Biostatistics & Population Health

Cost-effectiveness analysis: QALYs and ICER

Clinical Overview and When to Suspect Cost-Effectiveness Concerns

— A question asks whether to adopt a new technology, screening test, or drug at a population/health-system level

— A stem provides two interventions with cost and QALY data and asks which is preferred

— A vignette involves insurance coverage decisions, formulary placement, or value-based care

— Public health screening rollouts (HPV, lung cancer LDCT, colonoscopy intervals)

— $50,000/QALY (historical benchmark)

— $100,000–$150,000/QALY (contemporary, ICER/WHO-aligned upper bounds)

— Below threshold → cost-effective; above → not cost-effective at that WTP

— Cost-benefit analysis (both costs and outcomes in dollars)

— Cost-minimization (assumes equivalent outcomes, picks cheaper)

— Cost-utility analysis (technically CEA using QALYs/DALYs as the utility outcome — the most common Step 3 framing)

Cost-effectiveness analysis (CEA) is the formal economic evaluation that compares the incremental cost of an intervention to its incremental health benefit, most often measured in quality-adjusted life years (QALYs).

The summary statistic is the incremental cost-effectiveness ratio (ICER) = (Cost_new − Cost_comparator) / (Effect_new − Effect_comparator), expressed in dollars per QALY gained.

When to "suspect" CEA reasoning is relevant on Step 3:

Common willingness-to-pay (WTP) thresholds used in US analyses:

CEA is distinct from:

Perspective matters: societal (includes lost productivity), payer, or healthcare system. Step 3 stems usually imply payer/societal.

Board pearl: If a stem gives you costs and QALYs for two strategies and asks "which is most cost-effective," you must compute the ICER between the two relevant alternatives, not raw cost or raw QALY alone. The cheapest option is not automatically the answer; the highest-QALY option is not automatically the answer either — the ratio against a threshold decides.

Presentation Patterns and Key History — How CEA Appears in Stems

— A table with Strategy A, Strategy B (± C), each row showing total cost and QALYs (or life-years); the question asks which is preferred at a stated WTP.

— A screening or vaccination policy vignette: "Should the health department fund universal HPV vaccination for boys?" with supporting cost/QALY estimates.

— A new drug adoption scenario: novel oncology agent vs standard, with marginal survival benefit but very high cost.

— A comparator framing question: which is the appropriate comparator (usual care, do-nothing, or current standard)?

— "Per QALY gained"

— "Incremental cost-effectiveness ratio"

— "Willingness-to-pay threshold"

— "Dominant strategy" — one option is less costly AND more effective (always preferred; no ICER needed)

— "Dominated strategy" — one option is more costly AND less effective (eliminated before ICER calculation)

— "Extended/weak dominance" — a strategy has a higher ICER than a more effective strategy; also eliminated

— Time horizon (lifetime vs 5-year) — chronic disease interventions need lifetime horizons

— Discount rate (typically 3% annually in US analyses for both costs and benefits)

— Model type: decision tree (short-term), Markov model (chronic disease with health states)

Typical Step 3 stem shapes:

Key phrases that should trigger CEA thinking:

Historical "history" of the analysis matters:

Key distinction: A dominant strategy is identified by inspection, not by ICER. If Drug A costs less and yields more QALYs than Drug B, A dominates B — do not calculate an ICER, simply pick A. Students lose this point by reflexively computing ratios. Always scan for dominance first, eliminate dominated options, then compute ICERs on the remaining frontier.

Physical Exam Findings — Structural Components of a CEA

Because CEA is methodological, the "exam" equivalent is identifying the anatomy of a well-conducted analysis:

— Direct medical costs: drugs, hospitalizations, clinician visits, devices, lab tests

— Direct non-medical costs: transportation, caregiver expenses (societal perspective)

— Indirect costs: lost productivity, premature mortality wages (societal only)

— Costs are typically reported in a reference-year US dollars and discounted to present value

— QALY = years of life × health-state utility weight (0–1)

— Utility weights derived from instruments like EQ-5D, SF-6D, HUI, or time trade-off / standard gamble methods

— 1 year in perfect health = 1.0 QALY; 1 year in a state valued at 0.7 = 0.7 QALYs; death = 0

— Negative utilities (states worse than death) are possible but rare

Numerator — Costs:

Denominator — Effectiveness in QALYs:

