Factors Contributing to the High Prevalence of Multidrug-Resistant Tuberculosis Among Previously Treated Patients: A Case

Abstract

Setting: Multidrug-resistant tuberculosis (MDR-TB) has emerged as a serious global public health problem. In China, the risk factors for MDR-TB have not been systematically evaluated. Objective: To identify risk factors associated with MDR-TB among previously treated patients in China. Design: A case–control study was carried out. Cases were selected from previously treated MDR-TB patients who were resistant to both isoniazid and rifampin, and controls were selected from previously treated TB patients who were sensitive to isoniazid and rifampin (non-MDR-TB). Information was collected from the registration database and a structured questionnaire. Results: A total of 61 cases and 50 controls were recruited. A multivariate analysis showed that the family annual per-capita income ≤7,000 Yuan (odds ratio [OR]=3.238; 95% confidence interval [CI]: 1.270–8.252), no history of fixed dose combinations (FDCs) in anti-TB treatment (OR=4.027; 95% CI: 1.457–11.129), and adverse reactions in the course of TB treatment (OR=3.568; 95% CI: 1.402–9.085) were independent predictors of MDR-TB. Moreover, among the TB patients who had adverse reactions, quitting the treatment was shown as a risk factor for MDR-TB (p=0.009). Conclusion: In the control of MDR-TB among previously treated patients, lower socioeconomic groups, the expanding use of FDCs, and improving adherence to treatment by implementing Directly Observed Therapy Short Course-Plus (DOTS-Plus), strictly should become a priority that requires strong commitment and collaboration among health organizations.

Introduction

Tuberculosis (TB) is becoming a worldwide problem due to the recurrence of the disease, the appearance of drug-resistant strains, and coinfection of TB and human immunodeficiency virus (HIV).²⁹ Multidrug-resistant tuberculosis (MDR-TB), defined as resistance to at least isoniazid and rifampin, has emerged as a global public health problem.²⁷ The World Health Organization (WHO) has documented that MDR-TB is becoming extensively widespread today. According to the most recent data, among all incident TB cases worldwide in 2008, 3.6% were estimated to have MDR-TB.

Almost half of the global MDR-TB cases were reported from China and India.³¹ China has the second largest number of MDR-TB cases in the world.²⁹ The National Anti-Tuberculosis Drug Resistance Survey in 2007 showed that, an estimated 120,000 new MDR-TB cases emerge annually in China, accounting for ≈24% of MDR-TB worldwide.¹⁴ In the past decade, many studies have evaluated the role of the Beijing genotype of Mycobacterium tuberculosis in global TB epidemic. In Europe, 85% of MDR-TB and extensively drug-resistant TB cases were caused by Beijing strains, while only 6–7% of drug-susceptible TB cases were caused by Beijing strains.⁶ As the name suggests, Beijing genotype strains are particularly prevalent in China.²⁵ In 2009, a nationwide drug resistance survey in China found that 5.7% of new cases of TB and 25.6% of retreated cases were infected with MDR-TB.³¹

Controlling MDR-TB is challenging, because it is difficult to diagnose and treat.⁵ In addition, it requires the use of second-line drugs that are difficult to acquire and much more toxic and expensive than the first-line regimen.⁸ Thus, MDR-TB is increasingly becoming a serious threat to TB control,^27,33 and the recognition of extensively drug-resistant TB, that is MDR-TB strains with resistance to any fluoroquinolone and one or more of the three injectable second-line anti-tuberculosis drugs, namely, kanamycin, amikacin, and capreomycin,³² has further highlighted this threat.^32,33 In China, MDR-TB rates have been reported,³¹ and previous anti-tuberculosis treatment is a widely reported risk factor for MDR-TB in many studies,^{2,7,9,13,19,20,26} with reported rates varying between 20% and 80%,^7,9 whereas other associated risk factors have not been fully described, especially among patients who have been treated in China.

