Browsing by Subject "India"
Now showing 1 - 10 of 10
- Results Per Page
- Sort Options
Item Assessing Water Fluoride Levels in Rural Rajasthan, India(2014-02-04) Mehta, Kajal; Sreeramoju, PranaviINTRODUCTION: Skeletal and dental fluorosis are crippling diseases associated with consumption of water with excess fluoride. The diseases manifest with symptoms such as severe staining, pitting, enamel damage and cracking of teeth in dental fluorosis to stiffness, joint pain, crippling, kyphosis, invalidism, and GI complications in skeletal fluorosis. Fluorosis is an endemic public health issue in many developing regions around the world, including areas of India. A study conducted in Rajasthan, India shows the prevalence of dental fluorosis to be 70.6-100% in village children and 68-100% in village adults. Prevalence of skeletal fluorosis was 5.2% among children and ranged from 7.4% to 37.7% in adults. The present study hopes to address the underlying problem of fluorosis in Rajasthan, India by analyzing water quality in the region. METHODS: This study is a retrospective observational study based on water data collected over a ten month period by the Jal Bhagirathi Foundation (JBF) in Jodhpur, India for a European Union funded water monitoring project. The water data are from the nearby Jalore, Barmer and Jaisalmer Districts, with water samples from all drinking water sources including handpumps, deep bore wells, tube wells, government run pipelines, ponds, lakes, and reverse osmosis water purification systems. The water samples were collected by trained JBF staff in sterile 1L containers, and the water quality testing was carried out by the on-site chemistry lab manager. Fluoride levels were tested within 24 hours of water sampling using a benchtop multiparameter meter and a fluoride electrode in 100 mL water. Additional tests for color, odor, turbidity, pH, alkalinity, total dissolved solutes, and presence of fecal bacteria were performed on each sample. RESULTS/OUTCOMES: The data were analyzed for frequency of fluoride levels above the recommended level of 1 parts per million (ppm). Of the water samples tested for fluoride concentration, 57 of 156 samples (36.5%) showed concentrations above 1ppm. The mean fluoride concentration of all tested water samples was .81ppm, with a standard deviation of .73ppm. CONCLUSION: The results demonstrate over one-third of the sampled drinking water is potentially toxic to the health of its consumers. This incidence is particularly alarming in the arid conditions of Rajasthan where rural inhabitants have limited water resources, leading them to rely heavily on these harmful supplies. These results provide an impetus to approach the problem by increasing community education of unsafe drinking water sources and methods of purification, including home systems and community-wide solutions like reserve osmosis plants.Item Association of Basic Maternal Factors and Maternal Autonomy with Malnourishment among Children 6-36 Months Old in an Urban Slum in Bangalore, India(2016-04-29) Sudanagunta, Sindhu; Baldridge, Courtney; Reisch, Joan; Niwagaba, LillianBACKGROUND: Child malnutrition is a significant problem in many developing countries with much of the burden falling on economically and historically underprivileged societies. Chronic undernutrition leads to decreased height for age (stunting) and contributes to reduced intellectual capacity, morbidity, and mortality. Acute undernutrition results in decreased weight for height (wasting) and can decrease a child's reserve for fighting illnesses and thriving. Recent attention on psychosocial dynamics has encouraged researchers to focus on maternal factors and its effect on child malnourishment. OBJECTIVE: Maternal autonomy - the freedom and ability to control the resources available to the mother to care for her child - are likely important factors influencing child health outcomes, especially in an economically disadvantaged, culturally segregated slum society. METHODS: To examine this relationship in an urban slum in Karnataka, India, a cross-sectional observational study was conducted on 199 mother-child dyads. Anthropometric measurements were collected from children between 6-36 months of age. A 34 question survey uncovering issues of basic maternal factors, decision-making abilities, and domestic violence was administered to the mothers. RESULTS: Stunting was present in 32% of children (7.5% were severely stunted) and wasting was present in 23% (6.5% were severely wasted). Logistic regression models were used to test association between maternal factors and risk of having stunted or wasted children. Number of people in the household and mother's education were significantly associated with stunting. This breakdown shows that both maternal education and more members in the household were independently protective against stunting. The only maternal autonomy factor significantly associated with stunting and wasting was her ability to decide on