Sociocultural aspects of delays in diagnosis among tuberculosis-diabetes comorbid patients in Satara, India: Its implications for the implementation of the national framework for joint tuberculosis-diabetes collaborative activities.

BACKGROUND: Tuberculosis(TB) and Diabetes comorbidity is an emerging public health problem in India. Delays in diagnosing TB or Diabetes would lead to adverse outcomes among comorbid patients, and attempts must be made to reduce these delays. Against this background, the study has been undertaken to clarify the role of sociocultural factors in determining diagnostic delays for TB and Diabetes among comorbid patients. METHODS: A cross-sectional cultural epidemiological survey of the randomly selected 180 TB-Diabetes comorbid patients was carried out. The study examined sociocultural factors of delayed diagnosis of TB and Diabetes among urban and rural TB-Diabetes comorbid patients registered under TB-Diabetes collaborative activities under the National TB Elimination Programme (NTEP) in the Satara district of Maharashtra by using a semi-structured interview schedule. The patterns of distress (PDs) and perceived causes(PCs) of TB and Diabetes were compared with patients' and providers' diagnostic delays of TB and Diabetes based on prominence categories. In addition, the relationship between PDs and PCs as explanatory variables and TB and Diabetes diagnostic delays as outcome variables were assessed using stepwise multiple logistic regression. RESULTS: Of the 180 TB-Diabetes comorbid patients, the proportion of men was higher, and they were 4.7 times more likely to get a delayed Diabetes diagnosis. Those who reported side effects of drugs and stigma reduced social status as the PDs were 2-3 times more likely to delay reaching TB facilities/providers (patients' diagnostic delay). Those who perceived inadequate diet and mental-emotional stress as the causes of TB were about three times more likely to reach the TB providers/facilities after two weeks. Also, those who perceived TB as a cause of punishment for prior deeds were two times more likely to reach TB facilities/providers after two weeks. Patients who reported fever and chest pain as the symptoms of TB were two times more likely to delay the diagnosis of TB. Patients who reported tobacco consumption, unhealthy lifestyles, thoughts, worries, tension, and germs or infection as perceived causes of TB were about two times more likely to be diagnosed after two weeks. Patients who reported excessive thirst as a diabetes symptom were about two times more likely to get delayed >2 weeks to reach diabetes facilities/providers. Patients who perceived environmental/occupational exposure as the cause of Diabetes were two times more likely to reach the diabetes facilities/providers after two weeks. Patients who reported excessive thirst and stroke as the physical problems of Diabetes were 3.2 and 9.6 times more likely to get delayed in the diagnosis of Diabetes (providers' diagnostic delay). Patients who perceived violation of taboo or misbehaviour as the perceived cause of Diabetes were 6.7 times more likely to get a delayed diagnosis of Diabetes. CONCLUSIONS: The sociocultural factors associated with TB and Diabetes diagnostic delays among comorbid patients are essential considerations in the evolving context of implementing TB-Diabetes collaborative activities. Therefore, acknowledging sociocultural factors concerning delayed diagnosis and minimising delays would strengthen joint TB-Diabetes collaborative activities under the National framework locally and nationally.

Predicting over-the-counter antibiotic use in rural Pune, India, using machine learning methods.

OBJECTIVES: Over-the-counter (OTC) antibiotic use can cause antibiotic resistance, threatening global public health gains. To counter OTC use, this study used machine learning (ML) methods to identify predictors of OTC antibiotic use in rural Pune, India. METHODS: The features of OTC antibiotic use were selected using stepwise logistic, lasso, random forest, XGBoost, and Boruta algorithms. Regression and tree-based models with all confirmed and tentatively important features were built to predict the use of OTC antibiotics. Five-fold cross-validation was used to tune the models' hyperparameters. The final model was selected based on the highest area under the curve (AUROC) with a 95% confidence interval (CI) and the lowest log-loss. RESULTS: In rural Pune, the prevalence of OTC antibiotic use was 35.9% (95% CI, 31.6 to 40.5). The perception that buying medicines directly from a medicine shop/pharmacy is useful, using antibiotics for eye-related complaints, more household members consuming antibiotics, and longer duration and higher doses of antibiotic consumption in rural blocks and other social groups were confirmed as important features by the Boruta algorithm. The final model was the XGBoost+Boruta model with 7 predictors (AUROC, 0.934; 95% CI, 0.891 to 0.978; log-loss, 0.279) log-loss. CONCLUSIONS: XGBoost+Boruta, with 7 predictors, was the most accurate model for predicting OTC antibiotic use in rural Pune. Using OTC antibiotics for eye-related complaints, higher consumption of antibiotics and the perception that buying antibiotics directly from a medicine shop/pharmacy is useful were identified as key factors for planning interventions to improve awareness about proper antibiotic use.