Comparator: must be the current standard of care or next-best alternative — comparing to placebo when an active standard exists overstates cost-effectiveness

Time horizon: long enough to capture all relevant costs/benefits (lifetime for chronic disease)

Discounting: 3%/year for costs and QALYs in US analyses (per Second Panel on Cost-Effectiveness in Health and Medicine)

Sensitivity analysis: one-way, multi-way, or probabilistic sensitivity analysis (PSA) with Monte Carlo simulation; outputs cost-effectiveness acceptability curves (CEACs)

Board pearl: If a stem reports an ICER without mentioning sensitivity analysis or discounting, the analysis is incomplete. Step 3 favors answers that explicitly account for parameter uncertainty via PSA before recommending widespread adoption — point estimates alone are insufficient for policy.

Diagnostic Workup — Calculating an ICER Step by Step

— C₁, E₁ = cost and effectiveness of new intervention

— C₀, E₀ = cost and effectiveness of comparator

— Units: $/QALY gained (or $/life-year, $/case prevented in non-utility CEA)

— Strategy A (standard): $20,000; 5.0 QALYs

— Strategy B (new drug): $50,000; 5.6 QALYs

— Incremental cost = $50,000 − $20,000 = $30,000

— Incremental QALY = 5.6 − 5.0 = 0.6 QALY

— ICER = $30,000 / 0.6 = $50,000/QALY

— At WTP = $100,000/QALY → cost-effective (adopt B)

— At WTP = $30,000/QALY → not cost-effective (keep A)

1. Order strategies by increasing effectiveness (QALYs)

2. Eliminate strictly dominated strategies (higher cost, lower QALYs than another)

3. Calculate ICERs sequentially on the remaining "efficient frontier"

4. Eliminate weakly dominated strategies (ICER higher than the next more-effective option)

5. Compare final ICERs against the WTP threshold

— A lower ICER is better (more health per dollar)

— A negative ICER is meaningless on its own — you must inspect whether it's because the intervention is cost-saving and more effective (dominant — adopt) or costlier and less effective (dominated — reject)

— Average cost-effectiveness ratios (C/E) are not the same as incremental ratios; Step 3 wants ICER

Formula: ICER = (C₁ − C₀) / (E₁ − E₀)

Worked example:

Multi-strategy comparisons (3+ options):

Interpretation pitfalls:

Step 3 management: When a vignette gives you three strategies, always rank by effectiveness first, eliminate dominance, then compute sequential ICERs. Skipping the dominance check is the most common reason for a wrong answer.

Diagnostic Workup — Advanced Concepts: QALY Construction and Utilities

— A treatment that extends life from 5 to 8 years at a utility of 0.8 yields: (8 × 0.8) − (5 × 0.8) = 2.4 incremental QALYs

— A treatment that improves utility from 0.5 to 0.8 over 10 years yields: 10 × (0.8 − 0.5) = 3.0 incremental QALYs, even with no life extension

— Standard gamble (SG): patient chooses between certain health state and a gamble of perfect health vs death; gold standard theoretically

— Time trade-off (TTO): how many years of life would you give up to be in perfect health?

— Visual analog scale (VAS): simplest but least theoretically grounded

— Multi-attribute instruments: EQ-5D-5L (most common in US/EU), HUI-3, SF-6D

— DALY (disability-adjusted life year) = years of life lost + years lived with disability; DALYs are bad (you want to minimize them)

— QALYs are good (you want to maximize them)

— WHO/global health analyses often use DALYs; US/HTA analyses use QALYs

— May systematically disadvantage older adults and disabled persons (lower baseline utilities → fewer QALYs to gain)

— US federal law (ACA Section 1182) prohibits Medicare from using QALY-based thresholds to deny coverage

— Alternatives: equal-value life-years (evLYG), healthy years equivalents (HYE)

QALY mechanics:

Utility elicitation methods:

DALYs vs QALYs:

Critiques of QALYs (relevant to ethics chunk):

Board pearl: A drug that prolongs survival in advanced cancer often has a high ICER because the gained life-years occur in low-utility health states (e.g., utility 0.4 during chemotherapy) — modest QALY gain combined with high drug cost frequently pushes ICERs above $150,000/QALY despite a survival benefit.

