Treatment outcomes of various types of tuberculosis in Pakistan, 2006 and 2007

PDF version

A.A. Chughtai,¹ C.R. MacIntyre,² Y.A. Wang,³ Z. Gao 2 and W. Khan⁴

residency, I saw a 8 years old kid with a gangrene almost half of his foot. His parents refused amputation and believed there is a way it could get back to normal. We tried to convince them there is no other way but amputation. They refused to listen. They took him to many doctors but of course there were no other solution. They came back 1 week later with the gangrene extending to his ankle. If we did the amputation first time that would have been a forefoot debridtement. This time it will be a below kneww amputation. I wanted to punch his fater in the face

حصائل معالجة مختلف أنماط السل في باكستان 2006-2007

أبرار أحمد شوجاتي، شانديني راينا ماكلينطير، أليكس وانج، زان هاي جاو، واثق محمود خان

الخلاصـة: إن قياس حصائل المعالجة أمر هام لنجاح برامج مكافحة السل. وتهدف هذه الدراسة للتعرُّف على حصائل الأنماط المختلفة لحالات السل التي سُجِّلت في باكستان على مدى عامَيْن مع مقارنة تلك الحصائل بين مختلف الولايات والأقاليم في باكستان. وهي دراسة استعادية أترابية تَمَّت فيها مراجعة التقارير حول حصائل معالجة السل وشملت 694 349 حالة سل رئوي مسجلة في باكستان خلال عامَيْ 2006 و2007، وبلغ عدد الحالات التي عولجت بنجاح 154 309 (%88.4)، وكان معدّل النجاح أعلى في الحالات الجديدة الإيجابية اللطاخة وأخفض في حالات إعادة المعالجة. أما المقارنة بين الولايات والأقاليم فقد أظهرت أن نجاح المعالجة أعلى بقدر يُعْتَدُّ به إحصائياً في 4 ولايات من أصل 8 ولايات. وخلصت الدراسة إلى أنه ينبغي تحسين معدّلات نجاح المعالجة ولاسيَّما في حالات إعادة المعالجة؛ وأنه ينبغي على البرنامج الوطني لمكافحة السل مراجعة البرامج في الأقاليم والولايات وأن يستفيد من دروس البرامج ذات الأداء الجيِّد، كما ينبغي دراسة العوامل المتعلقة بالمريض والتي قد تؤثِّر على حصائل المعالجة.

ABSTRACT Measuring treatment outcome is important for successful tuberculosis (TB) control programmes. The purpose of this study was to examine the outcomes of various types of TB cases registered in Pakistan over a 2-year period and compare those outcomes among the different provinces and regions of the country. A retrospective, cohort study was conducted in which TB treatment outcome reports were reviewed. Of the 349 694 pulmonary TB cases registered in Pakistan during 2006 and 2007, 309 154 (88.4%) were treated successfully. Treatment success was significantly higher in new smear-positive cases and lower in retreatment cases. Among the provinces and regions, treatment success was significantly higher in 4 out of 8 provinces. Treatment success needs to be improved, particularly in retreatment cases. The national TB control programme should review the provincial and regional programmes and learn lessons from well-performing programmes. Patient factors that may affect the treatment outcome should be also studied.

Résultats thérapeutiques pour les différents types de tuberculose au Pakistan en 2006 et 2007

RÉSUMÉ Mesurer les résultats thérapeutiques est important pour la réussite des programmes de lutte antituberculeuse. L'objectif de la présente étude était d'évaluer les résultats thérapeutiques des différents types de cas de tuberculose enregistrés au Pakistan au cours d'une période de deux ans et de comparer ces résultats avec ceux des différentes provinces et régions du pays. Une étude de cohorte rétrospective a été menée au cours de laquelle les rapports sur les résultats de traitement de la tuberculose ont été examinés. Sur les 349 694 cas de tuberculose pulmonaire enregistrés au Pakistan en 2006 et 2007, 309 154 (88,4 %) ont été traités avec succès. Le taux de guérison était nettement supérieur pour les nouveaux cas à frottis positif et inférieur pour les cas de retraitement. En outre, au niveau des provinces et des régions, il était nettement plus élevé dans 4 des 8 provinces. Le taux de guérison doit être amélioré, notamment pour les cas de retraitement. Le programme national de lutte antituberculeuse devrait revoir les programmes dans les provinces et les régions et tirer les enseignements des programmes performants. Les facteurs concernant le patient qui pourraient affecter les résultats du traitement doivent aussi être étudiés.

