About one-third of people given medicine to lower their cholesterol levels stop taking it or switch drugs because they do not seem to work or have side effects, according to a study by Massachusetts scientists.
The study, to be published in Thursday's New England Journal of Medicine, found that in real life, adverse side effects and ineffectiveness are far more common than the results suggested by scientific tests to evaluate such drugs.Results from scientific studies "have an overly optimistic success rate" that are "distorting the judgment of both practicing physicians and the policy makers who assess the cost-effectiveness of these drugs," said the research team, led by Susan Andrade of the Harvard School of Public Health.
The findings are not unexpected. In formal drug tests, researchers use strict criteria to select the subjects to be used in the tests. Anyone with complicated health problems or patients who are less likely to take the medicine to which they're assigned, are often excluded from such tests.
The Andrade team used the records from 2,369 patients at two Massachusetts health maintenance organizations to determine how many people actually stayed on the anti-cholesterol drugs they were originally given.
They found the overall dropout rate among the real-life HMO patients was about 32 percent, compared to a dropout rate of about half that in the relevant scientific studies.
In some cases, the differences were dramatic. While only about four percent of volunteers testing the drug niacin dropped out from a scientific study, the dropout rate was 46 percent in real life, the researchers found.
Niacin is produced by a variety of manufacturers.
The real-life dropout rates were also high for the cholesterol-lowering drugs like gemfibrozil, sometimes sold by Parke-Davis as Lopid, and bile acid sequestrants such as Questran, made by Bristol-Myers Squibb Co.