Seminar: Financial Strategies for Quality Improvement

Title: ‘Paying private obstetricians to provide emergency obstetric care to poor women: The Chiranjeevi Yojana public private partnership Program in India

Do popular financial strategies for quality improvement, such as performance based pay approaches, encourage effective clinical decision making and quality services or instead foster a system that rewards ‘patterns of behaviour’ among providers? How should we view them, given there has been little research into the effectiveness of provider incentives in improving the quality of care provided?

At this seminar, held at DFID Whitehall, on 22 July 2015, our guest speaker, Professor Ayesha De Costa, Associate Professor and Senior Lecturer at the Karolinska Institutet in Sweden, delivered a presentation to explore financial strategies for quality improvement in more depth.

View the presentation

Presentation by Ayesha de Costa at DFID Whitehall, July 2015

Find out more

Speaker bioPresentation summaryDiscussion summaryExplore further

Professor Ayesha De Costa is Associate Professor and Senior Lecturer at the Karolinska Institutet in Sweden. She has been principal investigator and coordinated multi country collaborations – randomised trials and non randomised studies in Asia and Africa. She has active research collaborations in India, Uganda, South Africa, Sweden, Oman, the United Kingdom, China and Finland. Her areas of research experience are: Pragmatic randomised trials, infectious diseases, epidemiology, maternal health, health systems.

To find out more click to view her staff profile page.

Background and context:
Prof. De Costa outlined the significance of the peri-partum period as the time with the highest risk of mortality and morbidity.  Globally governments are focusing on this period and there has been a push to increase facility based deliveries and access to emergency obstetrics care (EmOC) (at minute 1.19).  This is intended to ensure that the recognised 15% of deliveries which will have complications take place in a facility with the resources to manage them.  However this strategy’s success is reliant on good quality EmOC being available; many public sector services are sub-optimal, underfunded and under resourced and will require significant investment and time to develop (at minute 2.40).

The design of the Chiranjeevi Yojana (CY) public private partnership programme:
The CY programme was designed to respond to the Gujrat context (at minute 10.57) and to facilitate access to underutilised capacity within the disproportionately large private sector to increase the uptake of facility deliveries by vulnerable women (defined as living below the poverty line or of tribal origin, at minute 10.26).  The scheme is a demand side financed scheme which is funded solely by the state and which provides financial incentives to private providers.  The aim is to “harness” the private sector for public health outcomes and to provide interim services whilst the capacity of the public sector is developed.  Further information on the rationale is provided in the presentation (at minute 11.46).

A snapshot of the allocation of incentives is provided within the presentation (at minute 28.30).  The scheme has been developed with the participation of a number of stakeholders to help reduce the risk of unnecessary, and potentially more lucrative, procedures being undertaken; a fixed payment based on the expected proportion of normal (85%), complicated (8%) and caesarean section (CS) (7%) deliveries is made regardless of the actual deliveries undertaken.

Progress to date:
Progress to date is measured against outputs and some outcomes. Some data included in the presentation include that since 2007, 1 million deliveries have been undertaken within the CY programme.  The proportion of all facility deliveries has increased from 40% in 2001 to 90% in 2010 (at minute 13.39).  This has been driven by an increase in the use of private facilities; the CY programme contributed approximately 15% of private deliveries in 2010.  Approximately one third of those women eligible for a CY delivery used the opportunity; it is not known where the remaining two-thirds delivered.

CS rates in the CY programme average 6% per annum although it is not possible to determine if this is level meeting the actual level of need (at minute 29.21).  Facility deliveries in the public sector have remained relatively constant at approximately 15% between 2001 and 2010; there has been no significant change in the proportional access of public services.  There have been some concerns that the programme may weaken the public sector but there has been shown not to be the case (at minute 33.04).

Assessing the overall impact of the programme is acknowledged as challenging due to the low level of maternal deaths, the range of variables which can contribute to a death and the lack of disaggregation of mortality data by facility (CY or non-CY).

Ensuring quality of care:
The presentation outlined three main elements for monitoring and ensuring quality of care:

Using the health systems framework:  Within this the programme measures the outputs and outcomes within the results chain (at minute 17.12).  These include:

  • Structural elements: meeting a set number of minimum criteria (eg. size, equipment, appropriate staffing) (at minute 19.03).
  • Level of EmOC: ability to offer blood transfusion and/or CS in comparison to non-CY private providers and the public sector (at minute 21.04
  • Human resources: availability of key staff in comparison to non-CY private providers and the public sector (at minute 21.45). The presentation also notes the role of skilled birth attendants (SBAs) and that the in-house training and monitoring of the SBAs may contribute to increased accountability for the provision of quality care; this is being researched further (at minute 23.07).
  • Supplies and equipment: availability of equipment such as ultrasound and access to blood banks and laboratories, in comparison to non-CY private providers and the public sector.

