Universal Health Care Coverage: Title and subTitle BreakA Potential Hybrid Solution

The United States spends a greater share of its gross domestic product on health care than other nations, yet more than 15.9% of its population lacks health insurance coverage.¹ The US health measures are no better (and are often worse) than those of other nations.² This combined inequity and inefficiency creates a moral imperative and opportunity for change. Some argue for a single-payer system, others for more reliance on the market and insurance subsidies. Apart from causing political gridlock, neither strategy would effectively both guarantee coverage and constrain costs. A hybrid approach combining universal risk pools, mandated coverage with income-based subsidies, and a restructured payment mechanism has the potential to improve both equity and efficiency.

Medicare, which provides near-universal coverage to US residents 65 years and older, is the prototypical single-payer model and routinely exhibits the problems of the model. Although permitted to arbitrarily set fees, Medicare has found it difficult to do so effectively. Across-the-board fee changes elicit broad-based political reaction; narrowly focused changes draw sub rosa special-interest lobbying.^{3 - 5} While patient advocacy groups, often supported by industry and specialty societies, encourage coverage for specific services, significant co-payments result in 91.8% of Medicare beneficiaries offsetting such incentives through supplemental insurance.⁶

Rather than market discipline, Medicare is subject to political manipulation and bureaucratic rigidity. Single-payer advocates envisioning an equitable and efficient health care system idealistically disregard the example of Medicare and the ethos of the US political system. An approach guaranteeing universality while minimizing political interference is necessary.

Most Americans younger than 65 years obtain health insurance through employment-related plans. Devised to sidestep wage and price controls established during World War II, the continued popularity of employer-sponsored plans is largely due to the tax subsidies they accrue; more than $200 billion, or 35% of the premium for 2006.⁷ There are substantial gaps in coverage, however, especially for part-time and “contract” workers. Many firms do not offer coverage, and the proportion of employees enrolling when coverage is offered has been shrinking.^{8 - 12} The subsidized employer contribution rarely covers the full premium; the net cost is often too high for younger workers, whose expenses rarely surpass their out-of-pocket premium.^{9 - 13} Employment-based plans do not pool potential contributions from all wage earners in a family or those with multiple jobs. Nonetheless, many reform proposals focus on incentives or mandates for employers to offer coverage, ignoring problems of employers sponsoring specific plans.

In most European countries, employers and employees contribute a share of payroll to an insurance pool covering everyone. In the United States, employers select 1 or several plans and then tailor them in idiosyncratic ways. The overall premium reflects the benefit structure and the average covered expenditures of those choosing to enroll. A new coverage and payment mechanism must be devised that aggregates employer contributions across multiple workers and jobs, avoiding the fragmentation and inefficiencies of sponsored plans.

Medical costs are highly concentrated in a small fraction of the population.¹⁴ This provides a rationale for insurance and might lead to voluntary enrollment if the need for care were purely random. Health problems, however, increase with age, and most plans' costs reflect the needs of older enrollees, raising premiums for younger enrollees. Individuals with chronic illnesses predictably have higher than average costs (even for their age), so insurers demand higher (often unaffordable) premiums. Premiums reflecting “average cost” combined with voluntary enrollment encourages those expecting to be healthy to decline coverage, especially if a safety net provides emergency and hospital care. Those likely to be sick will face ever-increasing premiums, and voluntary coverage will continue to decline. Without precise and accurate risk adjustment—a difficult prospective task—mandating coverage and allowing choice of plans is unsustainable, because plans will seek to enroll the healthy and avoid the sick.

Instead, a single risk pool is needed to cover the types of high-cost events that cause voluntary models to fail. These include costs associated with hospitalizations and expensive outpatient facility–based services, as well as with chronic illness. Based on enrollees in a set of employment-based health plans, the first category accounts for roughly 4% of all episodes of care but 34% of all costs; the second, about 21% of care episodes and 28% of costs. The remainder—38% of all costs—is spread over 75% of all care episodes, mostly for minor acute-care problems and some preventive care. (J. O’Connor, Ingenix Inc, written communication, June 2006. From a data set of individuals with employer-based coverage during 2003-2004 [N=796 635]. Other samples and episode groupers may yield somewhat different results but are unlikely to alter the major findings with respect to concentration of costs.) Coincidentally, individuals with private insurance pay out of pocket for 30% to 35% of their medical costs due to coverage exclusions, deductibles, coinsurance, and balance-billing.¹⁵ This suggests there is already roughly enough money in the current insurance system—it simply needs to be more purposefully allocated.

