The U.S. Department of Health and Human Services recently projected that health care spending will double to $3.6 trillion by 2014, consuming 18.7 percent of the nation's economy. Inflation is expected to add up to 30 percent over the same period. Why does health spending continue to rise at a level that far outpaces inflation? There are as many expert opinions about this as there are experts.
Increases in health expenditures per capita across different countries are actually fairly similar—averaging about 3 percent a year adjusted for overall inflation. So it's likely that the basic drivers are new and better technology—which can at times lower cost but may also improve outcomes at higher cost—and the cost of labor in a service-intensive industry. But the U.S. does have much higher costs than other countries, and we might achieve savings by learning from international experience.
For too long, U.S. employers and policy officials seeking to reduce costs have focused on the "demand side," looking for short-term demonstrable savings, the quick fix. But if we continue to do what is expedient, we will get the same unsatisfactory outcomes: more and more complexity and fragmentation and spending that continues to grow rapidly as a percent of the nation's economy. It is time to devote serious research, demonstration, and policy development to the "supply side" of the health care market—a strategy that has worked in many other countries.
Here are 10 approaches that show promise for reducing health care expenditures. None of these would be easy to implement or painless for those directly affected. Each is a real-world example of what's possible if we put our minds to making changes that not only will save money but, in some cases, will likely save lives as well.
1. Reduce hospitalization of patients with high-cost conditions.
Ten percent of all Americans account for 70 percent of health care costs, or $1.2 trillion in 2005. But many hospitalizations and hospital readmissions could be averted through proper monitoring of patients with chronic conditions. Want some specifics?
Congestive heart failure is the leading cause of hospitalization among Medicare patients. Approximately 20 percent of heart failure patients are readmitted within 30 days of hospital discharge and half are readmitted within a year. But if these patients were given information on self-care, and if their conditions and medications were properly monitored, between one-quarter and one-half of these readmissions could be prevented. One study found that annual health care costs for frail elderly patients could be cut by 36 percent if advanced practice nurses saw patients and their families in the hospital and in their homes. The basic problem is that insurance often fails to cover the services required to achieve these savings. This model of "transitional care" is being tested by Aetna in a Medicare chronic care improvement project. If results are favorable, advanced practice nurses or patient education services for high-risk patients with congestive heart failure should become covered Medicare services, conditional on ongoing monitoring of quality and costs.
Asthma is another condition for which monitoring patients at home can save money. One study found that emergency room use and hospitalization of high-risk asthmatic children could be nearly eliminated with use of a simple handheld computer. The "Asthma Buddy" prompts patients to answer questions about their condition, peak flow rate, and medication use and then uploads this information on a daily basis to a trained medical monitoring team.
Monitoring of certain medications, including anti-coagulant and cholesterol-lowering drugs, can greatly improve patient adherence and long-term outcomes. One study found that, in just one year of pharmacist team monitoring, the number of high-risk patients with controlled cholesterol increased from 53 percent to 84 percent.
Another study finds that, in 2001, a third of diabetes patients were hospitalized two or more times for diabetes or related conditions, including many hospitalizations that could be avoided with better management of chronic care.
2. Reduce variation in charges for patients with similar conditions.
John Wennberg and colleagues at Dartmouth Medical School have found wide variation in Medicare charges for patients with the same condition, largely due to variation in the number of physicians involved, specialty consultations, and lengths of stay in intensive care units. This leads to increased spending, and it needn't be so.
For example, among the 25 Pennsylvania hospitals that treat the greatest number of heart attack patients, total charges for management of this condition vary from $11,000 to more than $88,000. Such variations are too large to be explained by geographic variations in wages or other inputs. There were no statistically significant differences in mortality rates between high-cost and low-cost hospitals. Pennsylvania is one of the few states that posts information on individual hospital charges and mortality rates adjusted for case-mix severity on a public Web site. Other states could and should follow this lead, including an explanation of how charges and outcomes are adjusted for case-mix severity and geographic differences in the cost of providing care.
Medicare, Medicaid, and private insurers should pool data on provider charges and patient outcomes, establish median total charges for patients with certain conditions, and set maximum payment caps based on these rates instead of on units of service (e.g., hospital days, physician visits, procedures, or consultations). It makes no sense for insurers to pay widely different rates for comparable care and similar outcomes.
3. Reduce overuse of medical procedures.
There is extensive discussion about over-utilization of health care services, but little attention given to measuring the problem. Treatment for the same condition varies widely among providers, fueling increased costs. There are ways to address this.
Educating patients and including them in health care decisions can reduce overuse of medical treatments that are known to vary widely across the country. Systematically informing patients about risks and benefits of different treatments, such as surgery versus medical management, should be a condition for approval of procedures such as prostate surgery, spinal fusion surgery, bariatric surgery, treatment of gastric esophageal reflux, and end-of-life care. This could be done through patients and doctors' use of medical decision-making tools.
We can also look for efficiencies—and improved care—for patients taking multiple medications. Compared with patients in other countries, Americans are much more likely to be taking four or more prescription drugs. Yet, about one-quarter report that their physicians have never reviewed their complete drug regimen. Systematic and periodic review of medication use for nursing home residents, patients leaving hospitals, as well as patients living at home would reduce over-prescribing and drug interactions.
4. Stop paying for medical errors.
Medicare pays hospitals more when patients experience complications, even if those complications are caused by preventable medical errors such as hospital-acquired infections, falls, or medication errors. These "co-morbidity" adjustments should be eliminated in establishing payment rates. The savings could be substantial.
