The 'Real' Side-Effect Cost of Statins: Sorting Fact from Fiction
Michael O'Riordan
March 13, 2014
LONDON, UK — For patients treated with statins, only a very small minority of side effects reported can be directly attributed to the drugs, according to the results of a new study[1]. In primary prevention, the risk of developing diabetes with statins is real, but just one in five cases of new-onset diabetes suspected to be caused by statins is likely caused by the drugs, while the risk of developing diabetes in secondary prevention is offset by the reduction in cardiovascular deaths, report investigators.
Speaking with heartwire , the researchers, including lead and senior authors Dr Judith Finegold and Dr Darrel Francis (National Heart and Lung Institute, London, UK), said that in clinical practice, patients frequently report side effects such as muscle aches, fatigue, and gastrointestinal distress when started with statin therapy, and this poses a problem for doctors looking to keep patients on the drugs.
However, Francis said the rationale for the analysis stems from the problem that physicians have in communicating the true risk of side effects to their patients. The package insert for statins lists a multitude of possible side effects, a list that contains nearly every side effect reported in clinical trials, including those reported in the placebo arms, but physicians have no way to determine whether those side effects are caused by the medication, would have happened anyway, or are derived from the nocebo effect, a term given to side effects experienced by patients if they anticipate a medication might be harmful.
"It was frustrating to have no reliable information to give to patients quite rightly asking about side effects from statins," said Francis. "If the patient has a side effect, how can I tell them that it's probably not from the tablet? I know from reading the trials, but how is the patient supposed to know? All the patient has is the strip of paper that comes with the tablets and lists every imaginable side effect. Having read that, why would anyone in their right mind take the tablet for an asymptomatic condition? How is the patient supposed to know which of those are significantly increased by the drugs, decreased by drugs, or left unchanged?"
The study is published online March 13, 2014 in the European Journal of Preventive Cardiology.
Side Effects Equally Common in Placebo Arm
To address the proportion of symptomatic side effects in statin-treated patients caused by the medication, the researchers analyzed randomized, controlled clinical trials that compared statin therapy with placebo for primary and secondary cardiovascular disease prevention. In total, 14 primary-prevention studies with 46 262 participants and 15 secondary-prevention studies with 37 618 participants were included in the analysis. To calculate the risk for symptomatic side effects, the absolute increase in risk observed in the placebo arm was subtracted from the risk observed in the statin arm.
In primary prevention, statin therapy increased the absolute risk of diabetes by 0.5% and decreased the risk of mortality by 0.5%. In secondary prevention, treatment with statin therapy decreased the risk of death by 1.4%. In both primary and secondary prevention, statin therapy was associated with an asymptomatic, 0.4% absolute increase in liver enzymes.
"We found that the majority of side effects in primary- and secondary-prevention patients were as common in the placebo arm as in the statin arm, with the exception of asymptomatic liver enzyme elevations and an increase in diabetes in the primary-prevention population," Finegold told heartwire .
In calculating the proportion of symptoms that were not attributable to the statin, the researchers estimated 80% of the new diagnoses for diabetes were not attributed to the medication. As Finegold pointed out, even though one in five cases of new-onset diabetes is attributed to the drug, there has been no clinical trial to date where the use of statins in diabetic patients caused harm. In fact, statins in this patient population have always been shown to save lives. For Francis, he points out that the absolute increase in new-onset diabetes in primary prevention was 0.5%, the same as the absolute reduction in the number of deaths.
"To express this in a simplistic but memorable way, for every primary-prevention patient in whom we cause diabetes, we save one life, prevent two heart attacks, and half a stroke," said Francis. "To me, that is a good deal. But a patient may feel otherwise, and we must respect their preference and just focus on making sure we have given them correct information."
Information from Industry
To heartwire , Finegold and Francis said they would like to see the side-effect profile of drugs, including statins, written in a way patients can understand. If the package insert listed commonly reported side effects, as well as the likelihood that such events were caused by the drugs, patients might be more likely to choose to persist with medication rather than give up.
"If you're a patient and you feel any new symptom, the first thing you do is open the drug box and look at the side effects," said Finegold. "And that, to be honest, seems to list every symptom that has ever been reported in the statin and placebo arms of the clinical trials. There is very little reliable side-effect information given to patients. If you read that leaflet you would assume it was due to the medication." For physicians, they face the same problem, with little information to rely on in terms of determining real vs fictitious side effects reported in the clinical trials.
"We think it's very important that just as we calculate benefit using scientific randomized controlled trials, we should be equally careful to make only true statements about side effects. Ultimately, patients should decide using genuine information," said Finegold.