major household purchases. The correlation, however, is inverse between stunting and wasting. Mothers who were more autonomous in this scenario were more likely to have children who were wasted, but less likely to have children who were stunted. CONCLUSION: In the psychosocial context of an urban slum in Karnataka, India, maternal autonomy poses an interesting inverse relationship between stunting and wasting suggesting the need for more research on this topic.Item Comparison of Pre-Transplant Criteria and Outcomes for Living Donor Kidney Transplant Programs in India and the United States(2015-01-26) Bansal, Sukriti; Raja, Hari; Rajora, Nilum; Kher, VijayBACKGROUND: One of the greatest obstacles to treatment of end stage renal disease globally is organ donor shortage. While some nations (i.e. the US), have primarily cadaveric organ donors, developing nations rely heavily on living donors. This project is a comparison of two kidney transplants programs -- one in the US & one in India -- looking at the pre-transplant criteria of each & assessing the patient outcomes. METHODS: This is a cohort study of living donor kidney transplant patients from St. Paul University Hospital in Dallas, TX & kidney transplant patients from Medanta the Medicity in Gurgaon, India. Data for India was collected from a database of all patients who underwent a kidney transplant at Medanta, selected for patients who fit the following criteria: one cohort of patients had been transplanted the previous month (N=29), one cohort had been transplanted one year prior the date of the study (N=29), & one cohort had been transplanted 3 years prior (N=13). Information from the database was used to calculate patient & graft survival rates for the relevant time periods. Data for St. Paul was obtained from the Scientific Registry of Transplant Recipients, which already had the calculated 1 month, 1 year, & 3 year patient and graft survival rates. Information on pre-transplant criteria was obtained from the transplant teams at each respective institution. RESULTS: The majority of medical pre-operative criteria between the two programs are identical. One significant difference is ABO-incompatible transplants are performed at Medanta, while at St. Paul ABO-incompatible donor/recipient pairs are referred for paired donation. Medanta requires all living donors to be related, while St. Paul will accept unrelated donors. The patient survival rates for St. Paul are 100% (1 mo, N=32), 95.23% (1 yr, N=32) and 85.71 % (3 yr, N=21). Graft survival rates are 100% (1 mo, N=32), 95.24% (1 yr, N=32), and 81.82% (3 yr, N=22). The patient survival rates for Medanta are 100% (1 mo., N=29), 100% (1 yr, N=29), and 100% (3 yr, N=13). Graft survival rates are 100% (1 mo. N=29), 100% (1 yr, N=29), and 100% (3 yr, N=13). For all patients transplanted at Medanta, the overall patient survival rate was 98.40% (N=874) and the overall graft survival rate was 98.51% (N=874). Corresponding data wasn't available for St. Paul. CONCLUSION: While it appears that the 3 year survival rates are better for Medanta than for St. Paul, there is a limitation on making conclusions because this data does not encompass the entire program at Medanta. Further study is needed to truly assess if there is a significant difference. The overall conclusion is that transplant programs in both settings have successful outcomes.Item Defining Practices, Outcomes, and Barriers to Gestational Diabetes Mellitus Screening at a Large, Urban Indian Community Hospital(2019-04-02) Balijepally, Ramya; Chang, Mary; Rajora, Nilum; McGarry, MaryBACKGROUND: Gestational diabetes mellitus (GDM) is defined as a glucose intolerance of varying severity with onset or first recognition during pregnancy. Uncontrolled GDM is linked to various pre-and postpartum complications and long term maternal health issues. Complications of pregnancy due to GDM include abortion, preterm labor, polyhydramnios, oligohydramnios, and fetal death. Complications to the fetus include fetal macrosomia, fetal malnutrition, defects of the neural tube, and cardiac anomalies such as ventricular septal defects and atrial septal defects. In 2010, the International Association of Diabetes and Pregnancy Study Groups (IADPSG) estimated that the prevalence of gestational diabetes mellitus (GDM) in Southeast Asia was 25%. Among the various populations globally, Indians have a higher frequency of GDM proving the need for an efficient screening process. This was an observational study conducted at large, urban community hospital in India. The purpose of this study was to define the patient population, GDM screening practices, and reasons why some women did not get screened for GDM. METHODS: This was an observational study conducted at a community hospital in south India. The timeframe of the study was from April 2017 to April 2018. The data was collected through patient interview and chart review. The patient's demographic information, risk factors for GDM, course