The objective of our study was to identify specific risk factors associated with MDR-TB among previously treated patients in China so that patients with high risk of MDR-TB can be identified early and policy recommendations to reduce the occurrence of MDR-TB can be made.

Study Population and Methods

Study design

A case–control study was conducted at Hangzhou Red Cross Hospital, Zhejiang province, that is, Zhejiang Provincial Tuberculosis Hospital (ZPTH): cases were previously treated MDR-TB patients who were resistant to both isoniazid and rifampin, and controls were previously treated TB patients who were infected with M. tuberculosis, which is sensitive to isoniazid and rifampin (non-MDR-TB). Cases and controls were identified from retreated TB patients who were registered in the hospital between May 2010 and July 2011 based on drug susceptibility testing (DST). Information was collected retrospectively from cases and compared with controls (non-MDR-TB).

Inclusion criteria

All enrolled patients were diagnosed sputum smear-positive pulmonary TB (PTB) during the survey period. A sputum smear-positive status was assigned according to WHO/International Union Against Tuberculosis and Lung Disease (IUATLD) definitions.³⁰ Previously treated TB cases were defined as patients who had a previous history of treatment with anti-tuberculosis drugs for more than 1 month.³⁴ Previously treated cases included relapses, treatment after failure, and treatment after default. MDR-TB was defined as resistance to at least both isoniazid and rifampin using DST.⁵ DST was done by the indirect proportion method on the 7H10 Middlebrook medium.

Cases were MDR-TB patients identified by searching microbiology laboratory records, inpatient diagnostic databases, and TB reporting systems. Controls were sensitive TB (M. tuberculosis sensitive to rifampin and isoniazid) patients obtained by selecting the preceding or subsequent sensitive patients for each MDR-TB case. Where the medical records of these control patients were unavailable, we selected the next available appropriate patient.

Data collection

The following information was obtained on a standardized form based on vigorous interview and validated review of the patient's medical record whenever possible: sex, age, occupation, ethnicity, residence status, working status, education levels, presence of cavities on chest radiographs, sputum smear test results, family income, health insurance, TB treatment history, TB family history, TB contact history, symptoms at first diagnosis of TB, current alcohol use, current smoking, medical and health conditions, TB knowledge, etc.

Data entry and statistical analysis

Data were double entered into a database using EpiData3.1 software. The two databases were compared to ensure accuracy. Data were exported into a SAS 9.1 (SAS, Inc., Cary, NC) to carry out univariate and multivariate analysis. Univariate analysis was performed to screen risk factors for MDR-TB. To adjust for multiple covariates, we used a multivariate logistic regression method. Variables included in the final multivariate model were chosen a priori on the basis of statistical criteria of p-value ≤0.20, and on the basis of the biological plausibility of their association with MDR-TB. The multivariate model uses a stepwise selection method with SLE=0.15 and SLS=0.20 to identify independent factors associated with MDR-TB. Crude and age- and sex-adjusted models were calculated using Mantel–Haenszel odds ratios (OR) and 95% confidence intervals (CI).

Protection of human subjects

This research project was approved by the Chinese Ethics Committee for TB Operational Research, Chinese Center for Disease Control and Prevention and the Institutional Review Board of Shandong University School of Public Health. Written informed consents were obtained from all patients before the investigation was performed for their information to be stored in the hospital database and used for research.

Results

Study population

A total of 111 previously treated TB patients were enrolled in the study, including 61 (55.0%) MDR-TB and 50 (45.0%) non-MDR-TB. Among the 111 patients, 77 (69.4%) were male and 34 (30.6%) were female, the age ranged from 17 to 82 years with a mean of 44.47±17.04 years (SD); 38 (39.2%) were relapsed cases, 42 (43.3%) were treatment failure cases, 17 (17.5%) were treatment after default cases. Among the 61 MDR-TB cases, 45 (73.8%) were male and 16 (26.2%) were female, the age ranged from 17 to 82 years with a mean of 43.59±16.29 years (SD). All MDR-TB cases suffered from PTB, four (6.6%) of whom had concomitant extrapulmonary TB. Of these 50 controls, 32 (64.0%) were male and 18 (36.0%) were female, the age ranged from 18 to 81 years with a mean of 45.54±18.02 years (SD) (Table 1).

