WWW. MEDSCIMONIT .COM Clinical Research © Med Sci Monit, 2004; 10(2): CR62-67 PMID: 14737045 CR62 Factors affecting delays in diagnosis and treatment of pulmonary tuberculosis in a tertiary care hospital in Istanbul, Turkey Döndü Güneylioglubdefg, Adnan Yilmazabcdefg, Sevinc¸ Bilginbdf, Ümmühan Bayrambf, Esen Akkayaace Department of Pulmonology, SSK Süreyyapas¸a Center for Chest Diseases and Thoracic Surgery, Istanbul-Turkey Source of support: none. Summary Background: To investigate delays in the diagnosis and treatment of inpatients with smear-positive pul- monary tuberculosis and to identify factors affecting these delays. Materials/Methods: 204 hospitalized patients with smear-positive pulmonary tuberculosis were identified. The clinical files of the patients were analyzed and questionnaires were created. Results: Mean application interval was 31.4 days, mean referral interval was 22.1 days, mean diagnosis interval was 3.3 days, and mean initiation of treatment interval was 1.4 days. Patient delay was present in 34.8 percent of the patients. The application interval was shorter for patients hav- ing an index case for tuberculosis (p=0.039) and for those with good economic status (p<0.005). 167 patients (81.9%) had institutional delay. The referral interval was longer for female patients than for male patients (p=0.015). The most common causes of institutional delays were a low index of suspicion for tuberculosis, health care system delays, and underuti- lized chest X-ray examinations. One hundred and three patients (50.5%) had delays in diag- nosis and 51 patients (25.0%) had delays in treatment. The most frequent reason for diagnos- tic delay was health care system delays (35.9%). Conclusions: There were several delays in the diagnosis of tuberculosis patients. For an effective tuberculo- sis control, efforts should be made to reduce these delays. Physicians and the public should be educated about tuberculosis. Health care system and laboratory delays should be improved. key words: delays • smear-positive • pulmonary tuberculosis • diagnosis, treatment Full-text PDF: http://www.MedSciMonit.com/pub/vol_10/no_2/3261.pdf Word count: 2347 Tables: 5 Figures: 1 References: 26 Received: 2002.11.07 Accepted: 2003.07.03 Published: 2004.02.01 Author’s address: Dr. Adnan Yilmaz, Maltepe Zümrütevler Atatürk Cad. Abant Apt. No:30 81530 Istanbul, Turkey, e-mail: elifim@rt.net.tr Authors’ Contribution: A Study Design B Data Collection C Statistical Analysis D Data Interpretation E Manuscript Preparation F Literature Search G Funds Collection CR63 Med Sci Monit, 2004; 10(2): CR62-67 Güneylioglu D et al – Delays in pulmonary tuberculosis diagnosis and treatment CR BACKGROUND Prior to the twentieth century, tuberculosis was one of the major causes of death in both developed and devel- oping countries [1]. During the twentieth century it con- tinued to be a major public health problem worldwide. In 1993, the World Health Organization (WHO) declared a state of global emergency for tuberculosis due to the steady increase of the disease worldwide [2]. It is estimated that by the year 2005 12 million cases of tuber- culosis will be identified in the world annually, a 58% increase from the 7.5 million estimated for 1990 [3]. It was reported that 19% to 43% of the world’s population was infected with Mycobacterium tuberculosis [4]. One of the main objectives in any tuberculosis control program is to reduce tuberculosis transmission in the community through early detection of tuberculosis cases and prompt implementation of a full course of therapy [5]. This is especially important in the case of untreated smear-posi- tive patients, who are the main sources of infection in a community [5,6]. Delays in the diagnosis and start of effective treatment of tuberculosis patients result in a prolonged period of infectivity in the community and