during pregnancy, method of GDM screening used, mode of delivery, maternal, and neonatal outcomes were collected by a simple questionnaire. When accessible, data collected through patient interview was verified by the patient's chart. The inclusion criterion was post-partum women who delivered at Apollo General Hospital. The exclusion criterion for study participants was women with pre-GDM. FINDINGS: 55 interviews were conducted. The mean age at delivery was 24.5 years (n = 55). The mean gestational age that prenatal care was established was at 4.7 months (n = 54). Of the 42 of the 55 patients (76%) who were screened for GDM, 5 patients (9%) were diagnosed with GDM. All 42 patients were screened using the WHO 1999 criterion. Only 7 of the 42 patients were screened between 24 and 28 weeks. The mean gestational month that GDM screening occurred at the hospital was at 5.9 months (n = 40). Of the 13 patients who were not screened, 9 did not get screened because of physician recommendation. 4 were not screened because of lack of patient knowledge. INTERPRETATION: The incidence of GDM in this population was 9%. Although the screening method was standardized, the rate of screening was not universal (76%). Most women were not screened between 24 and 28 weeks, which is recommended by the American College of Obstetricians and Gynecologists (ACOG). Based on this study, future research should explore if there are any harmful consequences from not screening between 24 to 28 weeks in this population and consider educational outreach opportunities.Item Defining Screening Practices for Gestational Diabetes Mellitus at a Large, Urban Indian Community Hospital(2018-01-23) Iyengar, Meera; Balijepally, Ramya; Lingvay, IldikoBACKGROUND: The aim of this project was to outline screening practices for gestational diabetes mellitus (GDM) at Apollo General Hospital in Hyderabad, India to understand if GDM was being reliably diagnosed. Current screening guidelines for GDM advocate using an oral glucose challenge test in all high-risk women. Indian women have a 11-fold increased risk of developing GDM, demonstrating the need for a consistent and reliable screening practices. HYPOTHESIS: If Apollo General Hospital in Hyderabad, India is administering the gold standard screening method for GDM, then pregnant women coming to Apollo should receive an oral glucose tolerance test during their pregnancy. METHODS: A chart review of all deliveries at Apollo General Hospital from its opening in 2012 to 2016 was completed. Information from each chart regarding the hospitalメs screening practices for GDM including procedures, results, and follow-up practices was collected. RESULTS: GDM screening rates were high (92%), but not universal. Of the 208 patients reviewed, only 51% received a GTT or GCT sometime during pregnancy. An equal percentage of patients received either an RBS (36%) or GTT or GCT (36%) as their first screening test. The remaining patients received an FBS (21%) or PPBS (0.5%) as their first screening test. 8% were not screened for gestational diabetes mellitus during their pregnancy. 10 out of 208 reviewed patients were diagnosed with GDM, of whom 5 received either a GTT or GCT (45%). The prevalence of GDM at Apollo was 4.8% compared to global prevalence of 6%. CONCLUSIONS: Overall screening for GDM is high at Apollo General hospital. However, the screening practices vary considerably per patient, and only around 50% of women had received the gold standard test during their pregnancy. In addition, the prevalence of GDM at Apollo was lower than the global prevalence of 6%. This combined with the fact that Indian women have a 11-fold increased risk of developing GDM argue that RBS, FBS, PPBS or HbA1c may not be adequate tests to screen for GDM. Inconsistent screening practices prove the need for standardization and inquiry into patient and provider factors that influence the current variability in testing, as undiagnosed GDM can have serious consequences for both the mother and child.Item Improving Nutritional Counseling in Hyderabad, India: A Pilot Study(2019-03-29) Kotamraju, Swetha; Reed, W. Gary; Patterson, Abigail M.; Phelps, Mary E.BACKGROUND: If not addressed before the age of five, pediatric undernutrition can lead to irreversible long term health effects, including death. Mild to moderate pediatric undernutrition specifically contributes to 83% of malnutrition-related deaths in children. India has one of the world's largest prevalence rates for mild to moderate pediatric undernutrition. In India, improving nutritional counseling for caregivers of undernourished children is the most effective and sustainable way to reduce the prevalence of pediatric undernutrition. LOCAL PROBLEM: Staff at the Niloufer Hospital nutritional counseling center in Hyderabad, India were concerned that their counseling sessions were not aligned with international guidelines for these sessions and consistent between patients. It was unknown how reliable these