Passive case finding: Diagnostic approaches, cost and delay in TB management in Dar es Salaam and Pwani regions, Tanzania.

Abstract

Background and objectives: TB case detection in Tanzania is mainly through passive case finding, however, it has remained below the WHO and national target of 70%. Several studies have investigated different strategies to increase TB case finding and how to reduce TB diagnostic and treatment delay. A study conducted in Brazil showed that the probability of having TB did not depend on cough duration. Moreover, interventions aimed at integrating passive TB case finding in other clinics like antenatal clinics have proven to be acceptable and have also been recommended in Malawi and South Africa. Active case finding for TB revealed a significant number of undiagnosed TB cases among women attending PMTCT clinics in South Africa. Still, delay in diagnosis and treatment is a challenge, and case detection remains low, especially among women. Therefore, this thesis is based on studies that were conducted to determine the following: i) the proportion of smear positive TB among patients with cough attending OPD, FP and MCH clinics in Dar es Salaam, regardless of the duration of cough (paper I and II) ii) the extent of, and factors responsible for delay in TB case detection in Pwani region (paper III) iii) the cost associated with TB healthcare seeking for the current passive case finding strategy in Dar es Salaam (paper IV). Study design and Setting: We conducted a cross sectional hospital based study in Dar es Salaam and Pwani region between 2007 and 2008. The two regions are adjacent to each other and are located along the coastal area of Tanzania, and share cultural values. Dar es Salaam and Pwani have a total population of 2,497,940 and 889,154 respectively. Study participants and data analysis: Study participants for detection of pulmonary tuberculosis among patients with cough attending OPD (paper I) involved patients aged 5 years and above. Study participants for paper II were women aged 15 years and above attending MCH and FP clinics with a cough. We actively asked for a cough, all study participants from FP and MCH clinics. All participants included in paper I and II were screened for TB by sputum microscopy, regardless of the duration of cough. Study participants for paper III were smear positive TB patients diagnosed to have TB within three months prior to the commencement of the study. The study for paper IV involved patients aged 15 years and above and who have just been diagnosed to have TB. We trained all research assistants. Questionnaires were pre tested. The outcome variables included: diagnosis of smear positive TB, magnitude of TB diagnostic delay and risk factors associated with delay as well as patients and households cost before TB diagnosis. We explored possible associations between cough duration and smear results, clinic of diagnosis, place of first presentation and number of visits made prior to diagnosis. The main methods for analysis were cross tabulation and bivariate logistic regression. We applied the following statistical tests to determine estimated proportions and risk factors: odds ratios, means and medians, Pearson Chi- square, Wald statistic and 95% CI. Results: The study involved 3,372 patients: 2274 patients with cough attending OPD; 749 women with cough attending FP and MCH clinics; 226 smear positive pulmonary tuberculosis patients from DOTS clinics and 123 TB patients who have just been diagnosed to have TB. The probability of being smear positive does not depend on cough duration. Out of 2274 patients with cough, regardless of the duration, and attending OPD, 2214 (97.4%) remembered their cough duration. One thousand nine hundred and seventy three patients (89.1%) coughed for two weeks or more compared to 241 (10.9%) patients who coughed for less than 2 weeks. Of those who coughed for two weeks or more, 250 (12.7%) had smear positive PTB, and of those who had coughed for less than two weeks, 21 (8.7%) had smear positive PTB. There was no statistically significant difference in the proportion of smear positive tuberculosis among the two comparison groups (Pearson Chi-Square 3.2; p = 0.074). The proportion of smear positive TB patients among women with cough attending FP and MCH clinics was 3.8%. Out of 749 women with cough, regardless of duration, attending FP and MCH clinics, 529 (70.6%) were from MCH clinics. Six hundred and sixteen (82.2%) patients had coughed for less than two weeks compared to 133 (17.8%), who had coughed for two or more weeks. Among 616 TB suspects, 14 (2.3%) were smear positive TB patients, and of the 133 who had coughed for two or more weeks, 13 (9.8%) were smear positive TB patients. Risk factors associated with smear positive results included attended more than one visit to any facility prior to diagnosis (OR = 6.8; 95%CI 2.57–18.0) and having HIV/AIDS (OR = 4.4; 95%CI 1.65–11.96). Long duration of cough was not a risk factor for being smear positive (OR = 1.6; 95%CI 0.59–4.49). Out of 226 patients enrolled for studying delays in TB detection in Pwani region, results were available for 206. The majority (66.5%) of the patients were males. Mean age for males and females was 37.3 and 33.7 years, respectively. Mean (SD) total delay before initiation of treatment was 125.5 (98.5) days (median 90 days). Out of 206 patients, 79 (38.4%) delayed to seek TB healthcare. Health facility delay was observed among 121 (58.7%) patients. Risk factors for patients’ delay was poor knowledge that chest pain may be a TB symptom (OR= 2.9; 95%CI 1.20- 7.03) and the belief that TB is always associated with HIV/AIDS (OR=2.7; 95%CI 1.39-5.23). Patients who presented with chest pain were (OR= 0.2; 95%CI 0.10-0.61) less likely to delay compared to those with no chest pain. One hundred and twenty three TB patients were enrolled for studying patient and household costs for TB diagnosis. The majority (53.7%) were males. The mean age for male and female patients was 34.8 and 37.0 (14.8) years, respectively. Before TB diagnosis, overall, TB patients lost 46% of their mean monthly income. At individual level, poor patients had relatively higher mean loss of income (15 USD, 52%) of mean monthly income compared to less poor patients (45 USD equivalent to 46%) of mean monthly income. Thirty eight (39.2%) patients stopped working due to TB illness with an average of 77 days off. Before diagnosis, overall, tuberculosis caused on average 25 patients lost work days. Among patients who stopped working, 17 (47.7%) were head and the main bread earner for their households whereas, 5 (55.6%) were poor compared to 12 (41.4%) less poor. At the household level, overall household mean monthly income dropped by 28%. Poor households had relatively higher loss of monthly income compared to less poor. In-direct cost contributed to almost two thirds of the total cost. While poor households spent 150% of their mean monthly income, less poor spent 43%. Conclusions: • Detection of smear positive PTB among patients who coughed for less than two weeks, was as high as for those who coughed for two weeks or more. • Almost four percent of women with cough attending MCH and FP were smear positive TB patients. • There is a considerable delay in TB diagnosis and treatment in Pwani, mainly due to patients’ delay. • The economic burden of TB health care seeking is high, especially for the poorest households.