Risk Factor Identification of Multi-Drug Resistant Tuberculosis

Background: Multi-drug resistant tuberculosis (MDR-TB) is a serious health problem with high morbidity and mortality. The identification of risk factors is needed to prevent the incidence TB-MDR. Objective: To identify the risk factors of TB-MDR in the Special Region of Yogyakarta. Method: This study used a case control design to assess the risk factors for the incidence of MDR-TB. Patients diagnosed with MDR-TB compared to non-MDR TB patients were then compared to exposure to the risk factors studied. The results of the study were analyzed by calculating the mean and proportion. Next, Chi-square tests and analysis of odds ratio (OR) with 95% confidence interval (CI) were conducted using a 2x2 dummy table as an aid with multivariate logistic regression analysis. Result: There were 53 patients with MDR-TB as the case group and 106 non-MDR TB patients as controls. There were no significant differences between the two groups regarding the characteristics of age, sex, place of residence, and type of work (p>0.05). The multivariate analysis found the risk factors associated with an increased incidence of MDR-TB were a history of previous TB treatment (p<0.001; OR 31.82; 95%CI: 9.40-107.77) and surrounding TB (p<0.01; OR 4.45; 95%CI: 1.45-13.70), while other factors that were not significantly related to the incidence of MDR-TB included body mass index, distance of home to health facilities where taking medicine, smoking, drinking alcohol, history of BCG vaccination, education, income, medication adherence, and comorbidities (p> 0.05). Conclusion: The risk factors found to be associated with an increased incidence of MDR-TB in this study were the presence of surrounding TB patients and a history of previous TB treatment.


BACKGROUND
It is estimated that the number of patients with TB in Indonesia is around 10% of the total number of TB patients in the world. At the same time, the dual immunity of TB germs to anti-TB drugs is increasingly a problem due to cases that have not been successfully cured 1  MDR-TB is caused by bacteria that is resistant to isoniazid and rifampicin which are the most effective anti-TB drugs, both as a result of primary infection and due to treatment in previous patients with TB. One of the causes of immunity (secondary resistance) is the result of mismanagement of patients with TB. Drug resistance is associated with a history of previous treatment. The possibility of resistance in patients with a history of previous treatment is as much as 4-fold, whereas for the occurrence of MDR-TB, it increases by 10 times or more compared to patients who have never been treated 1 .
Theoretically. there are five factors that are considered to play a main role in causing the 'outbreak' of MDR-TB, namely: (1) Inadequate treatment (causing resistant Mycobacterium tuberculosis mutants); (2) Patients who are slowly diagnosed with MDR, thus becoming a source of continuous transmission; (3) Patients with incurable MDR-TB which will continue transmission; (4) Patients with MDR TB despite being treated continuously but with inadequate drugs resulting in doubling of resistant mutants; and (5) HIV co-infection which facilitates primary and secondary resistance.
Research in several countries identified a variety of risk factors for MDR-TB, including previous TB treatment, poor adherence to treatment, inadequate regimens and positive BTA examinations at the end of the second or third month of treatment. Patient characteristics such as HIV coinfection, alcohol abuse, and young age are also believed to influence the occurrence of drug resistance even though information about it is still limited and controversial. The World Health Organization (WHO) includes age as a TB risk factor 3 .
MDR-TB is one of the biggest problems with TB prevention and eradication with an incidence that increases significantly every year. Therefore, certain risk factors that increase the incidence of MDR-TB need to be investigated in order to reduce the cost, morbidity and mortality of MDR-TB. This study aimed to identify risk factors for MDR-TB associated with the incidence of MDR-TB in the Special Region of Yogyakarta.