sessions were. Therefore, the aim of this study was to increase the reliability of the evidence-based practice for nutritional counseling given at a nutritional center in Hyderabad, India by 25% by March 2017. Reliability was defined as how often counseling sessions followed 100% of international guidelines appropriate for a given patient. METHODS: The quality improvement methodologies of DMAIC (Define-Measure-Analyze-Improve-Control) and PDSA (Plan-Do-Study-Act) were used. The design and improvement of the interventions was carried out with PDSA cycles within the larger DMAIC methodology. During the define phase, a project charter was established and it was found that locally adapted WHO-UNICEF counseling guidelines were the best practice for this setting. During the measure phase, observation of counseling sessions showed that the sessions were given orally by nurses, without visual aids, and were inconsistent from patient to patient. A study to measure reliability and compliance of current practices and future intervention was created. Compliance is the percentage of topics covered out of the total topics listed in the age-appropriate guidelines for a given session. Guidelines were grouped into three age groups: under six months, six months to 35 months and three years to five years. Baseline reliability and compliance scores were collected in the analyze phase. During the improve phase, checklists and subsequently, flipcharts, were created as interventions to improve counseling using multiple PDSA cycles for each intervention. Reliability and compliance scores were collected for sessions using each intervention. INTERVENTIONS: The first intervention created was a checklist, a low-cost tool shown to improve long-term compliance with standardized medical processes. The content of the final checklists was adopted from the WHO-UNIFEF guidelines that served as a standard for this setting. A checklist was made for each of the three age groups. The second intervention created was a flipchart, which is a booklet with illustrations and corresponding talking points on opposite pages. It is the most common visual aid used in nutritional counseling worldwide. The final version of the flipcharts was created by using infographics adopted from a UNICEF nutritional counseling flipchart for India and supplemental images created by the local staff. The talking points of the flipchart were based on the checklists created earlier in the study. A flipchart was created for each of the three age groups. RESULTS: Counseling reliability did not increase with either intervention and was 0% for all age groups during all phases. However, counseling compliance increased with both. All results are reported for the age groups in the following order: under six months, six months to 35 months, and three years to five years. The average counseling compliance during the baseline phase were 20.6% (SD=4.1), 24.2% (SD= 8.2), and 28.9% (SD = 5.2). The average counseling compliance during the checklist phase were 56.8% (SD = 7.8), 57.8% (SD = 13.8), and 57.7% (SD = 10.5). The average counseling compliance during the flipchart phase were 64.6% (SD= 7.3), 57.8% (SD = 10.3), and 70.8% (SD = 7.8).The largest increase in average compliance was between the baseline and flipchart phase and was 44.0%, 33.9%, and 41.9%. A one-way ANOVA with post-hoc comparisons compared the effect of the interventions on compliance during each phase. For the youngest and oldest group, the mean compliance during the checklist and flipchart phase was found to be significantly different than for the baseline phase and from each other. For the middle group, the mean compliance during the checklist and flipchart phase was found to be significantly different than for the baseline phase but not from each other. CONCLUSION: This pilot study showed that both checklists and flipcharts can improve compliance in this setting though neither helped achieve the target reliability. Flipcharts were a more successful and advantageous intervention than the checklists for increasing compliance. Additionally, this study demonstrated that quality improvement framework used was an effective model to develop sustainable interventions in low resource settings. Further studies are needed to determine if these interventions can be improved and expanded to wider use.Item Improving Nutritional Knowledge of Caregivers in Hyderabad, India: A Pilot Study(2019-04-02) Tinger, Sophia Tibe; Patterson, Abigail M.; Reed, W. Gary; Phelps, EleanorBACKGROUND: India has the highest prevalence of underweight children under 5 in the world. While lack of access to food is one contributing factor to undernutrition, a caregiver's nutritional knowledge is a more important determinant. Caregivers can improve their nutritional knowledge through nutritional counselling, which is already one of the standard management options for treating children with mild to moderate malnutrition. Therefore, improving nutritional counselling for