The risk pool should cover everyone who might consume health care—with the possible initial exceptions of Medicare beneficiaries and those receiving care through the Departments of Defense or Veterans Affairs. A universal pool eliminates the adverse selection problem inherent in voluntary coverage. Whether public or semi-independent like the Federal Reserve, the pool must be designed to avoid the pressures to which a single payer is susceptible: special-interest pleadings and preference for the status quo. The risk pool transparently spreads and reallocates risk. Payments from the pool are structured to give consumers, clinicians, and payers a stake and influence in how resources should be used and paid for, with incentives for efficiency and quality.

For inpatient and similar services, the pool would build on the Medicare model for payments based on diagnosis related groups (DRGs), expanding DRGs to include the professional fees and services associated with an episode of care, usually a hospitalization. Clinicians and facilities would create formal or informal care delivery teams (CDTs) to treat patients and allocate funds among themselves. Using increasingly refined measures of quality over time, the pool would update payments for health care episodes to reflect the average costs incurred by those CDTs with better than average outcomes. Internal efforts by CDTs to improve efficiency will likely result in better care at lower cost.

Although payments from the risk pool would reflect both the resources used by those CDTs providing above-average outcomes and the wage rates in a CDT's locality, some clinicians and facilities might want higher fees and some patients might want services with little demonstrated effect on health outcomes. Supplemental payments could be made by voluntarily purchased insurance, with premiums reflecting the preferences enrollees have for such care rather than their underlying risk, which would be offset by the pool. For instance, Medicare data show enormous differences across small geographic areas and hospitals in intensive-care use by patients in the last 6 months of life.^{16 - 19} When voluntary coverage payments increase within a geographic area or medical subspecialty, the pool would be alerted that adjustments in payments are needed to reflect new services or shortages of clinicians.

The cost of ongoing management of chronic illness (acute exacerbations are included in the expanded DRG payments) would be offset by monthly payments from the pool. These would cover the cost of the clinical services, drugs, and devices used by clinicians managing similar cases and achieving better than average outcomes. Risk factors within a chronic condition, such as cancer stage, or New York Heart Association severity score for congestive heart failure, would be used to adjust payments and outcomes. Organizing pool payment around conditions, rather than specific services, would allow more flexibility in treatment, simplified billing, and a much greater focus on outcomes and improving processes.

Creating CDTs to accept episode-based payments is a challenge, but it is surmountable, given DRG payments to hospitals and the growing role of hospitalists. Allowing supplemental coverage and targeting pool payments to the amounts needed for high-quality care should encourage the various clinicians and health care organizations to cooperate. Paying for care in the ambulatory setting is more complex, as it involves numerous clinicians who may not work together and a wide range of diagnostic and referral services as well as pharmaceuticals. An intermediary, with administrative flexibility but not overwhelming market power, would accept the monthly payments from the pool and, in turn, pay various clinicians and health care facilities. Health insurers currently play such an intermediary role for Medicare, bearing no risk but handling all claims within defined geographic regions.

The risk pool eliminates the necessity for employment-based group plans. Instead, payment intermediaries could be organized in a more purposeful way—around the physician. A physician would choose 1 carrier (or intermediary) to handle receivables from patients and other carriers. The carrier would offer patients a range of plans with varying deductibles, co-payments, and breadth of preferred-provider networks. By handling all services provided or recommended by the physician, the carrier would have data on the physician's practice style: eg, long office visits and few referrals, or short visits with many tests.

Premiums charged to the patient would reflect the net effects of coverage choices, negotiated fees, network, and practice style, after taking into account expected payments from the risk pool for major acute episodes or chronic care. Transparent pricing would show the expected overall payments for care, the share borne by the pool and the patient, and the carrier's retentions. Carriers would pay for out-of-network services and bill for appropriate co-payments. They would have no reason to interfere with physicians' practices; differences in efficiency and fees would simply be reflected in the premiums quoted and out-of-pocket costs. More importantly, carriers would compete for physicians based on their effectiveness in managing receivables and in providing information on clinical practice styles and quality-improvement efforts.

