A Nuanced View of Gender Based Health Financing

In one of my courses on Health Insurance which I used to offer in the second year of PGP at IIM Ahmedabad, we used to discuss one interesting reading on "risk selection and risk adjustment" (see the reference below). The principal theme was using information about age, gender and health status to predict outcomes in health insurance and thereby pricing and approaches of predictive modeling.

In this particular reference, we had this great graph presenting actual spending by age and gender in the US context in 2009. Each point in the graph below presented one-year average health spending per capita as per age and gender. The spending in very early years (babies) are higher, and also women have higher medical expenses during their reproductive age group than males at the same reference age group. The per capita spending on health is more than double for females in age groups ranging between 25 and 35 years. This non-linear pattern can also be captured through modeling process.

Source: Ellis and Layton (2014)

The data and graph presented above also introduces us to understand why there is need for gender-based HCF policies in any economic system. In India HCF to a large extent has tried to address this issues and programmes such as NRHM, now NHM, allocated significant resources to meet reproductive and child health (RCH) challenges.

However, we are not sure whether the same is true for health insurance claims settlement. Though we do not have much information on gender based health insurance claim settlements, the data provided by RG Jeevandayee Arogya Yojana in Maharashtra provide some insights in gender based claim settlement for populations belonging to lower income groups. Here the gender disparity in claims looks quite obvious. The number of claims for females is 28% less than that of males and in terms of the amount of claims, it is 39% less. This difference is not explained by sex-ratio which is 929 females per 1000 males in Maharashtra. The RGJAY insurance scheme is also enrolled on family basis in 35 districts of the state and therefore has no enrollment bias towards males. There are 491 hospitals in the network.

The study of Insurance Information Bureau of India (IIBI) also indicate that for the year 2013-14, the data received from all insurers indicate that amount of claims paid to males is 72% and that to females is 28%.

Gender Disparity of Claims in

RG Jeevandayee Arogya Yojana (RGJAY)

Source: https://www.jeevandayee.gov.in/RGJAY/FrontServlet?requestType=

CommonRH&actionVal=HomepageStats&pageName=5 (accessed on 9 March 2017)

As gender has significant influence on how people will respond in a complex and context-specific situations, gender should be at the core of health financing policies. Some important learnings from this are:

• health finance is not gender neutral and therefore health care financing (HCF) policies need to be considerate of this fact,
• gender can be a key to stratification to assess risks and therefore developing gender-based health financing policies,
• gender affects how we live, work and relate to each other at all levels, including in relation to the health finance,
• household decision-making and health seeking behaviour and health financing policies around gender will be critical as it affects access to and utilization of health services.

As a first step to make policy makers and other stakeholders aware of gender impact, data and information must first be disaggregated by gender and make it an integral part of gender analysis into health financing research and analysis. Disaggregating data by gender is critical as aggregated datasets can mask many critical differences affecting policies in a unfavourable way. It would be useful to have gender-wise and age-wise data of settlement of claims and its analysis. Agencies such as IRDA and other government agencies engaged in presenting and analyzing the insurance data need to provide the information to monitor the gender based health care financing trends.

Ellis R P and T J Layton (2014). Risk Selection and Risk Adjustment, in Encyclopedia of Health Economics, ed. A J Culyer, Volume 3 pp. 289-297, Elsevier Inc. 

Government Health Insurance: Need to Navigate Equity Objective

Developing targeted health insurance options by pooling the risks and providing protection from catastrophic expenditures to population belonging to lower income groups has been one important HCF policy response in many countries. The central ministry and various state governments in India have developed and implemented RSBY, and other government supported health insurance (GSHI) programme targeted at BPL and other populations. 

As per the IRDA annual report of 2015-16, about 273 million persons are covered by various GSHI schemes in India. This may be considered as one of the major intervention globally bringing a vast pool together in a targeted manner to address health finance challenge. The total annual spending on these schemes is about Rs 2,500 crores per annum for which the contribution comes from both central and state governments. The net incurred claim ratio of these schemes for the year 2015-16 has been around 109 percent.