Risk Stratification — Interpreting ICERs Against Thresholds

— < $50,000/QALY: high value, broadly cost-effective

— $50,000–$100,000/QALY: intermediate value

— $100,000–$150,000/QALY: upper bound of acceptable value

— > $150,000/QALY: low value

— NE quadrant (costlier, more effective): trade-off — apply ICER vs WTP

— SE quadrant (cheaper, more effective): dominant — adopt

— NW quadrant (costlier, less effective): dominated — reject

— SW quadrant (cheaper, less effective): trade-off — apply ICER vs WTP (rare; e.g., de-implementation)

— Childhood vaccinations: cost-saving (dominant)

— Statin for secondary CV prevention: ~$10,000–25,000/QALY (high value)

— LDCT screening for lung cancer in eligible smokers: ~$50,000–80,000/QALY

— Hemodialysis: historically ~$80,000–130,000/QALY

— Many novel oncology agents: >$200,000/QALY

— CEA informs policy/coverage; individual patient decisions still rest on clinical benefit, preferences, and shared decision-making

— A non-cost-effective intervention may still be clinically indicated for a specific patient

WTP threshold tiers commonly cited in US literature (per ICER, the Institute for Clinical and Economic Review):

Cost-effectiveness plane (four quadrants based on ΔCost and ΔEffect):

Examples of well-known interventions' ICERs (approximate, illustrative):

Population vs individual decisions:

CCS pearl: If a CCS-style scenario hints at resource stewardship (formulary committee, ACO setting, capitated payment), favor the lower-ICER strategy when clinical outcomes are equivalent, and document the shared decision-making conversation when cost considerations influence the plan.

Pharmacotherapy Analogy — First-Line "Regimen" for Conducting CEA

When the "drug" is the analytic method itself, the recommended first-line CEA workflow (per the Second Panel on Cost-Effectiveness in Health and Medicine, 2016) is:

— Population, intervention, comparator, outcomes (PICO)

— Specify perspective — societal (recommended reference case) and healthcare sector

— Long enough to capture all material consequences (often lifetime for chronic disease)

— Direct medical, direct non-medical, productivity (societal)

— Convert to a single reference year using medical CPI

— QALYs preferred; report life-years and intermediate outcomes (events prevented) as secondary

— 3% annually for both costs and QALYs (US standard); 5% in some European analyses

— Decision tree for short, deterministic problems

— Markov/state-transition model for chronic disease with recurring events

— Microsimulation for individual-level heterogeneity

— Apply dominance rules before reporting

— Deterministic (one-way, tornado diagrams): identifies influential parameters

— Probabilistic (PSA): assigns distributions to all parameters, runs Monte Carlo (typically 10,000 iterations), generates CEAC

Define the decision problem:

Choose the analytic horizon:

Identify and measure costs:

Identify and measure effects:

Apply discounting:

Build the model:

Calculate ICER vs comparator(s):

Conduct sensitivity analysis:

Report transparently using CHEERS 2022 checklist

Step 3 management: When asked which step is most likely to change the conclusion of a CEA, the answer is usually sensitivity analysis on the most uncertain parameter — typically utility weights, treatment effect size, or drug cost — not the base-case point estimate. Robustness across plausible parameter ranges, not the headline ICER, is what supports policy adoption.

"Procedural" Application — Health Technology Assessment and Coverage Decisions

— ICER (Institute for Clinical and Economic Review) — independent US nonprofit; publishes value-based price benchmarks

— NICE (UK) — uses explicit threshold £20,000–£30,000/QALY; below → recommend; above → restrict or reject

— CADTH (Canada), IQWiG (Germany), PBAC (Australia)

— Medicare cannot use a per-QALY threshold to deny coverage (statutory prohibition under ACA §1182)

— Private payers, PBMs, and value-based contracts increasingly use ICER-style analyses

— Inflation Reduction Act (2022) allows Medicare drug price negotiation but explicitly excludes QALY-based price-setting

— CEA answers "is it worth it?" (efficiency)

— BIA answers "can we afford it?" (affordability over 1–5 years for a defined population)

— Both are needed; a drug can be cost-effective per QALY but unaffordable at population scale (e.g., hepatitis C DAAs at launch)

— Price at which ICER = WTP threshold

— Example: if a drug yields 0.4 incremental QALYs vs standard, at $100,000/QALY WTP the value-based incremental price is $40,000

— Shared decision-making aids increasingly incorporate out-of-pocket cost alongside efficacy

— Step 3 vignettes may test financial toxicity awareness — asking about drug cost as a barrier to adherence

Health Technology Assessment (HTA) translates CEA into policy — coverage, formulary, and pricing decisions:

US-specific constraints:

Budget impact analysis (BIA) vs CEA:

Value-based pricing:

Patient-level applications:

Board pearl: When a stem describes a cost-effective drug that nonetheless causes patient non-adherence due to copay, the best Step 3 answer is typically switch to a financially accessible alternative or engage a patient assistance program, not "continue current therapy and counsel on importance" — adherence-adjusted effectiveness drives real-world QALY gain.

