Skip to:

Millions could have incorrect statin, aspirin & blood pressure prescriptions

getty-patient.jpg

Photo credit: 
Getty Images

More than 11 million Americans may have incorrect prescriptions for aspirin, statins and blood pressure medications, according to a study led by Stanford Health Policy's Sanjay Basu and other researchers.

Their findings are based on an updated set of calculations — known as pooled cohort equations, or PCEs — that are used to determine the risk of a heart attack or stroke. 

The PCEs are the foundation for cardiovascular-disease-prevention guidelines in the United States. They help physicians decide whether to prescribe aspirin, blood pressure or statin medications, or some combination of these, by estimating the risk a patient may have for a heart attack or stroke. Most physicians calculate a patient’s risk using a CE web calculator or a smartphone app; the equations are also built into many electronic health records so that a patient’s risk is automatically calculated during an office visit.

But there has been debate over whether the PCEs are based on outdated data and therefore putting some patients at risk for over- or under-medication.

“We found that there are probably at least two major ways to improve the 2013 equations,” said Basu, MD, PhD, assistant professor of primary care outcomes research at the School of Medicine and a core faculty member at Stanford Health Policy. “The first was well known: that the data used to derive the equations could be updated.”

For example, he said, one of the main data sets used to derive the original equations had information from people who were 30-62 years old in 1948, and who would, therefore, be 100 to 132 years old in 2018 — that is, likely dead. The older equations were often estimating people’s risk as too high, possibly by an average of 20 percent across risk groups.

“A lot has changed in terms of diets, environments and medical treatment since the 1940s,” Basu said. “So, relying on our grandparents’ data to make our treatment choices is probably not the best idea.”

Basu is the senior author of the study published in the Annals of Internal Medicine. The lead author is Steve Yadlowsky, a graduate student in electrical engineering at Stanford. 

Furthermore, the researchers found that the old data may not have had a sufficient sample of African-Americans. For many African-Americans, physicians may have been estimating the risks of heart attacks or strokes as too low.

“So while many Americans were being recommended aggressive treatments that they may not have needed according to current guidelines, some Americans — particularly African-Americans — may have been given false reassurance and probably need to start treatment given our findings,” Basu said.

The researchers have updated the PCEs with newer data in an effort to substantially improve the accuracy of the cardiovascular risk estimates. The National Institutes of Health — specifically the National Heart, Lung and Blood Institute — maintains and updates the cohort data. The research team applied to the NIH and received its formal approval to use and newer data from them for this project, as well as funding to update the equations. 

The other authors include Rodney A. Hayward and Jeremy B. Sussman from the University of Michigan; Robyn L. McClelland from the University of Washington; and Yuan-I Min of the University of Mississippi.

The second problem with the equations, the authors found, is that the statistical methods used to configure the 2013 equations caused errors which the research team detected and fixed. A second improvement to the equations, the authors found, was to update the statistical methods used to derive the equations.

“We found that by revising the PCEs with new data and statistical methods, we could substantially improve the accuracy of cardiovascular disease risk estimates,” the authors wrote.

Basu is also a member of Stanford Bio-X and the Child Health Research Institute