Don't Cut Research: Use It to Cut the Deficit

Congress should consult the vast body of medical literature and academic experts before making costly policy decisions.
This post was published on the now-closed HuffPost Contributor platform. Contributors control their own work and posted freely to our site. If you need to flag this entry as abusive, send us an email.

While Congress has averted a government default, any deficit deals Congress comes up with will
undoubtedly slash "discretionary" programs including education, medical research and health care for the poor and disabled. Cutting debt is crucial, but instead of broadly cutting these programs Congress could use a scientific approach to eliminate those that are useless or ineffective. Here are just a few examples of costly, ineffective, or potentially unsafe programs that paying attention to science could have helped us avoid:

Paying Physicians for Performance:

One popular policy pays doctors extra compensation to meet quality of care standards, such as
measuring or treating high blood pressure. This approach is embraced by hundreds of public and private health care programs nationwide (exceeding an estimated $20 billion in annual costs) but studies show it accomplishes very little. Recently my colleagues and I conducted the largest and longest study of such a program -- the $2 billion dollar program in the UK that pays physicians to improve care for patients with high blood pressure and other illnesses. [Serumaga et al, BMJ] Under that policy British physicians could earn an additional 25% of salary for meeting certain performance standards. Yet our published findings did not show even a tiny effect of that approach on the health of 400,000 patients with high blood pressure. It turns out that money alone will not improve care. For doctors, like other professionals, it's the intangibles that count -- non-monetary factors such as trusted colleagues and up-to-date clinical information. Moreover, giving doctors extra cash doesn't address the other half of the doctor-patient relationship -- patients -- who by staying on their medicines, themselves are the most effective agents in lowering their blood pressure.[Osterberg et al, New Eng J Med]

My team is not alone -- international scientific reviews have also found little evidence to support this monetary approach. [Cochrane Systematic Review] Some studies found that paying for performance encourages unethical behavior by incentivizing doctors to avoid treating the sickest patients who are less likely to reach the performance targets. Nevertheless, this costly and ineffective approach to improving health care is about to be expanded as part of our national health reform legislation.

Paying Teachers for Better Student Grades

Just as with doctors, policymakers have been trying to link teachers' pay to questionable performance goals -- in this case student test results. Recently, the US Department of Education
spent billions of dollars in grants to states that encouraged linking teacher compensation to student performance on standardized tests. But scientific studies have shown the error of this path. In 2009, a large randomized controlled experiment (the gold standard of research) in several hundred schools found that the promise of extra payments had no effect on student effort or performance. [Fryer et al NBER] And just as with doctors, these policies penalize teachers who serve the neediest students. What's more, teachers "encouraged" by test-linked compensation spend too much time administering practice tests and too little encouraging creativity and experimentation.

Rushing $30 billion on Health Information Technology to all US doctors

Perhaps the most wasteful and ineffective policy in the 2009 federal stimulus law was a requirement that by 2014 physicians throughout the US adopt electronic health records with "decision support" (e.g., alerts to reduce duplicate drugs). In lobbying for the legislation, the president and the commercial electronic records industry have claimed that the program would improve health, reduce deaths, and lower costs. Admittedly, the Veterans Administration system and other "home grown" systems have made modest improvements in care by incorporating such practices as reminding doctors when it's time for certain patients to get certain tests. But the most rigorous research spanning several decades have shown that the commercial software systems that will make up the lion's share of the $30 billion in federal investments do not improve health or save money. [Black et al, PLoS] In fact, a growing number of studies and FDA reports have shown software glitches in these record systems can actually interfere with patient care. In 2005 researchers at the Children's Hospital of Pittsburgh documented a three-fold increase in deaths among very sick children due to a problem with their electronic records system that prevented nurses from ordering timely life-saving medicines. [Han et al, Pediatrics]. Other studies have documented a lack of compatibility among the hundreds of available systems, making it difficult to share patient information across platforms. This problem has caused unsuspecting doctors to duplicate drug orders and cause dangerous drug overdoses. [Soumerai and Avery, Huffington Post] Paying attention to numerous such studies could have prevented legislators from including this costly, and ineffective program in the 2009 health care legislation.

These three examples represent a tiny fraction of wasteful policies that are unsupported by, or
even contrary to scientific evidence. But there is hope, if policymakers are willing to pay more attention to science. Congress should consult the vast body of medical literature and academic experts before making costly policy decisions, such as the premature outlays on electronic health records. And if studies have not been done to evaluate certain policies, Congress should fund them. Admittedly, good research isn't cheap. But it's a lot more cost-effective than mandating huge public expenditures that turn out to be boondoggles.

Stephen B. Soumerai is Professor of Population Medicine at Harvard Medical School.

Popular in the Community