Table 1.

Demographical Characteristics of Cases and Controls

	MDR-TB (n=61)		Non-MDR-TB (n=50)
Characteristics	No.	%	No.	%	χ²	p-Value
Sex					1.234	0.267
Male	45	73.8	32	64.0
Female	16	26.2	18	36.0
Age					1.043	0.307
≤42 years	34	55.7	23	46.0
>42 years	27	44.3	27	54.0
Ethnicity					–	1.000^a
Han	60	98.4	49	98.0
Other	1	1.6	1	2.0
Occupation					1.615	0.204
Farmer or rural migrant worker	28	45.9	17	34.0
Other	33	54.1	33	66.0
Education					–	0.176^a
Primary school and below	22	36.1	24	48.0
Secondary and high school	36	59.0	21	42.0
College and above	3	4.9	5	10.0
Marital status					0.606	0.436
Married/cohabit	41	67.2	37	74.0
Single/divorced/widow	20	32.8	13	26.0
Registration categories^b					2.926	0.231
Relapsed cases	25	44.6	13	31.7
Treatment failure cases	24	42.9	18	43.9
Defaulters	7	12.5	10	24.4

Fisher's exact test.

Fourteen cases missing.

MDR-TB, multidrug-resistant tuberculosis; –, not available.

Risk factors for MDR-TB

In univariate analysis, the body mass index (BMI) (p=0.045), education (p=0.176), family annual per-capita income (p=0.014), type of housing (p=0.092), family per-capita housing area (p=0.006), exposure to dust working environment (p=0.092), immediate family members suffering from TB (p=0.185), symptoms of night sweats at the first diagnosis of TB (p=0.001), symptoms of appetite loss at the first diagnosis of TB (p=0.034), taking anti-TB drugs supervised by somebody (p=0.108), the number of times of anti-TB treatment before the present one (p=0.014), the history of fixed-dose combinations (FDCs) in anti-TB treatment (p=0.001), adverse reactions in the course of TB treatment (p=0.003), and quitting treatment after adverse reactions among the patients who had adverse reactions (p=0.009) were significantly associated with MDR-TB (p<0.20). Since this step is just for selecting factors for the next multivariate analysis, p<0.20 was considered significant. Among the TB patients who had an adverse reaction, quitting the treatment was significantly associated with MDR-TB (p=0.009) (Table 2).

Table 2.

Univariate Analysis of Risk Factors for MDR-TB Among Previously Treated Patients in Hangzhou, China, May 2010 Through July 2011

