita in each region. For a programme of active case-finding, therefore, maximum willingness-to-pay is a function of the number of DALYs averted through the intervention: WTPj 5 SjDj
where WTPj is the maximum willingness-to-pay for an active case-finding programme in region j, Sj is GNP per capita in region j, and Dj is the total number of DALYs averted through active case-finding in region j. The maximum willingness-to-pay per person screened in a programme may be calculated simply by dividing WTPj by the number of individuals screened.
RESULTS Table 1 summarizes the benefits of the four active case-finding strategies that have been evaluated in this study, compared with the baseline DOTS scenario. One 12-year cycle of active case-finding based on a symptomatic screen could reduce the number of new cases of tuberculosis between 1998 and 2050 by 17 million and the number of deaths by 7 million. Ninety-five per cent of the cases and deaths averted through this strategy would be in Asia and SSA. By incorporating MMR rather than simply asking about
S11
respiratory symptoms, an active case-finding strategy would prevent an additional 13 million cases and 6 million deaths over the same time period. The magnitude of these benefits is enormous compared to the benefits that may be gained by pursuing other health interventions. Continuous cycles of active case-finding in which one-seventh of the adult population in each region is screened each year would yield even greater benefits. Continuous symptomatic screening could reduce tuberculosis incidence by 64 million cases and prevent 27 million deaths in the next halfcentury. Using continuous MMR would prevent an additional 36 million cases and 17 million deaths. Figure 2 depicts the projected trends in mortality by region under the four active case-finding scenarios, compared to the baseline DOTS scenario. Society’s willingness-to-pay for these programmes (in 1998 US dollars) is shown in Table 2 by region and for the world. For these calculations, the number of DALYs averted and the costs of each strategy have been discounted to present value terms. Using per capita income as the cutoff for defining a cost-effective intervention, it is apparent that society should be willing to spend enormous resources on active case-finding programmes, because the potential benefits are so great. For a single cycle of active case-finding using a symptomatic screen, society should be willing to pay up to $71 billion; for a single cycle of MMR-based screening, this sum rises to $134 billion. Table 2 also provides estimates of the willingness-to-pay for programmes of repeated cycles of active case-finding. These larger figures are less informative, as society is unlikely to commit today to pay for continuous cycles of active case-finding over the next half-century. If society did commit to one cycle of active screening, however, willingness-to-pay for another incremental cycle would likely remain high if evaluated at the end of the first cycle. From the perspective of a decision-maker or a tuberculosis programme manager, the willingness-topay per person screened through active case-finding may be more informative (Table 3). These figures vary widely across regions. In EME and FSE, it is very unlikely that an active screening programme could be
Table 1 Tuberculosis cases and deaths averted by region through various strategies based on active case-finding, incremental on baseline DOTS, 1998–2050 Cases averted (‘000s) Region EME FSE LAC/MEC Asia SSA World
Deaths averted (‘000s)
ACFSym1
ACFMMR1
ACFSym
ACFMMR
ACFSym1
ACFMMR1
ACFSym
ACFMMR
22 78 803 8854 6929 16685
42 139 1465 15908 12567 30121
56 144 2163 32471 29105 63939
104 244 3612 51814 44366 100140
5 19 286 3713 3300 7323
11 37 548 6880 6132 13609
14 35 764 13168 12861 26843
28 65 1367 21779 20186 43425
EME = Established Market Economies; FSE = Formerly Socialist Economies of Europe; LAC = Latin America and the Caribbean; MEC = Middle Eastern Crescent; Asia = China, India, and Other Asia and Islands; SSA = Sub-Saharan Africa; ACFSym1 = single-cycle active case-finding, symptomatic; ACFMMR1 = single-cycle active case-finding, using MMR; ACFSym = continuous active case-finding, symptomatic; ACFMMR = continuous active case-finding, using MMR.
