ICMR Study Unveils Simple TB Risk Prediction Tool that Can Identify High-Risk Patients at Diagnosis
Advertisement
The researchers developed a web-based prediction calculator capable of estimating both early mortality risk and the probability of death within one year.
Indian Council of Medical Research (ICMR) has developed and unveiled a simple risk prediction web-based tool that can estimate a tuberculosis patient's risk of death at the time of diagnosis using routinely available clinical indicators, enabling earlier intervention and reducing TB mortality in resource-constrained settings.
Published in BMJ Open, the study was carried out by researchers from the ICMR-National Institute of Epidemiology (NIE), Chennai, in collaboration with the Tamil Nadu TB programme.
The team analysed data from 55,971 adult TB patients notified through public health facilities across Tamil Nadu between July 2022 and June 2023.
Their findings showed that 7.4% of patients died within one year of diagnosis, while nearly 68% of those deaths occurred during the first two months.
Based on these observations, the researchers developed a web-based prediction calculator capable of estimating both early mortality risk and the probability of death within one year.
“This initiative has shown that it is possible to bring down TB-related deaths remarkably by following scientifically designed tools and methods. Severe illness can be quickly identified through triaging (preliminary assessment), and patients can be promptly admitted after diagnosis,” Dr Hemant Deepak Shewade, senior medical scientist at ICMR-NIE and author of the study, said.
Unlike models that rely on extensive laboratory investigations or multiple database variables, the calculator uses basic clinical measurements that are routinely recorded during the first patient assessment.
These include body mass index (BMI), oxygen saturation, respiratory rate, pedal oedema, and whether the patient can stand without support. The model also incorporates age, sex, disease site, previous TB treatment history, and microbiological confirmation to improve prediction accuracy.
According to the researchers, the simplified approach performed almost as effectively as more complex prediction models built using numerous variables later available in the government's Ni-kshay TB database.
Because the required information is collected during diagnosis, the tool can support immediate triaging without waiting for additional investigations.
The authors have recommended that severe illness indicators become part of routine TB assessments nationwide. They have also suggested that the prediction calculator could support resource-constrained healthcare settings by enabling data-driven decisions on hospital admission, closer monitoring, and intensive follow-up for patients identified as high risk.
However, the researchers cautioned that the tool has not yet undergone external validation outside the state and has not been adopted for nationwide implementation by the Central TB Division under the National Tuberculosis Elimination Programme.
Stay tuned for more such updates on Digital Health News