caregivers can be an effective way to combat undernutrition. OBJECTIVE: The aim of this study was to increase caregiver nutritional knowledge gained after receiving diet counseling at Niloufer Hospital, in Hyderabad, India, by 25% by March 2017 through the use of quality improvement tools. METHODS: Checklists and flipcharts were designed using PDSA cycles and DMAIC, two frequently used quality improvement tools. These interventions were then sequentially implemented in two-week increments over a period of six weeks. Caregivers of children 5 and under were given short surveys both before and after counselling. Change in knowledge scores were calculated from the difference in the pre- and post-counselling survey scores. This was done for three different age groups over each phase of the project: Baseline Phase, Checklist Phase, and Flipchart Phase. This data was then analyzed using a one-way ANOVA, as well as a post-hoc analysis with a student-Newman-Keuls test to confirm the ANOVA. RESULTS: The aim was achieved for all age groups when comparing the change in knowledge scores from the final phase (Flipchart Phase) to the Baseline Phase. For the Under 6 Months age group, the change in knowledge scores were 5.24, 10.17, and 12.20 for the Baseline, Checklist, and Flipchart phases respectively. The percent increase from the Baseline Phase to the Flipchart phase was 132.8%. For the 6 to 35 Months group, the change in knowledge score was 9.25 in the Baseline Phase, 8.14 in the Checklist Phase, and 18.86 in the Flipchart Phase, a 103.9 percent increase from the Baseline Phase. Finally, there was a change of knowledge score of 3.57 in the Baseline Phase, 6.25 in the Checklist Phase, and 29.17 in the Flipchart Phase for the 3-5 Years age group. This age group showed a 717.1 percent increase between the Baseline Phase and the final phase of the study. There were no statistically significant findings in this pilot study. However, there was a positive trend showing the flipcharts were the most effective intervention. CONCLUSION: This pilot study demonstrated that these simple-to-create and low-cost interventions, especially the flipcharts, can improve knowledge gained through counselling in low resource settings. Future studies are needed to determine if these interventions can be improved and expanded for wider use.Item Live Donor Renal Transplantation in India: Outcome and Comparison of Different Induction Therapies with a Focus on Gender Bias in Live Donor Renal Transplantation(2018-03-23) Khan, Maryam Idrees; Nwariaku, Fiemu; Rajora , Nilum; Tanriover , BekirBACKGROUND: As of 2014, an estimated 9% of the global population aged 18+ years was affected by diabetes. The World Health Organization (WHO) also estimated around 2.5% of deaths were attributed to diabetes in 2012 and more than 80% of those deaths occurred in low-middle income countries. It is apparent that diabetes and its complications are becoming a global issue as an increasing common, preventable, non-communicable disease. Along with cardiovascular disease, blindness, and neuropathy, end stage renal disease (ESRD) is one of the serious complications that can develop as a result of diabetes. Diabetes is the leading cause of ESRD in both developed countries like the United States and developing countries like India. India is a particularly interesting country to observe given their vast population base, rapid growing economy, genetic predisposition to diabetes and increased insulin resistance. It is estimated that 100,000 patients develop ESRD each year in India with diabetes as the main underlying cause (44% of all ESRD cases). Once a patient develops ESR, renal replacement therapy (RRT) is required to sustain life. RRT consists of three options: 1) hemodialysis (HD), 2) peritoneal dialysis (PD), or 3) renal transplant (RT). Of the three options, renal transplant is considered the best in terms of quality of life and cost effectiveness, but only about 5% of Indian patients with ESRD end up receiving RT. Most RT in India come from living donors rather than cadaveric donors like in the United States. Induction therapy with interleukin-2 receptor alpha chain (IL2-RA) is recommended as a first line agent in LRT however comparative outcomes of induction therapy remains controversial in Indian LRT population. OBJECTIVE: To evaluate patient survival and allograft function in LRT with a specific focus on the Indian population between 2010 and 2014 and to access the impact of different induction therapies on the outcomes of Indian LRT patients. METHODS: A single center (Medanta Medicity, Gurgaon, India) dataset was retrospectively studies for patients receiving LRT from 2010 to 2014 (N=901) to compare effectiveness of IL2-RA to other induction options (no-induction and rabbit anti-thymocyte globulin [r-ATG]). IL2-RA and no induction were chosen for immunologically low risk patients. R-ATG was primarily given to the recipient with PRA>20% and HLA mismatch >5 antigen out of 