Health-care costs are out of control in our nation. Prescription drug costs represent the fastest-growing component of these expenditures. A recent, highly publicized report on a cholesterollowering drug illustrates the complexity of this issue.
On Nov. 10, news reports highlighted a study in which administration of the statin Rosuvastatin (Crestor) to adults with normal levels of low-density (bad) cholesterol significantly reduced both their cholesterol levels and their incidence of heart attacks and strokes. Adults were accepted into the study if they had an elevated serum level of high-sensitivity C Reactive Protein (hs-CRP).
Many cardiologists consider hs-CRP as a marker for coronary artery disease. Half of the subjects received a daily placebo; half received a daily 20 mgm Crestor dose. A study designed to continue for four years was stopped after an average of 1.9 years when monitors noted a significant reduction of cardiovascular events in the Crestor group.
Upon hearing the report, I read the study online in the New England Journal of Medicine. The article was released at the time as the media blitz. Here are my concerns:
The media picked up the report before physicians and medical researchers had a chance to read and to critique the scientific article. Given the media coverage of the report, I suspect many physicians have been swamped by requests for hs-CRP testing and prescriptions for Crestor from patients with normal cholesterol levels.
Of the 14 co-authors of the study, 10 acknowledge at the end of the article that they receive support — research grants and/or lecture and consulting fees — from Crestor’s manufacturer, AstraZeneca. This pattern of financial relationships between the authors of medical papers and the companies who produce the agents upon which the reports are based is quite common. This presents a potential conflictof-interest for many clinical studies.
Because medical journals are so dependent for their revenue on advertising from drug-company ads, can these publications impartially assess articles that deal with pharmaceutical agents?
The exact role of hs-CRP in clinical medicine is unclear. When elevated, the test reflects inflammation, which may originate in blood vessels, as well as joints, lungs, and other bodily tissues. Thus, in any one patient, an elevated hs-CRP may have multiple causes.
Some insurance companies regard hs-CRP as an investigational test and will not pay for its use. One of the authors of the Crestor study is listed as a co-inventor of tests for inflammation that include hs-CRP. The patents for these discoveries are held by Brigham and Women’s Hospital in Boston. On-line sources quote a $20 fee per hs-CRP test.
This does not include the costs of obtaining the necessary blood sample or the administrative overhead associated with the test.
Most clinical studies of drugs compare the drug to a placebo. There are relatively few studies that compare results among drugs within the same class. Thus, in the current study we are given no insights into whether the same effects could have been realized with another, less expensive statin.
The cost implications are substantial, as shown by these comparisons from Drugstore. com:
Crestor 20 mgm. 100 tablets cost $345.30
Simvastatin (generic of Zocor) 20 mgm. 100 tablets $82.19
Lovastatin 20 mgm. 100 tablets $66.64.
No detailed cost-benefit data accompanied the study.
The article referred to a slight increase in the incidence of diabetes mellitus in the Crestor group. What exactly does this mean? Would this incidence increase over time? What is an adequate length of time for a clinical study to run? Are there cholesterol levels below which therapy should be stopped?
No one knows.
If we are ever to control medical costs, we must have a more rational system for introducing new therapies for general use. The Food and Drug Administration must have sufficient, independent staff to evaluate new medications and new uses for medications already on the market.
Before a medication is approved, a detailed cost, benefit and risk assessment must be completed independent of the manufacturer. We must also employ independent panels to advise us on the appropriate use of tests, such as hs-CRP. Hoopla must not be permitted to interfere with scientific assessment of diagnostic tests and prescription drugs.
E-mail Clif Cleaveland at