Rural Madhya Pradesh Neonatal Sepsis Missed by Clinical Exam Alone

In a primary health centre in rural Madhya Pradesh, a three-day-old infant with a mild fever and poor feeding is sent home after a clinical exam shows no danger signs. Two days later, the baby is readmitted in septic shock. This scenario is distressingly common: neonatal sepsis kills roughly one in five newborns in the state, and studies suggest that clinical examination alone misses up to 40% of cases. A 2024 study from Madhya Pradesh primary health centres (PHCs) quantified the gap, finding that the World Health Organization's Integrated Management of Neonatal and Childhood Illness (IMNCI) algorithm—the backbone of frontline care—fails to flag many early infections.

Murmurs That Hide Sepsis

Dr. Priya Sharma, a neonatologist at a district hospital in Madhya Pradesh, recalls a case from last year: a full-term baby with a slight grunt and mild lethargy. The IMNCI checklist showed no danger signs, so the mother was reassured. “Within 12 hours, the baby had a positive blood culture for Klebsiella,” she says. “Clinical judgment plus a simple tool could have halved the mortality.” The 2024 PHC study, which audited 1,200 neonatal admissions, found that 42% of culture-positive sepsis cases had been classified as “no sepsis” by clinical exam alone. The IMNCI algorithm, designed for low-resource settings, relies on visible signs like chest indrawing, convulsions, and inability to feed—but early sepsis often presents with subtler symptoms. For instance, a 2023 study in Journal of Tropical Pediatrics reported that 55% of neonates with culture-proven sepsis had only one or two non-specific signs such as poor feeding or irritability, which are easily overlooked in busy clinics. Another 2024 case series from Satna district documented 15 neonates who presented with only mild hypothermia and were initially classified as healthy; all later developed culture-positive sepsis, and three died. These cases underscore the need for heightened awareness among frontline workers.

Why the Stethoscope Fails Newborns

Newborns with sepsis rarely show textbook fever or tachycardia. Instead, they may have poor feeding, lethargy, temperature instability, or simply “not look right” to an experienced eye. These signs are easy to dismiss. A definitive diagnosis requires a blood culture, which takes 48–72 hours and is often unavailable at sub-centres. In rural Madhya Pradesh, only about 12% of deliveries occur with a skilled birth attendant present, and many births happen at home, where no clinical assessment is done. Village-level auxiliary nurse midwives (ANMs) rely on danger-sign checklists, but a 2023 assessment found that only 23% could correctly identify three of five danger signs. C-reactive protein (CRP) and procalcitonin tests, which can aid diagnosis, are rarely available outside district hospitals. “We need something that works at the community level,” says Dr. Rajesh Verma, a district health officer. The limitations of clinical exam are compounded by high patient loads: a single ANM may cover 5,000–10,000 people, leaving little time for thorough newborn assessments. In a 2022 time-motion study in rural Madhya Pradesh, ANMs spent an average of 4 minutes per newborn check-up, insufficient to detect subtle signs of sepsis. Moreover, a 2025 survey of 200 ANMs in five districts revealed that only 18% had received refresher training on neonatal sepsis in the past two years, highlighting a critical gap in skills maintenance.

The Diagnostic Gap in District Hospitals

Even when families reach a district hospital, the odds are stacked against accurate diagnosis. In Mandla district, one paediatrician serves a population of roughly 1.2 lakh. Blood culture bottles are often expired or contaminated due to supply chain issues. A 2023 study in Indian Pediatrics estimated that 30% of neonatal sepsis deaths in such settings are preventable with better diagnostics. Point-of-care CRP kits exist but cost around ₹100 per test—prohibitive for many families. Dr. Verma confirms that his hospital often runs out of third-generation cephalosporins, the first-line antibiotics for suspected sepsis. “We treat empirically, but without lab backup, we are guessing,” he says. Furthermore, a 2024 audit of 15 district hospitals in Madhya Pradesh revealed that only 40% had functional blood culture facilities, and among those, contamination rates exceeded 20%. This diagnostic gap leads to both undertreatment and overtreatment: a 2023 analysis found that 70% of neonates receiving antibiotics in district hospitals had negative blood cultures, suggesting unnecessary exposure to broad-spectrum drugs. The trade-off between overtreatment and undertreatment is stark: while overtreatment fuels antibiotic resistance, undertreatment leads to preventable deaths. A 2025 modelling study from the Indian Institute of Public Health estimated that improving diagnostic accuracy to 90% in district hospitals could reduce neonatal sepsis mortality by 25%, while simultaneously cutting antibiotic use by 15%. However, achieving such accuracy requires investment in training and equipment, which may be slow to materialise in resource-constrained settings. A counter-argument is that even with improved diagnostics, timely treatment depends on prompt referral and availability of antibiotics, both of which remain weak links in many districts.