Ensuring appropriate and safe care:

  • Preventing unnecessary procedures: as noted above, is an aspect of ensuring quality of care.
  • Accountability: for providing safe services and for protecting reputations drives good practices associated with quality of care.  However, the presentation notes some accusations of “cherry picking” where complex cases may be referred (which may work both ways).

Quality in the process of care:

  • Technical and interpersonal quality of care: ensuring high clinical standards of care in line with protocols etc along with respectful and non-discriminatory care.  These are harder to assess within the programme as they require observation and interaction; access to private facilities for these purposes can be challenging.  In addition there are no standard reporting systems or monitoring mechanisms (at minute 31.25).  This is recognised as a drawback.

Challenges:

  • Measuring and monitoring progress and determining impact:  The programme’s measurement is focused on outputs and outcomes.
  • Progressive attrition of private obstetricians: potentially due to the interface with the state including high levels of administration and trust issues and, in the past, with dissatisfaction over remuneration levels, amongst other reasons (at minute 24.50).
  • As noted above the financing system has an inherent risk of inducing unnecessary demand for CS, which has been mitigated by the financing mechanism.

The following is a précis of the discussion that was had between those present:

CY programme design:
Prof De Costa confirmed that the CY is funded solely by the State Government of Gujurat via taxation. The original rationale for CY emerged firstly from an acknowledgement that the public sector was not able to provide EmOC for the most vulnerable women because of a number of structural constraints. The decision to partner with the private sector was a pragmatic one. It was facilitated by the incumbent health secretary of the time, who was in support of the idea as well as academic support from influential institutions like the Indian Institute of Management. It was helped by the fact that the national government allowed significant flexibility in the approach that States took in tackling maternal mortality, hence the idea in Gujarat was given space to take shape.

Investing in the public sector has been raised on occasion by different stakeholders and varies with the perspective of the incumbent health minister/ secretary.  Improving the public sector is a longer-term structural issue that is not going to move forward at any pace. CY provides the State with an opportunity to harness the private sector capacity as a mid term strategy to meet its responsibility to ensure that the most vulnerable women receive care.  Investment in other areas along the continuum of care (opposed to the peri-partum period) have not been considered; the investment has been focused on the period which evidence shows has the highest mortality. So although there would be some benefits e.g. to focusing on ante natal care the best investment would be on performance of facilities.

Additionally it is easier to have a demand side financing program for a single discrete event.  The programme targets obstetricians, rather than SBAs, as its aim of CY is to have all women deliver at a facility so that the potential 15% of women with complications (which cannot be predicted) are in an adequately resourced environment should something happen.  In reality this means an EmOC facility with an obstetrician rather than relying on SBAs.

Although the programme is intended to be free of charge for vulnerable women out of pocket expenses were incurred by women. On average the subsidy received was around 67% in comparison to what women paid in private facilities outside the program. So the cost would be much greater had they opted into non CY private providers.

One criticism of the program early on was that it was very mother focused, so another element was introduced called ‘Bal Sakha’, which involved the services of private paediatricians to examine neonates of these mothers. The payment model was similar to the private obstetricians. However utilisation has been somewhat low.

CY implementation:
It is unlikely that the programme is contributing to the severe difficulties the public sector is experiencing with attracting and retaining obstetricians, (by providing an incentive to stay outside the public sector) as the public sector has long-standing fundamental issues which constrain obstetricians from taking up available vacancies.  Also the financial incentive provided by CY is too small to generate the massive disparity in public and private obstetrician employment. It is also unlikely that there is any potential conflict of interest for obstetricians working in both private and public because the public sector has such few obstetricians employed. Also the private obstetricians in the CY run their own establishment where they are required to be full time service providers.

There isn’t a strong sense of prestige for obstetrician of being involved but it is unlikely that this affects the attrition rates.  It may have an effect on how strongly attracted obstetricians are to joining / being involved but there is evidence to suggest that it depends more on other factors such as age.  The programme’s research shows that younger obstetricians tend to be more interested as they are building their practices and are keen to build up a client base. On the other hand there are more established obstetricians, who serve a particular market segment. They expressed reservations about joining the scheme as they feared going ‘downmarket’.

