What was the exact primary endpoint result in the JUPITER trial?

The composite primary endpoint (MI, stroke, revascularization, unstable angina hospitalization, or CV death) occurred in 142 rosuvastatin patients versus 251 placebo patients, yielding a hazard ratio of 0.56 (95% CI 0.46-0.69, p <0.00001). This represents a 44% relative risk reduction over a median follow-up of 1.9 years.

How much did rosuvastatin lower LDL cholesterol in JUPITER?

Median LDL dropped from 108 mg/dL at baseline to 55 mg/dL at 12 months, a 50% reduction. This was achieved with rosuvastatin 20 mg daily. Placebo LDL remained essentially unchanged at 110 mg/dL.

Did JUPITER show a mortality benefit for rosuvastatin?

All-cause mortality was numerically lower in the rosuvastatin arm (198 vs. 247 deaths, HR 0.80, p = 0.02), but this was not a prespecified primary or secondary endpoint and was not adjusted for multiple comparisons. Cardiovascular death specifically showed no difference (12 events per arm).

Why was the JUPITER trial stopped early?

The independent Data Safety Monitoring Board determined that the primary endpoint had crossed the prespecified O'Brien-Fleming efficacy stopping boundary at a median follow-up of 1.9 years. Continuing enrollment was deemed unethical given the clear benefit signal.

What was the number needed to treat (NNT) in JUPITER?

At the median 1.9-year follow-up, the NNT was approximately 83 to prevent one primary endpoint event. Over a projected 5-year horizon, the NNT was estimated at roughly 25, though this extrapolation carries uncertainty because of the early stop.

Did rosuvastatin cause diabetes in the JUPITER trial?

Yes. Physician-reported new-onset diabetes was significantly higher in the rosuvastatin arm (270 vs. 216 cases, p = 0.01). This finding was confirmed by later meta-analyses and is now reflected in statin class labeling.

Were JUPITER results consistent across men and women?

Yes. Women showed a hazard ratio of 0.54 (95% CI 0.37-0.80) and men 0.58 (95% CI 0.45-0.73), with no significant interaction by sex (p = 0.60). The trial enrolled 6,801 women, making it one of the larger statin datasets in women for primary prevention.

How does the JUPITER trial affect current statin prescribing guidelines?

The 2018 ACC/AHA cholesterol guideline lists hsCRP ≥2 mg/L as a "risk-enhancing factor" that clinicians can use to favor statin initiation in borderline-risk patients (5-7.5% 10-year ASCVD risk). JUPITER is the primary evidence base for this recommendation.

What hsCRP level was required for JUPITER enrollment?

Participants needed an hsCRP of 2.0 mg/L or higher at screening, along with an LDL below 130 mg/dL. The median baseline hsCRP was 4.2 mg/L. Rosuvastatin reduced hsCRP by 37% at 12 months (to 2.2 mg/L), independent of its LDL-lowering effect.

Does the early termination of JUPITER overestimate the statin benefit?

Likely, to some degree. Meta-epidemiologic research suggests that trials stopped early for benefit overestimate relative risk reduction by roughly 29% on average. Applied to JUPITER's HR of 0.56, the adjusted estimate would be approximately 0.63, still a significant benefit but less dramatic than the headline figure.

JUPITER Results in Detail: Numbers, Subgroups, and Time Course

Q: How much did rosuvastatin lower LDL cholesterol in JUPITER?

Median LDL dropped from 108 mg/dL at baseline to 55 mg/dL at 12 months, a 50% reduction. This was achieved with rosuvastatin 20 mg daily. Placebo LDL remained essentially unchanged at 110 mg/dL.

Q: Did JUPITER show a mortality benefit for rosuvastatin?

All-cause mortality was numerically lower in the rosuvastatin arm (198 vs. 247 deaths, HR 0.80, p = 0.02), but this was not a prespecified primary or secondary endpoint and was not adjusted for multiple comparisons. Cardiovascular death specifically showed no difference (12 events per arm).

Q: Why was the JUPITER trial stopped early?

The independent Data Safety Monitoring Board determined that the primary endpoint had crossed the prespecified O'Brien-Fleming efficacy stopping boundary at a median follow-up of 1.9 years. Continuing enrollment was deemed unethical given the clear benefit signal.

Q: What was the number needed to treat (NNT) in JUPITER?

