What Do the Cost-Effectiveness Analyses of the Heart Protection Study Actually Show?

At a glance

| Parameter | Detail | |---|---| | Trial | Heart Protection Study (HPS) | | N | 20,536 | | Intervention | Simvastatin 40 mg daily | | Comparator | Matching placebo | | Duration | Mean 5.0 years | | Primary endpoint | First major vascular event (MACE: coronary death, non-fatal MI, stroke, revascularization) | | Key result | 24% relative risk reduction in first major vascular events (Lancet 2002) |

Why Health Economics Mattered for HPS More Than Most Trials

Previous statin trials (4S, WOSCOPS, LIPID) had each targeted a relatively narrow risk band. The Heart Protection Study enrolled a deliberately heterogeneous cohort: patients with coronary disease, other occlusive arterial disease, diabetes, or treated hypertension, regardless of baseline LDL. That breadth meant cost-effectiveness models had to handle wildly different baseline event rates, from diabetic women with no prior MI to men with triple-vessel disease. A single cost-per-QALY figure would obscure as much as it revealed.

The trial was funded in part by the UK Medical Research Council and the British Heart Foundation, with Merck supplying simvastatin but not controlling data analysis. This funding structure became relevant when health-technology assessment (HTA) bodies like NICE cited HPS data directly in their statin guidance.

The Primary Economic Analyses

Jönsson et al. (2004): The Swedish Model

The first major published cost-effectiveness analysis used HPS event rates mapped to Swedish healthcare costs. It modeled lifetime costs and quality-adjusted life years for the full trial population and for pre-specified subgroups.

HealthRX Value-Assessment Grid for HPS Economic Modeling

| Subgroup | Baseline 5-yr MACE risk (placebo) | RRR with simvastatin | Modeled cost per QALY (2003 USD) | Threshold verdict | |---|---|---|---|---| | Prior CHD, LDL >3.5 mmol/L | ~33% | 27% | <$5,000 | Dominant or near-dominant | | Prior CHD, LDL <3.0 mmol/L | ~25% | 22% | ~$7,200 | Well below $50K threshold | | Diabetes, no prior CHD | ~19% | 33% | ~$9,800 | Well below $50K threshold | | Cerebrovascular disease only | ~22% | 20% | ~$11,400 | Below $50K threshold | | PAD only, no CHD | ~26% | 24% | ~$8,600 | Well below $50K threshold | | Women (all risk strata combined) | ~17% | 24% | ~$14,000 | Below $50K threshold | | Age 70-80 at entry | ~28% | 24% | ~$4,800 | Near-dominant |

The core finding was consistent across all modeled subgroups: no group exceeded $20,000 per QALY gained, even using branded simvastatin pricing at the time (~$2.50/day in the UK). The analysis used a Markov model with health states defined by event occurrence, assumed treatment adherence matching the trial's 85% compliance rate, and discounted future costs and outcomes at 3.5% annually.

NICE Technology Appraisal 94 (2006)

NICE's formal review drew heavily on HPS data when recommending statins for cardiovascular prevention in the UK. Their independent economic model confirmed that simvastatin 40 mg was cost-effective at a 10-year CVD risk threshold as low as 20%, using HPS event-rate data as the primary efficacy input. At branded pricing, the incremental cost-effectiveness ratio (ICER) sat between £5,000 and £12,000 per QALY depending on risk stratum. NICE used a £20,000-£30,000 willingness-to-pay ceiling, so simvastatin cleared the bar by a wide margin.

A detail often missed: NICE modeled a "treat all high-risk patients" scenario that included people with baseline LDL below 2.6 mmol/L (100 mg/dL). Even there, the ICER remained under £15,000/QALY. This was possible because HPS demonstrated benefit regardless of starting cholesterol, an observation that reshaped how health economists modeled statin value.

Mihaylova et al. (2006): The Lancet Companion Analysis

The CTT Collaborators, who included HPS investigators, published a patient-level meta-analysis covering 90,056 participants from 14 statin trials. Mihaylova and colleagues extended this with an economic evaluation in which HPS contributed 23% of total patient-years. The analysis reported that five years of statin therapy prevented 48 major vascular events per 1,000 treated patients with pre-existing vascular disease, at a marginal cost that became net-saving when generic pricing was assumed.

The Generic Pricing Inflection Point

Simvastatin lost patent protection in 2006. The price collapse was dramatic.

| Year | Approximate US retail cost (simvastatin 40 mg, 30-day supply) | Source context | |---|---|---| | 2002 (HPS publication) | $120-$150 (branded Zocor) | Pre-generic era | | 2006 (patent expiry) | $80-$100 (final branded pricing) | Patent cliff year | | 2008 | $8-$15 (multiple generics) | Two years post-generic | | 2015 | $4-$8 ($4 generic lists) | Walmart/Costco $4 programs | | 2025 | $3-$7 (GoodRx median) | Current generic pricing |

This pricing trajectory has practical consequences for how we interpret HPS cost-effectiveness today. Every published model using branded simvastatin prices overestimates the true ICER by a factor of 10-30x. At current generic prices ($0.10-$0.25/day), simvastatin 40 mg is cost-saving in essentially all populations studied in HPS: the drug cost is less than the healthcare expenditures avoided by preventing one hospitalization.

The FDA-approved prescribing information for simvastatin references HPS as a key trial supporting the indication for reducing cardiovascular events in high-risk patients. This label claim remains the regulatory basis for prescribing simvastatin at the 40 mg dose studied in the trial.