1National Tuberculosis Control Programme, Islamabad, Pakistan (Correspondence to A.A. Chughtai: This email address is being protected from spambots. You need JavaScript enabled to view it.).
2School of Public Health and Community Medicine; 3School of Women and Child Health, University of New South Wales, Sydney, Australia.
4World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt.
Received: 22/11/11; accepted: 22/04/12
EMHJ, 2013, 19(6):535-541

Introduction

Tuberculosis (TB) is one of the major public health problems in Pakistan. The country ranks 6th among the high-burden TB countries and harbours 63% of the disease burden in the Eastern Mediterranean Region. The incidence of smear-positive TB cases is 97/100 000 and the incidence of all types of TB cases is 231/100 000 [1]. The Pakistan national TB control programme (NTP) adopted the World Health Organization (WHO) DOTS treatment strategy in 1995 [2]. After lapsing, the NTP was revived in 2001 and TB was declared a national emergency [3]. Since then efforts have been made to expand partnerships and bring all stakeholders on board in order to control this disease more effectively. In line with Millennium Development Goals the case detection rate for new smear-positive cases increased from 2.8% in 2000 to 75% in 2008. The treatment success rate (TSR) has also increased from 74% in 2000 to 87% in 2007 and been maintained to date, against the target of 85% [4].

Measuring treatment outcome is always important for successful TB control programmes. Although many factors affect the successful outcome of TB treatment, the type of TB is the most important predictor of treatment outcome. The TSR is low in multi-drug resistant and other retreatment TB cases, including cases of relapse, treatment after failure and treatment after default. Studies also show that the chances of being cured reduces as sputum grading increases [5] and this is particularly significant in retreatment cases [6]. TSRs are considered good in DOTS areas [7]. Globally, TB is reported to be higher in males [8,9], but treatment compliance is usually higher in females [10–14]. Treatment outcome has been also assessed on the basis of treatment periods [15–18] and among different regions [16]. In most of these studies, the TSR and cure rate increased over time [17,18]. In this national study in Pakistan the outcome of TB cases registered from 2006–07 was assessed on the basis of the time period, types of TB and different provinces/regions.

Methods

Study design

A retrospective cohort study was conducted to evaluate the factors associated with the outcome of TB cases registered in Pakistan. TB treatment outcome data of all provinces and regions was included in this study.

Data collection

TB case finding and treatment outcome reports were reviewed and all patients registered and reported during 2006– 07 were included in the study. A total of 409 366 TB cases were reported during this period, out of which 349 694 were pulmonary and 59 772 were extra­pulmonary. As the NTP does not declare the outcome of extrapulmonary cases only pulmonary TB cases were included in this study.

The NTP uses the WHO’s recommended recording and reporting system for TB surveillance. The WHO has developed a set of tools and these have been recently revised through a consultative process [19]. The NTP has adopted these revised recording and reporting tools, with some minor modifications, in the local context. The NTP has a set of nationally agreed nomenclature and definitions for various treatment outcomes of the TB patients. The definitions used in the programme are compatible with international recommendations. We used the standard treatment outcomes: cured, treatment complete, died, treatment failure, defaulted and transferred out.

The data were analysed by type of TB case (new smear-positive, new smear-negative, relapsed, treatment after failure and treatment after default), year (2006 and 2007) and region [Azad Jammu and Kashmir Province (AJK), Balochistan, Federally Administered Tribal Areas (FATA); Northern Areas (NA), North-West Frontier Province (NWFP), Punjab, Sindh and the Federal TB control centre].

Ethical issues

Ethical clearance was obtained from the human research ethics advisory panel of University of New South Wales, Australia. Permission was sought from the NTP in the Ministry of Health, Pakistan, to use de-identified TB data for this study.

Data entry and analysis

The TB treatment outcome data of all districts was entered into SPSS, version 17. All data were double-checked for errors and missing values. Means and frequencies were used for the description of various variables. Logistic regression was used to estimate the odds ratios (OR) and their 95% confidence intervals (CI), with the TB treatment result as an outcome. First, univariate analyses were performed to examine the effect of each variable on the different outcomes of TB. Then multivariate logistic models were constructed; in the Punjab province, new smear-positive TB cases and the year 2006 were reference categories. In all statistical tests, P-value < 0.05 was considered significant.