S12
The International Journal of Tuberculosis and Lung Disease
Figure 2 Projections of tuberculosis deaths by region, 1998–2030. The lines on each graph indicate projections under alternative scenarios, including the baseline DOTS strategy and four alternative extensions of DOTS: 1) a single cycle of active case-finding using a symptomatic screen (ACFSyml); 2) a single cycle of active case-finding using mass miniature radiography (ACFMMR1); 3) continuous active case-finding, symptomatic (ACFSym); and 4) continuous active case-finding using MMR (ACF MMR). Note that the graphs do not all use the same scale. (EME = Established Market Economies; FSE = Formerly Socialist Economies of Europe; LAC = Latin America and the Caribbean; MEC = Middle Eastern Crescent; SSA = Sub-Saharan Africa; Asia = China, India and Other Asia and Islands.)
implemented at costs lower than one to four dollars per person screened. In other words, a threshold criterion of GNP per capita implies that active case-finding would not be cost-effective in industrialized nations. Even if the threshold were raised to several multiples of GNP per capita, an active screening programme would not be affordable in EME and FSE. In Table 2 Maximum social willingness-to-pay for DALYs averted through alternative tuberculosis control strategies based on active case-finding
contrast, in LAC/MEC and Asia, society should be willing to pay more than $30 per person screened through an MMR-based strategy. Because of the extremely high prevalence of tuberculosis in SSA, the willingness-to-pay for active case-finding is an extraordinary $56 per person for a symptomatic screening programme and $104 per person for MMR.
Table 3 Maximum social willingness-to-pay per person screened for DALYs averted through alternative tuberculosis control strategies based on active case-finding
Maximum social willingness-to-pay (1998 US$, millions) Region EME FSE LAC/MEC Asia SSA World
ACFSym1
ACFMMR1
ACFSym
ACFMMR
922 553 11127 34014 24620 71236
1963 1081 21641 63768 45809 134262
1799 835 22392 87930 69707 182662
3662 1573 40873 148951 111880 306938
EME=Established Market Economies; FSE = Formerly Socialist Economies of Europe; LAC = Latin America and the Caribbean; MEC = Middle Eastern Crescent; Asia = China, India, and Other Asia and Islands; SSA = SubSaharan Africa; ACFSym1 = single-cycle active case-finding, symptomatic; ACFMMR1 = single-cycle active case-finding, using MMR; ACFSym = continuous active case-finding, symptomatic; ACFMMR = continuous active casefinding, using MMR.
Maximum social willingness-to-pay (1998 US$) Region EME FSE LAC/MEC Asia SSA World
ACFSym1
ACFMMR1
ACFSym
ACFMMR
1.58 2.26 15.43 16.41 55.87 17.54
3.37 4.41 30.01 30.76 103.95 33.05
0.78 0.90 6.36 9.87 28.06 10.06
1.59 1.69 11.62 16.72 45.04 16.91
EME = Established Market Economies; FSE = Formerly socialist Economies of Europe; LAC = Latin America and the Caribbean; MEC = Middle Eastern Crescent; Asia = China, India, and Other Asia and Islands; SSA = SubSaharan Africa; ACFSym1 = single-cycle active case-finding, symptomatic; ACFMMR1 = single-cycle active case-finding, using MMR; ACFSym = continuous active case-finding, symptomatic; ACFMMR = continuous active case-finding, using MMR.