One of the objectives of this intervention is to ensure equity in health care financing. The performance of these schemes across states is, therefore, of interest from equity viewpoint. The gross premiums collected indicate that 56% of these premiums are accounted for by four states which are Maharashtra, Tamil Nadu, Karnataka, and Gujarat. The rest of India accounts for 44% of the total premium. These four states which have 56% of premium share roughly account for 25% of the total population of the country. Maharashtra which has 9% of the population, has 31% share in premiums. IRDA annual report does not provide detailed data for all states in India, and therefore the analysis here is restricted in some sense.

Premium of GHIS, 2015-16 (Rs in crores)

Percent Premium and Percent Population of Selected States 2015-16

The above graphs raise some interesting research questions. The performance across states in implementing GSHI schemes is not uniform, and therefore one is not sure whether the equity objective of HCF is being fulfilled. Since the RSBY and GSHI schemes have resource sharing arrangement between the centre and the states, the fiscal position of some states may contribute to this variation in experience. Also, the state level priorities of health, institutional arrangements developed at the state level and administrative efficiency to implement this scheme may also contribute to this variation.  If there are clusters of states which share common structural characteristics such as administrative efficiency or institutional developments favoring the RSBY policy, then government needs to consider a differentiated approach to make sure other states implement the programme effectively. 

The preference of insurance companies to work in certain regions for the reasons of either supply-side constraints (not having adequate health infrastructure) or inadequate demand of private voluntary insurance (PVI). The data suggest that Maharashtra accounts for 32% of PVI insurance and 31% of GSHI schemes premium. States having a high penetration of health insurance may indicate a well-developed network of private providers.

Also, the issues such as what is the exact coverage and take-up of target population and premiums paid under GSHI schemes across various states in India, what is the claim settlement experience across states in India, is low take-up in certain states because of lack of awareness or supply side health issues in health, is there a case of cream-skimming at macro level, insurance companies focusing on better states.

At a time when policy makers, scholars, and thought leaders are concerned about high OOP expenditures on health, what is restraining the system to ensure equitable distribution of coverage? What does this mean for the health of people on margins and how are they struggling with this experience?

Health Budgets: Beyond the Piecemeal Approach

The report published in the Indian Express today is an interesting read. The Ministry of Finance (MoF) has asked the MoHFW to have a relook at the National Health Mission (NHM) budget of 2017-18 which has planned a renewal of the Mission. The NHM had asked for a little over Rs 33,500 crore for next year and a little over Rs 1,37,000 crore for the next five years. The news report suggests that Expenditure Finance Committee (EFC) of the MoF has raised the questions about the way programme has been implemented and asked the MoHFW to revised the proposal. The intervention of MoF is a welcome step.

The year 2016-17 was the last year of most of the national programmes which also happened to be the last year of the twelfth five-year plan. The programmes include the NHM, the NACP, RNTCP, etc. The annual plans for most of the national programmes are generally part of the strategic plans of the programme for next phase starting in April 2017.

There are three developments worth noting that changed since the preparation of last phase strategic plans of various programmes.

First, the NITI Ayog had proposed to come with 15-year vision plan and through this vision document work towards the social objective mission of UNDP's 2030 sustainable goals which has16 goals with 169 sub-goals. These goals provide a much larger focus on the development. So various strategic plans of programmes need to articulate a vision or plan to guide how these goals will be achieved and how much resources will be required to achieve this.

Second, 14th Finance Commission envisaged giving more untied funds to States with greater fiscal responsibility in implementing centrally-sponsored schemes, and as part of the financial devolution strategy, FC increased the states’ share in central taxes from 32 percent to 42 percent. The new strategic plans for next phases were expected to integrate the developments of larger finance devolution as recommended by the Finance Commission to states. Several programmes had also planned to implement the integration and convergence strategy.

Third, the budget speech of Finance Minister in 2015-16 had mentioned that the government would abandon the plan and non-plan distinction from 2017-18. The budgets of 2017-18 were expected to have new and better rationale for justifying the budgets.

