ICER (Incremental Cost Effectiveness Ratio)

The incremental cost effectiveness ratio (ICER) is the ratio of the difference in costs between two health interventions to the difference in their health outcomes, most commonly measured in quality adjusted life years (QALYs). It expresses the additional cost required to gain one additional unit of health benefit and serves as the primary metric for judging whether a new treatment offers good value for money.

The ICER sits at the center of health economic decision making in many health systems. HTA bodies use it to determine whether a new therapy, device, or diagnostic merits public funding. Pharmaceutical and medical device companies calculate ICERs as part of their value dossiers and pricing strategies. Clinicians encounter ICERs in guideline recommendations that distinguish first line therapies from alternatives reserved for patients who fail cheaper options. Because health care budgets are finite, the ICER provides a principled way to compare the value of different investments in health, even across unrelated disease areas.

The ICER is calculated using the formula: ICER = (Cost_new minus Cost_comparator) divided by (Effect_new minus Effect_comparator). The numerator captures the incremental cost of the new intervention, including drug acquisition, administration, monitoring, and any savings from avoided hospitalizations or adverse events. The denominator captures the incremental health gain, usually in QALYs but sometimes in life years gained or other natural units.

The resulting ratio is plotted on the cost effectiveness plane, a two dimensional graph where the x axis represents incremental effect and the y axis represents incremental cost. If the new intervention falls in the northeast quadrant (more effective and more costly), the ICER determines whether the extra cost is justified. If it falls in the southeast quadrant (more effective and less costly, or "dominant"), it is clearly preferred. The northwest quadrant (less effective and more costly, or "dominated") means the new intervention should be rejected. The southwest quadrant (less effective and less costly) requires the ICER to determine whether the savings justify the loss in effectiveness.

Decision makers compare the ICER against a willingness to pay (WTP) threshold, which varies by country and context. NICE in England has used a range of 25,000 to 35,000 pounds per QALY since April 2026 (previously 20,000 to 30,000 pounds). The United States has no official threshold, but benchmarks of $100,000 to $150,000 per QALY are used, for example by the Institute for Clinical and Economic Review. For an intervention in the northeast quadrant, an ICER below the threshold means it is deemed cost effective; in the southwest quadrant the rule reverses, because there the ICER is the saving per QALY given up.

Because ICERs are point estimates derived from models with uncertain parameters, analysts conduct sensitivity analyses. One way deterministic sensitivity analysis varies one parameter at a time to identify influential inputs. Probabilistic sensitivity analysis (PSA) simultaneously varies all parameters according to their distributions over thousands of simulations, generating a cloud of incremental cost and incremental effect pairs on the cost effectiveness plane. The results of PSA are summarized in a cost effectiveness acceptability curve (CEAC), which shows the probability that the intervention is cost effective across a range of WTP thresholds.

Calculate an ICER whenever you are comparing a new health intervention against an existing standard of care and need to express relative value in a way that decision makers can evaluate against spending thresholds. ICERs are standard in HTA submissions, formulary reviews, and clinical guideline development. They are less appropriate when comparing interventions with entirely different outcome measures that cannot be reduced to a common metric, or when budget impact rather than efficiency is the primary concern.

A common error is comparing the ICER against a threshold from a different country or context without adjusting for differences in health system costs and valuation methods. Another mistake is reporting a single deterministic ICER without conducting probabilistic sensitivity analysis, which fails to communicate the uncertainty around the estimate. Analysts sometimes also make the error of calculating an ICER when the new intervention is dominated (more costly and less effective), in which case the ICER is meaningless and the intervention should simply be rejected. Finally, some submissions ignore relevant cost offsets, such as reduced hospitalizations, which biases the ICER upward.

An economic evaluation compares a new biologic therapy for rheumatoid arthritis ($18,000 per year) against methotrexate ($2,000 per year). Over a 10 year model horizon, the biologic produces 6.8 QALYs compared to 6.2 QALYs for methotrexate, with total discounted costs (drug acquisition, monitoring and other care) of $180,000 and $85,000 respectively. The ICER is ($180,000 minus $85,000) divided by (6.8 minus 6.2), which equals $158,333 per QALY. Probabilistic sensitivity analysis shows a 35% probability of cost effectiveness at a $150,000 per QALY threshold, prompting the manufacturer to negotiate a confidential price discount with the payer.