Login
Subscribe
In the 1999 Edition of Worst Pills, Best Pills, we urged patients not to use donepezil (ARICEPT), made by Pfizer/Esai, because of a combination of adverse effects such as nausea, diarrhea and vomiting along with serious questions about clinically significant effectiveness. We quoted the prestigious Medical Letter, an independent evaluation for physicians of newly approved drugs, which stated “There is no evidence that use of...donepezil... leads to substantial functional improvement or prevents the progression of the disease.” Since then, Pfizer/Esai has run ads that claim the drug delays the time to nursing home placement. The claim is misleading and an exaggeration of the drug’s extremely modest efficacy. We wish to reinforce this recommendation and highlight the need for Food and Drug Administration (FDA) to take action: the ads should be halted and Pfizer/Esai should issue a retraction.
In order to boost the sales of this drug, for over a year Pfizer/Esai have run extraordinarily misleading and predictably highly successful full-page ads in leading medical journals such as the Journal of the American Medical Association and the Journal of the American Geriatrics Society, a journal widely read by geriatricians and others taking care of older patients. The ads proclaim that patients with Alzheimer’s disease who “persistently” use donepezil are able to stay in the community and avoid going into a nursing home for almost two years more than patients who have only limited treatment or no treatment with the drug (see advertisement for an example). The implication is startling. The families and other friends who are caring for patients with Alzheimer’s disease will have a substantial amount of extra time to have their loved ones in the community before they have to go into a nursing home if only the patients are lucky enough to be persistent users of donepezil.
If this conclusion were based on a randomized, placebo-controlled trial (the scientific gold standard for medical research) wherein 50 percent of the patients got donepezil and the other 50 percent of exactly the same kinds of patients were randomly assigned to be given a placebo, it would be a major public health breakthrough and a cause for celebration. Unfortunately, it is not based on such a study and is not any cause for celebration.
The study cited as the basis for this seemingly remarkable improvement is flawed. The group on donepezil and the group not on donepezil appear to have been as dissimilar as the proverbial apples and oranges rather than being identical in every way except for the drug they were being given, as would have been the case in a properly done randomized placebo-controlled study.
The major flaw is what has been referred to as “drop-out-bias”, a flaw in many previous studies claiming to prove meaningful improved social outcomes with drugs such as donepezil. The study used as the basis for the misleading ad was a follow-up to previous studies that were themselves randomized. The drop-out after this first phase occurs when patients were told whether they had been given a drug or a placebo. The bias occurs because people drop out based at least in part on how they felt they did in the RCT and this drop-out almost certainly occurs in an extremely unbalanced way.
How does this bias work? If you are told you got a placebo in the earlier study, you are more likely to volunteer for the continuation of the study, regardless of how you “did” on the placebo, because you would like to see how you might do on a “real” drug. If, instead, you were in the group who got drug treatment and you think your short-term improvement was caused by the drug, you will be more likely to continue taking the drug. Conversely, if after having taken the drug in the previous study you do not perceive an improvement, or even got worse, you are more likely to drop out. Thus, the drop-outs selectively occur in the group who got the drug, not the placebo, in the first part of the study and did not do so well. As a result, the overall short-term mix of that group who becomes the “persistent donepezil” user group is a selected group of responders. That this group subsequently did better than those who were not persistent users of donepezil is preordained.
Buried in the middle of the journal article cited in the ad as the source of this finding is the statement that since it is an observational study, (observing two different groups of people) “the current investigation....could not prove conclusively that taking effective doses of donepezil delayed nursing home placement.” The article further stated that the only way to prove this finding of delayed nursing home placement would be to do a randomized double-blind placebo-controlled prospective trial that would assure that the two groups were completely comparable. Similarly buried in the copy of the ads is the statement that in all “studies of this type, results [delayed nursing home placement] may be attributed to various factors.”
In addition to the drop-out bias described above, factors such as spousal relationship and willingness to stay in research are factors that have themselves (without drugs) been shown to improve patient outcome and lessen the chances of needing to go into a nursing home. Many of the characteristics of the caregiver and the interaction between the caregiver and the patient with Alzheimer’s disease could well be quite different in the group getting “persistent” treatment from the other group. Hence, exposure to donepezil may have had nothing to do with the delay to nursing home.
Indeed, a randomized, placebo-controlled study in the United Kingdom, known as AD (Alzheimer’s Disease) 2000, done by Dr. Richard Gray and his colleagues at the University of Birmingham Clinical Trials Unit, examining the effect of donepezil on outcomes such as nursing home placement, was presented almost two years ago at a medical meeting in Stockholm and has been submitted for publication. That study, lacking the flaws of the study used as the basis of the misleading ad, found no significant delay in nursing home placement in those using donepezil. There was also no improvement in the amount of time needing to be spent by the caregiver with the Alzheimer’s patient nor a slowing of the progress of the disability of the patient.
Thus, the conclusion that is the main theme of such ads, that the drug resulted in a substantial delay in the time to nursing home placement, is misleading because it is not supported by scientific evidence. These exploitative ads should have been stopped long ago. Where has the FDA been? Why have these misleading ads, clearly exploiting patients with Alzheimer’s disease and their families and friends through the medium of misleading doctors, not been stopped long ago?
As we have mentioned before, the past five years have seen the virtual demise of FDA enforcement of the laws and regulations governing prescription drug advertising. False and misleading advertising, typically overstating the benefits and/or understating the risks, constitutes a health hazard because it may result in a patient getting a prescription for a drug thought to be safer or more effective than it actually is. Therefore there is a relative danger inherent in such a “misprescription” compared to a safer drug that might have otherwise been prescribed. Consequently, FDA enforcement of prescription drug advertising, whether the kind directed at doctors or at patients, is a crucial part of the agency’s regulation of the pharmaceutical industry.
In 1998, the FDA stopped 157 illegal prescription drug ads but in 2003, they stopped only 24: an 85 percent decrease in enforcement. This is despite the lack of any evidence that the quality or accuracy of such ads has improved. In the face of a massive increase in prescription drug advertising since 1998 that would, if anything, argue for an increased number of FDA enforcement actions rather than the sharp decrease that has occurred. The continuation of these misleading ads for donepezil for almost one and one-half years without FDA intervention illustrates this problem of a lack of law enforcement very clearly.
What You Can Do
Do not use donepezil (ARICEPT) because, as the Medical Letter has stated, "There is no evidence that use of...donepezil...leads to substantial functional improvement or prevents the progression of the disease."