Special Populations — Elderly and Patients with Limited Life Expectancy

— Fewer remaining life-years → smaller absolute QALY gains from any intervention

— Lower baseline utilities (comorbidity, frailty) → each year contributes fewer QALYs

— Result: interventions can appear less cost-effective in older populations on a per-QALY basis

— Statins for primary prevention in adults >75: marginal QALY gain, ICER often exceeds $100,000/QALY

— Aggressive cancer screening in patients with <10-year life expectancy: rarely cost-effective; USPSTF stops most cancer screening at age cutoffs partly for this reason

— Dialysis initiation in nonagenarians: high cost, low utility, modest survival

— Some analyses use age-weighting or equal-value-of-life-years-gained (evLYG) to avoid systematic disadvantage

— ICER (the organization) reports both QALY-based and evLYG-based value benchmarks

— Drug dosing adjustments alter both cost (lower doses) and effectiveness (efficacy may be preserved or reduced)

— Increased adverse event risk shifts the cost side upward (hospitalizations) — can flip an ICER unfavorably

— Need for TDM, dose titration, and monitoring adds direct medical costs

— Time-to-benefit for many preventive interventions (e.g., tight glycemic control, intensive BP lowering) exceeds remaining life expectancy in frail elders → deprescribing is often the cost-effective choice

Older adults and patients with shortened life expectancy present methodological and ethical challenges for QALY-based analyses:

Examples:

Equity adjustments:

Renal/hepatic impairment considerations in CEA:

Frailty-adjusted decision-making:

Step 3 management: In an 85-year-old with limited life expectancy considering a costly preventive therapy with a 5–10 year time-to-benefit, the cost-effective and clinically appropriate decision is typically deprescribing or not initiating — both QALY math and clinical guidelines converge here.

Special Populations — Pediatrics, Pregnancy, and Equity Considerations

— Often highly cost-effective or dominant because benefits accrue over many decades

— Childhood vaccinations (MMR, HPV, pneumococcal) are repeatedly shown to be cost-saving (dominant — pay less and gain QALYs)

— Newborn screening programs (PKU, CCHD pulse oximetry, SCID) typically <$50,000/QALY

— Pediatric QALY measurement is methodologically challenging — utilities elicited from parents/proxies rather than children

— Maternal and fetal QALYs both considered; some analyses count fetal QALYs from viability or birth

— Examples: aspirin for preeclampsia prevention in high-risk pregnancies, GBS screening, prenatal aneuploidy screening — generally cost-effective

— RhoGAM for Rh-negative mothers — cost-saving

— Distributional CEA (DCEA) incorporates how costs and benefits are distributed across socioeconomic groups

— Equity weights can be applied to favor interventions that reduce disparities

— Increasingly required in Medicaid and public health analyses

— A cost-effective intervention is meaningless if it is inaccessible (rural geography, uninsured, language barriers)

— Implementation costs (community health workers, patient navigation) must be included in realistic CEA

— HPV vaccine for ages 9–14 (2-dose): ICER consistently <$30,000/QALY; expansion to age 26 also cost-effective; expansion to 27–45 has higher ICER and is shared decision-making

Pediatric interventions:

Pregnancy-related analyses:

Equity-informed CEA:

Vulnerable populations and access:

Pediatric vaccine pricing example:

Key distinction: A pediatric or vaccine intervention can be dominant (cost-saving) while an adult chronic-disease intervention with identical clinical benefit may be only cost-effective — the difference is the time horizon over which QALYs accrue. Recognizing dominance in childhood interventions is a frequent Step 3 testing point.