METHODS
This study used a case control design with MDR-TB patients as a case group and patients with non-MDR TB as a control group. The research subjects were patients who were over 18 years old, diagnosed with MDR-TB from January 2014 to December 2016 and patients living in the Special Region of Yogyakarta, while the control group were non-MDR TB patients who lived in one area with MDR-TB patients.
The inclusion criteria were: patients with MDR-TB and non-MDR TB patients, residing in the Yogyakarta Special Region, willing to be the subjects of research. The exclusion criteria were: patients living outside Yogyakarta Special Region and where the patient's residence was not found due to relocation. The study sample was selected using a total sampling method with 56 patients as the case group and the control group was twice the number of the case group.
TB researchers and/or programmers conducted the data retrieval by looking at data on TB 01 and questionnaires given to the research subjects. Anamnesis was done with a questionnaire on the research subjects' characteristics related to age, sex, height and weight, education, income, employment, residence, smoking, drinking alcohol, surrounding TB patients, BCG vaccination, history of previous TB treatment, adherence to taking medication, and comorbidities. Data were analyzed using the SPSS computer program.
The researcher guarantees the confidentiality of the data concerning the subjects of the study. Ethical clearance was issued by Medical and Health Research Ethics Committee (MHREC) from the Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada.

RESULTS
There were 56 cases of MDR-TB in the Special Region of Yogyakarta between January 2014 to December 2016, but only 53 cases were included in the study because 2 patients refused to take the study and 1 patient did not meet the inclusion criteria because he resided outside the Special Region of Yogyakarta. The ten criteria of suspected MDR-TB (Ministry of Health Republic of Indonesia, 2014) are: 1) TB patients failed treatment category; 2) Non-converted category 2 treatment TB patients (remain positive) after 3 months of treatment; 3) TB patients who have a history of non-standard TB treatment and use quinolones and secondline injection drugs for at least 1 month; 4) Patients with failed category 1 TB treatment; 5) TB patient who remains positive after 3 months of treatment category 1; 6) Relapse case of TB patients with categories treatment 1 or 2 ; 7) TB patients who return after loss to follow-up (negligent treatment/default); 8) Suspected TB who has a history of close contact with MDR TB patients; 9) TB-HIV coinfected patients who do not respond to drug administration; 10) Patients other than the above criteria (TB smear + new or TB Child). This picture below shows the numbers of each category of suspected MDR-TB. The most common criteria for suspected MDR-TB in the case group were MDR-TB caused by relapses of category 1 or 2. There were no suspicious criteria for MDR-TB in TB-HIV co-infection patients who did not respond to drug administration.
The characteristics of patients with MDR-TB and the non-MDR TB patients did not show significant differences.
Risk factors that show significant differences between patients with MDR-TB and non-MDR TB patients are education, income, medication adherence, drinking alcohol, presence of comorbidities, history of TB treatment, and surrounding TB.
Education, income, obedience in taking medicine, smoking, drinking alcohol, comorbidities, TB treatment history, and surrounding TB have a relationship with TB-MDR incidence.
The multivariate logistic regression analysis identified a history of TB treatment and the presence of surrounding TB patients were the significant risk factors for MDR-TB.

DISCUSSION
Age as a risk factor in this study was not related to the incidence of MDR-TB. The average age of MDR-TB patients is in the productive age of 42 y.o. Another study with similar results found that the prevalence of MDR-TB was more common at ages 36-64 y.o 4 . Some studies have shown that age factors have an independent relationship with MDR events and an increase in their proportions at  The number of male patients with MDR-TB was more than women even though it is not related to the incidence of MDR-TB. The same results were found in patients with MDR-TB in Europe. MDR-TB was more likely in men even though male sex was not a risk factor for MDR-TB 6 . In a study in Ethiopia, male was a risk factor for MDR-TB, while the same study in Nigeria showed the same results and found most treatment failures occurred in men. This is hypothesized because women are more disciplined and regularly take medicine 7 .
Low education is significantly associated with the incidence of MDR-TB. This result is the same as research conducted in Bangladesh 8 . Several results from other studies that analyze risk factors for MDR-TB are consistent with these results.
In this study, the income below the UMP was associated with the incidence of MDR-TB. Some studies also showed that family opinion is low and the absence of health insurance is associated with MDR-TB. Patients with low income may have limited access to health services and health costs 9 . In this study the work was not related to the incidence of MDR-TB. Gomes' research et al. showed the same results. Patients with low social status/income or low education were not significantly associated with the incidence of MDR-TB 10 .
Drug disobedience in this study was a risk factor associated with the incidence of MDR-TB. In a meta-analysis, the poor adherence to medication is a risk factor for MDR-TB in China 11 . Giving inadequate TB therapy will cause resistant mutants. This is what causes resistance to firstline OAT. Factors that cause inadequate TB therapy include long-term treatment of more than 6 months so that patients get bored, side effects of drugs so that treatment is not finished, provided the wrong medicine, improper dosage of drugs and/or procurement of medicines was cut off 12 .
Smoking was not a significant risk factor for MDR-TB in this study. Different results from other studies have shown a significant relationship between smoking and MDR-TB 8 . The number of patients with MDR-TB who have a history of smoking in this study is large, which is 59%. These results are similar to studies conducted in Georgia that found there were more cases of patients with MDR-TB who were smokers. Patients with MDR-TB who are smokers and properly treated for MDR-TB have an outcome risk 3 times worse than non-smokers 13 .