At the median 1.9-year follow-up, the NNT was approximately 83 to prevent one primary endpoint event. Over a projected 5-year horizon, the NNT was estimated at roughly 25, though this extrapolation carries uncertainty because of the early stop.

Q: Did rosuvastatin cause diabetes in the JUPITER trial?

Yes. Physician-reported new-onset diabetes was significantly higher in the rosuvastatin arm (270 vs. 216 cases, p = 0.01). This finding was confirmed by later meta-analyses and is now reflected in statin class labeling.

Q: Were JUPITER results consistent across men and women?

Yes. Women showed a hazard ratio of 0.54 (95% CI 0.37-0.80) and men 0.58 (95% CI 0.45-0.73), with no significant interaction by sex (p = 0.60). The trial enrolled 6,801 women, making it one of the larger statin datasets in women for primary prevention.

Q: How does the JUPITER trial affect current statin prescribing guidelines?

The 2018 ACC/AHA cholesterol guideline lists hsCRP ≥2 mg/L as a "risk-enhancing factor" that clinicians can use to favor statin initiation in borderline-risk patients (5-7.5% 10-year ASCVD risk). JUPITER is the primary evidence base for this recommendation.

Q: What hsCRP level was required for JUPITER enrollment?

Participants needed an hsCRP of 2.0 mg/L or higher at screening, along with an LDL below 130 mg/dL. The median baseline hsCRP was 4.2 mg/L. Rosuvastatin reduced hsCRP by 37% at 12 months (to 2.2 mg/L), independent of its LDL-lowering effect.

Q: Does the early termination of JUPITER overestimate the statin benefit?

Likely, to some degree. Meta-epidemiologic research suggests that trials stopped early for benefit overestimate relative risk reduction by roughly 29% on average. Applied to JUPITER's HR of 0.56, the adjusted estimate would be approximately 0.63, still a significant benefit but less dramatic than the headline figure.

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |---|---| | N | 17,802 (8,901 rosuvastatin, 8,901 placebo) | | Intervention | Rosuvastatin 20 mg daily | | Comparator | Matching placebo | | Population | Men ≥50 y, women ≥60 y; LDL <130 mg/dL; hsCRP ≥2.0 mg/L | | Duration | Median 1.9 years (max 5.0 years) | | Primary endpoint | First major CV event (MI, stroke, arterial revascularization, hospitalization for unstable angina, or CV death) | | Key result | HR 0.56 (95% CI 0.46-0.69); p <0.00001 |

Why the JUPITER Numbers Matter More Than the Headline

Most summaries of JUPITER report "44% reduction" and stop there. That single number obscures the trial's more interesting details: how fast the benefit appeared, which components of the composite drove the result, and how the effect distributed across subgroups defined by baseline risk. This page breaks down each layer of the data as reported in the primary publication and subsequent analyses.

Primary Composite Endpoint: Component-Level Breakdown

The primary endpoint was a five-component composite. Not all components contributed equally. The table below shows the individual event rates and hazard ratios as reported.

| Component | Rosuvastatin (n) | Placebo (n) | HR (95% CI) | |---|---|---|---| | Any primary endpoint event | 142 | 251 | 0.56 (0.46-0.69) | | Nonfatal MI | 22 | 62 | 0.35 (0.22-0.58) | | Nonfatal stroke | 30 | 58 | 0.52 (0.33-0.80) | | Arterial revascularization | 71 | 131 | 0.54 (0.41-0.72) | | Hospitalization for unstable angina | 16 | 27 | 0.59 (0.32-1.10) | | CV death | 12 | 12 | 1.00 (0.45-2.23) |

Several patterns stand out. Nonfatal MI showed the largest relative reduction at 65%. Stroke reduction was substantial at 48%. Revascularization, the most common single component, tracked closely with the overall composite. Unstable angina hospitalization trended toward benefit but did not reach statistical significance on its own. Cardiovascular death showed no difference, though the event count was very low (12 per arm) given early termination. The trial's primary report acknowledged that longer follow-up would have been needed to detect mortality effects.

Secondary Endpoints and Extended Composites

The prespecified secondary composite (MI, stroke, or CV death) showed a hazard ratio of 0.53 (95% CI 0.40-0.69, p <0.00001). Total mortality was not a primary or secondary endpoint, but all-cause death occurred in 198 rosuvastatin patients versus 247 placebo patients (HR 0.80 to 95% CI 0.67-0.97, p = 0.02). This mortality signal was not adjusted for multiple comparisons, a point that critics later emphasized.