health care workers [7–9]. These delays have been noted in both high- and low-prevalence countries [6,7,10–16]. Turkey has an estimated incidence of tuberculosis of 30 per 100,000 of the population. The SSK Süreyyapa l ba Center, located in Istanbul, is a tertiary care hospital for chest diseases and tuberculosis. This center is one of Turkey’s main health facilities and provides treatment to tuberculosis patients who have been referred by dis- trict health care facilities throughout the country. In this study, we aimed to investigate delays in the diagno- sis and treatment of hospitalized patients with smear- positive pulmonary tuberculosis and to identify the fac- tors affecting these delays. MATERIAL AND METHODS The present study was conducted at the SSK Sürey- yapa l ba Center for Chest Disease and Thoracic Surgery, consisting of 1600 beds. We reviewed the clinic records of all patients hospitalized between June and August 2001 and identified 204 newly diagnosed inpatients with smear-positive pulmonary tuberculosis. Cases with previous histories of tuberculosis treatment were excluded. Study design The clinical files of the cases were analyzed and a ques- tionnaire was created to obtain data by interview. For each patient, the following information was gathered: (1) sex, (2) age, (3) education level, (4) area of residence, (5) index case, (6) economic status, (7) first symptom, (8) first visiting physician, (9) symptom duration, (10) date of first doctor visit, (11) date of admission to hospital, (12) date of diagnosis, and (13) date of treatment initia- tion. The presence of cough, fever, night sweats, hemoptysis, weight loss, anorexia, fatigue, and dyspnea were the criteria for the onset date of symptoms. Patients were classified as ‘urban’ if they lived in a city or metropolis and ‘rural’ if they lived in the towns and villages surrounding these. The patients were catego- rized into three groups with respect to their economic status. When monthly income was below $200 (US), the patient’s economic status was regarded as poor. Monthly income was between $200 and $550 for patients with moderate economic status. Patients with good economic status had monthly incomes over $550. The following time intervals and delays were deter- mined for each patient: The patient’s application interval was defined as the time interval between the onset of symptoms and the first doctor visit. Intervals that exceeded 30 days were considered indicative of a patient delay [12]. The referral interval was defined as the time from the first doctor visit to admission [17]. With regard to our health care system, intervals that exceeded two days were considered indicative of an institutional delay [14]. The diagnosis interval was regarded as the time from admission to a positive acid-fast smear. Intervals that exceeded one day were considered indicative of a delayed diagnosis [14,18]. The treatment interval was the time from diagnosis to initiation of treatment. Intervals that exceeded one day were considered indicative of a delayed treatment [14,18]. Clinic delay was defined as the time from admission to initiation of treatment. Regar- ding the diagnosis and treatment intervals, those that exceeded two days were considered indicative of a delay. Doctor delay was defined as the time from the first doctor visit to initiation of treatment. Intervals that exceeded four days were considered indicative of a delay [14,19]. Figure 1 presents the time intervals and delays. The reasons for the delays were also evaluated. Statistical analysis The chi-square test, ANOVA test or Student-t test were used to assess differences between groups. RESULTS During the study period, 204 new smear-positive adult patients were evaluated. There were 122 male and 82 female. The mean age of the patients was 38.6 years (range 14–90 years). One hundred and eighteen