The strategic planning exercise of various programmes is generally expected to start one year in advance. For example, NACO started the exercise of developing a strategic plan by publishing a Mid-term Appraisal of NACP IV by NACO in August 2016. The findings and conclusions of this review were supposed to be integrated into the next phase strategic plan. However, we have not seen this strategic planning exercise moving any further. In the same manner, RNTCP strategic planning exercise started in September 2016, and final draft of the RNTCP programme have been shared with key stakeholders for comments during last week of February.

The new strategic plan of various programmes in the changed focused effort needed to ensure:

• consider all the three developments pointed out above
• cost-effectiveness of interventions
• a clear statement on state-level contribution to funding
• integration and convergence strategies leading to economies
• focus on prevention to ensure it reduces the future health burden
• district-level and state-level plans to ensure there is flexibility at district levels and whether the overall is well integrated and need-based

I am not sure whether various strategic plans meet these expectations, certainly not whether the interventions proposed are cost-effective or not.

* NHM’s health in doubt as Ministry told to revise cost, Indian Express, 9 March 2017

Why the Strategic Purchasing of Health Services by the Government?

For many years there was the belief that willingness to pay and ability to pay can guide the allocation of health care resources in any country. Development of the health care financing options was guided by these two principles. However, the collectivist view as articulated in Thatcherism era in the UK in the eighties articulated the principal that (1):

Adequate health care should be provided for all, regardless of their ability to pay, must be the foundation of any arrangement for financing the health services.

This collectivist view also gained prominence as the government systems failed to provide healthcare to all and the consequences of OPP expenditures on health care started affecting lives of a large section of the population. The challenges arose as the public health sector, which assumed responsibility for producing and distributing the services started crippling as they were not able to cope up with the demand or turned out to be inefficient.  The result was that people resorted to buying the services from the private sector which resulted in high OOP expenditures. The governments in such situation explore health financing options in the following ways:

1. Developing targeted health insurance options by pooling the risks and providing protection from catastrophic expenditures. The government in India and various states developed and implemented RSBY insurance programme targeted at BPL population group. As per the IRDA annual report of 2015-16, about 273 million persons are covered through various government-sponsored insurance schemes. This may be considered as one of the major policy reforms by bringing a very large pool together in a targeted manner to address health finance challenge. The annual spending by government on these schemes is about Rs 2,500 crores per annum.
   
2. Price controls by regulating the prices of selected drugs and interventions ensuring that such measures result in lower financing burdens and private providers do not take advantage of market imperfections. Healthcare access in India is affected with 70:70 paradox; 70 per cent of healthcare expenses are incurred by people from their pockets, of which 70 per cent is spent on medicines alone, leading to impoverishment and indebtedness (2). Since the spending on drugs and medicines is a significant part of total OOP expenditure on health, regulation of the price of essential medicines has been another policy option government has implemented. National Pharmaceutical Pricing Authority (NPPA) is responsible for fixing and revising prices of controlled bulk drugs and formulations and to enforce prices and availability of the medicines in the country. The NPPA also monitors the prices of decontrolled drugs. The annual turnover of the Indian pharmaceutical industry is estimated to be about 165,200 crores during the year 2014-15 of which the share of export is Rs. 78,792 crores. Nearly 680 formulations, spread over 27 therapeutic categories including HIV, diabetes, heart diseases, cancer are under price control.
   
3. Developing option of strategic buying by the government and providing these services to the people in need either at subsidized prices or no cost.  There is general belief that some of the services such as diagnostic and various investigative procedures can be produced efficiently by the private sector and governments can buy these services and make them available to people in need. These are not under price controls.

The initiatives of the Delhi state government provide several reasons for the optimism if the reforms of strategic purchasing of health services they have embarked on go well.  The option of strategic buying may also turn out to be preferable option to health insurance option.

