An indirect treatment comparison (ITC) estimates the relative effect of two treatments that have never been compared head-to-head in a clinical trial, using evidence from other trials.
In health technology assessment (HTA), this situation is common. A new medicine is usually tested against placebo or against one standard treatment, while HTA bodies want to know how it compares with every relevant option used in their country. Indirect comparisons help close that gap.
In simple terms, an indirect comparison answers the question: how does treatment A compare with treatment B, when there is no trial of A versus B?
Why are indirect comparisons important in HTA?
HTA decisions are organised around a PICO: the Population, the Intervention, the Comparators and the Outcomes. The comparators requested by the HTA body are often not the ones used in the pivotal trials.
Without comparative evidence against those comparators, the assessment faces a critical evidence gap, and the added value of the technology may be questioned.
This has become even more relevant with the EU Joint Clinical Assessment, where several PICOs can be requested for the same product. Read more in our article on the EU Joint Clinical Assessment.
What are the main methods?
Bucher method (anchored indirect comparison)
When two treatments have been compared with the same common comparator, for example placebo, their relative effect can be estimated through that common comparator. The randomisation within each trial is preserved. It is simple and transparent, but it only connects a small number of treatments.
Network meta-analysis (NMA)
A network meta-analysis combines direct and indirect evidence from a network of trials to compare several treatments at the same time. It can rank treatments and use all the available evidence, but it depends on the trials being similar enough to be combined.
Population-adjusted indirect comparisons (MAIC and STC)
When the trials differ in patient characteristics that affect the outcome, a standard comparison may be biased. Population-adjusted methods use individual patient data from one trial to adjust for those differences.
- Matching-adjusted indirect comparison (MAIC) reweights the individual patients of one trial so that their characteristics match the published average characteristics of the other trial.
- Simulated treatment comparison (STC) uses a regression model to predict the outcome of one treatment in the population of the other trial.
These methods are also used when there is no common comparator, for example with single-arm trials. These unanchored comparisons rely on stronger assumptions and are treated with more caution by HTA bodies.
Which assumptions need to be checked?
- Similarity. The trials should be comparable in the factors that modify the treatment effect, such as patient characteristics, disease severity, treatment duration and outcome definitions.
- Homogeneity. Trials comparing the same treatments should give consistent results.
- Consistency. In a network, direct and indirect evidence for the same comparison should agree.
If these assumptions are not met, the results may be biased. A good analysis reports how each assumption was assessed and how the remaining uncertainty affects the conclusions.
What do European guidelines say?
The Member State Coordination Group on HTA (HTACG) adopted a methodological guideline on direct and indirect comparisons in March 2024, together with a practical guideline for assessors. Both assume that the evidence synthesis starts from a properly conducted systematic literature review.
For companies, this means the choice of method has to be justified, the assumptions made explicit and the limitations reported transparently. An analysis that looks convenient but cannot be defended will lose weight in the assessment.
How to plan an indirect comparison
- Start from the PICO. Define the comparators and outcomes that the HTA body is likely to request.
- Run a systematic literature review. Identify all relevant trials for those comparators. Our article on systematic literature reviews in HTA explains why this step matters.
- Assess feasibility. Compare the trials in design, population, treatment duration, outcome definitions and timepoints before choosing a method.
- Choose the method. Use anchored methods when a common comparator exists; consider population adjustment when there are important differences in effect modifiers.
- Report transparently. Document every assumption, run sensitivity analyses and explain the limitations.
Frequently asked questions
What is the difference between an indirect comparison and a network meta-analysis?
A network meta-analysis is one type of indirect comparison. It combines direct and indirect evidence across a network of trials to compare several treatments at once.
When is a MAIC needed?
When the trials to be compared differ in patient characteristics that modify the treatment effect, and individual patient data are available for at least one of them.
Are indirect comparisons accepted by HTA bodies?
Yes, they are widely used when head-to-head trials are missing. Their weight in the decision depends on the quality of the evidence, the plausibility of the assumptions and how transparently the uncertainty is reported.
Final thoughts
Indirect treatment comparisons are often the only way to show how a new technology compares with the treatments that decision-makers care about. The method matters, but so does the preparation: a clear PICO, a thorough literature review and an honest assessment of the assumptions.
At Clevidence, we design and conduct indirect treatment comparisons, network meta-analyses and population-adjusted comparisons for HTA. See how we used indirect comparisons with network meta-analysis to support a reimbursement decision and how we applied MAIC and STC to address heterogeneity across trials in a rare genetic disease, or learn more about our evidence review and synthesis services.