patients (57.9%) had primary school education, 56 (27.4%) had secondary school education, 7 (3.4%) were university graduates and 23 (11.3%) had no formal edu- cation. One hundred and fifty-one patients (74%) lived in urban areas and 53 patients (26%) in rural areas. While 139 patients (68.1%) had no history of an index case for tuberculosis, there was a positive history in 65 cases (31.9%). Table 1 presents the values of the intervals. The mean values of the time intervals (SD) were 31.4 (38.9) days for the application interval, 22.1 (29.7) days for the referral interval, 3.3 (5.2) days for the diagnosis interval, and 1.4 (1.8) days for the initiation of treatment inter- val. The application interval was shorter than 30 days in 133 patients (65.2%) and longer than 30 days in 71 patients (34.8%). According to these results, 34.8% of the patients had patient delays. CR64 Med Sci Monit, 2004; 10(2): CR62-67Clinical Research Table 2 gives a sub-analysis of application interval and patient delays with respect to several factors. Age, sex, area of residence, and education level had no effect on the application interval. Significantly shorter mean appli- cation intervals were noted in patients having an index case for tuberculosis (p=0.04) and patients who had good economic status (p=0.03). The rate of cases having patient delay was lower among patients with hemoptysis than patients with other first symptoms (p=0.04). Factors associated with the referral interval are summa- rized in Table 3. The referral interval was shorter than three days in 37 patients (18.1%). This interval was between 3 and 10 days in 64 patients (31.4%) and was longer than 10 days in 103 patients (50.5%). One hun- dred and sixty-seven (81.9%) patients had institutional delay. Sub-analysis identified 200 reasons for institu- tional delay in these patients. While age, education level, residence area, and economic status had no effect on the referral interval, this interval was significantly shorter in male patients (p=0.015). Similarly, the rate of cases having institutional delay was significantly lower among male patients (p=0.003). Patients referred by a chest specialist had a significantly shorter referral inter- val than those referred by the other physicians (p=0.043), with a subsequent lower rate of institutional delay (p=0.038). Of the 167 patients with an institution- al delay, tuberculosis was not suspected in 82 patients (41%) at the time of their first visit. Fifty-one patients (25.5%) had delays in chest-X ray examinations. There were delays in the health care system in 47 (23.5%) patients. Economic status resulted in institutional delays in twelve patients (6%). The reason of institutional delay was not identified in eight patients (4%). Table 4 shows the distribution of the diagnosis intervals and initiation of treatment intervals with respect to days. One hundred and three patients (50.5%) had delays in diagnosis and 51 patients (25%) had delays in initiation of treatment. The mean (95%, CI) diagnosis interval was 3.7 days (2.5 to 4.0 days) in male patients and 2.4 days (1.9 to 2.9 days) in female patients (p=0.015). Age and education level had no effect on the diagnosis interval. Our health care system resulted in delayed diagnosis in 37 patients (35.9%). Reasons for delayed diagnosis included underutilized or delayed sputum examinations for acid-fast smears in 30 patients (29.1%). A low index of suspicion for tuberculosis by a chest physician was a reason of delayed diagnosis in 22 patients (21.4%). While laboratory delays were identi- fied in 9 patients (8.7%), no reason was determined for delayed diagnosis in 5 patients (4.9%). The mean initia- tion of treatment interval (95%, CI) was 1.5 days (1.2 to 1.8 days) in male patients and 1.3 days (0.9 to 1.7 days) in female patients (p>0.05). Age and education level had no effect on this