	MDR-TB (n=61)		Non-MDR-TB (n=50)
Variables	No.	%	No.	%	χ²	p-Value
Body mass index (BMI)					–	0.045^c
<18.5 kg/m²	25	41.0	17	34.0
Normal	35	57.4	26	52.0
≥24 kg/m²	1	1.6	7	14.0
Local residence time					0.409	0.522
≤25 years	33	54.1	24	48.0
>25 years	28	45.9	26	52.0
Family annual per-capita income (Yuan)					6.033	0.014
≤7,000	41	67.2	22	44.0
>7,000	20	32.8	28	56.0
Type of housing					2.841	0.092
Building	40	65.6	40	80.0
Brick or adobe bungalows	21	34.4	10	20.0
Family per-capita housing area (m²)					7.408	0.006
≤24	39	63.9	19	38.0
>24	22	36.1	31	62.0
Number of house windows					1.404	0.236
≤4	42	68.9	29	58.0
>4	19	31.1	21	42.0
Exposure to dust working environment					2.839	0.092
Yes	15	24.6	6	12.0
No	46	75.4	44	88.0
Immediate family members suffering from TB					1.756	0.185
Yes	10	16.4	4	8.0
No	51	83.6	46	92.0
History of smoking					0.087	0.768
Yes	31	50.8	24	48.0
No	30	49.2	26	52.0
History of drinking					1.571	0.210
Yes	24	39.3	14	28.0
No	37	60.7	36	72.0
Taking medication by yourself when had a disease without seeing a doctor					0.563	0.453
Often	9	14.8	5	10.0
Occasional or never	52	85.2	45	90.0
History of isoniazid use before TB treatment					0.050	0.823
Yes	10	16.4	9	18.0
No	51	83.6	41	82.0
History of rifampin use before TB treatment					0.033	0.856
Yes	9	14.8	8	16.0
No	52	85.2	42	84.0
Symptom of night sweats at the first diagnosis of TB					11.034	0.001
Yes	17	27.9	2	4.0
No	44	72.1	48	96.0
Symptom of appetite loss at the first diagnosis of TB					4.504	0.034
Yes	19	31.1	7	14.0
No	42	68.9	43	86.0
Time of TB symptoms lasting before first seeing a doctor					0.111	0.739
≤10 days	31	50.8	27	54.0
>10 days	30	49.2	23	46.0
Taking anti-TB drugs supervised by somebody					2.577	0.108
Yes	42	68.9	27	54.0
No	19	31.1	23	46.0
Number of times of anti-TB treatment before the present one^a					10.555	0.014
One	10	16.4	13	26.0
Two	25	41.0	14	28.0
Three and more	16	26.2	5	10.0
Unknown	10	16.4	18	36.0
History of FDCs in anti-TB treatment					11.798	0.001
Yes	24	39.3	36	72.0
No	37	60.7	14	28.0
Adverse reactions in the course of TB treatment					8.583	0.003
Yes	39	63.9	18	36.0
No or unknown	22	36.1	32	64.0
Behavior after adverse reactions in the course of TB treatment^b					–	0.009^c
Quitting treatment	30	76.9	5	35.7
Adhering to treatment	9	23.1	9	64.3
Quitting treatment one or more times in anti-TB treatment					0.573	0.449
Yes	19	31.1	19	38.0
No	42	68.9	31	62.0
Suffering or once suffered from other infectious disease					–	1.000^c
Yes	3	4.9	3	6.0
No	58	95.1	47	94.0

One time of treatment is defined as having taken anti-TB drugs consecutively for more than 2 months.

Four controls missing.

Fisher's exact test.

FDCs, fixed-dose combinations.

To address possible cofounders in this study, all variables shown significantly associated with MDR-TB in the univariate analysis were included in the multivariate logistic regression method. It was found that family annual per-capita income ≤7,000 Yuan (OR=3.238; 95% CI: 1.270–8.252), no history of FDCs in anti-TB treatment (OR=4.027; 95% CI: 1.457–11.129), and adverse reactions in the course of TB treatment (OR=3.568; 95% CI: 1.402–9.085) were independent predictors of MDR-TB (Table 3).

Table 3.

Multivariate Logistic Regression Analysis for Independent Risk Factors for MDR-TB Among Previously Treated Patients in Hangzhou, China, May 2010 Through July 2011

Variables	Crude OR (95% CI)	Adjusted OR (95% CI) ^a
Family annual per-capita income (Yuan)
≤7,000	2.609 (1.204–5.652)	3.238 (1.270–8.252)
>7,000	1	1
History of FDCs in anti-TB treatment
Yes	1	1
No	3.964 (1.776–8.850)	4.027 (1.457–11.129)
Adverse reactions in the course of TB treatment
Yes	3.151 (1.446–6.866)	3.568 (1.402–9.085)
No or unknown	1	1

Adjusted for all other factors included in the multivariate model.

OR, odds ratio; CI, confidence interval.