Diagnostic and investigative procedures (DIP) market has its challenges, and it is important to understand the characteristic of this market in the background of policy option of strategic purchasing. One of the characteristics on which the markets are typified is whether they are contestable. Contestable markets are one that has no exist/entry barriers. If in that market prices increase much beyond the average price level (and generate excess profits), potential rivals will enter the market to exploit this situation. The existing players will respond by bringing down the prices appropriate to yield normal profits. The perfect contestability will make it sure that even a single player can exhibit competitive behavior.  The way the DIP markets have grown in India with no control on the certificate of need (domain and geographic specific), this market is not perfectly contestable as there are huge sunk costs creating huge barriers to exit and entry. It is difficult to find a suitable buyer of assets once it is discovered the entity is not doing well. Compounded with this is the fact that most decisions to use a particular procedure is decided by the doctor and not by the patient leading to a situation of information asymmetry between doctors and patients. Given perverse incentives, this may lead to a situation of supplier-induced demand. It has been shown that given these imperfections, insurers are not able to solve this problem. For example, insurance gets seriously affected by fee for service system which increases the discretion and unnecessary clinical interventions. Given the contestability issues even having a large number of players, the competition will not necessarily lead to efficiencies and good practices. In private health care settings, in particular, the ability to pay takes precedence over the need, and DIP pricing policies give dominance to business interests over health care. 

How does one improve the contestability in the DIP market? The reforms that separate the role of purchaser and provider by formalizing relationships between buyers and sellers in the form of contracts which articulate the volume, price and quality characteristics of transactions helps in improving contestability in health markets (1). This is the reform UK embarked on through the creation of quasi-markets when District Health Authorities were assigned the role to assess local health care needs and purchase "cost-effective" treatments to meet these needs within available budget. Under the system, the larger primary care practices could also elect to hold budgets for the purchase of diagnostics, outpatient care, and non-emergency inpatient treatments. When government intervenes through the process of creating quasi-markets, they become aggregators of demand in some sense. The insurance system is not able to deal with the fragmented providers and have to be negotiate with them individually either through TPA or insurer themselves. The aggregation of demand provides an advantage of bargaining volume based prices increasing efficiency with which government resources are used. 

The pace and consequences with which the process of strategic purchasing unfolds in the system are always unpredictable. But its effects can, however, be significant as the strategic purchasing helps in aggregating the demand for health care, unbundling and outsourcing some components within the health production process, governments get in better  position to restructure the supply-side of the market by responding to needs of the population and strengthen developing mechanisms of greater integration of primary and hospital care. Any system is fraught with some challenges.There are several reasons for optimism for these reforms, but at the same time, some reasons for caution are well-known.

How does one instill accountability and responsibility of clinicians recommending the purchase of services? Are there clear identification of roles and rules that ensure (a) what is being purchased, (b) for whom it is purchased and (c) of what quality? How does the system protect from market price manipulations and inflation in costs? Given we have limited knowledge of cost-effectiveness analysis of various technologies available, the problems of measurability and contestability associated with expensive, complex and concentrated procedures may require a stronger regulatory environment and skilled contracting mechanisms before governments can rely on obtaining these services from the private sector (1). 

1. Alan Maynard (1991) Developing the Health Care Market. The Economic Journal, 101:1277-1286.
2. Golechha M (2015) Healthcare agenda for the Indian government. Indian J Med Res 141:151-153.

Making Cost Effectiveness Analysis of Health Policy Choices Mandatory

The application of Cost Effectiveness Analysis (CEA) methodology helps in making appropriate choices in health. At present, CEA has a marginal role in health policy making in India. Many health care interventions have been introduced in the country without doing appropriate and contextual CEA. Introduction of various interventions, drugs, and procedures have used CEA methodology in a very limited manner. We have not paid attention to developing adequate capacities in this area and as a result, capacity in this domain remains scarce. Indian health policies have not made it a point to include CEA as a necessary requirement of introducing new technologies and interventions. Over the years India has compromised introduction of new and expensive technologies without doing contextual CEA, and as a result, we have adopted practices which are responsible for passing on an increasing proportion of costs to patients and in the absence of insurance protection to a large section of population, this has led to high OOP expenditures on health.