interval. The distribution of time from the first doctor visit to ini- tiation of treatment with respect to days is shown in Table 5. The mean (95%, CI) interval was 26.7 days (22.5 to 30.8 days) and median interval was 15 days. There were doctor delays in 178 of 204 patients (87.2%). The mean interval (95%, CI) from onset of symptoms to initiation of treatment was 64.1 days (55.2 to 73.1 days). This interval suggested total delay. DISCUSSION We analyzed delays in the diagnosis and treatment of inpatients with smear-positive pulmonary tuberculosis and factors affecting these delays. The present study indicated that there were several delays between the onset of symptoms and initiation of treatment in our patients. These delays included patient delay, institu- tional delay, diagnostic delay, and delay in the treat- ment. Both patient and institutional delays result in increased infection risk for the population. Diagnostic and treatment delays result in increased infection risk for medical personnel [14]. An untreated smear-positive patient may infect on average more than 10 contacts onset of first visit to admission to diagnosis initiation of symptoms physician hospital treatment patient’s delay institutional diagnostic delay delayed treatment delay application interval referral interval diagnosis interval treatment interval doctor’s delay Clinic delay Figure 1. Components of the time from onset of symptoms to initiation of treatment and the delays. Interval Mean SD Median 95% CI Application Referral Diagnosis Treatment 31.4 22.1 3.3 1.4 38.9 29.7 5.2 1.8 17.5 11.0 1.5 1.0 26.1–36.7 18.1–26.2 2.5–4.0 1.2–1.7 Table 1. The values associated with several intervals (days). CR65 Med Sci Monit, 2004; 10(2): CR62-67 Güneylioglu D et al – Delays in pulmonary tuberculosis diagnosis and treatment annually [20]. A previous report indicated that at least 24% of the employees exposed to infectious tuberculosis in their office for 4 months were infected [13]. Delays for patients with pulmonary tuberculosis are more com- mon in developing countries [10,12,14]. A has previous report suggested that delays associated with pulmonary tuberculosis were common in Turkey [14]. Delays have also been reported in developed countries [7,11,13]. In our study, the median application interval was found to be 17.5 days. Seventy-one patients (34.8%) had patient delay. While many reports present a shorter median application interval than ours [21,22], longer median application intervals were noted in two previous studies [10,12]. This interval was between 0.3 weeks and 120 days in these reports [12,22]. A previous report indicated that 28.4 percent of the patients had patient delay and that the median application interval was 17.5 days [14]. The rate of patient delay was slightly higher in the present study than in our previous report. Similar median application intervals were determined in these series. This study showed that the rates of doctor delay were higher than those of patient delay. We noted that one hundred and seventy-eight patients (87.2%) had doctor delay and that the median doctor delay was 15 days. A previous report indicated that the rate of doctor delay was 88.2% and that the median doctor delay was 9 days [14]. Liam et al. reported that median doctor delay was 7 weeks [21]. Doctor delay included institutional delay, diagnostic delay and delayed treatment [14]. The rates of delay were 81.9% for institutional delay, 50.5% for diagnostic delay, and 25% for delayed treatment in the present study. These results indicate that the rate of institutional delay was more significant than those of delays in diagnosis and treatment. In a previous study, the rates of institutional, diagnostic and treatment delays were presented as 61.1%, 69.4% and 25.4%, respectively [14]. Pirkis et al. reported lower rates of delayed treatment than our