Non-communicable diseases (NCDs) are emerging as a significant health problem for tribal populations, aside from the established problems of infectious diseases, undernutrition, and healthcare access. The epidemiological transition, dietary changes, increased tobacco and alcohol use, occupational changes, decreased physical activity, obesity, and inadequate screening have contributed to the rising burden of NCDs in tribal populations. Evidence from India shows that non-communicable diseases in tribal areas are significant causes of mortality, and many cases of hypertension and diabetes mellitus go undiagnosed. In a multi-centric community-based study conducted in 12 tribal districts, 66% of deaths in individuals aged 15 years and older were due to NCDs, among which cardiovascular diseases were the leading cause of mortality in most of the study sites. A systematic review and meta-analysis of 42 studies also reported a significant prevalence of hypertension in the tribal population of India. Recent multicentric studies have highlighted the burden of type 2 diabetes and prediabetes in tribal populations, with hypertension, obesity, and smokeless tobacco use as significant risk factors. Artificial intelligence (AI) and machine learning (ML) can enable multidimensional risk assessment and personalized healthcare by analysing demographic, behavioural, anthropometric, clinical, and longitudinal data using algorithms such as the Random Forest, XGBoost, and Long Short-Term Memory (LSTM). Most current AI-based clinical models, however, were not designed with tribal populations in mind, raising questions about representation, population-specificity, biases, privacy, and interpretability. This integrative narrative review highlights the relationships between the increased burden of NCD, lifestyle and behavioural risk factors, AI-based risk prediction models and decision support systems, and healthcare for tribal populations. Based on the evidence, this review introduces a conceptual framework for a culturally sensitive Clinical Decision Support System (CDSS) designed to leverage tribal community-specific factors, such as lifestyle and behavioural data, along with clinical data, an AI-based predictive analytics module, and communityspecific clinical guidelines. The review argues that CDSS should be designed as an augmentation to clinical decision-making rather than a substitute for human expertise. Community health workers can play a critical role in this framework by bridging the gap between AI-assisted clinical decision-making and community health. The review suggests that AI-enabled tribal healthcare can evolve from predictive analytics to a closed-loop system that includes promoting appropriate lifestyle changes, improving screening, facilitating referrals, and monitoring. The framework can guide future research directions, including population-specific AI model development and empirical studies of AI-based healthcare delivery in tribal populations.

Keywords : Artificial intelligence; Clinical decision support system; Tribal communities; non-communicable diseases; Machine learning; Random Forest; XGBoost; LSTM; Lifestyle; Preventive healthcare; Equity; India.

Author : Sravanthi. V 1, Prof. T. Sobha Rani 2

Title : Artificial Intelligence-Enabled Clinical Decision Support for NonCommunicable Disease Prevention among Tribal Communities Using Lifestyle Behaviour: Integrating Risk Prediction and Healthcare

Volume/Issue : 2026;08(04)

Page No : 1-32