Results

A total of 349 694 pulmonary TB cases were included in the study: 153 063 cases were registered in 2006 and 196 631 cases in 2007. Among the total cases, most were new cases (153 590 smear-positive and 183 448 smear-negative); < 4% were retreatment cases (6839 relapsed, 1296 treatment after failure and 4521 treatment after default) (Table 1). A significant variation was seen in the type of TB cases registered in different provinces and regions.

Of the total cases registered during 2006 and 2007 123 975 (40.1%) cases were cured and 185 179 (59.9%) completed treatment. Thus 309 154 (88.4%) cases were classified as treatment successful (Table 1). Cure and treatment completion rates were high in new TB cases and low in retreatment cases (relapsed, treatment after failure and treatment after default). Death, failure and default rates were high in retreatment cases.

Univariate analysis showed a high TSR in smear-positive cases compared with all other types of TB (Table 2). New smear-negative cases were 14% less likely to be treated successfully than new smear-positive cases (OR 0.86, 95% CI: 0.84–0.88). Among the retreatment cases, TSRs were lowest for the treatment after default cases (OR 0.28, 95% CI: 0.26–0.30), followed by the treatment after failure (OR 0.30, 95% CI: 0.27–0.34) and relapsed cases (OR 0.60, 95% CI: 0.56–0.63). TSR significantly improved in 2007 compared with 2006 (OR 1.34, 95 CI: 1.32–1.37). Compared with Punjab province, more cases were successfully treated in the NA (OR 3.55, 95% CI: 2.90–4.30), AJK (OR 2.21, 95% CI: 1.98–2.50), NWFP (OR 1.51, 95% CI: 1.46–1.58) and FATA (OR 1.42, 95% CI: 1.28–1.58). In Baluchistan, Sindh and the Federal TB centre, the outcome was poor when compared with the Punjab (Table 2).

The death rate was significantly lower in new smear-negative cases (OR 0.84, 95% CI: 0.80–0.88) and higher in relapsed (OR 1.62, 95% CI: 1.43–1.83), treatment after failure (OR 2.00, 95% CI: 1.55–2.59) and treatment after default cases (OR 1.7, 95% CI: 1.46–1.98). The death rate was lower in 2007 compared with 2006 (OR 0.996, 95% CI: 0.995–0.997). Compared with Punjab, more patients died in AJK (OR 1.69, 95% CI: 1.47–1.93), Baluchistan (OR 1.62, 95% CI: 1.47–1.92) and Sindh (OR 1.20, 95% CI: 1.14–1.27) provinces.

After multivariate analysis with adjustment for the type of TB and the year of registration there were no major changes in the ORs.

Discussion

Declaring treatment outcomes is important for monitoring the quality of case management and the progress of TB control programmes. Data from 2006 and 2007 were analysed in this study. The treatment outcome was significantly higher in 2007 compared with 2006; TSRs increased and death rates decreased significantly in 2007. The TSR of new smear-positive cases was 74% in 2000 and this had increased to 91% by 2007. Data from the NTP Pakistan showed that the TSR of retreatment cases was 66% in 2002 and had increased to 79% in 2007 [20]. It is also interesting to note that the NTP was in the DOTS expansion phase from 2002 to 2005 and achieved 100% DOTS coverage in 2005. With this rapid DOTS expansion, the TSR also increased in that time period [20]. Some other studies also showed that the DOTS population coverage had a significant effect on the overall TSR and countries with full DOTS coverage had an 18% higher rate of successful treatment [21].

