Primary Health Care Financing Systems:
International Comparisons and Lessons

JOHN MINOT, ARIJIT SARKAR

Introduction

John Minot was raised in Austin, Texas, to Northern parents as a cultural transplant to the South. He attended Cornell University’s College of Arts and Sciences, graduating in 2006 with a double major in Government and Asian Studies. He spent the next three years at the Bethesda, MD consulting firm Pacific Bridge, which mostly assists medical firms with regulatory requirements for doing business in Asia. After leaving the firm in April 2009, he spent three months in Chennai, India, as an intern at the IKP Centre for Technologies in Public Health, researching primary healthcare financing systems worldwide. He is now enrolled in the Goldman School of Public Policy of the University of California at Berkeley, and expects to graduate as a Master in Public Policy in May 2011.

Arijit Sarkar is Vice-President, Health Care Solutions at ICTPH. His primary interest is to invest in and provide support to socially relevant and sustainable business models, with a particular focus on nascent and early stage entrepreneurships. He has two years of financial modeling experience. His previous role was that of a quantitative research associate at Lehman Brothers, where he built models of liquidity in public equity markets. He has a bachelor’s degree in Electrical Engineering from IIT Bombay.

Primary healthcare is one of the most immediately familiar aspects of health systems. Whether it is at a private practice, a government-run health center, or a health mission having a place to go for non-emergency complaints and regular checkups, it is an integral part of healthcare access.

Over the past 60 years, primary care has also been seen as a vital mode of intervention, though with varying justifications. As early as 1946, the Bhore Committee recommended the construction of a health system in India based around non-specialized primary services, which led to today’s Primary Health Centres (PHCs) and Sub Centres (SCs). In 1978, a commitment to primary care was affirmed internationally in the Alma Ata Declaration, as part of sustainably addressing the developing world’s health problems, especially infectious diseases. Thirty years later, this declaration has had mixed results, as free public primary care systems in the developing world, including India, generally continue to stagnate due to insufficient funding (Peters 2002).

In the developed world, primary care has gained respect recently for different reasons. In American discourse, one of the first research results in this direction was the finding of the Dartmouth Project that in areas with a greater proportion of primary care practitioners (PCPs) to population, Medicare outcomes improved. Since then, a wide variety of reports have shown increased primary care supply to improve mortality, infant mortality, low birth weight, and mortality from a wide range of causes, mostly noncommunicable diseases. These results have been quite robust, holding up after controlling for many confounding factors, including income, education, percentage of elderly, and even lifestyle factors like smoking and obesity (Starfield et al. 2005). They also extend beyond the United States. Across 18 OECD countries from 1970 to 1998, stronger primary care systems are correlated with lowered mortality, premature mortality, and premature mortality from a variety of specific conditions (Macinko et al. 2003). Primary care is also believed to lower costs by preventing many hospitalizations (Jamison et al., eds., 2006:1240).

In the developing world, the justification for primary care is less founded on experiential evidence and centers around the utility of available treatments for different diseases. In developed countries, infectious disease has been largely brought under control, and the more difficult cases – cancer, heart disease, and other chronic conditions – are considered the pressing task for health systems to address. However, primary care can potentially help make up a great deal of the health gaps in India and the developing world at large. As described in Disease Control Priorities in Developing Countries, interventions for maternal and childhood diseases, malaria, tuberculosis, HIV/AIDS, hypertension, etc. are highly cost-effective (in terms of disability-adjusted life years (DALYs) and well-adapted to prevention, treatment, and management at the primary level (Jamison et al., eds., 2006).

The health situation in India today also indicates the need for greater access to and quality of primary healthcare. The current structure of government-provided “free” care provision includes PHCs and SCs which are explicitly designed to prevent and treat basic health issues that only need medical professionals up to the level of general practitioner (GP) or family doctor. However, this system has enough shortfalls – in funding, human resources, governance, etc. – that the World Bank’s Human Development Network believes it would not be cost-effective to give the system the necessary resources to perform its intended mission, even if such money existed (Peters 2002). More promising will be new schemes that complement and supplement existing public and private care, conceivably eventually to supplant them if successful.

At the same time, while an extensive private health provision network exists in India, it may not be adequate to the task of providing primary care to all. It is particularly scanty in rural areas, and even has affordability problems. According to the National Sample Survey, in 2004, economic status was a significant factor in whether ailments were treated in any way; for rural residents surveyed with a monthly per capita expenditure (MPCE) of less than 225 rupees, 76% of spells were treated, whereas for those with an MPCE of 950 rupees and up, the figure was 89%. Of those rural Indians who received no treatment, 28% said it was for financial reasons, 32% said it was because the ailment was not considered serious (which is partly an issue of affordability as the “seriousness” of an ailment is weighed against the cost of care and lost income), and 39% gave various other reasons (National Sample Survey Organisation 2006).

Therefore, it is likely that India’s health problems call for a system that makes primary care available and affordable to all (probably together with secondary and tertiary care). For reasons of sustainability, if such a system is built, it will almost certainly take some form of contributions from its users, rather than solely public funding. It may also take advantage of the wide existing network of private primary care providers in order to put at arm’s length actual management of provision.

In any such system, with or without private providers, the exact method of financing is a complicated decision which significantly influences how actors make use of the system. It affects utilization, moral hazard, physician decisions, and more. Exact methods of financing are the topic of this white paper.

There are arguments that difficulty in affording primary care is not a major concern. Most of these arguments stem from the Rand Health Insurance Experiment (HIE), which found that health insurance plans varying by amount of cost-sharing (from 0% to 95%, with a maximum expenditure level) reduced utilization and costs but did not affect health outcomes overall. It is generally interpreted to mean that a large proportion of normal care – especially at the primary level – is sought largely for patients’ peace of mind, and discouraging it through cost-sharing therefore reduces incidence of care to when it is truly necessary (Newhouse 1993).

If this is indeed the case, it would imply the best policy solution is guaranteed catastrophic coverage, possibly paired with a medical savings account (MSA) to ensure that funds will exist to purchase normal care. However, there are a number of criticisms of this interpretation of the Rand HIE. In particular, analysis suggested that patients in high cost-sharing plans did not just cut out unnecessary care, but care generally whether necessary or not. Indeed, it is common for normal patients to be unable to determine the necessity of care on their own. The most relevant fact for applying the Rand HIE results to India is that one of Rand’s own analyses (Keeler 1992) showed that preventive care as a category – one of the most important functions of primary care for the Indian situation – markedly declined for those in high cost-sharing plans. Also, according to Hudman and O’Malley 2003, low-income children in high cost-sharing plans often did not receive care for several conditions very widespread in India, such as bronchitis, diarrhea, influenza, acute upper respiratory infection, gastroenteritis, etc. Finally, Normand 1994 observed that while the health of the experimental subjects as a whole was unaffected by cost-sharing, the health of those classified as “sick poor” (about 6% of the population) did decline. While this observation is vulnerable to charges of cherry-picking, the sick poor are the segment of the population who need effective and affordable care most of all.

There is not enough data for this white paper to conclusively point to the superiority of either the insurance model or the savings model for primary care. However, since most developed countries have converged on an insurance model where primary care is cheap or free, the bulk of the data presented will be on such systems. Singapore, and the achievements of its MSA model, will also be briefly discussed.

  Next Section: Objectives »