The recent example of cardiac stents and drugs which were brought under the purview of price control signal that payers may now question the transparency and force providers in making choices more apparent. This indicates an important role and need for CEA. We are now at tipping point. Indian Medical Association (IMA) report of the core-committee for revision of the national list of essential medicines (NLEM) in November 2015 had some interesting observations on this, and I hope they will be taken to their logical end. Among other things, the report suggested:

When more than one medicine are available from the same therapeutic class, preferably one prototype/medically best suited medicine of that class to be included after due deliberation and careful evaluation of their relative safety, efficacy, cost effectiveness.

The observations of the report also suggest that the price of total treatment be considered and not the unit price of medicine. This will require a massive effort of developing system of ascertaining cost of "care cycle" and not just some components of the treatment. This is an important and all stakeholders need complete understanding on this. Fixed Dose Combination (FDC) are not included unless the combination has unequivocally proven advantage over single compounds administered separately, in terms of increasing efficacy, reducing adverse effects and/or improving compliance. The medicine in NLEM will be based at primary, secondary and territory levels of health care according to treatment facilities and training, experience and availability of health care personnel at these levels. These are welcome observations from IMA. We need a strategic plan to ensure CEA capacities are created to achieve these goals.

The CEA analysis is an important instrument to inform various value-based decisions and by using the evidence produced by the CEA helps in giving access to critical medical technologies and interventions. CEA is a well-discussed approach that helps in quantifying the relative value of competing treatments by estimating the additional cost of generating an addition unit of health outcome. And based on this, CEA provides information that allows policy makers to judge whether the incremental benefit is worth the incremental cost. Of course, if we have to start integrating CEA, we have a repository of a lot of global experience in this area and initiating the process can be leveraged on this experience.

Indian Council of Medical Research (ICMR) recently took the lead in organizing a two-day workshop on March 3-4, 2017 for their senior scientists to apprise them about the CEA methodology. In the absence of a centralized agency responsible for conducting CEA, ICMR can play a pivotal role in taking this initiative forward and preparing the vision for next five years by describing areas where CEA could be initiated. Apart from having many advantages of this effort, the tasks would include standardization of methods, addressing various biases that may creep in such analysis, keeping the analysis objective and free from various vested interests, organization structure, and processes. Along with the capacity building, the strategy would also need developing collaborative linkages across various institutions and make the participation of health economists possible in this initiative.

The other stakeholders also have a vital role in the system. For example, private insurance market and IRDA must also contribute to this effort and institute framing guidelines for making CEA of interventions a mandatory requirement for all health interventions. People who buy voluntary insurance spend about Rs 22,000 crore in 2015-16 on covering their hospitalization risks. Government sponsored insurance schemes contribute about Rs 2,500 crores.  Both these put together cover about 36 crores of the population. For both these segments, insurance companies have settled claims to the tune of Rs 21,760 crores in 2015-16. The payers and those who pay premium have right to know whether their contributions are being spent on correct and cost effective medical procedures and interventions. In a system where insurance is growing at CAGR of 17 percent, CEA provides information which payers would need to judge the relative value of competing interventions and opportunity cost of these interventions. Insurers must insist on CEA information to inform their critical decisions.

What is wrong with Group Health Insurance Business?

Health insurance market in India regarding the gross premium collected during the year 2015-16 stood at Rs 24399 crore ($3.6 billion). The government sponsored insurance programme contributed about 10 percent of this premium, and 90 percent was the voluntary group and individual health insurance plans.

The number of people covered during the same year has been 35.90 crores of which 27.33 crores are covered by government insurance schemes which include RSBY and other state government run insurance schemes. Private voluntary insurance covers about 8.57 crore individuals.

The average unit premium of three broad classes of insurance presents an interesting picture. The per capita insurance premium is highest in individual voluntary schemes at Rs 3607 and group premium is about 60 percent of the individual premium. The differential between government sponsored insurance, and private individual voluntary schemes is to the extent of 40 times (see Figure below for comparison). The membership differential between these two schemes is about ten times.