Treatment outcome is generally poor in retreatment TB cases [1], particularly in cases of treatment after failure [22]. Confirming WHO data [1], the outcome of retreatment cases in our study was poorer than smear-positive cases. Among the retreatment cases, we found that successful outcome was worse for the cases of treatment after default (OR 0.33, 95% CI: 0.29–0.38) than treatment after failure (OR 0.28, 95% CI: 0.26–0.30) . TSRs were high in smear-positive cases and this was mainly due to the high default rate in other types of TB cases. New smear-positive cases are usually more severe than smear-negative cases, but less severe than retreatment cases. This may be a reason for the high number of deaths in retreatment cases compared with smear-positive cases. Among the retreatment cases, more cases of treatment after failure died (OR 1.88, 95% CI: 1.45–2.42) than treatment after default (OR 1.68, 95% CI: 1.44–1.95) and relapsed cases (OR 1.60, 95% CI: 1.40–1.80). These results also highlight the possibility of drug resistance among treatment after failure cases who are sputum smear-negative at the start of treatment and become smear-positive during the course of treatment. Failure to treat in category 2 also resulted in treatment failure (this includes smear-positive relapses, failure, treatment after default and other smear-positive patients who have received anti-TB treatment for more than 1 month without being registered). Consequently, it would appear that there are more chances of drug resistance in these cases. These cases must be offered drug susceptibility testing and 2nd-line TB drugs from the beginning of treatment. The Pakistan NTP is using an 8-month regimen and 2 drugs, isoniazid and ethambutol, in the continuous phase of category 1 cases (this includes new cases of pulmonary TB, whether smear-positive or -negative, extrapulmonary TB and other smear-negative and extrapulmonary patients). Studies have shown that a 6-month regimen, which contains rifampicin, has a good outcome and low relapse rate. Treatment failure and relapse is more common with an 8-month regimen [23]. Most high-burden TB countries have already shifted to the 6-month regimen containing rifampicin. It is therefore important to evaluate the efficacy of the 8-month regimen that does not contain rifampicin in Pakistan and, importantly, consider the 6-month regimen, which is more effective.

As a whole, the TSR was good in most of the regions except Sindh province and the Federal TB centre. Compared with Punjab, more cases were successfully treated in NA, AJK, NWFP and FATA. In Baluchistan, Sindh and the Federal TB centre, the outcome was poor compared with Punjab. The main reason for the low success rate seemed to be the high number of default cases. The population of the provinces, socioeconomic status and security situations are important factors to consider, as these can affect the default rate. At around 80 million people Punjab is the most populous province and contains half of the population of the Pakistan. The default rate was low in NA, AJK and FATA, presumably because the population is very low compared with Punjab. The population of NA is around 1 million and of FATA and AJK around 3 million. Furthermore, a poor law and order situation and low socioeconomic status may be a cause of high default rates. Both factors exist in Baluchistan, where the law and order situation is very poor and socioeconomic status is also low. The high default rate in Sindh is not easy to explain and the NTP should explore the reasons for this phenomenon. The Federal TB centre is a specialized TB institution, affiliated with the Federal government, and caters for cases not only from the Federal capital but also from neighbouring districts. As the Federal capital is not included in any province and region, the data were collected separately. The TSR was low and this may be due to poor case retention and a high default rate. Patients often come to the centre from remote areas and once they start improving they do not return. Interestingly, AJK had a good TSR but had a significantly higher number of deaths than other provinces. Although deaths were also lower in FATA and NA, this was not significant. Unfortunately, the causes of death cannot be assessed through the outcome data, as individual TB patients’ data are required for this purpose.

According to our knowledge, this was the first study in Pakistan to evaluate the outcome of TB on the basis of types and regions. Nevertheless it also had some unavoidable limitations. Firstly, we were restricted to using TB patients’ outcome reports. As individual patient data was not used in this study, many other variables which may also affect the treatment outcome—for example age, sex, sputum status and duration of symptoms—could not be accessed. Secondly, there was the possibility of misclassification, for example, patients were categorized as new or retreatment cases on their history of previous treatment. If patients are not categorized properly, there is a greater chance of unfavourable outcomes. Thirdly, the causes of deaths and default were not given in outcome reports and death and default are the main reasons for low treatment success. It is particularly important to know whether patients died due to TB or to other reasons. Fourthly, drug susceptibility and HIV status was not routinely tested before the start of treatment and this may have affected the outcome of TB. Finally, these data were mainly obtained from the public sector, as very few private providers sent data to the NTP. The NTP has recently started many public–private partnership ventures to involve the private sector in DOTS.

Conclusion

We found that the treatment outcome was generally high for new TB cases (smear-positive and smear-negative) and low for retreatment cases (relapsed, treatment after failure and treatment after default). Ways to improve TB case management need to be investigated to reduce the chances of recurrence and treatment failure, particularly in retreatment cases, perhaps by adopting the 6-month short-course regimen. The treatment outcome also varied among different provinces and regions. The NTP should review the provincial/regional programmes and learn lessons from well-performing programmes. Defaulting from treatment was linked to poor outcomes in some areas. Causes of the high death rate in AJK and Baluchistan need to be explored and contributing factors addressed promptly. It is also essential to conduct further studies to assess the effect of other patient variables on the outcome of TB treatment.