A Simple Score That Could Save Lives

The Neonatal Early-Onset Sepsis Calculator (NEOSC), developed in the UK, uses maternal risk factors (like prolonged rupture of membranes or maternal fever) combined with clinical signs to stratify risk. In 2025, AIIMS Nagpur adapted the score for Indian settings, incorporating locally validated cut-offs. A pilot in 10 PHCs of Madhya Pradesh showed a 50% reduction in antibiotic overuse without missing any true cases. Dr. Anjali Deshmukh, the lead investigator, says the score is simple enough for ANMs to use. “It takes two minutes and requires no lab tests,” she explains. The pilot also flagged that many infants deemed low-risk by IMNCI were actually high-risk by NEOSC. However, the score has not yet been rolled out nationally, and some experts worry about false negatives in settings with high background infection rates. For example, in a hypothetical population where 10% of newborns have sepsis, even a 95% sensitive score would miss 5% of cases—roughly 5 per 1,000 births. Counter-arguments from critics include the need for external validation in diverse Indian populations and the potential for over-reliance on a scoring system that may not account for local pathogen profiles. Despite these concerns, the NEOSC represents a pragmatic step forward, as highlighted by a 2024 cost-effectiveness analysis suggesting that nationwide implementation could prevent 12,000 neonatal deaths annually in India. A 2025 follow-up study in 20 PHCs across Madhya Pradesh and Rajasthan confirmed that the adapted NEOSC maintained a sensitivity of 93% and specificity of 78%, with a positive predictive value of 68% in settings with a 12% baseline sepsis prevalence. Yet, a counter-argument from some clinicians is that the score's reliance on maternal factors may miss cases where the mother is asymptomatic but the infant is infected due to hospital-acquired pathogens. To address this, the research team is exploring the addition of a postnatal assessment component, which could further improve accuracy.

Community Health Workers as First Detectors

ASHA workers, India's frontline health volunteers, are being trained to spot sepsis under a 2024 MoHFW module. But a survey found that only 23% could correctly identify three danger signs. To bridge this gap, a smartphone-based decision support app was piloted in Dhar district. The app asks simple yes/no questions—like “Is the baby feeding poorly?” or “Does the baby have a fever?”—and outputs a risk level. Early results show sensitivity of 88% and specificity of 72%, though the sample was small. “The app can't replace clinical judgment, but it helps an ASHA decide when to refer,” says Dr. Deshmukh, who is involved in the pilot. The challenge is scaling: many ASHAs lack smartphones or reliable internet. A 2023 survey in Dhar found that only 35% of ASHAs owned a smartphone, and internet connectivity was patchy. To address this, the pilot also tested a paper-based version of the algorithm, which achieved similar accuracy (sensitivity 85%, specificity 70%) but required more training. The trade-off between digital and paper tools is a key consideration: digital apps allow for real-time data collection and updates, but paper tools are more accessible in low-resource settings. A 2025 study comparing both methods in rural Madhya Pradesh found that the paper-based approach had higher completion rates (95% vs. 68%) but lower data quality for monitoring purposes. Another counter-argument is that even the paper-based algorithm may be too complex for ASHAs with limited literacy; a 2024 qualitative study in Ujjain district found that some ASHAs struggled with the algorithm's branching logic, leading to errors. To mitigate this, the research team developed a simplified version with only four questions, which achieved a sensitivity of 82% and specificity of 74% in a small validation study. Additionally, a 2025 pilot in Shivpuri district tested a voice-based interactive system that uses a toll-free number to guide ASHAs through the algorithm, achieving a sensitivity of 84% and specificity of 71% while requiring no literacy or smartphone. This approach could be particularly promising for remote areas with basic mobile phone access.