Per Capita Insurance Premium (INR) 

Data Source: IRDA Annual Report 2015-16

At the theoretical level, the dynamics of health insurance markets is well known. The insurers operate in situations where demand-supply conditions do not adequately reflect the real preferences, and because of information asymmetry, the market will always be in flux. The two key risks viz., adverse selection and moral hazard problems influence the pricing and performance of insurance schemes. Insurer's design of plans include conditionality to protect themselves from such hazards. The co-payment, deductibles, and exclusion clauses are some of the instruments used by the insurer.

Among other measures, the group insurance schemes are expected to have significant economies of scale and writing such policies may reduce the adverse selection problem by improving the risk pooling. One of the measures to monitor the sustainability of insurance pool is the incurred claim ratio, which is measured by dividing the net incurred claims by net premium collected. This ratio across the three segments of insurance present an interesting read.

Claim Ratio Across Different Health Insurance Segments for 2015-16

Data Source: IRDA Annual Report 2015-16

The claim ratio of group plans across all three types of insurers are higher. For public sector insurer the claim ration is 123 percent, where as for private and standalone insurers it is 98 percent and 87 percent respectively. The expected claims in group plan should have been lowest. This raises some interesting research questions:

• Do we lack appropriate risk rating in group health insurance plans?
• Are market failures in group-plans more severe than individual markets?
• Do good corporates retain employee health risk and do not buy insurance?
• Is there an overselling of group plans to meet the revenue target of the insurer as the reason and insurers take more risks?
• Are the group plans more liberal in using the preferred provider and receive less scrutiny of claim authorization and claim processing?

Number of Cardiac Stents Implanted in India

Data collected from various sources and estimation made for 2016 on cardiac stents implanted in India over the years is presented below.

We observe a spike in implantation data in 2010 and 2014, the two years where the increase has been significantly high. What was significant about these two years? The general discussion on this issue suggest that not all implantations are necessary. Knowing the fact that there is serious supplier-induced demand problem in general in health care, it may be worth examining how much of this implantation are necessary and how much fall in the category of really "not-needed"?  On financing side, it may be interesting to find how many patients paid the cost "out-of-pocket" and how many were insured? It is estimated that about 60 percent of these stents are imported. It may interesting to examine the political economy of stents. 

When it comes to health, we are really in a data starved situation and can not argue effectively on many policy issue. Some speculations are, however, inevitable. Indian medical system predominantly works on fee-for-service based payment method. In this context understanding, the impact of the price cap on stents would be an interesting area to research. Prediction of a few consequences include: (a) the price control will weed out unnecessary implantations because of removal of perverse incentives and thereby resulting in lower implantations - a desirable outcome, (b) hospitals will reallocate resources affecting lower volume of implantations because treatment of patients becomes less remunerative as compared to relative activities - called substitution effect, not desirable from overall societal viewpoint, (c) hospitals modulating and strategizing the service configuration and increasing prices of other services - price effect, non-desirable, (d) hospitals increasing the volume of implantations to compensate for loss of revenue - revenue effect, desirable and (e) the formation of cartels by companies to weaken the price control. On the last one there is an interesting study published on mitigating regulatory impact: the case of partial price controls on Metformin in India, Health Policy and Planning 2017. The study found that firms coordinated to increase the price of the regulated formulation in the period before regulation, which led to a higher ceiling price and this coordination was stronger among larger firms and for time-release formulations. The study presents anecdotal evidence to suggest that pharmaceutical trade associations facilitated coordination among firms, and conclude that partial price control of Metformin in India is, at best, a modest improvement over no regulation. 

Which effects will play out in actual practice are hard to predict. Some studies from insurance based markets in US suggest when the reimbursements in insurance are reduced, a 1% reduction lead to 0.40% increase in treatment volume. This is will happen when hospitals follow target revenue approach. What will be the effect of the price caps on practice is an important area of research. One also wonders that why Indian healthcare providers  have not been able to do what Aravind did in manufacturing of intraocular lenses at an affordable cost.