For most ambulatory care for minor acute and chronic illness, payment to clinicians is likely to be on a fee-for-service basis, with patients responsible for premiums and varying co-payments, depending on the plan selected. These premiums and co-payments should be subsidized for low-income enrollees. Large group practices might choose to accept payment in a more bundled form, allowing them even more flexibility in resource allocation, but would still be required to report measures of outcomes and quality.

Everyone would automatically be enrolled in the pool, and payment for major acute and chronic care would “follow the patient.” Funding of the pool by a broad-based tax, as for Medicare, is preferable but may be politically unacceptable. Substituting a payroll tax for current employer contributions creates transition problems, with windfalls for some employers and sharp cost increases for others, reflecting in part the average age and health of their workforce. Economists agree that, in the long run, employees bear the costs of employer-sponsored health insurance.²⁰ The tax subsidy makes benefits generally more attractive to workers, but the way premiums are set makes benefit coverage unattractive to some.

A small mandatory minimum employer contribution of 4% to 5% of payroll could be absorbed more easily than mandating that employers purchase insurance for all employees. To ease the transition from the current fragmented plan-based system, employers would buy into the pool at the average cost for people in a given age or geographic category. This would be attractive for firms with sicker enrollees. Firms with healthier than average (for their age) enrollees might join the pool because their transaction costs would be lower and the rate of increase of their costs reduced. Employers currently offering contributions greater than those required for the pool would be unlikely to eliminate them because of the 35% tax subsidy; this “coverage” could be used for minor acute and preventive care.

Public programs such as Medicaid would also join the pool. Because they often cover individuals because they are sick, these programs would turn over their full payments for major episodes and chronic illness in exchange for the pool taking on those obligations. Likewise, public funds for safety-net hospitals would be transferred into the pool, with clinicians and hospitals drawing funds as needed for their patients. Pool payments would be equalized, markedly improving access for those whose current plans pay fees so low that few clinicians are willing to participate.

The funds necessary for this approach are already available for those with health care coverage. Hidden charge shifts and public safety-net programs already pay for much of the costs of the uninsured. Currently among individuals reporting no insurance in the month prior to hospitalization, only 15% of their hospital expenses were borne by their families; the remainder was covered by plans and hospital and clinician write-offs (W. Yu, MA, Agency for Healthcare Research and Quality, written communication, December 14, 2006; based on the Medical Expenditure Panel Survey). By aggregating funds and lumping payments through the risk pool, hospitals and clinicians would no longer be forced to be involved with cross-subsidies or to take on substantial bad debt.

The risk pool would cover expenses for major acute and chronic care. The remaining expenditures, 75% of the episodes and 38% of the costs, could be paid for with a debit card or by using a health plan. Wealthy individuals might choose to pay out of pocket, but most would probably select a plan from a menu offering choices with respect to coinsurance, deductibles, and network breadth, and with premiums reflecting their physicians' practice styles. For individuals with low incomes, such premiums and out-of-pocket costs may deter getting needed care, including prevention and early intervention that could save the pool money in future expenses. An expanded income-based subsidy, analogous to the earned income tax credit, combined with a requirement that each person have evidence of ability to pay for minor and preventive care, would increase efficiency and equity. Some funds for such coverage are already included in Medicaid program expenditures.

This fundamental restructuring of the payment system would achieve both universal coverage and improved efficiency. Focusing attention on patient outcomes would free clinicians and hospitals to creatively explore ways to deliver care and eliminate payers' focus on fee constraints and micromanagement of clinical decisions. The government's role in the operation of the system would shift to ensuring information availability and transparency in payment. Government would maintain current employment-based subsidies (if a payroll tax is not substituted) and implement income-based redistribution for individuals with low incomes.

A collective risk pool would reallocate funds so all can access appropriate care regardless of their individual health status. The simplified system would eliminate unnecessary administration. Market-disciplined carriers would facilitate payment, provide information, and respond to patient preferences. Appropriate incentives would help ensure that health care expenditures are driven by informed patient and clinician decisions about the care needed to achieve high-quality outcomes.