study. [11] The median interval for diagnosis was 6.5 days in a previous report [7]. Taylor et al. [17] noted that the median referral interval was 18 days. When our results were compared with these reports, it was suggested that our patients had shorter median referral and diagnosis intervals, but longer treatment intervals. It is known that the length of the interval may be associ- ated with several factors [12,16,21,23–25]. Long et al. reported that the application interval was longer among women than among men [22]. A previous report noted longer patient delay in patients aged 45 years and over and in rural patients [12]. However, Liam et al. showed that sex [21], age, education level, and initial symptom had no significant association with patient delay. We ascertained lower application intervals for patients hav- ing an index case for tuberculosis and for those with good economic status. The rate of cases having patient delay was lower among patients with hemoptysis than patients with cough and other symptoms. We also found that sex, age, residence area, and education level had no significant effect on the application interval. The referral interval was shorter in males than in females. This inter- CR Application interval (days) Patient delay n (%) Mean (SD) Median 95% CI Yes No p Sex Male Female Age <45 ≥45 Residence Urban Rural Index case Yes No Education No education Primary Secondary University Economic status Poor Moderate Good First symptom Cough Hemoptysis Other 35.2 (42.9)* 25.7 (31.4)* 29.5 (38.3)* 34.5 (39.8)* 33.2 (43.2)* 25.8 (22.2)* 25.0 (37.5)** 33.6 (39.1)** 34.1 (44.8)*** 33.8 (39.5)*** 27.4 (31.3)*** 19.3 (18.1)*** 66.2 (52.5) 21.9 (24.9) 9.8 (9.1) 32.1 (38.0) 26.6 (43.1) 33.1 (36.6) 45 (36.9)** 26 (31.7)** 41 (32.5)** 30 (38.5)** 51 (33.8)** 20 (37.7)** 16 (24.6) 55 (39.5) 7 (30.4)* 47 (39.8)* 15 (26.8)* 2 (28.6)* 44 (83.0) #,## 25 (21.7) #, * 2 (5.6) ##, * 49 (37.1) ###,& 8 (20.0) ###,&& 14 (43.8) &,&& 77 (63.1) 56 (68.3) 85 (67.5) 48 (61.5) 100 (66.2) 33 (62.3) 49 (75.4) 84 (60.5) 16 (69.6) 71 (60.2) 41 (73.2) 5 (71.4) 9 (17.0) 90 (68.3) 34 (94.4) 83 (62.9) 32 (80.0) 18 (56.2) 18.0 15.0 17.0 19.0 16.0 20.0 13.0 20.0 15.0 16.0 16.0 15.0 52.0 15.0 8.0 20.0 10.0 20.0 27.4–42.6 18.7–32.5 22.8–36.2 25.5–43.4 26.4–40.2 19.6–31.9 15.7–34.3 27.1–40.1 14.7–53.4 8.4–48.9 19.0–35.8 2.5–36.1 51.7–80.7 17.3–26.5 6.7–12.9 25.6–38.6 12.8–40.4 19.8–46.2 * , **p>0.05 * , **p>0.05 * , **p>0.05 **p=0.039 * , *** p>0.05 # p<0.005 ## p<0.005 *p=0.026 ### p=0.055 & p>0.05 && p=0.04 Table 2. A sub-analysis of patient delay. CR66 Med Sci Monit, 2004; 10(2): CR62-67Clinical Research val was also shorter in patients who visited a chest spe- cialist. Age, education level, residence area, and econom- ic status had no significant effect on the referral interval. While Nakagawa et al. found that the referral interval was longer in female than in male patients [24], Taylor et al. determined that the median referral interval was similar between males and females [17]. It is known that health care providers and physicians first visited by patients affect the rates of doctor delay [12,21,25]. In our opinion, the degree of suspicion of tuberculosis differs among physicians, and this difference may be an impor- tant reason for difference among doctor delays. There are several differences among the studies with respect to delays. We conclude that the patient’s characteristics may result in these differences among studies. Many reasons for delays have been identified in previ- ous reports. Underutilized chest X-ray examinations and sputum smear examinations, a low degree of suspi- cion of tuberculosis, health care system and laboratory delays, as well as patient characteristics, such as age, sex, education level and socioeconomic status, are the most common reasons for doctor delays [7,12,14,16,18,26]. We determined similar reasons