Acknowledgements

I am thankful to the NTP Pakistan, for providing data for this project and giving permission to use that data for research purposes. Particularly I would like to acknowledge the support of Dr Noor Ahmad Baloch (national manager, NTP) and Dr Ejaz Qadeer (research coordinator, NTP).

Funding: None.

Competing interests: None declared.

References

1. Global tuberculosis report. Geneva, World Health Organization, 2010.
2. Strategic plan. Islamabad, National Tuberculosis Control Programme, Government of Pakistan, 2010.
3. Performance indicators. Pakistan National Tuberculosis Control Programme [online] (http://ntpgovpk/PerformanceIndicatorshtm, accessed 15 March 2013).
4. Revised national tuberculosis treatment guideline. Islamabad, National Tuberculosis Control Programme, Government of Pakistan, 2012.
5. Gopi P et al. Association of conversion and cure with initial smear grading among new smear positive pulmonary tuberculosis patients treated with category I regimen. Indian Journal of Medical Research, 2006, 123:807–814.
6. Mukherjee A et al. Outcomes of different subgroups of smear-positive retreatment patients under RNTCP in rural West Bengal, India. Rural and Remote Health, 2009, 9:926.
7. Dye C, Hosseini M, Watt C. Did we reach the 2005 targets for tuberculosis control? Bulletin of the World Health Organization, 2007, 85:364–369.
8. Ottamani S, Uplekar M. Gender and TB: points from routine records and reports. International Journal of Tuberculosis and Lung Disease, 2008, 12:827–828.
9. Weiss M. Gender and TB: socio-cultural aspects. International Journal of Tuberculosis and Lung Disease, 2008, 12:825–866.
10. Uplekar M et al. Attention to gender issues in tuberculosis control. International Journal of Tuberculosis and Lung Disease, 2001, 5:220–224.
11. Zellweger J, Coulon P. Outcome of patients treated for tuberculosis in Vaud county, Switzerland. International Journal of Tuberculosis and Lung Disease, 1998, 2:372–377.
12. Ditah IC et al. Monitoring tuberculosis treatment outcome: analysis of national surveillance data from a clinical perspective. Thorax, 2008, 63:440–444.
13. Lagrada L, Uehara N, Kawahara K. Analysis of factors of treatment completion in dots health facilities in metro Manila, Philippines: a case-control study. Kekkaku, 2008, 83:765–772.
14. Okanurak K, Kitayaporn D, Akarasewi P. Factors contributing to treatment success among tuberculosis patients: a prospective cohort study in Bangkok. International Journal of Tuberculosis and Lung Disease, 2008, 12:1160–1165.
15. Farah MG et al. Treatment outcome of new culture positive pulmonary tuberculosis in Norway. BMC Public Health, 2005, 5:14.
16. Faustini A, Hall A, Perucci C. Tuberculosis treatment outcomes in Europe: a systematic review. European Respiratory Journal, 2005, 26:503–510.
17. Bao Q-S, Du Y-H, Lu C-Y. Treatment outcome of new pulmonary tuberculosis in Guangzhou, China 1993–2002: a register-based cohort study. BMC Public Health, 2007, 7:344.
18. Hsu C-B et al. Preliminary evaluation of Taiwan’s tuberculosis DOTS strategy. Taiwan Epidemiology Bulletin, 2008, 24:230–253.
19. Revised TB recording and reporting forms and registers – version 2006. Geneva, World Health Organization, 2006 (WHO/HTM/TB/2006).
20. National Tuberculosis Control Program. Government of Pakistan [online database] (http://ntp.gov.pk, accessed 30 March 2013).
21. Obermeyer Z, Abbott-Klafter J, Murray C. Has the DOTS strategy improved case finding or treatment success? An empirical assessment. PLoS One, 2008, 3(3):e1721.
22. Mehra R et al. Study of relapse and failure cases of CAT I retreated with CAT II under RNTCP—an eleven year follow up. Indian Journal of Tuberculosis, 2008, 55:188–191.
23. Treatment of tuberculosis: guidelines for national programs. Geneva, World Health Organization, 2003.