Discussion

History of anti-tuberculosis treatment was the most widely reported risk factor for MDR-TB.^{2,7,9,13,19,20,26} Our study was designed for previously treated MDR-TB and non-MDR-TB patients, thus disentangling the predominant risk factor—history of anti-tuberculosis treatment for better evaluation of other potential predictors of MDR-TB. As for the risk factors for MDR-TB among cases without a history of TB treatment in China, future further studies are needed to clarify them.

The present study showed that a lower family annual per-capita income, having no history of FDCs in anti-TB treatment and adverse reactions in the course of TB treatment were independent predictors of MDR-TB. This is in agreement with observations in many studies that the impact of poverty and crowded housing status were associated with the development of MDR-TB.^10,16,17,22 In the present study, smaller family per-capita housing area was also found significantly associated with MDR-TB in univariate analysis (Table 2), although this factor failed to emerge as a significant independent predictor of MDR-TB in multivariate logistic regression analysis. This might be attributable to a Type II error and/or to the intercorrelation between the various predictor variables. It is theoretically possible that the patients of poorer economic status may have limited access to medical treatment and because of lack of money they were less willing to seek medical attention when they had suspicious symptoms, causing delayed diagnosis. This is in line with findings in some studies that delayed diagnosis is a risk factor for MDR-TB.¹⁰ These findings suggest that the prevalence of MDR-TB among TB patients may be poverty related, therefore, these lower socioeconomic groups should be the highest priority for MDR-TB prevention efforts.

In the present study, no history of FDCs in anti-TB treatment was significantly associated with MDR-TB. Effective treatment of TB patients with short-course multidrug chemotherapy is the cornerstone of the modern approach to the control of the disease. To emphasize this principle, WHO and the IUATLD recommend the use of FDC formulations of the essential anti-tuberculosis drugs as one further step to ensure adequate treatment of patients and prevent further emergence of drug-resistant TB.^24,35,36 Some studies suggest that FDCs simplify the prescription of drugs and the management of drug supply and may also limit the risk of MDR-TB arising as a result of inappropriate drug selection and monotherapy.^1,15,36 FDCs can also reduce the number of tablets to be taken to make treatment easier, and minimize the probability of splitting the doses or of taking only some of the drugs in the regimen, thus improving the patients' compliance with treatment.³ Similar results have been observed in our study. This highlights the importance of expanding use of FDCs in the control of MDR-TB.

In the present study, adverse reaction in the course of TB treatment was found to be significantly associated with MDR-TB in the multivariate model. Furthermore, among the TB patients who had adverse reaction, quitting the treatment was shown as a risk factor for MDR-TB. This finding is perhaps due to the fact that adverse reaction is a main reason for quitting TB treatment, that is to say, poor compliance to treatment is a direct cause for the development of MDR-TB. A study performed by Liang et al. in China reported that the side effects of TB treatment were perceived by the interviewees as influencing factors of MDR-TB,¹² and many studies have reported that poor compliance to treatment increased the chances of development of MDR-TB.^11,18,21 Similar results have been observed in the present study. These findings emphasize the important role of strong supervision of treatment by service providers and the greater need to provide support to patients taking treatment to maximize compliance to TB treatment guidelines. In 1995, the Directly Observed Therapy Short Course (DOTS) strategy for TB control was recommended by WHO and is hailed as one of the most cost-effective health interventions to date. In some areas, however, its success is threatened by the rise of MDR-TB. To address the problem of MDR-TB, WHO in collaboration with its international partners has recommended the DOTS-Plus strategy for MDR-TB, which not only grants access to high-quality, low price second-line anti-TB drugs, but can also maximize compliance to TB treatment guidelines for MDR-TB patients.²⁸ Therefore, DOTS-Plus strategy should be thought the most effective approach to improve adherence to anti-MDR-TB treatment.