Table 3. The Results of Multivariate Logistic Regression Analysis of Risk Factors for MDR-TB
The consumption of alcohol in this study was not a significant risk factor with the incidence of MDR-TB.
Other studies provide different results, indicating alcohol as a risk factor for MDR-TB 14 .
Diabetes mellitus (DM) in this study was not significantly associated with the incidence of MDR-TB. The same results were shown in a study in Malaysia where DM was not significantly associated as a risk factor for MDR-TB 15 .
The results of this study are different from other studies which mention DM as an independent risk factor associated with the incidence of MDR-TB 10 . while it is 13% in patients with treatment. The prevalence is 10 times higher in cases of resistance of patients who have received prior treatment 3 .
The risk of drug resistance is increased in cases with a history of treatment. Accordingly, the frequency of MDR-TB increases in the risk group of patients with TB, namely chronic TB patients, TB treatment failure, TB relapse, and TB patients with a history of dropout / default. In addition to the group of patients with a history of treatment, resistance was also suspected in patients with TB who were not converted after initial/intensive treatment. and TB patients who had close contact with patients with MDR-TB. As a result, it is important in practice to identify whether there is a history of previous treatment and monitor treatment response to identify the presence or absence of conversion after initial/intensive treatment or after insertion 16 .
The BCG vaccine history was not significantly associated with the incidence of MDR-TB. Some studies provide the same results, indicating the BCG scale is not related to the incidence of MDR-TB 8 . BCG is effective in preventing miliary TB, severe pulmonary TB, and TB meningitis in children, but not for pulmonary TB in adults, especially in developing countries. This is due to the protective effect of the BCG vaccine given when the baby survives up to 10 years, but it is not known whether there is still a protective effect after this period 17 .
In this study a history of contact with TB patients was significantly associated with the incidence of MDR-TB. This study found the same results of several studies indicating contact history with MDR-TB patients as a risk factor that is significantly associated with the incidence of MDR-TB. The relationship of contact history with patients with MDR-TB and MDR-TB strains was caused by patients who developed primary resistant to drugs 18 . Other studies have identified significant differences in the history of close contact in the MDR-TB group and while the history in the non-MDR TB group was not significantly different 19,20 .

CONCLUSIONS
A history of previous TB treatment and the presence of surrounding TB patients are risk factors that are significantly associated with the incidence of MDR-TB.

Suggestions
Suggestions in this study are (1) Carrying out the slogan "Find Obedience to Heal" in all patients with TB to prevent transmission and resistance to OAT by increasing the discovery of TB patients with active selective case finding in the community by activating the role of health cadres; (2) Perform early detection of TB suspects by conducting microscopic examination of sputum in patients with cough for more than 2 weeks; (3) Tighter monitoring of TB treatment by optimizing the role of the Drugs Supervisor, especially in TB patients with DM and HIV/ AIDS; (4) Increasing Public Private Mix (PPM) starting with Hospital DOTS Linkage (HDL) activities involving hospitals. Based on consideration of the great potential in TB control programs, the aim of the PPM is to improve the quality of treatment for patients with TB in preventing MDR-TB cases, increasing access and even distribution of TB services, and reducing the cost burden for TB patients in Indonesia.