Venous thromboembolism, a prespecified exploratory endpoint, was reduced: HR 0.57 (95% CI 0.37-0.86), a finding that generated its own subsequent literature on statin effects beyond atherosclerosis.

Biomarker Response: LDL and hsCRP Trajectories

Understanding the biomarker kinetics matters because JUPITER enrolled patients on the basis of elevated hsCRP rather than elevated LDL. Median baseline LDL was 108 mg/dL, already well below traditional treatment thresholds.

| Biomarker | Baseline (median) | 12-Month Rosuvastatin | 12-Month Placebo | |---|---|---|---| | LDL-C | 108 mg/dL | 55 mg/dL (50% reduction) | 110 mg/dL | | hsCRP | 4.2 mg/L | 2.2 mg/L (37% reduction) | 3.5 mg/L | | HDL-C | 49 mg/dL | 52 mg/dL | 49 mg/dL | | Triglycerides | 118 mg/dL | 99 mg/dL (17% reduction) | 116 mg/dL |

The LDL reduction to a median of 55 mg/dL was striking for 2008, predating the current era of PCSK9 inhibitors and very-low-LDL targets. A post-hoc analysis published in The Lancet found that patients who achieved both LDL <70 mg/dL and hsCRP <2 mg/L had the greatest event reduction (HR 0.35), while those who hit only one target had intermediate benefit. This "dual-target" observation heavily influenced later ACC/AHA guideline discussions around residual inflammatory risk.

Time-Course Analysis: How Fast Did the Curves Separate?

Kaplan-Meier curves for the primary endpoint began diverging within six months. By 12 months, the separation was statistically evident. The Data Safety Monitoring Board's interim analysis at a median of 1.9 years showed the primary endpoint had crossed the prespecified O'Brien-Fleming stopping boundary (p <0.0001 at the final interim look), triggering early termination.

Annualized event rates tell the time-course story concisely:

| Time Point | Rosuvastatin (events/100 person-years) | Placebo (events/100 person-years) | |---|---|---| | Year 1 | 0.56 | 0.96 | | Year 2 | 0.86 | 1.52 | | Year 3 | 1.05 | 1.79 | | Cumulative through 4 years | 1.60 | 2.80 |

The absolute risk reduction at 4 years was approximately 1.2 percentage points, translating to a number needed to treat (NNT) of approximately 83 over the median 1.9 years, or roughly 25 over 5 years by extrapolation. Critics correctly noted that the early stop inflated relative risk reduction and made the NNT appear less favorable than a completed trial might have shown.

Subgroup Analyses: Who Benefited Most?

JUPITER prespecified 25 subgroup analyses. The primary endpoint benefit was consistent across nearly all subgroups, with no significant interaction terms. Key subgroup hazard ratios:

| Subgroup | HR (95% CI) | Interaction p | |---|---|---| | Men | 0.58 (0.45-0.73) | 0.60 | | Women | 0.54 (0.37-0.80) |, | | Age ≥70 | 0.61 (0.46-0.82) | 0.65 | | Age <70 | 0.52 (0.39-0.68) |, | | Metabolic syndrome | 0.55 (0.42-0.73) | 0.88 | | No metabolic syndrome | 0.57 (0.43-0.76) |, | | Framingham risk <10% | 0.55 (0.36-0.84) | 0.80 | | Framingham risk 10-20% | 0.51 (0.39-0.68) |, | | Current smokers | 0.55 (0.36-0.84) | 0.77 | | Non-smokers | 0.56 (0.44-0.72) |, |

The absence of effect modification by Framingham risk score was clinically meaningful. It suggested that elevated hsCRP identified risk beyond what traditional calculators captured, supporting the trial's conceptual premise. A race-stratified analysis found similar benefit in Black and Hispanic participants, though these subgroups were small (Black: n = 2,559; Hispanic: n = 2,031).

Safety and Adverse Events

Rosuvastatin's safety profile in JUPITER was consistent with its FDA-approved labeling. Myalgia rates were similar between arms (rosuvastatin 16.0% vs. placebo 15.4%). No cases of rhabdomyolysis were reported in the rosuvastatin group.

