New Delhi, July 7 (PTI) Tamil Nadu has set a precedent as the first state in India to implement a predictive model designed to estimate the likelihood of fatalities among adults afflicted with Tuberculosis. This innovative model has been seamlessly integrated into the state's existing TB SeWA application, which aids in triaging patients right at the point of diagnosis.
The predictive tool, developed by the National Institute of Epidemiology (NIE) under the Indian Council of Medical Research (ICMR), was launched last week. Its core objective is to minimize the time interval between diagnosis and hospital admission for critically ill TB patients, thereby further lowering mortality rates, according to Dr. Asha Frederick, Tamil Nadu's State TB Officer.
This new functionality is an enhancement of the state's current TB SeWA (Severe TB Web Application) system, operational since 2022 under Tamil Nadu's differentiated care model, termed Tamil Nadu - Kasanoi Erappila Thittam (TN-KET). Dr. Frederick explained the model's framework to PTI.
Through TN-KET, healthcare professionals perform a triage assessment for each newly diagnosed adult TB patient, evaluating variables such as body mass index (BMI), pedal oedema, respiratory rate, oxygen saturation, and the capacity to stand unassisted. These variables identify very severe undernutrition, respiratory challenges, or poor physical condition.
Once data is inputted into TB SeWA, the application determines whether the patient is categorized as severely ill. Under the TN-KET framework, individuals identified as severely ill—owing to extreme undernutrition or respiratory distress—are prioritized for hospital admission, as noted by Dr. Frederick.
The TB SeWA system, until now, flagged patients as 'severely ill' utilizing these variables, assisting healthcare workers in prioritizing inpatient care, stated NIE Director Dr. Manoj Murhekar.
"The new feature progresses a step further," Dr. Murhekar remarked, "it calculates and displays the predicted probability of death for adult TB patients."
This objective risk percentage supersedes any subjective judgments regarding the seriousness of the condition and prompts frontline staff to expeditiously arrange hospital admissions for critically ill TB patients at the point of diagnosis. He explained that the probability of death ranges from 10% to as high as 50% among severely ill patients based on the conditions present.
Conversely, patients not categorized as 'severely ill' experience a sharply reduced predicted death probability of merely 1-4%, elucidated Dr. Hemant Shewade, senior Scientist at NIE.
A review of three years' data indicates that approximately 10-15% of adults with TB in Tamil Nadu are identified as severely ill at diagnosis, Dr. Shewade reported to PTI.
"This clear risk estimation ensures the sickest patients are admitted to hospitals without delays," Dr. Shewade emphasized. He further detailed that, while the average time from diagnosis to admission for severely ill patients under TN-KET is one day, around 25% of such patients encounter an admission delay ranging from three to six days.
"These delays will be mitigated with the enhanced model," Dr. Shewade stated. "It will motivate healthcare workers to make immediate referral decisions, ensuring better outcomes. Over time, this could also demonstrate a reduction in TB patient mortality in the state. Data shows two-thirds of TB deaths occur within the first two months post-diagnosis."
The development of the predictive model stemmed from data on nearly 56,000 TB patients diagnosed in public health facilities across Tamil Nadu between July 2022 and June 2023.
Dr. Frederick noted the accuracy of TN-KET's five triage variables—when predicting TB fatalities—was on par with baseline data collected in India's national TB portal, Ni-kshay. Dr. Shewade added that acquiring baseline data in Ni-kshay takes about three weeks, which is too delayed for predictive use, while Tamil Nadu's five triage variables are captured within a day.
Currently, all 2,800 public health facilities in Tamil Nadu—from primary health centers to medical colleges—utilize the TB SeWA application complemented by a paper-based triage tool. "Tamil Nadu is uniquely implementing systematic documentation and usage of these five triage variables for guiding patient management," Dr. Frederick asserted.
According to an NIE study, after implementing TN-KET over three years, there has been a notable reduction in the gaps within the care cascade, with two-thirds of districts documenting declines in TB fatality rates.
ICMR-NIE scientists highlighted that this initiative sets a crucial benchmark for other states, where TB deaths, particularly early fatalities, persist as a significant challenge despite the provision of free diagnosis and treatment.
(Only the headline of this report may have been reworked by Editorji; the rest of the content is auto-generated from a syndicated feed.) 
