Provider Experiences with AI Clinical Decision Support in eSanjeevani
- 1ICMR-National Institute for Research in Digital Health (NIRDH), New Delhi, India
Abstract
BackgroundClinical decision support systems (CDSS) are increasingly embedded in national telemedicine services, however how different stakeholders who use them experience these tools, particularly in low-and middle-income countries, is less explored. In the assisted model of Indias eSanjeevani service, spoke health workers enter structured symptoms and hub clinicians review a symptom summary and an artificial intelligence (AI)-generated differential diagnosis.
ObjectiveTo describe role-specific experiences of eSanjeevanis CDSS functions and to estimate how acceptance coexists with specific reported difficulties within each role.
MethodsWe analyzed a program feedback web survey completed between 11 July and 29 July, 2025 by hub clinicians and spoke providers. Prespecified item profiles covered review of summaries and suggested diagnoses, the perceived relation of suggestions to the clinicians own diagnosis, ease of use and intention to use (hub), and symptom-capture utility, question quantity, question relevance, summary presentation and satisfaction (spoke). Within-role contrasts were reported as absolute differences in acceptance with Newcombe 95% CIs. Short open comments underwent content analysis.
ResultsThe cohort comprised 190 hub and 662 spoke respondents from 13 and 15 states (21.4% and 19.3% of eligible providers). Among hub respondents, 76.8% regularly reviewed the symptom summary and 58.9% the suggested diagnoses; 72.6% judged suggestions matched (32.6%) or appropriate but different (40.0%), 7.9% mostly inappropriate, and 11.1% reported that suggestions were mostly not generated. Overall, 85.3% intended to keep using the tools. Affirmative intention was 98.4% when suggestions matched, 66.7% when they were mostly inappropriate (difference vs matched or appropriate-23.2 percentage points, 95% CI-48.5 to-4.0) and 61.9% when mostly absent (-25.7, 95% CI-47.1 to-7.3); technical issues, reported by 38.4%, made little difference (-2.8, 95% CI-14.1 to 7.3). Among spoke respondents, 63.4% rated question quantity appropriate, with miscalibration in both directions (15.6% too many; 16.6% too few); 89.9% found the questions relevant and 91.8% found the summary clear. Satisfaction was 75.3% and did not differ materially between those reporting too many questions and those reporting an appropriate number (-2.7, 95% CI-12.8 to 6.0). Comments located spoke difficulties in question burden and missing complaints, and hub difficulties in freezing and in the scope and availability of suggestions. Estimates were stable across cohort, ambiguity and state-exclusion sensitivity analyses.
ConclusionsAcceptance of eSanjeevanis CDSS was high in both roles. For hub clinicians it tracked the perceived quality and availability of AI suggestions rather than technical friction; spoke workers accepted the question burden, but its two-directional miscalibration points toward complaint-specific, adaptive questioning. These provider reports do not measure diagnostic accuracy. They identify log-based and audit-based indicators that the next evaluation cycle should measure directly.
Related
- Community health workers experiences with a digital clinical decision support system for longitudinal hypertension and diabetes care in rural Lesotho: a mixed-methods study
- A qualitative evaluation of provider decision-making and patient experiences with 1-step and 2-step gestational diabetes (GDM) screening
- Decision-support modelling for liver paired exchange integrating clinical thresholds and operational feasibility
- Real-world evaluation of a multilingual agentic retrieval-augmented clinical AI framework for evidence-grounded medical decision support
- Extracting smoking history from clinical notes for lung cancer screening decision support: comparing a structured-judgment model with general-purpose large language models
BibTeX
@misc{bhagat2026provider,
title = {{Provider Experiences with AI Clinical Decision Support in eSanjeevani}},
author = {P. Bhagat and S. Rana and K. N. Kaur and S. Sivasubramanian and V. Garg and S. Khalikar and T. Dutta and A. Chander and S. Sood and H. Singh},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.12.26365316},
url = {https://www.medrxiv.org/content/10.64898/2026.10.12.26365316v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Provider Experiences with AI Clinical Decision Support in eSanjeevani
AU - P. Bhagat
AU - S. Rana
AU - K. N. Kaur
AU - S. Sivasubramanian
AU - V. Garg
AU - S. Khalikar
AU - T. Dutta
AU - A. Chander
AU - S. Sood
AU - H. Singh
PY - 2026
DA - 2026/10/12
PB - medRxiv
DO - 10.64898/2026.10.12.26365316
UR - https://www.medrxiv.org/content/10.64898/2026.10.12.26365316v1
AB - BackgroundClinical decision support systems (CDSS) are increasingly embedded in national telemedicine services, however how different stakeholders who use them experience these tools, particularly in low-and middle-income countries, is less explored. In the assisted model of Indias eSanjeevani service, spoke health workers enter structured symptoms and hub clinicians review a symptom summary and an artificial intelligence (AI)-generated differential diagnosis. ObjectiveTo describe role-specific experiences of eSanjeevanis CDSS functions and to estimate how acceptance coexists with specific reported difficulties within each role. MethodsWe analyzed a program feedback web survey completed between 11 July and 29 July, 2025 by hub clinicians and spoke providers. Prespecified item profiles covered review of summaries and suggested diagnoses, the perceived relation of suggestions to the clinicians own diagnosis, ease of use and intention to use (hub), and symptom-capture utility, question quantity, question relevance, summary presentation and satisfaction (spoke). Within-role contrasts were reported as absolute differences in acceptance with Newcombe 95% CIs. Short open comments underwent content analysis. ResultsThe cohort comprised 190 hub and 662 spoke respondents from 13 and 15 states (21.4% and 19.3% of eligible providers). Among hub respondents, 76.8% regularly reviewed the symptom summary and 58.9% the suggested diagnoses; 72.6% judged suggestions matched (32.6%) or appropriate but different (40.0%), 7.9% mostly inappropriate, and 11.1% reported that suggestions were mostly not generated. Overall, 85.3% intended to keep using the tools. Affirmative intention was 98.4% when suggestions matched, 66.7% when they were mostly inappropriate (difference vs matched or appropriate-23.2 percentage points, 95% CI-48.5 to-4.0) and 61.9% when mostly absent (-25.7, 95% CI-47.1 to-7.3); technical issues, reported by 38.4%, made little difference (-2.8, 95% CI-14.1 to 7.3). Among spoke respondents, 63.4% rated question quantity appropriate, with miscalibration in both directions (15.6% too many; 16.6% too few); 89.9% found the questions relevant and 91.8% found the summary clear. Satisfaction was 75.3% and did not differ materially between those reporting too many questions and those reporting an appropriate number (-2.7, 95% CI-12.8 to 6.0). Comments located spoke difficulties in question burden and missing complaints, and hub difficulties in freezing and in the scope and availability of suggestions. Estimates were stable across cohort, ambiguity and state-exclusion sensitivity analyses. ConclusionsAcceptance of eSanjeevanis CDSS was high in both roles. For hub clinicians it tracked the perceived quality and availability of AI suggestions rather than technical friction; spoke workers accepted the question burden, but its two-directional miscalibration points toward complaint-specific, adaptive questioning. These provider reports do not measure diagnostic accuracy. They identify log-based and audit-based indicators that the next evaluation cycle should measure directly.
ER -