for delays. A low degree of suspicion of tuberculosis was one of the most com- mon causes for delay among our patients. Chest X-ray and sputum smear examinations were underutilized by physicians in our patients with pulmonary tuberculosis. The present data suggest that the health care system and laboratory delays result in high rates of institutional and diagnostic delays. Poor economic status was an important reason of patient delay in our series. CONCLUSIONS In conclusion, the present study suggests that our patients with smear-positive pulmonary tuberculosis had several delays from the onset of symptoms to initiation of treatment. Because untreated smear-positive patients are the main sources of infection, delays in the diagnosis and treatment of these patients increase the risk of disease transmission in the community. These delays are also associated with a prolonged period of infectivity for medical personnel. According to our data, doctor delay was more significant than patient delay, and institutional delay was the most important component of doctor delay. The low index of tuberculosis, underutilized chest Days n%n% Diagnosis interval Diagnosis interval 0–1 2–10 >10 101 93 10 153 50 1 49.5 45.6 4.9 75.0 24.5 0.5 Table 4. The distribution of diagnosis and treatment intervals with respect to days. Days n % 0–4 5–10 11–20 >20 26 53 36 89 12.8 26.0 17.6 43.6 Table 5. The distribution of the time interval between the time of the first doctor visit and initiation of treatment with respect to days. Referral interval (days) Institutional delay n (%) Mean (SD) Median 95% CI Yes No p Sex Male Female Age <45 ≥45 Residence Urban Rural Education No education Primary Secondary University Economic status Poor Moderate Good First visit Chest specialist Practitioner Other 19.1 (28.7)* 26.6 (30.8)* 20.5 (26.9)* 24.8 (33.1)* 21.2 (27.8)* 22.6 (30.5)* 29.6 (37.2)** 21.3 (30.5)** 19.8 (21.9** 14.1 (21.3)** 25.1 (34.6)* 21.5 (27.8)* 20.3 (28.3)* 11.4 (20.6)** 23.1 (26.8)** 25.3 (33.0)** 91 (74.6)*** 76 (92.7)*** 104 (82.5)*** 63 (80.8)*** 123 (81.5)*** 44 (83.0)*** 21 (91.3)* 95 (80.5)* 45 (80.4)* 6 (85.7)* 41 (77.4)*** 95 (82.6)*** 31 (86.1)*** 26 (68.4)* 47 (83.9)* 94 (85.5)* 31 (25.4) 6 (7.3) 22 (17.5) 15 (19.2) 28 (18.5) 9 (17.0) 2 (8.7) 23 (19.5) 11 (19.6) 1 (14.3) 12 (22.6) 20 (17.4) 5 (13.9) 12 (31.6) 9 (16.1) 1 6 (14.5) 7.5 14.5 11.0 11.0 11.0 11.0 19.0 8.0 11.0 4.0 15.0 10.0 11.0 6.0 14.0 14.0 14.0–24.2 19.8–33.4 15.8–25.1 17.4–32.3 13.4–29.0 17.7–27.1 13.5–45.7 15.5–27.1 13.9–25.6 5.5–33.8 15.5–34.6 16.3–26.6 10.7–29.9 4.7–18.2 15.9–30.3 19.1–31.6 *p=0.015 ***p=0.003 * , ***p>0.05 * , ***p>0.05 * , **p>0.05 * , ***p>0.05 **p=0.043 *p=0.48 Table 3. Factors associated with the referral interval. CR67 Med Sci Monit, 2004; 10(2): CR62-67 Güneylioglu D et al – Delays in pulmonary tuberculosis diagnosis and treatment X-ray and sputum smear examinations among physi- cians, and health care system and laboratory delays were the most common causes of doctor delay. Several efforts should be made to reduce these delays for the sake of tuberculosis control. Physicians and the public should be educated about tuberculosis. Health care system and lab- oratory delays should be minimized. These efforts can reduce delays in pulmonary tuberculosis treatment. We conclude that delays are important reasons for the increased period of infectivity, and that decreasing these delays will help tuberculosis control. REFERENCES: 1. Bloom BR: Tuberculosis: back to a frightening future. Nature, 1992; 358: 538-39 2. World Health Organization: TB a global emergency. WHO/TB/94.177. Geneva: World Health Organization, 1994 3. Centers for Disease Control and Prevention. Estimates for future global tuberculosis morbidity and mortality. MMWR Morb Mortal Wkly Rep, 1993; 42: 961-64 4. Sudre P, Dam GT, Kochi A: Tuberculosis: a global overview of the situation today. Bull World Health Org, 1992; 70: 149-59 5. Enarson DA, Rieder HL, Arnadottir T, Trebucq A: Tuberculosis guide for low income countries. 