The United States faces unfathomably difficult decisions as it tries to cut back on the $2.5 trillion spent annually on health insurance and health care services. Should the U.S. continue to spend "whatever it takes" to help a premature infant survive? Is it cost-effective to give a 99-year-old woman a pacemaker? Answering these questions is not easy, as one person's wasteful expenses are another person's essential expenses. To help at least discuss these issues, let's consider the medical costs and societal benefits associated with treating babies who are born three months prematurely.
Writing in The New York Times
, Jane Brody told a very compassionate story
of an 11-year-old girl who was born after just 25 weeks gestation, weighing only 13.5 ounces, but who is now, amazingly, an accomplished writer and illustrator. Brody did not cite the costs of the child's neonatal intensive care, other than to say that the infant spent the first five months of her life in a Falls Church, Va., hospital. (It is difficult to find data on such costs; one hospital in Rhode Island
estimated the daily expense at around $2,000, which means that a five-month stay would cost upwards of $300,000.) The Times
article concludes with a comment by Dr. Michele Walsh, a neonatologist in Cleveland, who says that although it is expensive to maintain "million-dollar babies," it becomes very cost-effective over time: "There is a return on investment when they get out into the work force and pay taxes."
Being an economist, I was struck by this explicit link between the cost and benefits. Even though the United States is on the verge of implementing a health care reform that explicitly calls for "bending the cost curve," only rarely does the issue of "cost effectiveness" arise when considering whether, for example, to take measures to save the life of a child. And, when the issue does arise, it is usually generates howls of outrage, such as occurred in 2006, when a U.K. research group recommended that doctors not routinely resuscitate extremely premature babies (those born before 22 weeks of gestation). The recommendation
was based on data showing that such babies rarely survived --- only 1 percent of those born between 22 and 23 weeks left the hospital.
More recently, researchers at the University of Leicester conducted an in-depth study
of survival rates of extremely premature babies born between 1994-2005 who were treated in neonatal intensive care units. Sadly, they found that none of the 150 babies born at 22 weeks survived.
In light of these statistics, should extremely intensive treatment be denied such babies? Most of us would say no. And American pediatricians
would agree. They examined this issue and came to a slightly different recommendation than their U.K. counterparts. Advances in medical technology create a tension between providing care that might result in survival -- but with a significant deterioration in quality of life (cerebral palsy, blindness and mental retardation are the most common conditions) -- and of providing non-intensive care that might result in death. The pediatricians recommended that even with low survival rates, physicians and parents should jointly determine whether to start or continue intensive care -- but they must take the child's best interest into account. As the doctors noted, "It is inappropriate for life-prolonging treatment to be continued when the condition is incompatible with life or when the treatment is judged to be harmful, of no benefit, or futile."
How can doctors decide which treatments are beneficial and which are not? One relatively recent approach is known as comparative-effectiveness research (CER). Under CER, researchers examine the outcomes of various forms of treatment and obtain a relative ranking of each one. The federal stimulus bill enacted in mid-2009 provided $1.1 billion in funds for CER on the argument that it is the best way to address the unsustainable growth in health care expenditures. CER has strong support in some areas. For example, in November a group of economists recommended that health care reform legislation increase funding for research into treatment effectiveness, and it appears the Senate was listening. The current Senate health reform bill provides funding for a Center for Health Outcomes and Research Evaluation.
Comparative effectiveness research has merit. It might lead, for example, to deciding to put a patient on a rather expensive pacemaker as a lower-cost alternative to spending months in the hospital --- even if the patient is 99 years old. Although some may question the benefits of giving an elderly patient an expensive pacemaker, few are suggesting that such patients should just live out his or her natural life span.
However, comparative effectiveness research is not uncontroversial. Rep. Tom Coburn (R-Okla.) argues that this process will lead to rationing by government bureaucrats, who will make decisions that belong in the hands of physicians, patients and families. Writing in The Wall Street Journal, Coburn argued that the Independent Medicare Advisory Board and the CER panels created in the House's health reform bill could ration care based on cost. Such a characterization doesn't sound far removed from Sarah Palin's "death panels."
