The JUPITER (Justification for the Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin) study shows that if healthy older adults with normal cholesterol and high C-reactive protein (CRP) levels take rosuvastatin (Crestor) daily for two years, they are likely to experience some benefit.1 However, the study does not show whether the regimen is cost-effective or more effective than standard care. A head-to-head trial comparing the JUPITER approach with counseling for a healthy lifestyle and a routine global cardiac risk assessment is needed. Prior experience preventing diabetes with medication or lifestyle modification comes to mind. The Prevention of Diabetes Study showed that metformin (Glucophage) had no advantage compared with the lower cost and greater benefit of lifestyle modification.2
First, let us take a look at some of the details of the JUPITER study. It was a randomized, double-blind, placebo-controlled, multicenter trial conducted at 1,315 sites in 26 countries; each site contributed an average of about 13 patients. It is difficult to ensure high-level quality control over 1,315 sites with small patient numbers. The study also may have commercial bias. In addition to the study being sponsored by the manufacturer of rosuvastatin, the principal investigator has numerous ties to pharmaceutical companies and holds patents on the use of high-sensitivity CRP levels in evaluating cardiovascular risk. All potentially eligible patients underwent a four-week run-in phase during which they received placebo. This creates a selection bias in favor of compliant patients, which will inf late the benefit of the drug compared with its use in clinical practice. Furthermore, there was a 0.55 percent difference between groups in the main outcome of all-cause mortality after two years (number needed to treat [NNT] of 182 patients for two years to prevent one death). Because the study included a run-in period to screen out non-compliant patients, the NNT is likely to be higher in clinical practice.
There was also a lengthy list of exclusion criteria for study participation: previous or current use of lipid-lowering therapy, previous or current use of post-menopausal hormone therapy, evidence of hepatic dysfunction, elevated creatine kinase or creatinine levels, diabetes, uncontrolled hypertension, cancer within five years before enrollment, uncontrolled hypothyroidism, any inflammatory or autoimmune condition, recent alcohol or drug abuse, or another medical condition that might compromise study completion. This means that 2,000 adults in the relevant age group would have to be screened to identify 250 patients with a combination of low levels of low-density-lipoprotein cholesterol and high levels of high-sensitivity C-reactive protein.
Table 1 presents the estimated costs of physician visits, blood tests, and medication over two years to prevent one premature death using the treatment and study group in the JUPITER study. The amount totals more than $800,000 per premature death prevented. If extended to younger, healthier patients, the NNT and cost per premature death are even higher. However, before using cheaper generic statins or aspirin, studies are needed to determine their benefit.
|Intervention||Patients affected||Cost per patient||Total cost||Comment|
|High-sensitivity CRP blood test||2,000||$50||$100,000||Adding high-sensitivity CRP to standard laboratory panel each year at a cost of $25 per test, either for screening or follow-up†|
|Physician visits||250||$155||$38,750||Assumes one additional visit and three visits with higher complexity ($80 + $25 + $25 + $25)|
|Rosuvastatin (Crestor)||250||$2,608 to $2,772||$652,000 to 693,000||$1,304‡ to $1,386§ per year for Crestor, 20 mg|
|Overall cost: $790,750 to $828,750|
The widely reported results of the JUPITER study sound impressive, but are they important? The following are key reasons why this study may not be clinically important. First, and foremost, we already have an approach that works: high-quality standard care that focuses on lifestyle modification and global cardiac risk assessment. Even with the noteworthy results of this trial, the results of the healthy lifestyle studies (HALE [Healthy Ageing: a Longitudinal study in Europe],3 ARIC [Atherosclerosis Risk in Communities Study],4 Health Professionals Follow-up Study,5 and Nurses’ Health Study6) are even better. The JUPITER study shows a 44 percent reduction in cardiovascular deaths, whereas the healthy lifestyle studies show a 50 to 60 percent reduction. Lifestyle interventions are also far cheaper to implement.
Not surprisingly, rosuvastatin appeared to offer some clinically significant benefit. However, the benefit falls far short of other approaches, particularly lifestyle interventions. This is a point that family physician, John Abramson, MD, has extensively documented in his book, Overdo$ed America: The Broken Promise of American Medicine.7 Furthermore, previous data show that statins may be beneficial in healthy military personnel,8 but they clearly are not cost-effective. A persuasive cost-effectiveness analysis that includes alternative treatment modalities is needed before we can reasonably commit to such a primary prevention strategy using a prophylactic drug. Routine high-sensitivity CRP measurement should not be universally adopted until it has been formally compared with simple global cardiac risk assessment.9
The biggest problem with the JUPITER study is that it suggests that physicians continue to test and treat. First, though, we need to take time to think. One of the things to ponder is this: Couldn’t we do something better for patients than measuring everyone’s CRP levels and treating those with elevated values? Instead, we should wait for a study comparing CRP measurement to routine coronary risk factor assessment, including studies on cost-benefit analysis. We already know that lifestyle changes are effective and, therefore, we should focus on innovative ways to assist patients in making these changes.