The bad news: as you have probably noticed, prescription drug spending is up. The good news: in one of his last papers, the late Nobel laureate and Stanford professor Kenneth Arrow has recommendations on how to slow it down.
Spending on prescription meds was $324.6 billion in 2015. After some years of slower growth, it rose 12 percent in 2014 and 9 percent in 2015. The Department of Health and Human Services projects it will average annual growth over six percent over the next decade.
How can drug spending be slowed?
Source: National Health Expenditures.
Professor Arrow and economists Kamran Blair and Alan Sorensen find that the proliferation of new information technology to more physicians could increase the speed and propensity with which they prescribe generic drugs.
In a new National Bureau of Economic Research working paper, they conclude that further proliferation of this information technology can substantially reduce aggregate drug spending without compromising the “careful medical decision-making” of physicians. The authors suggest that use of the drug reference database could lead to better matching between patients and the drugs available.
Arrow and his colleagues combine data on prescriptions and the use of an electronic drug reference database for more than 125,000 physicians. They use this dataset to analyze how the use of this information technology influences physician response to the introduction of new cholesterol drugs.
Some of the variation in the adoption of new cholesterol drugs by physicians is explained by differences in access to information through the use of the electronic drug reference database. The database provides physicians with detailed information about FDA-approved drugs, such as clinical guidelines.
While some of the information is broadly available, a feature of the database is information on retail pricing and formulary status, which influence prescription patterns. Information about new drugs is added at approximately the same time they are made available for commercial prescription, which is important for the authors’ analysis.
Price was not a factor influencing adoption of the database, as a free version with the core features was readily available. A more robust version had annual fees below $200. Instead, convenience was the main reason physicians gave for adopting the database as many of them reported it saved them substantial amounts of time.
Even with these factors, use of the database was low. In the average month, only about 24 percent of physicians were registered database users. About 13 percent actually used the database to look up one of the cholesterol drugs considered. Considerable room for growth exists when it comes to utilizing this new information source.
This prompts the authors to ask: Why do some doctors adopt the technology while others do not? It turns out that one of the factors influencing adoption was when doctors graduated from medical school, as recent grads learned about the database and were more likely to have adopted it. For older physicians, adoption was influenced by the share of other doctors in their zip code who used it. As the composition of doctors shifts more towards those who had learned of the database in medical school, the share of physicians using it should also increase.
Their conservative estimate is that a physician prescribes a new generic drug two weeks sooner after they start using the database, relative to their previous baseline. Controlling for factors that might influence how quickly physicians adopt the information technology, the estimate climbs to two months.
The effect differs for generics relative to branded drugs: while database adoption leads doctors to increase the likelihood of prescribing both old and new generic drugs, it reduces the likelihood of prescribing new branded drugs.
Physicians who use the database also prescribe a more diverse set of drugs. Adoption of the reference database led physicians to raise the number of unique drugs prescribed each month by a statistically significant amount and correspondingly reduced the market concentration of the drugs they prescribed. The authors suggest the increase in diversity might reflect that physicians, armed with more robust information, are better able to match their patients to available therapies.
More widespread adoption of this database technology could empower physicians to provide better treatment to patients while also reducing the amount spent on prescription drugs. As more doctors hear about the database in medical school or from other physicians, the share of them using it will grow over time. This could deliver substantial benefits to physicians and patients without resorting to regulations or mandates.
Charles Hughes is a policy analyst at the Manhattan Institute. Follow him on twitter @CharlesHHughes.
Interested in real economic insights? Want to stay ahead of the competition? Each weekday morning, E21 delivers a short email that includes E21 exclusive commentaries and the latest market news and updates from Washington. Sign up for the E21 Morning Ebrief.