# Comparative effectiveness
> Comparative effectiveness assesses how two or more interventions perform relative to each other for a defined patient population, using direct trials, indirect comparisons or real-world studies.
Source: https://www.visfo.health/glossary/comparative-effectiveness
Updated: 2026-08-16T21:55:00.428492+00:00

What is comparative effectiveness?
Comparative effectiveness is the direct or indirect comparison of two or more healthcare interventions to determine which may produce better outcomes for a defined patient population. Interventions can include medicines, devices, procedures or models of care, while outcomes may cover efficacy, safety, quality of life, treatment persistence, healthcare use and other patient-relevant effects.

Evidence may come from head-to-head clinical trials, network meta-analyses or real-world studies. The conclusion is specific to the population, comparator, outcome, setting and follow-up period assessed; it is not a universal ranking of treatments.

Why does comparative effectiveness matter in market access and HEOR?
Comparative evidence helps payers, health technology assessment bodies and guideline developers judge how an intervention performs against current treatment options. For [market access](/glossary/market-access) teams, it can affect coverage, reimbursement, treatment positioning and the strength of claims about added value.

Within [HEOR](/glossary/health-economics-and-outcomes-research-heor), relative treatment effects also inform estimates of costs and health outcomes. Weak or poorly matched comparisons create uncertainty that can carry through into submissions, economic analyses and decisions about resource allocation.

How is comparative effectiveness assessed in practice?
Teams first define the decision question: the target population, interventions, relevant comparators, outcomes, treatment setting and time horizon. They then identify the best available evidence and select a method suited to the evidence network and the decision being made.

- Head-to-head trials directly compare interventions within the same study.
- Network meta-analysis combines direct and indirect trial evidence across a connected set of comparators.
- Observational analyses use [real-world evidence](/glossary/real-world-evidence-rwe) to compare outcomes in routine care, usually with methods intended to reduce confounding and selection bias.
- Adjusted indirect comparisons may be used when studies differ in their patient characteristics or when individual patient data are available for only part of the evidence base.

Results should report the relative effect, its uncertainty and the assumptions behind the analysis. Sensitivity, subgroup and scenario analyses help show whether conclusions change when evidence or methodological choices vary.

What is the difference between direct and indirect comparative evidence?
Direct evidence comes from patients assigned to or observed on the treatments being compared within the same study. It usually reduces problems caused by differences between separate study populations, outcome definitions and periods of follow-up, although trial design and execution still affect validity.

Indirect evidence estimates a comparison through one or more common comparators. For example, two treatments may each have been studied against usual care but not against each other. This approach can fill an evidence gap, but its credibility depends on the studies being sufficiently similar and on assumptions such as consistency between direct and indirect evidence.

What can make a comparative effectiveness result misleading?
A technically correct analysis can still answer the wrong question if it uses an irrelevant comparator, an unrepresentative population or outcomes that do not matter to decision-makers. Differences in baseline risk, prior treatment, outcome measurement, follow-up and treatment switching can also distort comparisons.

In observational research, treatment selection is not random, so differences between groups may reflect patient characteristics or clinical practice rather than treatment effects. Statistical adjustment can reduce measured confounding but cannot guarantee that all bias has been removed. Good reporting therefore separates observed findings from causal claims and explains residual uncertainty.

How does comparative effectiveness differ from related evidence concepts?
Comparative effectiveness establishes how outcomes differ between interventions. It is narrower than an overall evidence strategy and does not by itself establish cost effectiveness, affordability or budget impact. Relative effects may instead become inputs to a [health economic model](/glossary/health-economic-model), alongside costs, utilities, disease progression and other assumptions.

It also differs from a single efficacy estimate, which describes whether an intervention works under the conditions studied without necessarily showing how it compares with every relevant alternative. Comparative effectiveness connects those treatment-specific findings to the choices facing clinicians, payers and patients.