Conceptually, total health care costs can be contained or decreased only by some combination of the following:

• Decreasing use of health care services
• Decreasing reimbursement for services that are used
• Decreasing overhead (payor, provider, or both)

Some strategies adversely affect access to care or outcomes; others may improve care. Evaluating different strategies is difficult, partly because accurately measuring patient-centered health outcomes (eg, morbidity and mortality, quality-adjusted life years [QALY]) tends to be expensive and to require large numbers of patients and long follow-up periods. As a result, most measures used to assess health care quality reflect processes (how care was delivered) rather than outcome. How well these process measures predict ultimate health outcomes is not always clear.

Decreasing Use of Health Care Services

Many strategies can decrease the use of health care services. Many involve limiting access to care (aimed at unnecessary care but sometimes affecting necessary care), but some limit need by improving health.

Limiting access to health care: Traditionally, limiting access has been the strategy used to limit health care costs.

Insurance companies may limit access to care by denying coverage to people likely to need care (eg, those with preexisting conditions) and by dropping coverage of heavy users (rescission).

Government may tighten eligibility criteria for medical assistance programs.

Payors may increase out-of-pocket costs, providing an economic incentive for patients to limit their own health care use. For example, payors may

• Limit the type and number of visits that are reimbursed (eg, mental health care, physical therapy)
• Increase deductibles and co-payments
• Decrease allowable amounts for covered procedures
• Establish or decrease lifetime maximum expenditures

These strategies probably affect outcomes because evidence indicates that many patients avoid necessary as well as unnecessary care. For example, women may avoid screening (eg, Papanicolaou testing, mammography) and subsequently present with late-stage cancer; at-risk patients may avoid influenza vaccination.

By erecting administrative hurdles to care (eg, requiring approval for tests, referrals, and procedures; having complex enrollment procedures and regulations), payors, although not technically denying care, decrease use by a small amount.

State agencies may limit issuance of construction permits for new facilities and laboratories (called certificates of need).

Limiting access to health care can cause problems. For example, when people denied access become seriously ill (which is more likely when routine care is lacking), they are often treated in a hospital on an emergency basis. This care is largely uncompensated (not paid for by patient, insurance, or other source), increasing the burden on people who pay into the health care system, and may be more expensive than if routine care had been provided.

Eliminating unnecessary care: Unnecessary care is easy to define (care that does not improve patient outcome) but often difficult to recognize and still more difficult to eliminate. First steps include conducting more and better studies of comparative effectiveness and cost-effectiveness, so that best practices can be identified. Comparative effectiveness studies can evaluate areas other than drugs, such as effects of exercise, of physical therapy, and of different providers, systems, settings of medical care, and reimbursement systems. Education and monitoring of providers may decrease practice variation and increase cost-effectiveness. Eliminating the economic incentive for providing more intensive care (fee-for-service model) by using prospective payment systems (see below) and pay-for-performance models may encourage providers to eliminate cost-ineffective care processes.

Better coordination of services among providers (eg, by closer communication and use of universally readable electronic medical records) may make evaluation and treatment more efficient (eg, by eliminating duplication of tests).

Encouraging palliative hospice care, when appropriate, may help decrease use of costly, often technology-intensive, cure-directed care.

Improving health: Increased use of relatively inexpensive preventive services (eg, screening, diagnosis, and treatment of diabetes, hypertension, and hyperlipidemia; screening for breast and colon cancer) may decrease the subsequent need for expensive treatments (eg, for MI, stroke, or late-stage cancer). However, preventive measures may not decrease costs for a given private insurance company because savings are often not realized for many years; by that time, many patients have switched insurance plans. In the US, people stay with a given insurance company for an average of about 6 yr (usually determined by how often they change jobs)—too short to realize a savings via preventive care.

Strategies to increase preventive care include

• Incentives to increase the number of primary care physicians (who can often provide appropriate screening measures and help prevent complications)
• Pay-for-performance measures that financially reward adherence to preventive care guidelines
• Elimination of co-payments for preventive services
• Free preventive services, particularly for needy people

Whether care management programs that attempt to improve patient adherence to treatment plans and clinician adherence to guidelines can improve outcomes or reduce costs (eg, of potentially avoidable hospitalization or complications) is unclear; some studies do not show a benefit.

Decreasing Reimbursement for Care Used

Even when health care is provided, strategies can be used to limit payments.

Lower fees: Payors (government and private) may negotiate lower fees with institutions and providers or simply dictate such fees. In the US, reimbursement rates established by Medicare and Medicaid tend to influence rates paid by other plans, sometimes decreasing reimbursement.

Increased use of primary care: Measures may help increase the use of less costly primary care vs specialty care. For example, in the patient-centered medical home model, primary care practitioners coordinate and integrate all aspects of medical care, including specialty and interdisciplinary care, in various settings (eg, home, hospital, long-term care facility). Many authorities think that this model can decrease unnecessary specialty care, duplicative care, and care that may be inappropriate for the individual's health goals (eg, palliation rather than diagnosis).