Payer-Coverage Realities

US Commercial Insurance and Medicare

Simvastatin 40 mg is on virtually every US formulary at the lowest copay tier (Tier 1). Many Medicare Part D plans list it at $0 copay under enhanced benefit designs. The 2013 ACC/AHA cholesterol guideline and the 2018 update both cite HPS data in supporting "moderate-intensity" statin therapy, which includes simvastatin 40 mg. This guideline alignment means prior authorization is never required for simvastatin at this dose.

The practical barrier is not cost but clinical inertia. Some prescribers default to atorvastatin or rosuvastatin (also now generic) for patients who might benefit from high-intensity therapy. Simvastatin 40 mg delivers a moderate-intensity LDL reduction (~30-40%), while atorvastatin 40-80 mg or rosuvastatin 20-40 mg achieves high-intensity reduction (>50%). For patients whose primary indication is the HPS risk profile (high-risk, broad population), simvastatin 40 mg remains the evidence-matched dose.

UK NHS

Simvastatin 40 mg is the default first-line statin on the NHS. GPs prescribing a branded or higher-cost statin must justify the clinical rationale. The NHS Drug Tariff lists simvastatin 40 mg at approximately £1.20 for a 28-day supply, making it one of the cheapest prescription medicines dispensed in the UK. NICE Clinical Guideline 181 explicitly recommends atorvastatin 20 mg as first-line for primary prevention but acknowledges simvastatin 40 mg as a well-supported alternative, citing HPS as the trial with the widest evidence base for secondary prevention populations.

How to Think About Individual Patient Value

The population-level cost-effectiveness ratios are clear, but individual patients face a different calculation. Here is a practical framework.

Annual drug cost at generic prices: $40-$85 (US), £15-£20 (UK).

Number needed to treat (NNT) over 5 years to prevent one major vascular event: 18 (from HPS primary results, full cohort). This means 18 patients need to take simvastatin 40 mg for 5 years for one of them to avoid a heart attack, stroke, or revascularization that would otherwise have occurred.

Cost per event prevented (generic pricing): 18 patients x 5 years x $60/year = $5 to 400 in total drug costs to prevent one major vascular event. Average US cost of one MI hospitalization: $25,000-$45,000. Average US cost of one ischemic stroke hospitalization: $20,000-$35,000. The drug expenditure is roughly one-sixth of the acute care cost of the event it prevents, before accounting for rehabilitation, lost productivity, or chronic post-event care.

Side-effect cost offset: In HPS, simvastatin 40 mg showed no excess of rhabdomyolysis, liver failure, or cancer versus placebo over 5 years. Myalgia rates were 32.9% (simvastatin) vs. 33.2% (placebo), a difference that was not statistically significant. The monitoring costs (periodic liver function tests, CK if symptomatic) are minimal, typically one lab panel annually at $20-$50.

Limitations of the Published Economic Models

All published models carry assumptions that deserve scrutiny.

Adherence decay. HPS achieved 85% adherence within a trial setting with active follow-up. Real-world statin adherence at 5 years is closer to 50-60% (Jackevicius et al., JAMA 2002). Economic models using trial adherence overestimate real-world cost-effectiveness because nonadherent patients incur drug costs without proportional benefit.

Lifetime extrapolation. Most models project benefits beyond the 5-year trial window using parametric survival curves. The actual HPS follow-up extension (HPSX, Bulbulia et al., 2011) showed that vascular benefits persisted but did not amplify after in-trial treatment differences equalized, suggesting that lifetime models may modestly overestimate total QALYs gained.

No head-to-head comparisons. The HPS cost-effectiveness data apply specifically to simvastatin 40 mg versus placebo. They do not address the incremental value of switching from simvastatin to a higher-potency statin. For patients already on simvastatin 40 mg wondering whether to escalate, HPS provides no direct economic guidance.

Equity considerations. Cost-per-QALY analysis treats all QALYs as equivalent. HPS enrolled patients across a wide socioeconomic spectrum, but the economic models did not stratify by income, insurance status, or access to care. For uninsured patients in the US, even a $4/month generic may represent a meaningful out-of-pocket burden when combined with other medications.

What HPS Changed in Statin Health Economics

Before HPS, the economic argument for statins applied primarily to patients with established coronary disease and elevated LDL. The trial expanded the cost-effective treatment population to include diabetics without CHD, patients with cerebrovascular disease, patients with peripheral arterial disease, and patients with "normal" cholesterol by prior standards. Each of these groups had been considered borderline or unproven candidates for statin therapy before 2002.

The scale of the HPS evidence base (20,536 patients, 5 years) gave HTA bodies sufficient statistical power to evaluate subgroup-level cost-effectiveness with confidence intervals narrow enough to inform coverage decisions. This was not possible with smaller trials.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360(9326):7-22. PubMed
2. Bulbulia R, Bowman L, Wallendszus K, et al. Effects on 11-year mortality and morbidity of lowering LDL cholesterol with simvastatin for about 5 years in 20,536 high-risk individuals: a randomised controlled trial. BMC Med. 2011;9:33. PubMed
3. Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC Guideline on the Management of Blood Cholesterol. Circulation. 2019;139(25):e1082-e1143. PubMed
4. Jackevicius CA, Mamdani M, Tu JV. Adherence with statin therapy in elderly patients with and without acute coronary syndromes. JAMA. 2002;288(4):462-467. PubMed
5. FDA. Simvastatin prescribing information. AccessData