Complications and Adverse Outcomes — Pitfalls in CEA Interpretation

— Wrong comparator: comparing new drug to placebo when active standard exists → overstates value

— Short time horizon: truncating before benefits or harms fully manifest

— Omitting adverse events: ignoring cost of side effects inflates apparent value

— Selective costing: including cost-saving downstream effects but not implementation costs

— Industry funding bias: sponsored CEAs more often report ICERs below WTP threshold

— Favorable analyses more likely to be published

— CHEERS 2022 checklist standardizes reporting and reduces bias risk

— Lower discount rates favor interventions with long-delayed benefits (vaccines, prevention)

— Higher rates favor near-term interventions

— Standard 3% should be tested in sensitivity analysis (0%, 5%)

— Treating WTP threshold as a hard cutoff ignores opportunity cost and distributional concerns

— Threshold should reflect marginal productivity of the health system — what is displaced when the new intervention is funded

— CEA is a decision aid, not a categorical denial mechanism

— Patient-level clinical judgment remains primary

— Wide confidence intervals around ICER often span both cost-effective and non-cost-effective regions

— CEAC showing probability of cost-effectiveness at each WTP is the appropriate uncertainty summary

— RCT-based effectiveness inputs may overstate real-world benefit (adherence, selection)

Common methodological flaws to recognize in stems:

Publication and reporting bias:

Discount rate manipulation:

Threshold misuse:

Confusing CEA with rationing:

Statistical uncertainty:

Real-world vs trial efficacy gap:

Board pearl: When a CEA reports a strikingly favorable ICER, the first quality check on Step 3 is to ask "what was the comparator?" — many overstated analyses compare against placebo or no treatment when standard care exists, artificially inflating incremental benefit and depressing the ICER.

When to Escalate — When CEA Findings Should Change Practice

— Intervention is dominant (cost-saving and more effective) — adopt without ICER

— ICER well below WTP and robust in PSA (>80% probability of cost-effectiveness on CEAC)

— Multiple independent analyses converge

— Aligns with clinical guideline recommendations (Class I/IIa)

— Intervention is dominated

— ICER far above WTP across all sensitivity analyses

— Clinical benefit marginal or uncertain

— ICER near threshold, PSA shows ~50% probability of cost-effectiveness

— Decision should depend on budget impact, equity, and clinical urgency

— P&T committee review for formulary additions

— ACO/payer review for population-level adoption

— Specialty society guideline integration

— CMS National Coverage Determination (NCD) for Medicare — note that CMS uses "reasonable and necessary" rather than explicit ICER thresholds

— When a patient cannot afford a cost-effective therapy → patient assistance program, manufacturer copay card, switch to therapeutic alternative, social work consult

— Document financial toxicity assessment in the chart — increasingly a quality metric

High-confidence adoption signals:

High-confidence rejection signals:

Equipoise / context-dependent:

Escalation pathways in real systems:

Individual escalation:

CCS pearl: In a CCS-style scenario where a patient is non-adherent due to drug cost, the correct action sequence is: (1) verify adherence barrier, (2) consult social work / pharmacy, (3) switch to formulary-preferred or generic alternative with equivalent efficacy, (4) arrange follow-up to confirm adherence and reassess outcome. Continuing the unaffordable drug without addressing cost is incorrect.

Key Differentials — Related Economic Evaluation Methods

Distinguish CEA from its same-category cousins:

— Used when outcomes are equivalent between alternatives (e.g., two generic statins)

— Picks the lowest-cost option

— Rarely valid in practice because true outcome equivalence is uncommon

— A subtype of CEA using QALYs (or DALYs) as the effectiveness measure

— When Step 3 says "CEA," most often means CUA

— Allows comparison across diseases and interventions (a unique strength of QALYs)

— Both costs and outcomes monetized (e.g., statistical value of life)

— Yields net monetary benefit (NMB) or benefit-cost ratio

— Used heavily in environmental/regulatory economics; less common in clinical medicine

— Avoids QALY but raises ethical concerns about assigning dollar values to life

— Reports costs and multiple outcomes separately, leaving the decision-maker to weigh

— Useful when no single utility measure captures all relevant outcomes

— Short-term (1–5 year) financial impact for a specific payer/population

— Complements CEA; does not replace it

— NMB = (E × WTP) − C

— Positive NMB → cost-effective; useful for ranking many interventions

Cost-minimization analysis (CMA):

Cost-utility analysis (CUA):

Cost-benefit analysis (CBA):

Cost-consequence analysis (CCA):

Budget impact analysis (BIA):

Net monetary benefit (NMB):

Key distinction: CEA/CUA compares interventions using a ratio ($/QALY); CBA compares using net dollars. CEA preserves health outcomes in natural units (QALYs) and is generally preferred in clinical medicine because clinicians and patients find it more interpretable than pricing a life-year in dollars.