Measures to increase the supply of primary care physicians have been proposed. They include increasing reimbursement for primary care, shifting more government funding of residency programs to primary care training, and enhancing the attractiveness of primary care among medical students, although how the last strategy could be implemented is unclear.

Prospective payment systems: In these systems, providers are paid a fixed amount regardless of how much care is provided. The amount may be based on a specified episode of care or be a fixed annual reimbursement per patient. For example, some Medicare reimbursement is based on diagnosis-related groups (DRGs); in such cases, Medicare pays a fixed amount based on the diagnosis. In capitated systems, providers are paid a fixed annual amount to provide health care for patients regardless of the services used.

Prospective payment systems reward less expensive care (and thus usually use of fewer services), in contrast to fee-for-service systems, which reward use of more services. However, prospective payment creates an economic disincentive to care for complex patients (eg, those who have multiple disorders or who are seriously ill) and may inhibit provision of necessary care. Because a decrease in the amount of care provided has the potential to decrease quality of care, quality control systems (eg, professional review organizations) are often also established.

Denial of claims: In the US, unlike in most of the developed world, insurance carriers routinely deny a significant percentage of claims for services delivered to patients. In one study in California, the denial rate averaged about 30% in 2009; some of the claims were paid after appeal, but appealing a claim is quite costly in time and effort for patients, providers, and payors alike.

Competition: Competition among providers for patients and among insurance companies for subscribers is thought to encourage lowering of charges (eg, by those who charge more than their competitors for a similar service). However, the ultimate consumers (ie, patients) usually do not know providers' charges in advance, and if they know, they often cannot act on this knowledge (eg, because patients are often limited to certain providers and limited in their ability to judge quality of care). Also, because the cost of medical care is subsidized for most consumers (eg, through employer-paid health insurance, tax deductions, and flexible spending accounts or medical savings accounts), consumers have less incentive to price shop than for most other purchases. Thus, competition is most effective in lowering costs and maintaining quality when it is among large organizations. For example, insurance companies can compete for contracts from employers such as corporations or the government; providers such as practitioner organizations and hospitals can compete for contracts with insurance companies.

Competition has some disadvantages. It results in multiple systems of claim submission and evaluation, which require more time from providers, their clerical staff, or both. Also, processes such as eligibility determination, referrals, co-payments, and coding must be coordinated between a large number of incompatible insurance company systems. Thus, competition increases the clerical (administrative) burden of the overall health care system.

Decreased drug costs: Using generic drugs or, when appropriate, more cost-effective brand-name drugs can help decrease drug costs. Strategies include

• Educating providers about cost-effective drug use
• Restricting drug marketing
• Establishing formularies and using pharmacy benefit managers
• Allowing the government to negotiate drug prices for patients covered by government insurance
• Allowing importation of drugs purchased from other countries to the US

Negative effects on medical research: In many academic medical centers, income from clinical practice has enabled physicians and institutions to participate in medical research. Similarly, income from drug sales supports pharmaceutical research. Thus, decreased reimbursement for care and drug sales may cause a decline in medical research. If other sources (eg, government or private grants) are used to fund research, these funds must be considered as health care costs and thus may offset savings realized from decreasing reimbursement.

Decreasing Overhead

Overhead is health care payments that do not go to health care providers (eg, administrative costs, malpractice insurance, corporate profits in for-profit hospitals and insurance companies).

Decreasing payor overhead: Government health care plans in developed countries (including the US) and private health plans outside the US have overhead costs that usually represent 3 to 5% of total costs (ie, ≥ 95% of all health care funds go to the delivery of health care). However, in the US, private insurers have overhead costs of about 20 to 30%, partly because these insurers need staff to do extensive underwriting (identifying and rejecting applicants likely to require costly care, including those with preexisting conditions or a high likelihood of developing disorders), to evaluate claims for denial, and to adjudicate appeals by providers; they also typically need to show a profit. No evidence indicates that these activities and their higher administrative costs improve clinical care or outcomes.

Strategies that may help minimize overhead costs include

• Increased use of standardized electronic health records
• Increased use of government plans and possibly not-for-profit plans, which have lower overhead than for-profit plans

Competition among payors is thought to encourage increased administrative efficiency, but it also increases the incentives to deny claims and coverage (which itself requires an extensive bureaucracy).

Decreasing provider overhead: Any payor reform that eliminates the need for the many billing and claims personnel who manage the billing of multiple payors and negotiate appeals and justify claims will decrease provider overhead. For example, some countries that have multiple insurance companies vying for business (eg, Germany, Japan) require the following:

• The payment amounts and rules are the same for all insurance companies.
• In many cases, payors are required to pay all provider bills.
• The cost of the same service is the same throughout the country.

Although malpractice costs are a small fraction of overall costs, malpractice costs for certain physicians can consume a considerable part of their annual income. Reforms that significantly decrease the number of suits and settlements should eventually lower premiums and greatly benefit these physicians; such reforms may also decrease the use of unnecessary, defensive medicine.

Last full review/revision February 2010 by Amal Trivedi, MD, MPH

Content last modified February 2012