Rethinking the role of active case-finding
DISCUSSION There has been considerable controversy over the appropriate role of active case-finding. The WHO, for many years, has recommended passive case-finding as the primary strategy for detecting tuberculosis.22 The widely-held view that active case-finding should not be pursued in high- or low-income countries is based on one interpretation of a number of empirical studies. From a cost-effectiveness perspective, this interpretation may be too pessimistic. A brief review of these studies is critical to interpreting the results of the present analysis. Studies of active case-finding in the 1960s, in Canada, Czechoslovakia and the Netherlands, and in the 1970s in Japan,23,24 found that only 12% to 24% of new smear-positive cases were detected by repeated active case-finding. Because in these countries nearly all cases of tuberculosis would eventually be diagnosed, the primary benefit expected from active casefinding was to diagnose cases earlier, perhaps before progressing to smear-positive. In these studies, however, the proportion of cases diagnosed that were smear-positive did not appear to increase as a result of active screening. The dominant interpretation of these findings is well summarized by Toman:25 It has been proved that the early detection of all cases with smear-positive pulmonary tuberculosis—the most dangerous sources of infection—by means of periodic mass radiography is impracticable, even when it is repeated at short intervals. The great majority of sputum-smear-positive cases develops in a shorter time than the shortest feasible interval between two mass radiography survey rounds. Moreover, 90% of patients with rapidly progressive pulmonary tuberculosis have objective symptoms, such as cough, fever, loss of weight, sputum, and haemoptysis. These symptoms develop rather soon after the onset of the disease, prompting the patient to seek medical advice. That is why most smear-positive tuberculosis cases are detected outside (usually earlier than) the periodic case-finding campaigns by the regular health services that the patient can consult whenever he feels ill.
In a series of studies in Kenya, various extensions to passive case-finding have been investigated.15,26–31 Many experts have interpreted these studies as evidence that active case-finding should not be pursued in high prevalence, low-income countries for two reasons: 1) for every smear-positive case detected, nearly 100 individuals with respiratory symptoms had to have sputum examinations; and 2) 80% of smearpositive individuals had sought care through the health services. What this interpretation does not acknowledge, however, is the large number of new smear-positive cases detected in the Kenya studies. In Machakos and Kirinyaga Districts, these studies showed that screening the population by asking household heads if anyone in the household had respiratory symptoms yielded smear-positive cases at a
S13
rate of 47 per 100 000 population in Machakos and 41 per 100 000 in Kirinyaga. The vast majority of these smear-positive cases had not been diagnosed with tuberculosis, although over 80% had reported contact with the health services. Thus far, operations research studies aimed at improving the diagnostic capabilities of the peripheral health care system have not been successful.31 The final argument that has been used against active case-finding is that because of the relatively low yield of sputum examinations, the costs and volume of work are impractical for low-income countries. Our analysis, however, clearly indicates that the potential benefits of active case-finding may be enormous. The greatest benefits would be in places with high prevalence, low case-detection and moderate to high treatment completion, such as in China. The estimates of $30 as the maximum willingnessto-pay per person screened using MMR in LAC/MEC and Asia, and over $100 in SSA, suggest that active case-finding may be highly cost-effective in some settings. It is very likely that the actual cost per person screened would be considerably less than $30 in Asia and certainly far less than $100 in SSA. In LAC/MEC, it is possible that active screening using MMR could be implemented in some countries for less than $30, but in a number of higher-income countries in that region, active screening may not be cost-effective, given the higher average costs of labor. Certainly, it would not be possible to implement active case-finding in EME and FSE for one to four dollars, which is the maximum WTP in these regions; the conclusion that active case-finding