Our study has some limitations. First, to reduce recall bias, the cases and controls were only selected from retreated TB patients who recently visited the hospital. In addition, since this was a hospital-based case–control study, patients who did not show up in the hospital might be different from participants in our study, which could cause selection bias. Second, we could not identify enough number of controls who were both sensitive to isoniazid and rifampin because many retreated patients were resistant to either isoniazid or rifampin. So, the sample size was relatively small and we may have low power to detect some potential risk factors. Third, since DST had not been carried out extensively when the patients were treated initially, we could not have information pertaining to the initial drug susceptibility of the previously treated MDR-TB cases. Fourth, although previous treatment is the strongest risk factor of MDR-TB, other risk factors such HIV coinfection have also been reported.^4,23 However, in the present study, all enrolled patients were HIV negative, a better-designed research should be done on a sample of relevant patients that includes and distinguishes between HIV-positive and HIV-negative patients. Despite these, the results of this study contribute to a growing literature identifying possible risk factors for MDR-TB among Chinese patients.

Conclusion

Apart from the well-described risk factors of previous anti-tuberculosis treatment, lower economic status, no history of FDCs in anti-TB treatment, and poor adherence to treatment can be used to predict the development of MDR-TB. In the control of MDR-TB, lower socioeconomic groups, expanding use of FDCs, and improvement of adherence to treatment by implementing DOTS-Plus strictly should become priorities that require strong commitment and collaboration among health organizations.

Footnotes

Acknowledgments

The authors thank the project staff in the Chinese Center for Disease Control and Prevention, Zhejiang Provincial Center for Disease Control and Prevention, Zhejiang Red Cross Hospital, and Shandong University for implementing this project. The study was supported by the Zhejiang Multidrug-Resistant Tuberculosis Prevention Dots-Plus Key Technology Project (WKJ2008-2-018) and the National Science and Technology Major Project (2008ZX10003-007).

Authors' Contributions

K.W. collected the data, conducted the statistical analysis, and drafted the article. S.C. helped with the study design, field work, and interpretation of the results. X.W. and J.Z. helped to interpret the results and modify the manuscript. X.W. and P.H. collected data and helped to write and modify the article. L.W., S.J., J.L., and H.Y. helped with the study design and interpretation of the results. Y.P. and L.W. administered study field and collected the data. W.M. conceived the project concept, assisted with the data interpretation, and helped to write the article. All the authors read and approved the final article.

Disclosure Statement

The authors declare that they have no competing interests.

References

Aguilar

, Garay

, Villatoro

, Ramirez

, Villatoro

, et al. 2005. Results of a national study on anti-mycobacterial drug resistance in El Salvador. Int. J. Tuberc. Lung Dis., 9:514–520.

Antunes

M.L.

, Aleixo-Dias

, Antunes

A.F.

, Pereira

M.F.

, Raymundo

, et al. 2000. Anti-tuberculosis drug resistance in Portugal. Int. J. Tuberc. Lung Dis., 4:223–231.

Blomberg

, Spinaci

, Fourie

, and Laing

2001. The rationale for recommending fixed-dose combination tablets for treatment of tuberculosis. Bull. World Health Organ., 79:61–68.

Caminero

J.A.

2010. Multidrug-resistant tuberculosis: epidemiology, risk factors and cases finding. Int. J. Tuberc. Lung Dis., 14:382–390.

Crofton

, Chaulet

, and Maher

. 1997. Guidelines for the management of drug-resistant tuberculosis. Available at http://whqlibdoc.who.int/hq/1997/WHO_TB_96.210_(Rev.1).pdf (Accessed 31 July 2013 ).

Devaux

, Manissero

, Fernandez de la Hoz

, Kremer

, van Soolinqen

, et al. 2010. Surveillance of extensively drug-resistant tuberculosis in Europe, 2003–2007. Euro Surveill. 15:pii:19518.

Drobniewski

, Eltringham

, Graham

, Magee

J.G.

, Smith

E.G.

