Skip to content
Back to search results

Quality-Adjusted Life Year (QALY)

Introduced by
Alan Williams (Health Economics Research Centre, Oxford University)
Year
1985
Type
Method
Family
health utility measurement
Font size

What does it do?

A QALY measures health benefit as utility weight (0 = death, 1 = perfect health) multiplied by time lived.

Developed by Alan Williams in 1985, QALYs enable comparison of disparate health interventions on a common metric. Used globally by health technology assessment bodies—NICE (UK), HAS (France), CADTH (Canada), WHO—to decide which treatments deserve public funding.

The intuition behind the method

Health policy requires a single currency to compare treatments: drugs for cancer, cardiac surgery, mental health interventions, preventive programs. A QALY captures two dimensions: how long you live AND how well. An intervention that extends life 2 years in perfect health yields 2 QALYs; an intervention that improves a disease-ridden quality of life from 0.4 to 0.7 utility for 1 year yields 0.3 QALYs. This allows health economists to rank investments by bang-for-buck: cost per QALY gained. Utility weights are not assumed—they come from population surveys using validated instruments (EQ-5D is most common: 5 dimensions, 3-5 severity levels each).

History

Alan Williams introduced QALYs in 1985 as a normative framework for healthcare resource allocation (Williams, A. (1985). Economics of coronary artery bypass grafting. BMJ, 291, 326–329). EQ-5D was developed by the EuroQol Group (1985–1990) to standardize utility measurement across Europe (Brooks, R. (1996). EuroQol: the current state of play). NICE adopted QALY as its core metric in 1999 (UK National Institute for Clinical Excellence guidance). SF-6D, derived from the SF-36, was developed in the late 1990s as an alternative (Brazier et al., 2002). HUI (Canada, Feeny et al. 1992) and 15D (Finland) followed. In 2009, there was growing criticism of QALY for disability discrimination, prompting debate (Amartya Sen, disability rights scholars). By 2020, most major HTAs use QALY as primary measure, with health equity and equity-weighted variants (e-QALY) emerging in some jurisdictions.

When to use it

When a health intervention (drug, device, procedure, public health program) affects both length and quality of life. QALY analysis is mandatory for health technology appraisal (HTA) submissions in most developed countries. Use when comparing treatments across disease areas (diabetes vs cancer), or when quality-of-life improvements matter as much as mortality reduction. Not appropriate for: interventions with non-health benefits (education, employment); acute single events (one-time surgery outcome); or settings where equity concerns override efficiency (QALY is utilitarian, not egalitarian).

Limitations and considerations

  • Utility elicitation is time-consuming and subject to response bias; VAS and TTO produce different utility scores for the same health state.
  • QALY assumes additive time × utility model; assumes constant utility over time; ignores duration effects (1 year at 0.5 utility ≠ same as 0.5 years at 1.0).
  • Does not capture equity: a QALY to a healthy person equals a QALY to a severely disabled person, violating fairness intuitions in some contexts.
  • Health utility varies by culture, age, comorbidity; transferring UK EQ-5D tariffs to low-income countries may misrepresent local preferences.
  • Disability scholars argue QALY discriminates against people living with chronic conditions (assigns lower baseline utility, reducing calculated benefit of interventions for disabled groups).

Source conditions do not replace independent scientific review or validation by an analysis engine.

Strengths

  • Integrates mortality and morbidity into one metric, enabling cross-disease comparison and resource allocation at system level.
  • Utility weights derived empirically from population preferences (not clinician opinion), anchoring valuation in patient/public values.
  • Transparent, auditable methodology: NICE and other HTAs publish detailed QALY models, allowing replication and critique.
  • Accepted internationally by regulators and payers, reducing negotiation friction for market access.

Applications

  • Pharmaceutical submission to NICE, HAS, CADTH: cost-effectiveness model predicts QALYs gained vs current therapy, informing pricing and formulary decisions.
  • Public health program evaluation: estimating QALYs from smoking cessation, vaccination, or weight loss interventions to justify population investment.
  • Clinical trial design: coprimary endpoint in some RCTs is both clinical efficacy (e.g., time to progression) and quality-adjusted survival (QALYs via EQ-5D questionnaire at follow-up visits).
  • Hospital resource allocation: competing surgical programs (transplant, oncology, orthopedics) ranked by cost per QALY to guide capital budgeting.
  • Personalized medicine: utility profiles by genetic subgroup, justifying precision oncology cost for specific mutation carriers.

How it works

How it works

1

Step 1: Recruit a representative sample of the general public, patient population, or healthcare professionals. Step 2: Administer a utility elicitation instrument. EQ-5D (5-item, 10 minutes): asks mobility, self-care, usual activities, pain/discomfort, anxiety/depression on 3-level (1 = no problem, 3 = extreme) or 5-level scale; UK tariff converts responses to utility 0–1. SF-6D (6-item, derived from SF-36): physical functioning, role limitation, social function, pain, mental health, vitality; algorithm produces utility. HUI (Health Utilities Index, Canada): 8 dimensions, 5–6 levels each. Time Trade-Off (TTO): ask patient: 'Would you trade X years in current health state for Y years in perfect health?'; utility = Y/X. Visual Analogue Scale (VAS): simple thermometer 0–100, less rigorous but quick. Step 3: Validate responses for internal consistency and repeatability.