would not be cost-effective in EME and FSE is consistent with the study results discussed above. The cost-effectiveness thresholds presented in this paper are based on the plausible claim that society, in order to save a year of healthy life, should be willing to spend what the average individual consumes in a year. Given this assumption, the expected benefits from active case-finding programs should warrant incredibly large investments, particularly in SSA and Asia. Compared with existing national health expenditures, these WTP figures are impressive. For the developing regions combined, we estimate that the health benefits of a single cycle of MMR should merit an investment of up to $131 billion over 12 years, while total health expenditures in these regions have been calculated at around $172 billion per year.32 While it is unlikely—particularly in Asia and SSA—that the actual costs of these screening programs would even approach the threshold WTP, it is clear that such an ambitious effort to reduce mortality and disability would need to be funded from incremental resources. These incremental resources would have to come from increases in national health expenditures, reallocation of existing national health resources from less cost-effective efforts, or from Overseas Develop-
S14
The International Journal of Tuberculosis and Lung Disease
ment Assistance. This analysis demonstrates that the benefits of active case-finding could be so large that society should be willing to invest considerable resources to achieve these gains. This analysis of potential DALYs averted through active case-finding and willingness-to-pay for these benefits is based on a series of conservative assumptions that may actually underestimate the willingnessto-pay per person screened. Income per capita is expected to increase in all regions,33 so that WTP for averting a DALY should increase substantially over the next decades. The underestimate of WTP is greater for the continuous cycles of active case-finding than for the single cycle, as the benefits of the former extend farther into the future. Many experienced experts on tuberculosis control believe that pursuing active case-finding is likely to have few, if any, benefits and may, in fact, be counterproductive. In low-prevalence populations, our models and other studies concur that the number of extra smear-positive cases detected would be relatively small. In high-prevalence, low-income populations, on the other hand, critics of active case-finding argue that it is too expensive and would distract scarce managerial attention away from implementing DOTS. In an environment where the budget for tuberculosis control activities is fixed, we affirm strongly that efforts should focus first on improving treatment completion rates through the implementation of DOTS. Our analysis, however, shows that in highprevalence populations such as SSA and Asia, the implementation of active case-finding as an extension to DOTS may yield such enormous benefits that society should be willing to increase the funding for tuberculosis control significantly. Further work will be required to estimate the actual costs of implementing active case-finding programmes in SSA, Asia and perhaps LAC/MEC. Given that these programmes are likely to be highly cost-effective, both bilateral and multilateral donors and governments in these regions should be willing to consider large investments in these extensions to the DOTS strategy. References 1 World Health Organization. Global Tuberculosis Programme. WHO report on the tuberculosis epidemic 1997. WHO/TB/ 97.224. Geneva; World Health Organization, 1997. 2 Murray C J L, Lopez A D. Evidence-based health policy — lessons from the Global Burden of Disease Study. Science 1996; 274(5288): 740–743. 3 Murray C J L, Lopez A D. Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet 1997; 349: 1498–1504. 4 The WHO/IUATLD Global Project on Anti-Tuberculosis Drug Resistance Surveillance. Anti-tuberculosis drug resistance in the world. WHO/TB/97.229. Geneva; WHO Global Tuberculosis Programme, 1997. 5 Broekmans J F. Control strategies and programme management. In: Porter J D H, McAdam K P W J, eds. Tuberculosis: back to the future. Chichester; John Wiley and Sons, 1994: 171–192.
