Nutritional assessment and counselling of tuberculosis patients at primary care in India: do we measure up?

Screening for Severe Illness at Diagnosis Has the Potential to Prevent Early TB Deaths: Programmatic Experience From Karnataka, India

Despite TB being a potentially fatal disease, severity is not systematically assessed at the start of drug-susceptible TB treatment. We document our experience screening people for severe illness at diagnosis/notification in program settings and the potential impact on reducing early TB deaths.

Due to the coronavirus disease (COVID-19) pandemic and its associated response, TB deaths increased for the first time in a decade. In any potentially fatal illness, an assessment of severity is essential. This is not systematically done for adults with TB, mostly due to a lack of policy and/or limited availability of diagnostic and clinical capacity. We developed a screening tool using simple and easily measurable indicators that can be used by paramedical TB program staff to quickly identify people with severe illness. During October–November 2020 in Karnataka, India, the paramedical program staff from 16 districts screened people with TB (aged ≥15 years) notified by public facilities for “high risk of severe illness,” which was defined as the presence of any of the following indicators: (1) body mass index (BMI) ≤14.0 kg/m²; (2) BMI ≤16.0 kg/m² with bilateral leg swelling; (3) respiratory rate >24/minute; (4) oxygen saturation <94%; (5) inability to stand without support. In this cohort study, we determined the incidence of program-recorded early deaths (within 2 months) and its association with high risk of severe illness. Of 3,010 people with TB, 1,529 (50.8%) were screened at diagnosis/notification, of whom 537 (35.1%) had a high risk of severe illness. There were 195 (6.5%, 95% CI=5.7, 7.4) early deaths: 59 (30.2%) within a week and 100 (51.3%) within 2 weeks of treatment initiation. The incidence of early deaths was significantly higher among those with high risk of severe illness (8.9%) at diagnosis compared to those without (3.8%) [adjusted relative risk: 2.36 (95% confidence interval=1.57, 3.55)]. To conclude, early deaths were especially high during the first 2 weeks and strongly associated with a high risk of severe illness at diagnosis/notification. Screening for severe illness should be explored as a potential strategy to end TB deaths.

Screening People with Tuberculosis for High Risk of Severe Illness at Notification: Programmatic Experience from Karnataka, India

Due to limited availability of diagnostics and capacity, people with tuberculosis do not always undergo systematic assessment for severe illness (requiring inpatient care). In Karnataka (south India), para-medical programme staff used a screening tool to identify people at 'high risk of severe illness', defined using indicators of very severe undernutrition, abnormal vital signs and poor performance status (any one): (i) body mass index (BMI) ≤ 14.0 kg/m² (ii) BMI ≤ 16.0 kg/m² with bilateral leg swelling (iii) respiratory rate > 24/min (iv) oxygen saturation < 94% (v) inability to stand without support. Of 3020 adults notified from public facilities (15 October to 30 November 2020) in 16 districts, 1531 (51%) were screened (district-wise range: 13-90%) and of them, 538 (35%) were classified as 'high risk of severe illness'. Short median delays in screening from notification (five days), and all five indicators being collected for 88% of patients, suggests the feasibility of using this tool in programme settings. However, districts with poor screening coverage require further attention. To end tuberculosis deaths, screening should be followed by referral to higher facilities for comprehensive clinical evaluation, to assess the need for inpatient care. Future studies should assess the validity (especially sensitivity in picking severely ill patients) of this screening tool.

Tuberculosis deaths are predictable and preventable: Comprehensive assessment and clinical care is the key

The goal of reducing tuberculosis (TB) mortality in the END TB Strategy can be achieved if TB deaths are considered predictable and preventable. This will require programs to examine and address some key gaps in the understanding of the distribution and determinants of TB mortality and the current model of assessment and care in high burden countries.

Most deaths in high-burden countries occur in the first eight weeks of treatment and in those belonging to the age group of 15–49 years, living in poverty, with HIV infection and/or low body mass index (BMI). Deaths result from extensive disease, comorbidities like advanced HIV disease complicated with other infections (bacterial, fungal, bloodstream), and moderate-severe undernutrition. Most early deaths in patients with TB, even with TB-HIV co-infection, are due to TB itself.

Comprehensive assessment and clinical care are a prerequisite of patient-centered care. Simple independent predictors of death like unstable vital signs, BMI, mid-upper arm circumference, or inability to stand or walk unaided can be used by programs for risk assessment. Programs need to define criteria for referral for inpatient care, address the paucity of hospital beds and develop and implement guidelines for the clinical management of seriously ill patients with TB, advanced HIV disease and severe undernutrition as co-morbidities. Programs should also consider notification and audit of all TB deaths, similar to audit of maternal deaths, and address the issues in delays in diagnosis, treatment, and quality of care.