6. Patient paper charts were analyzed for dates not included in the Medanta database which included follow-up dates with corresponding creatinine levels (at 3 months, 6 months, 1 year, and last follow up), date and type of rejection if applicable, graft loss and death. Patients included in the data set had their last follow up at Medanta within the last 6 months from the time data was collected. The patient data was used to calculate rejection rate, graft failure, and hazard ratio (HR) for overall graft failure. The main outcomes were the risk of acute rejection at one-year and overall allograft failure (graft failure or death) post-transplantation through the end of follow-up. RESULTS: Similar Kaplan Meier curves for overall graft survivals were observed among induction categories. Rejection rate was higher in no-induction and IL2-RA groups (~25%) compared to r-ATG induction. On univariate Cox analysis, compared to no-induction therapy, overall allograft failure was similar among induction categories. Most of the rejections were borderline or Banff Type I acute cellular rejections. CONCLUSION: Compared to no-induction therapy, IL2-RA induction was not associated with better outcomes in Indian LRT recipients. R-ATG appears to be an acceptable and possibly preferred induction alternative for IL2-RA in high rejection risk Indian patients.Item Outcome of Different Induction Therapies in Living Donor Renal Transplant in Indian Population: A Single Center Experience(2016-01-19) Lowther, Megan; Khan, Maryam; Kher, Vijay; Parekh, Justin; Tanriover, Bekir; Rajora, NilumBACKGROUND AND OBJECTIVES: Induction therapy with interleukin-2 receptor antagonist (IL2-RA) is recommended as a first line agent in living donor renal transplantation (LRT). However, comparative outcomes of induction therapy remains controversial in Indian LRT population. DESIGN, SETTING, PARTICIPANTS, AND MEASUREMENTS: A single center (Medanta Medicity, Gurgaon, India) dataset was retrospectively studied for patients receiving LRT from 2010 to 2014 (N=901) to compare effectiveness of IL2-RA to other induction options (no-induction and rabbit anti-thymocyte globulin [r-ATG]). IL2-RA and no-induction were chosen for immunologically low risk patients. RATG was primarily given to the recipient with PRA>20% and HLA mismatch > 5 antigen out of 6. The main outcomes were the risk of acute rejection at one-year and overall allograft failure (graft failure or death) post-transplantation through the end of follow-up. RESULTS: Donor, recipient and transplant characteristics of three induction categories are shown in Table 1. Similar Kaplan Meier curves for overall graft survivals were observed among induction categories, shown Figure1. Rejection rate was higher in no-induction and IL2-RA groups (~25%) compared to r-ATG induction. On univariate Cox analysis, compared to no-induction therapy, overall allograft failure was similar among induction categories. Most of the rejections were borderline or Banff Type I acute cellular rejections. CONCLUSIONS: Compared to no-induction therapy, IL2-RA induction was not associated with better outcomes in Indian LRT recipients. r-ATG appears to be an acceptable and possibly the preferred induction alternative for IL2-RA in high rejection risk Indian patients.Item Patient Experience under India's Revised National Tuberculosis Control Programme in Pondicherry(2018-04-03) Bian, Louis; Sreeramoju, Pranavi; Perl, Trish M.; Barnes, ArtiWhile India's Revised National Tuberculosis Control Programme (RNTCP) has been a success on a statistical and national scale, awareness of the program and tuberculosis (TB) still remains low in many areas of India. In Puducherry (previously known as Pondicherry), a survey of 50 RNTCP patients reveals that only 40% had heard of TB before diagnosis, only 16% suspected that they had TB, and only 10% had heard of directly observed treatment short-course (DOTS). Women were more likely than men to have heard of TB (50% vs 35.3%, respectively) and DOTS (25% vs 2.9%, respectively), likely due to targeted TV advertisements during the day when men are out working. As expected, patients with more years of schooling were more likely to have heard of TB. Men were more likely to report missing doses, but women were more likely to report side effects. As income and years of schooling increased, the likelihood of seeking private treatment also increased likely due to less crowding and faster treatment in private settings. Most patients were happy with their experienced under RNTCP, major complaints included side effects of taking the pills, missing work in order to obtain pills, and facing the stigma of having TB. Reported compliance was high overall, but 44% of patients reported missing at least one dose during the course of their treatment. While RNTCP has come a long way in reducing the morbidity and mortality of TB, it still has a long way to go especially in population awareness and prevention of multi-drug resistant TB.