4 th ed., Paris, France: International Union Against Tuberculosis and Lung Disease, 1996 6. Salaniponi FML, Harries AD, Banda HT et al: Case seeking behav- iour and diagnostic processes in patients with smear-positive pul- monary tuberculosis in Malawi. Int J Tuberc Lung Dis, 2000; 4: 327-32 7. Mathur P, Sacks L, Auten G et al: Delayed diagnosis of pulmonary tuberculosis in city hospitals. Arch Intern Med 1994; 154: 306-10 8. Iseman MD: A leap of faith: what can we do to curtail intra institu- tional transmission of tuberculosis? Ann of Int Med, 1992; 117: 251-53 9. Menzies D, Fanning A, Yuan L, Fitzgerald M: Tuberculosis among health care workers. N Engl J Med, 1995; 333: 92-98 10. Lawn SD, Afful B, Acheampong JW: Pulmonary tuberculosis: diag- nostic delay in Ghanaian adults. Int J Tuberc Lung Dis, 1998; 2: 635-40 11. Pirkis JE, Speed BR, Yung AP et al: Time to initiation of anti-tuber- culosis treatment. Tuberc Lung Dis, 1996; 77: 401-06 12. Wandwalo ER, Morkve O: Delay in tuberculosis case-finding and treatment in Mwanza, Tanzania. Int J Tuberc Lung Dis 2000; 4: 133-38 13. MacIntyre CR, Plant AJ, Hulls J et al: High rate of transmission of tuberculosis in an office: impact of delayed tuberculosis. Clin Infect Dis 1995; 21: 1170-74 14. Yilmaz A, Boga S, Sulu E et al: Delays in diagnosis and treatment of hospitalized patients with smear-positive pulmonary tuberculo- sis. Respir Med, 2001; 95: 802-05 15. Counsell SR, Tan JS, Dittus RS: Unsuspected pulmonary tubercu- losis in a community teaching hospital. Arch Intern Med, 1989; 149: 1274-78 16. Sherman LF, Fujiwara PI, Cook SV et al: Patient and health care system delays in the diagnosis and treatment of tuberculosis. Int J Tuberc Lung Dis, 1999; 3: 1088-95 17. Taylor WR, Zumla A: Referral times of Vietnamese refugees with tuberculosis in camps in Hong Kong. Int J Tuberc Lung Dis, 2000; 4: 350-55 18. Rao VK, Iademarco EP, Fraser VJ, Kollef MH: Delays in the suspi- cion and treatment of tuberculosis among hospitalized patients. Ann Intern Med, 1999; 130: 404-11 19. Calder L, Gao W, Simmons G: Tuberculosis: reasons for diagnostic delay in Auckland. N Z Med J, 2000; 113: 483-85 20. Styblo K: Epidemiology of tuberculosis; selected papers. The Royal Netherlands Tuberculosis Association, 1991; 24: 53-54 21. Liam CK, Tang BG: Delay in the diagnosis and treatment of pul- monary tuberculosis in patients attending a university teaching hos- pital. Int J Tuberc Lung Dis, 1997; 1: 326-32 22. Lienhardt C, Rowley J, Manneh K et al: Factors affecting time delay to treatment in a tuberculosis control programme in a sub- Saharan African country: the experience of the Gambia. Int J Tuberc Lung Dis, 2001; 5: 233-39 23. Long NH, Johansson E, Lonnroth K et al: Longer delays in tuber- culosis diagnosis among women in Vietnam. Int J Tuberc Lung Dis, 1999; 3: 388-93 24. Yamasaki-Nakagawa M, Ozasa K, Yamada N et al: Gender differe- nce in delays to diagnosis and health care seeking behaviour in a rural area of Nepal. Int J Tuberc Lung Dis, 2001; 5: 24-31 25. Enkhbat S, Toyota M, Yasuda N, Ohara H: Differing influence on delays in the case-finding process for tuberculosis between general physicians and specialists in Mongolia. J Epidemiol, 1997; 7: 93-98 26. Steen TW, Mazond GN: Pulmonary tuberculosis in Kweneng District, Bostwana: delays in diagnosis in 212 smear positive patients. Int J Tuberc Lung Dis, 1998; 2: 627-34 CR . Yamasaki-Nakagawa M, Ozasa K, Yamada N et al: Gender differe- nce in delays to diagnosis and health care seeking behaviour in a rural area of Nepal. Int J Tuberc. tuberculosis diagnosis and treatment X-ray and sputum smear examinations among physi- cians, and health care system and laboratory delays were the most common causes