Key Differentials — Other-Category Concepts Often Confused with CEA

Distinguish CEA from non-economic study designs and metrics that students conflate:

— Clinical efficacy metric, not an economic one

— NNT × per-patient cost ≈ rough cost per event prevented (but not per QALY)

— Lower NNT is better; does not account for cost or quality of life

— Efficacy = ideal-conditions benefit (RCT)

— Effectiveness = real-world benefit (pragmatic trial, observational)

— CEA should use effectiveness estimates when possible; using efficacy overstates value

— Compares clinical outcomes of two active interventions; does not require cost data

— CEA builds on CER by adding cost

— PCORI (Patient-Centered Outcomes Research Institute) funds CER in the US but is statutorily prohibited from using QALYs as the primary measure

— Health-system payment models linking reimbursement to outcomes

— Informed by CEA but operationally distinct

— Process or outcome benchmarks; track care quality, not cost-effectiveness directly

— Observational data (claims, EHR, registries) increasingly feed CEA inputs

Number needed to treat (NNT):

Effectiveness vs efficacy:

Comparative effectiveness research (CER):

Pharmacoeconomics: umbrella discipline encompassing CEA, CUA, CMA, CBA, BIA applied to drugs

Value-based care / pay-for-performance:

Quality measures (HEDIS, MIPS):

Real-world evidence (RWE):

Board pearl: A vignette describing "cost per case prevented" is doing a non-utility CEA (using cases prevented as the effect measure). This is acceptable but less comparable across diseases than $/QALY — you cannot compare $/case-of-stroke-prevented to $/case-of-cancer-prevented directly, but you can compare $/QALY across both. The QALY's portability is its defining methodologic strength.

Secondary Prevention — Applying CEA to Long-Term Care Decisions

CEA is heavily applied to secondary prevention, which is where Step 3 frequently tests it:

— High-intensity statin: ICER <$25,000/QALY — high value, near-universal adoption

— Aspirin for established CAD: cost-saving (dominant) in most analyses

— PCSK9 inhibitors in very high-risk patients: original launch ICERs >$300,000/QALY; price negotiations now bring closer to $50,000–100,000/QALY

— Cardiac rehabilitation post-MI: ICER <$50,000/QALY, often underutilized despite cost-effectiveness

— ICD for primary prevention in EF ≤35%: ~$30,000–70,000/QALY depending on subgroup

— Metformin in T2DM: dominant/cost-saving

— SGLT2 inhibitors and GLP-1 RAs in T2DM with CV/renal disease: ICERs variable, often $50,000–150,000/QALY; falling as generics emerge

— ACEi/ARB in diabetic nephroprotection: highly cost-effective

— Benefits accrue over decades — discounting reduces apparent value

— Why 2.5–3% rather than higher rates is conventional

Cardiovascular secondary prevention (post-MI, post-stroke):

Diabetes secondary prevention:

COPD/asthma: smoking cessation interventions (counseling + pharmacotherapy) are dominant — pay less, gain more QALYs

Cancer survivorship: surveillance imaging frequency increasingly evaluated; over-screening can be cost-ineffective

Discharge medication reconciliation programs: ICER consistently <$50,000/QALY by reducing readmissions

Discount considerations for prevention:

Step 3 management: On post-MI discharge, the most cost-effective bundle combines aspirin + high-intensity statin + beta-blocker + ACEi/ARB + cardiac rehab referral + smoking cessation counseling — virtually every element is dominant or high-value (<$50,000/QALY). Missing rehab referral is both a quality-of-care gap and a cost-effectiveness gap on Step 3 vignettes.

Follow-Up — Monitoring CEA Assumptions Over Time

— Drug prices fall when generics or biosimilars enter (ICERs improve)

— Comparator standards change — yesterday's standard is today's outdated comparator

— New evidence alters effectiveness estimates

— Population epidemiology shifts (prevalence, baseline risk)

— Utility weights evolve with new measurement methods

— HTA bodies (NICE, ICER) typically revisit major decisions every 3–5 years

— Formulary decisions reviewed annually by P&T committees

— Coverage policies updated with new pivotal trials

— Real-world adherence (claims data, MPR — medication possession ratio)

— Outcome registries to validate trial-derived effectiveness estimates

— Cost trends (drug price index, hospital cost growth)

— Equity audits — are benefits reaching all subgroups?