6 Murray C J L, DeJonghe E, Chum H J, Nyangulu D S, Salomao A, Styblo K. Cost effectiveness of chemotherapy for pulmonary tuberculosis in three sub-Saharan African countries. Lancet 1991; 338: 1305–1308. 7 China Tuberculosis Control Collaboration. Results of directly observed short-course chemotherapy in 112 842 Chinese patients with smear-positive tuberculosis. Lancet 1996; 347: 358–362. 8 World Bank. World development report 1993: investing in health. New York: Oxford University Press (World Bank); 1993. 9 Kochi A, Nunn P, Dye C, Tayler E. Global burden of disease [letter]. Lancet 1997; 350: 142. 10 Murray C J L, Salomon J A. Using mathematical models to evaluate global tuberculosis control strategies. Cambridge; Harvard Center for Population and Development Studies, 1998. 11 Raviglione M C, Snider D E, Kochi A. Global epidemiology of tuberculosis: morbidity and mortality of a worldwide epidemic. JAMA 1995; 273: 220–226. 12 World Health Organization. Global Tuberculosis Programme. Tuberculosis—a global emergency: case notification update, February 1996. WHO/TB/96.197. Geneva; World Health Organization, 1996. 13 Murray C J L, Lopez A D. Global health statistics: a compendium of incidence, prevalence and mortality estimates for over 200 conditions. Cambridge, MA; Harvard University Press, 1996. 14 Murray C, Sawert H. Tuberculosis in the United States: strategies for developing and validating models (Health Transition Working Paper Series). Cambridge; Harvard Center for Population and Development Studies, 1994. 15 Aluoch J A, Swai O B, Edwards E A, et al. Study of case-finding for pulmonary tuberculosis in outpatients complaining of a chronic cough at a district hospital in Kenya. American Review of Respiratory Diseases 1984; 129: 915–920. 16 Gothi G D, Narayan R, Nair S S, Chakraborty A K, Srikantaramu N. Estimation of prevalence of bacillary tuberculosis on the basis of chest X-ray and/or symptomatic screening. Indian J Med Res 1976; 64(8): 1150–1159. 17 Murray C J L. Rethinking DALYs. In: Murray C J L, Lopez A D, eds. The global burden of disease. Cambridge, MA; Harvard University Press, 1996: 1–89. 18 Jamison D T, Mosley W H, Measham A R, Bobadilla J L, eds. Disease control priorities in developing countries. New York; Oxford University Press (World Bank), 1993. 19 Bobadilla J–L, Cowley P, Musgrove P, Saxenian H. Design, content and financing of an essential national package of health services. Bull World Health Organ 1994; 72: 653–662. 20 Frenk J, Lozano R, Gonzalez-Block M A, et al. Economia y salud: propuestas para el avance del sistema de salud en Mexico. Informe final. Mexico, DF; Fundacion Mexicana para la Salud, 1994. 21 Murray C J L. Towards an analytical approach to health sector reform. In: Berman P, ed. Health sector reform in developing countries: making health development sustainable. Cambridge, MA; Harvard University Press, 1995. 22 World Health Organization. Ninth report of the WHO Expert Committee on Tuberculosis. WHO Technical Report Series, No. 552. Geneva; World Health Organization, 1974. 23 Krivinka R, Drapela J, Kubik A, Dankova D. Epidemiological and clinical study of tuberculosis in the district of Kolin, Czechoslovakia. Bull World Health Organ 1974; 51: 59–69. 24 Meijer J, Barnett G D, Kubik A, Styblo K. Identification des sources d’infection. Bull Int Union Tuberc 1971; 45: 5–54. 25 Toman K. Tuberculosis case-finding and chemotherapy: questions and answers Geneva; World Health Organization, 1979. 26 Nsanzumuhire H, Lukwago E W, Edwards E A, Stott H, Fox W, Sutherland I. A study of the use of community leaders in case-finding for pulmonary tuberculosis in the Machakos district of Kenya. Tubercle 1977; 58: 117–128. 27 Aluoch J A, Karuga W K, Nsanzumuhire H, Edwards E A, Stott H, Fox W. A second study of the use of community lead-
Rethinking the role of active case-finding
ers in case-finding for pulmonary tuberculosis in the Machakos district of Kenya. Tubercle 1977; 59: 223–243. 28 Nsanzumuhire H, Aluoch J A, Karuga W K, et al. A third study of case-finding methods for pulmonary tuberculosis in Kenya, including the use of community leaders. Tubercle 1981; 62: 79–94. 29 Aluoch J A, Edwards E A, Stott H, Fox W, Sutherland I. A fourth study of case-finding methods for pulmonary tuberculosis in Kenya. Trans Roy Soc Trop Med Hyg 1982; 76: 679–691. 30 Aluoch J A, Swai O B, Edwards E A, et al. Studies of case-finding for pulmonary tuberculosis in outpatients at 4 district hospitals in Kenya. Tubercle 1985; 66: 237–249.
S15
31 Aluoch J A, Oyoo D, Swai O B, Kwamanga D, Agwanda R, Edwards E A. A study of the use of maternity and child welfare clinics in case-finding for pulmonary tuberculosis in Kenya. Tubercle 1987; 68: 93–103. 32 Murray C J L, Govindaraj R, Musgrove P. National health expenditures: a global analysis. Bull World Health Organ 1994; 72: 623–37. 33 International Monetary Fund. World economic outlook. Washington; International Monetary Fund, 1995.