Politics aside, how do economists evaluate the cost effectiveness of medical technology?
To answer this question, let's again consider the many difficult questions that arise when treating low-birth-weight babies. The number of premature births is large and growing. In 2007, 546,547 U.S. infants (12.7 percent of all live births) were classified as pre-term, meaning they were born before 37 weeks of gestation. Among these preemies, 88,069 births (2.04 percent) occurred before 32 weeks. Both figures are 20 percent higher than they were a decade ago. Low birth weight is a major health concern for premature infants. Nearly 65,000 infants weighed less than 3 pounds 4 ounces at birth. Doctors advise that extending the gestation period by one week can bring tremendous health benefits to the infant; unfortunately, doctors have limited techniques to extend pregnancy and are thus faced with improving the technological support given to these children once they are born.
How much does it cost to care for those half-million premature babies born each year in the United States? Business Week reported the National Academy of Sciences' calculation: about $26 billion.
What is the economic benefit? That's difficult to say, but in a study published a decade ago, economists David Cutler and Ellen Meara estimated that technological advances added 12 years in life expectancy for each low-birth-weight baby. Thus, although neonatal technology is expensive, the authors calculated that the rate of return in these cases exceeds 500 percent. (The study excluded extremely low-birth-weight infants.)
Behavioral economist Richard Thaler addressed
the cost-benefit issue in the context of the recent recommendation that women under age 50 stop having annual mammograms. Thaler noted that evaluating this recommendation requires cost-benefit calculations that most of us are reluctant to make. For example, for every breast cancer death that early screening prevents, between four and 20 women will undergo unnecessary radiation, chemotherapy or mastectomies as a result of false positives the mammograms identify. Is the value of the one life saved worth more than the medical costs for the four to 20 women treated unnecessarily? Certainly, the breast cancer survivor would argue that it is. But the women whose quality of life may have significantly deteriorated due to the unnecessary treatment might provide a different answer.
For another issue
, consider the case of Avastin, a drug used to treat colon cancer. Patients typically take it for about 11 months and gain an additional five months in life expectancy relative to other treatments. One year of Avastin treatment costs about $50,000 per patient, or $2.5 billion for the roughly 50,000 Americans who take the medication for these purposes. Are five months of extra life worth the cost? The maker of Avastin, Genentech, wants to make the drug available to treat breast cancer, but at twice the cost. With numerous women suffering from breast cancer and thus potentially benefiting from Avastin, is this additional expense one that society will accept without questioning its cost effectiveness?
Many difficult questions arise when considering the billions of dollars spent on medical procedures, especially if one treatment diverts resources from another. Is the $26 billion spent each year treating low-birth-weight babies a better use of resources than the $56 billion the National Institute of Health estimates is spent treating cancer each year? Do we even have to make a choice -- isn't America wealthy enough to afford whatever medical care we need? If so, are we expected to provide a blank check to the health services industry?
The answer is no. Many Americans -- even those who have insurance --- find it difficult to pay for costly procedures. For example, a 2005 survey by USA Today, the Kaiser Family Foundation and the Harvard School of Public Health found that not only were 28 percent of adults unable to pay all of their medical expenses, 62 percent of those in difficulty had health insurance. Doctors estimate that patients may have to pay $20,000 out-of-pocket for their Avastin. With the typical household earning just $50,000 a year, that clearly would be a problem for many Americans.
What's not at all clear is how we, as a society, should decide what medical expenditures are worth the cost. But it is clear that until we draw lines somewhere, runaway spending will continue. With health care expenditures accounting for one out of every six federal dollars spent and projected to grow much higher, if we are unable to say no to increased demands on our health care system, then we will have to cut back in other areas. Those cuts might occur in two other "untouchable" areas -- defense and Social Security. Or, alternatively, we may have to raise taxes. None of these decisions will be easy to make.