— Discuss out-of-pocket cost openly during prescribing — financial toxicity is a quality measure

— Use generic substitution, 90-day fills, mail-order pharmacy, manufacturer assistance as appropriate

— Document the shared decision-making conversation

— Diabetes Prevention Program (DPP) — covered by Medicare as a cost-effective intervention (ICER <$15,000/QALY for delayed diabetes onset)

— Cardiac and pulmonary rehab — underutilized despite cost-effectiveness; track referral rates as a quality metric

Why CEA conclusions expire:

Reassessment cadence:

Monitoring parameters in practice:

Counseling patients about cost:

Rehab and population programs:

Board pearl: A CEA published in 2015 showing an oncology drug at $200,000/QALY may now be at $80,000/QALY because the drug went generic or because price negotiation lowered acquisition cost. ICERs are not static — Step 3 expects you to recognize that current price and current comparator drive the conclusion, not the historical published ratio.

Ethical, Legal, and Patient Safety Considerations

— Discriminates against disability: a year lived with disability counts as less than a year of perfect health, potentially undervaluing interventions for disabled persons

— Age bias: older adults gain fewer QALYs simply by having fewer years remaining

— Aggregation hides distribution: total QALY gain can mask which subgroups benefit

— ACA §1182 prohibits the Patient-Centered Outcomes Research Institute (PCORI) and HHS from using QALY-based thresholds to deny Medicare coverage

— Inflation Reduction Act (2022) Medicare drug price negotiation explicitly excludes QALY-based pricing

— Several state laws (e.g., Oklahoma, Texas) restrict Medicaid use of QALYs

— Disability advocates have successfully challenged QALY-based coverage denials

— Patients have a right to know out-of-pocket costs before consenting to therapy

— Financial informed consent is increasingly considered part of valid consent for expensive interventions

— Withholding cost information that leads to non-adherence is a patient safety issue

— Discharging a patient on a $2,000/month medication without verifying affordability or arranging follow-up is a recognized patient safety failure

— Always check formulary, copay, and patient assistance before discharge prescribing

— Use teach-back to confirm patient understands cost and adherence plan

— Use distributional CEA or equity weights when interventions disproportionately benefit advantaged groups

— Document and disclose conflicts of interest in industry-sponsored CEAs

Ethical critiques of QALYs:

US legal/policy landscape:

Informed consent and shared decision-making:

Transitions of care — Step 3 patient safety vignette:

Equity and procedural justice:

Mandatory reporting: not applicable to CEA itself, but reporting adverse events affects cost inputs

Step 3 management: Before discharge on a high-cost medication, the safe and ethically required steps are: verify insurance coverage, check copay, offer assistance program enrollment, confirm patient understanding, and arrange 1–2 week follow-up to assess adherence. Skipping these steps is the most testable patient-safety failure linked to cost.

High-Yield Associations and Rapid-Fire Clinical Facts

ICER formula: (C₁ − C₀) / (E₁ − E₀); units $/QALY

WTP thresholds (US): $50K (high value), $100K (intermediate), $150K (upper bound)

NICE threshold (UK): £20,000–£30,000/QALY

QALY = years × utility weight (0–1); perfect health = 1.0; death = 0

DALY is bad (minimize); QALY is good (maximize)

Dominant = cheaper AND more effective → adopt without ICER

Dominated = costlier AND less effective → reject

Discount rate: 3%/year for costs and QALYs in US analyses

Reference perspective: societal (preferred) + healthcare sector (Second Panel 2016)

Reporting standard: CHEERS 2022 checklist

Most cost-effective categories: childhood vaccines, smoking cessation, aspirin in CAD — frequently dominant

Frequently low-value: late-stage oncology drugs with marginal survival, intensive screening in elderly with limited life expectancy

Sensitivity analysis types: one-way, multi-way, probabilistic (PSA → CEAC)

Markov models: chronic disease with recurring health states

NMB = (E × WTP) − C; positive → cost-effective

Value-based price: price at which ICER = WTP

CEAC: plots probability of cost-effectiveness across WTP values

Medicare cannot use QALY thresholds (ACA §1182, IRA 2022)

ICER organization (Boston-based): publishes US value benchmarks

Budget impact analysis ≠ CEA: answers "can we afford it?" not "is it worth it?"