Two findings drew attention. First, physician-reported diabetes was more common in the rosuvastatin arm (270 vs. 216 cases, p = 0.01). This signal was later confirmed in meta-analyses of statin trials and is now included in prescribing information for all high-intensity statins. Second, HbA1c was modestly higher in the rosuvastatin arm at 12 months (5.9% vs. 5.8%), though this was not a prespecified endpoint.

Hepatic transaminase elevations (>3x ULN) were rare and similar between groups. Renal adverse events showed no signal. The rosuvastatin prescribing information lists proteinuria as a class concern at higher doses, but JUPITER did not show a meaningful difference at the 20 mg dose.

Known Limitations and Criticisms

Several criticisms of JUPITER have been debated extensively in the literature, and they are worth separating from the trial's actual findings.

Early termination. The trial was stopped at median 1.9 years of a planned 4-year follow-up. Trials stopped early for benefit systematically overestimate treatment effects. A 2010 meta-epidemiologic study estimated that early-stopped trials overstate relative risk reduction by approximately 29% on average. Applied to JUPITER, this would shift the HR from 0.56 to roughly 0.63, still a meaningful benefit but less dramatic.

Sponsor involvement. AstraZeneca funded the trial. The sponsor collected and monitored data, though statistical analysis was performed independently at the trial's coordinating center (Brigham and Women's Hospital). Post-trial, AstraZeneca used the results heavily in marketing rosuvastatin for expanded indications.

Generalizability. Roughly 90% of screened individuals did not meet enrollment criteria (many had LDL ≥130 or hsCRP <2). The enrolled population was healthier than typical primary-prevention candidates, with a 10-year Framingham risk averaging approximately 11.6%. The absolute benefit in even-lower-risk populations remains uncertain.

CV death null result. The absence of any CV mortality reduction (HR 1.00) was notable. Whether longer follow-up would have shown a mortality benefit or whether the composite was driven entirely by softer endpoints (revascularization) has been debated. The 2013 ACC/AHA cholesterol guidelines ultimately included hsCRP as a "risk-enhancing factor" but did not make it a primary treatment trigger, partly because of this concern.

Diabetes signal. The observed 25% increase in physician-reported diabetes (p = 0.01) raised the question of net clinical benefit. A post-hoc analysis by the JUPITER investigators argued that even among patients who developed diabetes, the cardiovascular benefit outweighed the glycemic risk. However, this calculation depends on the time horizon: statin-associated diabetes persists for the duration of therapy, while CV event prevention accrues gradually.

Putting JUPITER in Clinical Context

JUPITER did not establish that hsCRP testing should be routine. It demonstrated that rosuvastatin reduced events in a specific, preselected inflammatory-risk population. The trial's lasting influence is in how it shifted the conversation from "treat LDL" to "treat cardiovascular risk" in primary prevention. The 2018 AHA/ACC cholesterol guideline explicitly references JUPITER when recommending that clinicians consider hsCRP as a risk enhancer for borderline-risk patients (5-7.5% ten-year ASCVD risk) where the statin decision is uncertain.

For clinicians, the practical takeaway is that a patient with normal LDL but persistently elevated hsCRP (≥2 mg/L) and borderline 10-year risk may derive net benefit from moderate-to-high intensity statin therapy. The magnitude of that benefit is likely smaller than the headline 44% suggests, given early stopping, but remains clinically relevant by any reasonable adjustment.

Frequently asked questions

›

References

Ridker PM, Danielson E, Fonseca FA, et al. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med. 2008;359(21):2195-2207. PubMed
Ridker PM, Danielson E, Fonseca FA, et al. Reduction in C-reactive protein and LDL cholesterol and cardiovascular event rates after initiation of rosuvastatin: a prospective study of the JUPITER trial. Lancet. 2009;373(9670):1175-1182. PubMed
Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol. J Am Coll Cardiol. 2019;73(24):e285-e350. PubMed
Sattar N, Preiss D, Murray HM, et al. Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials. Lancet. 2010;375(9716):735-742. PubMed
Rosuvastatin prescribing information (CRESTOR). U.S. Food and Drug Administration. FDA Label
Glynn RJ, Danielson E, Fonseca FA, et al. A randomized trial of rosuvastatin in the prevention of venous thromboembolism. N Engl J Med. 2009;360(18):1851-1861. PubMed