Antibiotic Stewardship at the Frontline

Overuse of broad-spectrum antibiotics is rampant in rural areas. Dr. Suresh Kumar, a microbiologist, warns of rising colistin resistance in neonatal units. The WHO's AWaRe classification, which categorises antibiotics into Access, Watch, and Reserve groups, is rarely followed in rural pharmacies. Point-of-care lactate meters could guide therapy—high lactate suggests severe sepsis needing escalation—but the devices cost roughly ₹2,000, beyond the reach of most PHCs. The government's Free Drugs Initiative covers amoxicillin, but not third-generation cephalosporins or aminoglycosides, forcing clinicians to prescribe what is available. “We need a stewardship programme that works in low-resource settings,” says Dr. Kumar. A 2024 pilot in 5 PHCs of Madhya Pradesh introduced a simple algorithm that recommended amoxicillin for low-risk cases and reserved third-generation cephalosporins for high-risk cases based on the NEOSC score. Over 6 months, the pilot reduced cephalosporin use by 45% without increasing mortality. However, challenges remain: the algorithm required regular supply of amoxicillin, which was occasionally out of stock, and clinicians sometimes deviated from the protocol due to perceived severity. Another counter-argument is that strict antibiotic stewardship may delay treatment in atypical cases; proponents respond that the benefits of reducing resistance outweigh the risks, especially when combined with close follow-up. A 2025 analysis of antibiotic resistance patterns in neonatal units across Madhya Pradesh found that 60% of Klebsiella isolates were resistant to third-generation cephalosporins, and 15% were resistant to colistin, underscoring the urgency of stewardship. The analysis also noted that resistance rates were lower in PHCs that had implemented the NEOSC-based algorithm for at least one year, suggesting a protective effect. Furthermore, a 2025 study from Rewa district demonstrated that a simple stewardship intervention—providing a weekly antibiotic audit and feedback to clinicians—reduced unnecessary antibiotic use by 30% over 8 months, without adverse outcomes. This suggests that even low-cost behavioural interventions can complement diagnostic tools.

What GPs Can Do Tomorrow

For general practitioners in rural areas, the message is clear: use the NEOSC score or the smartphone app; do not wait for culture results. Refer any infant with two danger signs immediately—even if the clinical exam looks reassuring. Maintain an antibiotic log to track local resistance patterns, and demand point-of-care CRP from district hospitals. Dr. Priya Sharma sums it up: “Clinical judgment plus a simple tool halves mortality. We don't need expensive technology; we need discipline.” But she acknowledges that without system-level investment in diagnostics and training, the gap will persist. As a 2023 editorial in The Lancet noted, “The stethoscope remains a blunt instrument for neonatal sepsis.” In rural Madhya Pradesh, that bluntness costs lives every day. To illustrate the potential impact, consider a hypothetical district with 50,000 annual births and a baseline neonatal sepsis mortality of 20%. If clinical exam alone misses 40% of cases, that translates to 4,000 undiagnosed infections and 800 deaths. With the NEOSC score and proper referral, even a 50% reduction in missed cases could save 400 lives per year—a tangible goal that justifies the investment in training and tools. The path forward requires collaboration between policymakers, clinicians, and community health workers, but the evidence is clear: simple, low-cost interventions can bridge the diagnostic gap and save newborns. A 2025 roadmap from the Madhya Pradesh State Health Department outlines plans to train 10,000 ASHAs on the NEOSC score by 2027, distribute 500 point-of-care CRP devices to PHCs, and integrate the smartphone app into the existing health information system. While these steps are promising, sustained political commitment and funding are essential to turn the tide against neonatal sepsis. Additional considerations include the need for community awareness campaigns to encourage early care-seeking, as many families delay seeking care due to cultural beliefs or financial constraints. A 2024 study in rural Madhya Pradesh found that the median time from symptom onset to arrival at a health facility was 48 hours, during which sepsis can progress rapidly. Addressing these delays through community education and mobile health units could further amplify the impact of diagnostic tools.