, et al. 2002. A national study of clinical and laboratory factors affecting the survival of patients with multiple drug resistant tuberculosis in the UK. Thorax, 57:810–816.

Espinal

M.A.

, Laszlo

, Simonsen

, Boulahbal

, Kim

S.J.

, et al. 2001. Global Trends in Resistance to Antituberculosis Drugs. N. Engl. J. Med., 344:1294–1303.

Faustini

, Hall

A.J.

, and Perucci

C.A.

2006. Risk factors for multidrug resistant tuberculosis in Europe: a systematic review. Thorax, 61:158–163.

10.

G.X.

, Wang

H.Y.

, Borgdorff

M.W.

, van Soolingen

, van der Werf

M.J.

, et al. 2011. Multidrug-resistant tuberculosis, People's Republic of China, 2007–2009. Emerg. Infect. Dis., 17:1831–1838.

11.

Karabay

, Otkun

, Akata

, Karlikaya

, Tuğrul

, et al. 2004. Anti-tuberculosis drug resistance and associated risk factors in the European section of Turkey. Indian J. Chest Allied Sci., 46:171–177.

12.

Liang

, Wu

, Gao

, Hao

, Liu

, et al. 2012. Factors contributing to the high prevalence of multidrug-resistant tuberculosis: a study from China. Thorax, 67:632–640.

13.

Lomtadze

, Aspindzelashvili

, Janjgava

, Mirtskhulava

, Wright

, et al. 2009. Prevalence and risk factors for multidrug-resistant tuberculosis in republic of Georgia: a population based study. Int. J. Tuberc. Lung Dis., 13:68–73.

14.

Ministry of Health of the People's Republic of China. Nationwide anti-tuberculosis drug resistant baseline surveillance in China (2007–2008) [in Chinese]. Beijing: People's Public Health Press, 2010.

15.

Moulding

T.S.

, Le

H.Q.

, Rikleen

, and Davidson

2004. Preventing drug-resistant tuberculosis with a fixed dose combination of isoniazid and rifampin. Int. J. Tuberc. Lung Dis., 8:743–748.

16.

Nair

D.M.

, George

, and Chacko

K.T.

1997. Tuberculosis in Bombay: new insights from poor urban patients. Health Policy Plan, 12:77–85.

17.

Portero

N.J.

, Rubio

Y.M.

, and Pasicatan

M.A.

2002. Socio-economic determinants of knowledge and attitudes about tuberculosis among the general population of Metro Manila, Philippines. Int. J. Tuberc. Lung Dis., 6:301–306.

18.

Pritchard

A.J.

, Hayward

A.C.

, Monk

P.N.

, Neal

K.R.

2003. Risk factors for drug resistant tuberculosis in Leicestershire—poor adherence to treatment remains an important cause of resistance. Epidemiol. Infect., 130:481–483.

19.

Ruddy

, Balabanova

, Graham

, Fedorin

, Malomanova

, et al. 2005. Rates of drug resistance and risk factor analysis in civilian and prison patients with tuberculosis in Samara Region, Russia. Thorax, 60:130–135.

20.

Shamaei

, Marjani

, Chitsaz

, Kazempour

, Esmaeili

, et al. 2009. First-line anti-tuberculosis drug resistance patterns and trends at the national TB referral center in Iran—eight years of surveillance. Int. J. Infect. Dis., 13:e236–e240.

21.

Sharma

S.K.

, Turaga

K.K.

, Balamurugan

, Saha

P.K.

, Pandey

R.M.

, et al. 2003. Clinical and genetic risk factors for the development of multidrug resistant tuberculosis in non-HIV infected patients at a tertiary care center in India: a case-control study. Infect. Genet. Evol., 3:183–188.

22.

Snider

D.E.

, and Castro

K.G.

1998. The global threat of drug-resistant tuberculosis. [Editorial]. N. Engl. J. Med., 338:1689–1690.

23.

Suchindran

, Brouwer

E.S.