Formula
Elicit health utility weights from patient sample using standardized instruments (EQ-5D, SF-6D, HUI, TTO, VAS, or time trade-off).
2

Formula: QALY = U × t, where U = utility weight (0 ≤ U ≤ 1) and t = time in years. Example: A patient with hypertension has utility 0.85 (85% of perfect health) for 5 years; QALYs = 0.85 × 5 = 4.25. For interventions, calculate incremental QALY: ΔQ = Q_intervention − Q_baseline. If treatment A costs £60,000 and gains 3 QALYs over 10 years, ICER (Incremental Cost-Effectiveness Ratio) = £60,000 / 3 = £20,000 per QALY. Multiple arms: weight by time at each health state. Discounting: future QALYs discounted (typically 3–5% annually) because society values health today more than health in 10 years.

Formula
QALY = Utility Score (0 to 1) × Time in Health State (years). Utility: dead = 0, perfect health = 1. Intermediate states reflect morbidity burden.
3

NICE threshold (UK): interventions costing ≤£20,000/QALY are considered cost-effective; £20,000–£30,000 is marginal; >£30,000 rarely approved unless exceptional circumstances. WHO threshold (developing countries): 1–3 times GDP per capita per DALY averted is cost-effective; equivalent to 1–3 times GDP per QALY in QALY framework. Quadrant analysis: (1) SE (south-east): cost-saving and more effective = clearly adopt; (2) NE (north-east): more costly, more effective = cost-effectiveness ratio vs threshold; (3) SW (south-west): cost-saving, less effective = rarely adopted (worse health gain); (4) NW (north-west): costly and less effective = reject. A non-dominated intervention (inside the acceptability frontier at threshold WTP) is recommended. Sensitivity analysis tests robustness to utility, discount rate, and time horizon assumptions.

Formula
Sum QALYs across life course or trial period. Compare intervention QALYs to standard care. Incremental QALY gain = (New QALYs) − (Baseline QALYs). Apply willingness-to-pay threshold (e.g., £20,000–£30,000/QALY for NICE) to determine cost-effectiveness.

Common pitfalls

  • Assuming utility is measured, not elicited: using clinician opinion or disease-specific scales (e.g., EORTC QLQ-C30 raw scores) instead of preference-weighted instruments. Always use EQ-5D, SF-6D, or HUI with published tariffs.
  • Forgetting to discount: failing to discount future QALYs at 3–5% means overvaluing distant benefits; most health systems require discounting.
  • Conflating QALY with quality of life (QoL): QALY is preference-weighted (what people trade off); QoL is descriptive (symptom burden, functioning). A patient may report high QoL (rich social life) but low QALY utility (severe pain).
  • Misinterpreting the threshold: believing £20,000/QALY is a hard rule rather than a guideline; NICE decision is multifactorial (uncertainty, equity, innovation premium, rarity).
  • Using inappropriate comparators: calculating QALY gain vs placebo instead of best current practice, inflating apparent benefit.

Frequently asked questions

How are utility weights assigned if a health state is not directly measured?

Mapping or extrapolation. If a trial uses disease-specific scales (e.g., EORTC QLQ-C30) but not EQ-5D, researchers use published mapping algorithms to convert disease scores to EQ-5D utilities. Alternatively, they conduct a small utility study in similar patients. Mapping reduces precision; direct measurement preferred.

What is the difference between QALY and DALY?

QALYs measure health gained (0 = death, 1 = perfect health); DALYs measure disease burden (0 = perfect health, 1 = death). QALY = utility × time (positive framing); DALY = YLL (years of life lost to death) + YLD (years lived with disability). NICE uses QALYs; WHO GBD uses DALYs. Inverse relationship: QALY gained ≈ DALY averted.

Why is discounting applied to QALYs in the future?

Time preference: society values immediate health more than delayed health (£1 today > £1 in 10 years). Standard rate in UK/Canada is 3–5% annually. A 1 QALY in year 10 is worth ~0.7 QALYs in net present value at 3% discount rate. Prevents overinvestment in interventions with distant, small benefits.

Is the EQ-5D utility valid for all populations?

No. EQ-5D tariffs vary by country (UK, France, Germany, US tariffs differ). EQ-5D is less sensitive in mild disease states and in pediatric populations. Cultural values, disability prevalence, and healthcare experience affect utility elicitation. When applying EQ-5D to a new population, a local valuation study is ideal; if not feasible, explicit sensitivity analyses comparing tariffs are required.

Can a treatment have negative QALYs (utility worse than death)?

In principle, yes, if a treatment extends life but creates a health state perceived as worse than death (utility < 0). In practice, most instruments cap minimum utility at 0 (equivalent to death) or slightly negative (rare, severe states). Interventions producing net negative QALYs are not recommended.

How is uncertainty in QALYs handled in cost-effectiveness analysis?

Sensitivity analysis: deterministic (vary utility ± 10%, discount rate ± 1%, time horizon ± years one at a time) and probabilistic (Monte Carlo simulation drawing 10,000 samples from distributions of all parameters). Acceptability curve plots probability that intervention is cost-effective across a range of willingness-to-pay thresholds. Scenario analysis tests structural assumptions (e.g., different utility instruments).

Related methods

Sources

  1. Kind, P. (1989). The EuroQol instrument: an index of health-related quality of life. In B. Teeling Smith (Ed.), Measuring health: a practical approach. Chichester: Wiley.

    Open publication reference
    Copied.
  2. Weinstein, M. C., & Stason, W. B. (1976). Hypertension: A Policy Perspective. Cambridge, MA: Harvard University Press.

    Open publication reference
    Copied.
  3. Brooks, R. (1996). EuroQol: the current state of play. Health Policy, 37(1), 53-72.

    Open DOI reference
    Copied.