With regard to regulation in the private sector, aside from the usual regulation (which is fairly minimal in general), providers and beneficiaries have to complete significant amounts of extra paperwork that needs to be submitted to the government to obtain the reimbursement due to the providers. This has acted as a disincentive: if any data is missing they may not be reimbursed (and then had to pay out of pocket). Providers have stated in interviews that they were less willing to engage because of this paperwork. The lack of trust that exists between the public and private sectors is ‘compensated for’ by control,which may not be optimal.

Judging success and future implications:
The monitoring of the programme has been primarily at output level i.e.  % of women delivering within facilities.  There has unfortunately not been an attempt to gather data on deaths in and out of the programme among eligible women. The State does not disaggregate maternal mortality data between CY and non CY facilities. So it has been hard to judge what the outcomes have been. This is a major frustration. However there has been work undertaken by colleagues who did community surveys among eligible populations to study maternal deaths by program and non-programme. That unpublished data shows a much lower number of deaths in the CY program.

It can often be easier to assess outcomes for still births and newborns than for mothers in MNH services.  With regard to still births, there is ongoing work to research this. The programme has data on this but the numbers are small. With regard to neonatal deaths, 72 deaths were identified among around 2500 births. The programme’s analysis indicates that there were no significant differences in neonatal deaths between CY and non CY facilities; and interestingly no differences compared to deliveries at home. However 72 deaths is a small number to say anything very conclusively.

The judging of the success of the CY programme may have been influenced by the political economy of health in India; it began in Modi’s early tenure in Gujarat. The political influence that Modi subsequently wielded probably did not have any affect (if any) on CY in its early stages, as he was only elected as Prime Minister of India quite recently. However, as mentioned before, the position and drive of previous health ministers/secretaries was key to the design uptake of the scheme over time. This changed with changes in these key drivers (secretaries).

In terms of implications for other states / national policy, CY could not be applied nationally as it depends on a number of contextual factors, not least the strength (and depth at the district level) of the private sector. Gujarat has a large private sector obstetrician presence, going down to the sub district level.  In addition it has a certain ‘culture of entrepreneurship’ which is particularly prevalent there. Some States in the South might be good candidates for such a scheme but in many ‘backward’ States, it would probably be a tough act to follow as the preconditions are not there.

Conclusions and Implications
The CY programme offers an opportunity for the Gujurat State Government to deliver its responsibility to provide quality services for more vulnerable populations by using public funds to engage the private sector, rather than investing in public sector capacity.  It raises interesting questions about the opportunities and challenges of partnerships with non-state actors and the value of financial strategies in improving access to health services, particularly for vulnerable populations.

It raises some interesting questions about determining the impact and effectiveness of such a strategy and highlights the challenges of taking a health systems approach to ensuring access to quality services.

The online resource discusses different financial strategies for quality improvement along with other categories of strategies such as organisational and regulation strategies.

This seminar relates to a number of key themes and issues outlined by the learning resource:

Topic / issue Related learning resource section:
Provider incentives Financial strategies (Section C)
EmONC Norms and standards for quality (Section D)
Skilled Birth Attendants (SBAs) Providing quality services: skilled birth attendants (Section B)
Direct & indirect maternal deaths Major causes of maternal and neonatal deaths (Section A)
Care throughout the peri partum period Continuity of care and the continuum of care (Section B)
Quality Concepts and Quality of Care Models of quality; Dimensions of quality; Quality of care (Section B)
Indicators of Quality Capturing structural dimensions; Capturing process dimensions; Capturing health outcomes (Section D)

The following resources provide more information about this specifci case study, and related examples:

  • The State-Led Large Scale Public Private Partnership ‘Chiranjeevi Program’ to Increase Access to Institutional Delivery among Poor Women in Gujarat, India: How Has It Done? What Can We Learn? – read full text
  • Considerations of private sector obstetricians on participation in the state led “Chiranjeevi Yojana” scheme to promote institutional delivery in Gujarat, India: a qualitative study – read full text
  • Gujarat’s Chiranjeevi Yojana – a difficult assessment in retrospect – read full text
  • Improving access to maternity services: an overview of cash transfer and voucher schemes in South Asia – read full text
Seminar: Financial Strategies for Quality Improvement was last modified: May 4th, 2016 by Adrian Bannister