, and Van Rie

. 2009. Is HIV infection a risk factor for multi-drug resistant tuberculosis?. A system review. PLoS One, 4:e5561.

24.

The promise and the reality of fixed-dose combinations with rifampicin: A joint statement of the International Union Against Tuberculosis and Lung Diseases and the Tuberculosis Programme of the World Health Organization. 1994. Int. J. Tuberc. Lung Dis., 75:180–181.

25.

van Soolingen

, Kremer

2009. Findings and ongoing research in the molecular epidemiology of tuberculosis [in Japanese]. Kekkaku, 84:83–92.

26.

Wang

, Fu

, Li

, Chen

, Liu

, et al. 2012. Drug-resistant tuberculosis in Zhejiang province, China, 1999–2008. Emerg. Infect. Dis., 18:496–498.

27.

WHO. Anti-tuberculosis drug resistance in the world: fourth global report. The WHO/IUATLD Global Project on Anti-tuberculosis Drug Resistance Surveillance; WHO/HTM/TB/2008.394. Available at www.who.int/tb/publications/2008/drs_report4_26feb08.pdf (Accessed 31 July 2013 ).

28.

WHO. DOTS-Plus: preliminary results and emerging issues: proceedings of the meeting of the Stop TB Working Group on DOTS-Plus for MDR-TB, Tallinn, Estonia, 10–12 April 2002. WHO/CDS/TB/2002.307. Available at http://apps.who.int/iris/bitstream/10665/67452/1/WHO_CDS_TB_2002.307.pdf (Accessed 31 July 2013 ).

29.

WHO. Global tuberculosis control: epidemiology, strategy, financing: WHO report 2009; WHO/HTM/TB/2009.411. Available at http://whqlibdoc.who.int/publications/2009/9789241563802_eng_doc.pdf (Accessed 31 July 2013 ).

30.

WHO. Global tuberculosis control-surveillance, planning, financing: WHO Report 2005; WHO/HTM/TB/2005.349. Available at http://library.cphs.chula.ac.th/Ebooks/Annual%20Report/WHO%20Report/WHO%20REPORT%202005.pdf (Accessed 31 July 2013 ).

31.

WHO. Multidrug and extensively drug-resistant TB (M/XDR-TB): 2010 Global Report on Surveillance and Response; WHO/HTM/TB/2010.3. Available at http://whqlibdoc.who.int/publications/2010/9789241599191_eng.pdf (Accessed 31 July 2013 ).

32.

WHO. Report of the meeting of the WHO Global Task Force on XDR-TB: Geneva, Switzerland, 9–10 October 2006; WHO/HTM/TB/2007.375. Available at www.who.int/tb/xdr/globaltaskforcereport_oct06.pdf (Accessed 31 July 2013 ).

33.

WHO. The global MDR-TB & XDR-TB response plan, 2007–2008; WHO/HTM/TB/2007.387. Available at http://libdoc.who.int/hq/2007/WHO_HTM_TB_2007.387_eng.pdf (Accessed 31 July 2013 ).

34.

WHO. Treatment of tuberculosis: guidelines for national programmes. Vol. 3rd ed. World Health Organization; Geneva: 2003. WHO/CDS/TB/2003.313. Available at http://whqlibdoc.who.int/hq/2003/who_cds_tb_2003.313_eng.pdf (Accessed 31 July 2013 ).

35.

WHO. Treatment of tuberculosis: guidelines for National Programmes. Vol. 2nd ed. Geneva: WHO, 1997.

36.

WHO. WHO Technical Report Series: guidelines for registration of fixed-dose combination medicinal products; 2005. Available at http://apps.who.int/prequal/info_general/documents/TRS929/WHO_TRS_929_annex5FDCs.pdf (Accessed 31 July 2013 ).

Factors Contributing to the High Prevalence of Multidrug-Resistant Tuberculosis Among Previously Treated Patients: A Case–Control Study from China