Comparative Effectiveness Research

Comparative effectiveness research (CER) compares the benefits and harms of alternative interventions for preventing, diagnosing, treating, or managing a condition, with the explicit aim of helping clinicians, patients, and policymakers make informed decisions. It emphasises real-world effectiveness rather than efficacy under idealised trial conditions, and head-to-head comparisons of options that are actually available in practice.

Definition

Comparative effectiveness research is the generation and synthesis of evidence that compares the benefits and harms of alternative methods to prevent, diagnose, treat, and monitor a clinical condition or improve care delivery, intended to inform decisions in real-world settings.

Scope

The entry covers the purpose and defining features of CER, the range of designs it uses (pragmatic trials and observational comparisons), the analytic challenges of non-randomised comparison, and its relationship to patient-centred outcomes. It is methodological in framing and gives no clinical or coverage recommendations.

Core questions

• Which of the available options works best, for whom, and under what circumstances?
• When should comparisons rely on pragmatic trials versus observational data?
• How are real-world effectiveness and idealised efficacy distinguished?
• How are confounding and selection handled in non-randomised comparisons?

Key concepts

• Effectiveness versus efficacy
• Head-to-head comparison of active options
• Pragmatic trials
• Observational and database comparisons
• Patient-centred outcomes
• Confounding by indication
• Propensity-score and doubly robust methods
• Heterogeneity of treatment effect
• Research prioritisation

Mechanisms

CER asks which of the genuinely available options is most effective in routine conditions, so it favours active comparators, clinically meaningful outcomes, and broadly representative patients. Evidence comes from pragmatic randomised trials, which preserve randomisation while relaxing the artificial constraints of explanatory trials (Ford & Norrie, 2016), and from observational comparisons using large datasets, where credible inference depends on controlling confounding by indication through propensity scores, doubly robust estimation, and related methods (Funk et al., 2011). Because comparisons are between active treatments rather than against placebo, and because effects may vary across subgroups, attention to heterogeneity of treatment effect is central. National bodies have defined CER's scope and set research priorities to address the most consequential decisions (Sox & Greenfield, 2009; Iglehart, 2009).

Clinical relevance

CER aims to inform real-world choices among interventions and to surface evidence relevant to patients typically excluded from explanatory trials. Understanding its designs supports appraisal of head-to-head evidence. This entry describes a research enterprise and is not a basis for individual diagnostic or treatment decisions or for coverage determinations.

Evidence & guidelines

The Institute of Medicine defined CER and set research priorities (Sox & Greenfield, 2009; Iglehart, 2009). Methodological references describe pragmatic trial design (Ford & Norrie, 2016) and the handling of confounding in observational comparisons (Funk et al., 2011). These sources are methodological and do not recommend specific treatments.

History

Although comparing treatments is as old as clinical research, CER acquired its name and priority in the United States around 2009, when the American Recovery and Reinvestment Act funded it and the Institute of Medicine defined its scope and priorities (Sox & Greenfield, 2009; Iglehart, 2009). The subsequent creation of the Patient-Centered Outcomes Research Institute institutionalised a patient-centred orientation, and pragmatic trial methods and large observational databases became its principal tools.

Debates

Pragmatic trials versus observational comparisons

Pragmatic randomised trials offer protection against confounding but can be costly and slow, while observational database studies are fast and broad but vulnerable to confounding by indication; the field debates when each is sufficient for decision-making.

Key figures

• Harold Sox
• Sheldon Greenfield
• John Iglehart
• Ian Ford

Seminal works

• sox-2009-iom
• iglehart-2009
• ford-2016

Frequently asked questions

How does comparative effectiveness research differ from a typical efficacy trial?

Efficacy trials test whether an intervention works under idealised conditions, often against placebo, while CER compares the real-world effectiveness and harms of active options that clinicians and patients actually choose between.

Does comparative effectiveness research require randomisation?

No. CER uses both pragmatic randomised trials and observational comparisons of large datasets; the latter require careful adjustment for confounding by indication to support credible conclusions.