Cost-saving ≠ cost-effective: cost-saving is a stronger claim (dominant)

HPV vaccine: ICER consistently <$30K/QALY ages 9–26; lower value 27–45

Statin for secondary prevention: ICER ~$10–25K/QALY

Cardiac rehab: ICER <$50K/QALY, underutilized

PCSK9 inhibitors: originally high ICER; improved with price reductions

Board pearl: If asked the single most important assumption in a CEA, the answer is almost always the comparator — wrong comparator invalidates the entire analysis regardless of how rigorous the rest of the methodology.

Board Question Stem Patterns

Recognizable Step 3 stems and the response framework:

— Stem provides cost and QALY for two strategies, names a WTP threshold

— Approach: compute (ΔC/ΔE), compare to WTP, choose accordingly

— Watch for dominance — if one strategy is cheaper and more effective, pick it without calculation

— Order by QALYs, eliminate dominated, compute sequential ICERs, apply weak dominance, compare to WTP

— Answer is usually not the highest-QALY option if its incremental ICER exceeds the threshold

— "Which best describes a QALY?" → year of life weighted by health utility (0–1)

— "What does the ICER represent?" → incremental cost per additional QALY

— Stem describes a CEA with a problem (wrong comparator, short horizon, missing AEs); pick the methodological critique

— Vignette describes formulary committee or insurer decision; pick the strategy with acceptable ICER + manageable budget impact

— Disability advocacy concern → recognize QALY discrimination critique, evLYG alternative, ACA §1182

— Patient cannot afford a cost-effective drug → choose switch to formulary alternative + patient assistance + follow-up, not "continue current regimen"

— CEAC shows 90% probability of cost-effectiveness at $100K/QALY → robust finding, support adoption

— Tornado diagram shows utility is the dominant driver → flag uncertainty in that input

Pattern 1 — Two-strategy ICER calculation:

Pattern 2 — Three-strategy ranking:

Pattern 3 — Conceptual definition:

Pattern 4 — Methodological flaw recognition:

Pattern 5 — Policy/coverage decision:

Pattern 6 — Equity/ethics framing:

Pattern 7 — Patient-level cost barrier:

Pattern 8 — Sensitivity analysis interpretation:

Step 3 management: When uncertain on a CEA stem, the safest defaults are: (1) check dominance first, (2) ICER between $50K–$150K/QALY = cost-effective at typical US WTP, (3) cheaper + more effective always wins, (4) when cost causes non-adherence, switch the drug rather than continue and counsel.

One-Line Recap

Cost-effectiveness analysis quantifies the value of a healthcare intervention as the incremental cost per quality-adjusted life year (ICER = ΔCost/ΔQALY), interpreted against a willingness-to-pay threshold (commonly $50,000–$150,000/QALY in the US) to guide population-level adoption decisions while respecting individual clinical judgment, equity, and patient affordability.

High-yield recap bullets:

Formula and units: ICER = (C₁ − C₀)/(E₁ − E₀), expressed in $/QALY; lower is better; compare to WTP threshold (US: $50K high value, $100K intermediate, $150K upper bound).

Dominance always trumps calculation: an intervention that is cheaper AND more effective dominates — adopt without ICER; an intervention that is costlier AND less effective is dominated — reject. Always scan for dominance before doing arithmetic.

QALY mechanics: years of life × utility weight (0–1); perfect health = 1.0, death = 0; standard US discount rate is 3%/year for both costs and QALYs; lifetime horizon for chronic disease.

Policy and patient-safety integration: Medicare cannot use QALY-based thresholds (ACA §1182, IRA 2022); QALYs have legitimate equity critiques (age, disability); at the patient level, financial toxicity is a safety issue — verify affordability, use generic/formulary alternatives, engage patient assistance programs, and schedule short-interval follow-up after discharge on costly medications.

Board pearl: On Step 3, the most common right answer when cost is in the stem is not the cheapest option and not the most effective option — it is the option whose ICER falls below the stated WTP threshold while remaining clinically appropriate and accessible to the individual patient.