BackgroundWhile diabetes mellitus (DM) is a known risk factor for tuberculosis, the prevalence among TB patients in India is unknown. Routine screening of TB patients for DM may be an opportunity for its early diagnosis and improved management and might improve TB treatment outcomes. We conducted a cross-sectional survey of TB patients registered from June–July 2011 in the state of Kerala, India, to determine the prevalence of DM.Methodology/Principal FindingsA state-wide representative sample of TB patients in Kerala was interviewed and screened for DM using glycosylated hemoglobin (HbA1c); patients self-reporting a history of DM or those with HbA1c ≥6.5% were defined as diabetic. Among 552 TB patients screened, 243(44%) had DM – 128(23%) had previously known DM and 115(21%) were newly diagnosed - with higher prevalence among males and those aged >50years. The number needed to screen(NNS) to find one newly diagnosed case of DM was just four. Of 128 TB patients with previously known DM, 107(84%) had HbA1c ≥7% indicating poor glycemic control.Conclusions/SignificanceNearly half of TB patients in Kerala have DM, and approximately half of these patients were newly-diagnosed during this survey. Routine screening of TB patients for DM using HbA1c yielded a large number of DM cases and offered earlier management opportunities which may improve TB and DM outcomes. However, the most cost-effective ways of DM screening need to be established by futher operational research.

BackgroundDespite being a recognized standard of tuberculosis (TB) care internationally, mandatory TB case notification brings forth challenges from the private sector. Only three TB cases were notified in 2013 by private practitioners compared to 2000 TB cases notified yearly from the public sector in Alappuzha district. The study objective was to explore the knowledge, opinion and barriers regarding TB Notification among private practitioners offering TB services in Alappuzha, Kerala state, India.Methods & FindingsThis was a mixed-methods study with quantitative (survey) and qualitative components conducted between December 2013 and July 2014. The survey, using a structured questionnaire, among 169 private practitioners revealed that 88% were aware of mandatory notification. All patient-related details requested in the notification form (except government-issued identification number) were perceived to be important and easy to provide by more than 80% of practitioners. While more than 95% felt that notification should be mandatory, punitive action in case of failure to notify was considered unnecessary by almost two third. General practitioners (98%) were more likely to be aware of notification than specialists (84 %). (P<0.01). Qualitative purposive personal interviews (n=34) were carried out among private practitioners and public health providers. On thematic framework analysis of the responses, barriers to TB notification were grouped into three themes: ‘private provider misconceptions about notification’, ‘patient confidentiality, and stigma and discrimination ’and ‘lack of cohesion and coordination between public and private sector’. Private practitioners did not consider it necessary to notify TB cases treated with daily regimen.ConclusionCommunication strategies like training, timely dissemination of information of policy changes and one-to-one dialogue with private practitioners to dispel misconceptions may enhance TB notification. Trust building strategies like providing feedback about referred cases from private sector, health personnel visit or a liaison private doctor may ensure compliance to public health activities.

Rationale India reports the largest number of multidrug-resistant tuberculosis cases in the world; yet, no longitudinal study has assessed factors related to treatment outcomes under programmatic conditions in the public sector. Objectives To describe demographic, clinical, and risk characteristics associated with treatment outcomes for all patients with multidrug-resistant tuberculosis registered in the Revised National Tuberculosis Control Programme, Kerala State, India from January 1, 2009 to June 30, 2010. Methods Cox regression methods were used to calculate adjusted hazard ratios with 95% confidence intervals (CIs) to assess factors associated with an unsuccessful treatment outcome. Measurements and Main Results Of 179 patients with multidrug-resistant tuberculosis registered, 112 (63%) had successful treatment outcomes (77 bacteriologically cured, 35 treatment completed) and 67 (37%) had unsuccessful treatment outcomes (30 died, 26 defaulted, 9 failed treatment, 1 stopped treatment because of drug-related adverse events, and 1 developed extensively drug-resistant tuberculosis). The hazard for unsuccessful outcome was significantly higher among patients who consumed alcohol during treatment (adjusted hazard ratio, 4.3; 95% CI, 1.1–17.6) than those who did not. Persons who consumed alcohol during treatment, on average, missed 18 more intensive-phase doses (95% CI, 13–22) than those who did not. Although many patients had diabetes (33%), were ever smokers (39%), or had low body mass index (47%), these factors were not associated with outcome. Conclusion Overall treatment success was greater than global and national averages; however, outcomes among patients consuming alcohol remained poor. Integration of care for multidrug-resistant tuberculosis and alcoholism should be considered to improve treatment adherence and outcomes.

Community-based active TB case finding (ACF) has become an essential part of TB elimination efforts in high-burden settings. In settings such as the state of Kerala in India, which has reported an annual decline of 7.5% in the estimated TB incidence since 2015, if ACF is not well targeted, it may end up with a less-than-desired yield, the wastage of scarce resources, and the burdening of health systems. Program managers have recognized the need to optimize resources and workloads, while maximizing the yield, when implementing ACF. We developed and implemented the concept of ‘individuals’-vulnerability-based active surveillance’ as a substitute for the blanket approach for population/geography-based ACF for TB. Weighted scores, based on an estimate of relative risk, were assigned to reflect the TB vulnerabilities of individuals. Vulnerability data for 22,042,168 individuals were available to the primary healthcare team. Individuals with higher cumulative vulnerability scores were targeted for serial ACF from 2019 onwards. In 2018, when a population-based ACF was conducted, the number needed to screen to diagnose one microbiologically confirmed pulmonary TB case was 3772 and the number needed to test to obtain one microbiologically confirmed pulmonary TB case was 112. The corresponding figures in 2019 for individuals’-vulnerability-based ACF were 881 and 39, respectively. Individuals’-vulnerability-based active surveillance is proposed here as a practical solution to improve health system efficiency in settings where the population is relatively stationary, the TB disease burden is low, and the health system is strong.