Assessing a digital technology-supported community child health programme in India using the Social Return on Investment framework.

An estimated 5.0 million children aged under 5 years died in 2020, with 82% of these deaths occurring in sub-Saharan Africa and southern Asia. Over one-third of Mumbai's population has limited access to healthcare, and child health outcomes are particularly grave among the urban poor. We describe the implementation of a digital technology-based child health programme in Mumbai and evaluate its holistic impact. Using an artificial intelligence (AI)-powered mobile health platform, we developed a programme for community-based management of child health. Leveraging an existing workforce, community health workers (CHW), the programme was designed to strengthen triage and referral, improve access to healthcare in the community, and reduce dependence on hospitals. A Social Return on Investment (SROI) framework is used to evaluate holistic impact. The programme increased the proportion of illness episodes treated in the community from 4% to 76%, subsequently reducing hospitalisations and out-of-pocket expenditure on private healthcare providers. For the total investment of Indian Rupee (INR) 2,632,271, the social return was INR 34,435,827, delivering an SROI ratio of 13. The annual cost of the programme per child was INR 625. Upskilling an existing workforce such as CHWs, with the help of AI-driven decision- support tools, has the potential to extend capacity for critical health services into community settings. This study provides a blueprint for evaluating the holistic impact of health technologies using evidence-based tools like SROI. These findings have applicability across income settings, offering clear rationale for the promotion of technology-supported interventions that strengthen healthcare delivery.

Evaluating the Interrater Agreement and Acceptability of a New Reference Tool for Assessing Respiratory Rate in Children under Five with Cough and/or Difficulty Breathing.

BACKGROUND: Manual assessment of respiratory rate (RR) in children is unreliable, but remains the main method to diagnose pneumonia in low-resource settings. While automated RR counters offer a potential solution, there is currently no gold standard to validate these diagnostic aids. A video-based reference tool is proposed that allows users to annotate breaths and distortions including movement periods, allowing the exclusion of distortions from the computation of RR measures similar to how new diagnostic aids account for distortions automatically. This study evaluated the interrater agreement and acceptability of the new reference tool. METHODS: Annotations were based on previously recorded reference videos of children under five years old with cough and/or difficulty breathing (n = 50). Five randomly selected medical experts from a panel of ten annotated each video. RR measures (breaths per minute, bpm) were computed as the number of annotated certain breaths divided by the length of calm periods after removing annotated distorted periods. RESULTS: Reviewers showed good interrater agreement on continuous RR {standard error of measurement (SEM) [4.8 (95%CI 4.4-5.3)]} and substantial agreement on classification of fast breathing (Fleiss kappa, κ 0.71). Agreement was lowest in the youngest age group [< 2 months: SEM 6.2 (5.4-7.4) bpm, κ 0.48; 2-11 months: 4.7 (4.0-5.8) bpm, κ 0.84; 12-59 months: 2.6 (2.2-3.1) bpm, κ 0.8]. Reviewers found the functionalities of the tool helpful in annotating breaths, but remained uncertain about the validity of their annotations. CONCLUSIONS: Before the new tool can be considered a reference standard for RR assessments, interrater agreement in children younger than 2 months must be improved.

Healthcare choices in Mumbai slums: A cross-sectional study.

Background: Informal urban settlements, known as slums, are the home for a large proportion of the world population. Healthcare in these environments is extremely complex, driven by poverty, environmental challenges, and poor access to formal health infrastructures. This study investigated healthcare challenges faced and choices made by slum dwellers in Mumbai, India. Methods: Structured interviews with 549 slum dwellers from 13 slum areas in Mumbai, India, were conducted in order to obtain a population profile of health-related socio-economic and lifestyle factors, disease history and healthcare access. Statistical tools such as multinomial logistic regression were used to examine the association between such factors and health choices. Results: Private providers (or a mixture of public and private) were seen to be preferred by the study population for most health conditions (62% - 90% health consultations), apart from pregnancy (43% health consultations). Community-based services were also preferred to more remote options. Stark differences in healthcare access were observed between well-known conditions, such as minor injuries, pulmonary conditions, and pregnancy and emerging challenges, such as hypertension and diabetes. A number of socio-economic and lifestyle factors were found to be associated with health-related decisions, including choice of provider and expenditure. Conclusions: Better planning and coordination of health services, across public and private providers, is required to address mortality and morbidity in slum communities in India. This study provides insights into the complex landscape of diseases and health providers that slum dwellers navigate when accessing healthcare. Findings suggest that integrated services and public-private partnerships could help address demand for affordable community-based care and progress towards the target of universal health coverage.

The power of data mining in diagnosis of childhood pneumonia.

Childhood pneumonia is the leading cause of death of children under the age of 5 years globally. Diagnostic information on the presence of infection, severity and aetiology (bacterial versus viral) is crucial for appropriate treatment. However, the derivation of such information requires advanced equipment (such as X-rays) and clinical expertise to correctly assess observational clinical signs (such as chest indrawing); both of these are often unavailable in resource-constrained settings. In this study, these challenges were addressed through the development of a suite of data mining tools, facilitating automated diagnosis through quantifiable features. Findings were validated on a large dataset comprising 780 children diagnosed with pneumonia and 801 age-matched healthy controls. Pneumonia was identified via four quantifiable vital signs (98.2% sensitivity and 97.6% specificity). Moreover, it was shown that severity can be determined through a combination of three vital signs and two lung sounds (72.4% sensitivity and 82.2% specificity); addition of a conventional biomarker (C-reactive protein) further improved severity predictions (89.1% sensitivity and 81.3% specificity). Finally, we demonstrated that aetiology can be determined using three vital signs and a newly proposed biomarker (lipocalin-2) (81.8% sensitivity and 90.6% specificity). These results suggest that a suite of carefully designed machine learning tools can be used to support multi-faceted diagnosis of childhood pneumonia in resource-constrained settings, compensating for the shortage of expensive equipment and highly trained clinicians.