India’s publicly financed insurance scheme: scope for revision – authors’ reply

We wish to thank Gupta H¹ for interest in our research article² and making important observations. We would like to provide the response to the points made in the Comment.¹

Basically, four points were raised in the Comment. First, what is the implication of upgradation of the district hospitals to medical colleges on the net financial benefit of Ayushman Bharat Pradhan Mantri Jan Aarogya Yojana (PMJAY). Second, the lack of effective gatekeeping and its potential reasons. Third, differential performance of State governments in utilizing public hospital infrastructure for treating COVID-19 patients using the PMJAY funds. Fourth, the heterogeneity in length of stay as well as use of expert elicitation (for length of stay) as opposed to actual observed data, which may affect the estimates of the present study to determine the financial benefit.

It is important to recognise the context of health financing situation in India, which has also been incorporated in the present analysis. The district hospitals continue to be funded through two routes: the supply side government funding through treasury route, as well as the demand-side additional payments received through the government funded health insurance schemes. The proposal of upgrading the district hospitals to medical colleges also envisages additional financing—to meet the shortfall of infrastructure and human resource shortfall—to be shared between the Central and State Governments.^3,4 As a result, neither the current analysis, nor the vision of the proposed strategy under the Ayushman Bharat Health Infrastructure Mission (PM ABHIM) assumes that the PMJAY funds may be used to bridge the shortfall. However, given that the facilities are likely to be upgraded, it may also entail higher utilisation, which may imply further increase in PMJAY revenue.

As mentioned by Gupta H¹, lack of effective gatekeeping by primary care facilities within public health system is a feature which is characteristic of several South Asian countries.⁵ Our study² also refers to this, however, from the point of view of gatekeeping certain secondary care services or packages under the PMJAY to the public sector district hospitals.6, 7, 8 A detailed exploration of reasons for ineffective gatekeeping by primary care facilities, leading to overcrowding of secondary and tertiary care facilities, and its implications, is definitely an important area of future research and needs further exploration. Similarly, the differential response of State Governments for using PMJAY for treatment of COVID-19 patients is also important for future research, which is beyond the scope of our analysis. We specifically chose 2019 as the year for analysis, to avoid any confounding due to influence of COVID-19 pandemic on extent and patterns of care utilisation under the PMJAY.

Gupta H¹ also reiterates an important issue of the lack of electronic health records to obtain patient level clinical information, an aspect which we highlighted in the limitations to our paper. In fact, this has been cited as an important bottleneck in several previous costing studies also.^9,10 In the context of our analysis, it is important to recognise that the net monetary benefit for district hospital was estimated for provision of both the medical and surgical care. For the medical packages, since the claims data of the PMJAY uses actual length of stay for patients admitted under the scheme, and hence calculate the claim amount paid, the present data limitation does not impose any problem. As per our additional analysis, provision of medical care constitutes 58.8% of claims volume at district hospital. For surgical package, we use of expert opinion driven length of stay to determine a case-based bundled cost. However, given the nature of case-based bundled provider payment, the claims amount paid to district hospitals for surgical packages is agnostic of the length of stay. Hence, any heterogeneity in the length of stay (which may arise because of variations in quality of care, or supplier behaviour, or variations in the clinical severity of case or other patient characteristics) is unlikely to influence the claims revenue generated at the district hospital and hence our analysis. Nonetheless, the availability of good quality electronic data at patient level can improve the estimates of cost of services, which may in turn further refine our estimates of net monetary benefit.

Finally, we agree with the observation in the letter about the collaboration of administrators and researchers, as well as the use of large administrative datasets “for the massive exercise so that we may make some sense by having a bird eye's view of the currently running scheme”. Bridging the gap between researchers and policy-makers has been recommended and demonstrated as a very important strategy for translation of evidence to policy.^11,12 Our study is a recognition of the significant importance of using routine data generated by the health programmes and schemes to draw meaningful inferences to further improve the programme design and implementation.

Contributors

SS, MPS and PB wrote and edited this Comment.

Declaration of interests

Dr Shankar Prinja is the former Executive Director (HP & QA) of National Health Authority, Ayushman Bharat PM-JAY, Government of India. All authors declare no other conflict of interest.