New analysis shows statins have "minimal" benefits

By Maryanne Demasi, PhD

The public health mantra about cholesterol has always been “the lower, the better.”

This has been reflected in expert guidelines which have called on doctors to aggressively lower their patient’s ‘bad’ LDL-cholesterol (LDL-C) with statin drugs to prevent heart disease.

However, our new analysis published in JAMA Internal Medicine (paywalled) challenges that notion. 

Over the years, influential researchers such as the Cholesterol Treatment Trialists (CTT) at Oxford University, have monopolised the scientific debate on statins. 

They have stated that for every 1mmol/L reduction in LDL-C, there is a 21% relative risk reduction in cardiovascular events.

But there are three problems with the various CTT analyses: 

1. The CTT researchers have acquired all the individual patient data (raw data) from published trials and will not share them with other researchers, thus preventing independent replication of their results. For this reason, our study used published trial data only.

2. The CTT promotes the view that statins can reduce “cardiovascular events” which is a composite endpoint with inherent problems. It combines objective outcomes such as death, heart attack and stroke (hard endpoints), together with subjective outcomes such as hospital admissions or revascularisations (soft endpoints). The soft endpoints are often based on a doctor’s ‘judgement call’ and therefore subject to bias. In some analyses, these can make up the majority of events so they can skew the overall result towards benefits for the less serious cardiovascular outcomes. To avoid this, we restricted our analysis to hard endpoints only (death, heart attack and stroke), making it less prone to bias.

3. The CTT and public health authorities often promote the benefits of statins in terms of a relative risk reduction (RRR), rather than an absolute risk reduction (ARR). The RRR is a more impressive statistic (demonstrated below) and can mislead the patient into believe the benefits are larger than they are. Therefore, our study reported both the RRR and ARR of statins on cardiovascular outcomes.

What did our analysis find?

We carried out a systematic review and meta-analysis of 21 statin trials involving 143,532 participants, using similar criteria to the CTT, and found no consistent relationship between lowering LDL-C with statins and death, heart attack or stroke. 

Statins are very effective at lowering LDL-C, but in some trials, that did not necessarily translate into a meaningful benefit for the patient. 

This contradicts the prevailing view, promoted by the CTT, that there is a strong “linear” relationship between lowering LDL-C and cardiovascular outcomes from statin therapy.

Our analysis also highlighted the significant difference in the relative risk reduction (RRR) and absolute risk reduction (ARR) of statin therapy on death, heart attack and stroke.

For example, if your baseline risk of having a heart attack is 2% and taking a drug reduces that risk to 1%, then in relative terms you halved your risk (50% RRR) which sounds impressive, but in absolute terms, you have only reduced your risk by 1% (ARR).

Our analysis showed that trial participants taking a statin for an average of 4.4 years, showed a 29% RRR in heart attacks, but the ARR was only 1.3%.

If this is not effectively communicated to a patient, can they make a fully informed decision about their treatment?  

It may also influence their calculated benefit:harm ratio if the patient is experiencing debilitating effects from the medication.

The graph shows just how deceptive it can be when patients are only told about the benefits of statins in terms of RRR (light blue bars) without the ARR (dark blue bars).

Reporting RRR without baseline risk has been described by experts in risk communication as “the first sin against transparent reporting”.

Selectively reporting the RRR is likely to lead individuals to overestimate the benefit of the drug, it can increases people’s willingness to receive a treatment, advise treatment, and pay to prevent the risk compared with ARR or other methods for communicating risk.

Our analysis combined all statin trials, as well as separated them on the basis of primary prevention (low risk) and secondary prevention (high risk) trials.

The maximum benefit of a statin over the trial period was achieved in people whose risk of heart attack was already very high (secondary prevention); i.e. there was a 2.2% ARR in heart attacks and a 0.9% ARR in deaths in high risk patients.

Remembering too, the vast majority of these trials were sponsored by statin manufacturers, as well as written, designed and analysed by researchers with ties to statin manufacturers.

Some have argued that the benefits may seem minimal in the first few years, but that the benefits may continue to accumulate over the life of the patient - often overlooking that the harms may also accumulate. 

Bottom line:

• JAMA Internal Medicine has published our new systematic review and meta-analysis on 21 statin trials involving 143,532 participants.

• Despite the widespread view promoted by public health, our new study found no consistent relationship between lowering LDL-C and death, heart attack or stroke, following statin therapy.

• Doctors are not effectively and transparently communicating cardiovascular risk to their patients, thereby not allowing informed decision-making.

• We concluded that the benefits of statins were minimal, and most of the trial participants who took statins, derived no clinical benefit.