In one survey, one-fifth of sicker adults reported medical mistakes or medication errors that resulted in serious problems. An estimated 44,000 to 98,000 people die of errors each year. One study estimates that errors on 18 patient safety indicators resulted in excess hospital charges of $9 billion a year.
HealthPartners, a large HMO in Minnesota, has announced that it will stop paying for 27 medical mistakes from a list developed by the nonprofit National Quality Forum. These are mistakes that should never happen, such as surgery on the wrong site or serious harm from contaminated drugs or medication errors. By refusing to pay for "never-should-happen" mistakes, HealthPartners is drawing a largely symbolic line between what is acceptable and what is not. The fact is that these mistakes occur relatively infrequently. Even so, infrequently is too frequent. Refusing payment sends the message to hospitals that there are financial consequences to egregious errors and that they would do well to make systemic changes, such as automated prescription order programs, to ensure patient safety.
5. Negotiate pharmaceutical prices.
Pharmaceutical prices in the U.S. are twice as high as drug prices in many other countries. The U.S. spends over $200 billion a year on prescription drugs. But there are ways to control costs while ensuring access to needed medications.
One study found that if the Medicare program negotiated prices to the same level as other major industrialized countries, the "doughnut hole" or gap in prescription drug benefits for beneficiaries could be filled in completely without additional spending. Other strategies include basing insurance reimbursement on the lowest cost-effective drug, or "reference pricing." However, it should be recognized that there are trade-offs. For example, research and development might be reduced, and pharmaceutical prices in other countries might increase, as prices in the U.S. fall.
6. Standardize insurance products to reduce administrative costs.
Private insurance companies have "overhead" of about 12 to 15 percent of revenues. Simplifying and standardizing private insurance could reduce administrative expenses. Hospitals, physicians, and other health care providers incur major administrative expenses as a result of variations across insurers and public programs in terms of benefits covered, payment regulations, conditions of provider participation, and coverage policies. Standardizing products and promoting common practices across all private and public insurers could save hospital and physician administrative costs.
7. Use evidence-based medicine guidelines to determine when a given test or procedure should be done.
According to a 2002 survey, twice as many coronary angioplasties are performed in the U.S. as in any other country. The U.S. has nearly twice as many MRIs per capita as the median across OECD countries. Establishing and following clinical guidelines for medical procedures and tests could achieve significant savings and promote higher-quality care.
For example, one hospital system implemented clinical guidelines encouraging obstetricians to induce labor after at least 39 weeks rather than earlier. As a result of these and other improvements, total maternal and neonatal variable costs decreased from $1,622 per case in January 2003 to $1,480 in the first half of 2004. This result was $300 better than expected based on historical trends.
8. Ensure that every patient has a regular provider who is responsible for prevention, management of chronic conditions, and coordination of care.
Three-fifths of uninsured patients have no regular doctor, making them much less likely to receive preventive care or have chronic conditions properly managed. About 50 percent of the uninsured receive their usual care from high-cost emergency rooms.
What's more, Americans are more likely to report difficulty in obtaining same-day appointments with physicians than residents of other nations. Emergency room use is associated with inability to obtain same-day appointments.
Many state Medicaid programs have demonstrated savings from primary care case management. Medicare and private insurers should follow this lead and ensure that all enrollees have a regular provider who assumes responsibility for accessibility of care, periodic preventive services, management of chronic conditions, and coordination of care across health care settings. But sometimes it is necessary to spend more to achieve savings, and enhanced payment for primary care may be required to ensure that a team of health professionals dedicated to achieving good outcomes is in place.
9. Reduce duplication.
Lack of care coordination is a pervasive problem in the U.S. health system. In one survey, 22 percent of sicker adults reported having duplicate tests performed by different health professionals. One-quarter of sicker adults reported that test results or medical information were not available when they saw physicians. Often duplicate tests or missing information result from antiquated paper-based information systems.
10. Implement modern information technology.
All of these savings would be easier to achieve if health care providers used modern information systems. Such systems would lower administrative costs, reduce medical errors, and make it easier to retrieve test results and review medications. Electronic medical records could give physicians timely access to complete medical histories, in many cases eliminating the need to hospitalize patients.
An electronic clearinghouse on insurance eligibility and claims would make it easier to establish patients' insurance status and enroll the uninsured in coverage that meets their needs. A multi-payer database on utilization of health care services also would help to ascertain provider quality and efficiency. It could be used to move toward a more competitive system of pricing, reducing the wide variation in payment for the same care.
Combined, these 10 supply-side approaches could generate substantial savings. Some would lower costs and improve care. None would require patients to forgo effective care or incur higher out-of-pocket costs.
These steps would result in changes in the health sector that would likely affect the profitability of the insurance and pharmaceutical sectors. Other steps would reduce incomes of specialist physicians, which might reduce the attractiveness of pursuing medicine as a career. Some changes would eliminate jobs in the health sector and construction industry, as hospitals are downsized or closed rather than renovated or expanded.
It is important that savings be re-deployed to cover the uninsured and improve long-term care to a growing frail elderly population. Consideration should be given to dedicating savings to a Coverage and Value Enhancement Trust Fund to ensure that this happens. Savings also could be invested in information systems, creating high-skill jobs in the health sector and information industry. Primary care physicians and advanced practice nurses who are willing to work with patients to improve health behavior and manage their chronic conditions could be better compensated for their time and results. Roles for nurses, social workers, and other health personnel could be expanded to improve child development screening and services to children with behavioral or developmental problems. Disparities in health status, access, and quality of care could be reduced for minority patients, including provision of language translation services. The net result would be a health system that is more responsive to all Americans, yields fewer errors, improves health, and is accessible to all.