C. Comparative Effectiveness Research and HTA

Comparative effectiveness research (CER) is the generation and synthesis of evidence comparing the benefits and harms of alternative technologies to prevent, diagnose, treat, and monitor diseases and other health care conditions in “real-world” settings in order to improve the delivery of health care (Federal Coordinating Council on Comparative Effectiveness Research 2009; Institute of Medicine 2009). The purpose of CER is to strengthen the evidence base that is used for assisting patients, other consumers, clinicians, health care managers, policymakers and others to make more informed health care decisions for individuals and populations.

Various attributes of what is known today as CER have been incorporated into research on the impact of health care technologies over the last several decades or more. Indeed, CER draws selected attributes from such inquiries and evaluations as RCTs, HTA, outcomes research, effectiveness research, and evidence based medicine. The emergence of CER as an explicit, coherent field of research in the early 2000s arose from a growing recognition of such factors as:

• Evidence of inappropriate use of health care technologies, including over-use, under-use, and improper use
• Evidence of large variations (geographic and other) in practice
• Insufficiency of evidence developed for market approval/clearance by regulatory agencies (e.g., FDA) to also support clinical and policy decisions; this is typically because such evidence:
  • emphasizes efficacy rather than effectiveness
  • is often not derived from controlled clinical trials (especially for many medical devices)
  • often lacks active comparators (e.g., where placebos are the only controls)
  • tends to exclude certain patient populations (those who are elderly, have multiple co-morbidities, etc.)
  • often is not derived from studies that enable subgroup analyses
• Insufficiently rigorous or absent evidence for the many technologies that are not subject to oversight by regulatory agencies (e.g., surgical procedures)
• Lack of evidence from “head-to-head” comparisons of alternative interventions for particular health problems
• Lack of evidence in “real-world” practice (efficacy vs. effectiveness)
• Continued steep increases in health care costs prompting interest in more efficient care delivery

The main attributes of CER are:

• Direct (“head-to-head”) comparisons of alternative interventions (rather than comparison with placebo or indirect comparisons)
• Applies to all types of technologies
• Measures effectiveness in real-world populations and health care settings
• Emphasizes health care outcomes (e.g., morbidity, mortality, symptoms, quality of life, adverse events) rather than surrogate or other intermediate endpoints
• Draws on variety of complementary research methods, data sources, and analytical tools
• Enables subgroup analyses to yield findings about different responses across patient types
• Includes emphasis on priority diseases and priority populations

In the US, there is no consensus on the role of economics, such as determining value for money, in the conduct of CER itself. Indeed, there are legislative constraints on the use of CER and cost-effectiveness analysis in making coverage decisions for the Medicare program and other health programs administered by the Department of Health and Human Services (Neumann 2012; Pearson 2010). Even so, the findings of CER and other evidence that may involve health economic analyses, such as cost-effectiveness analysis, is available for use by various analysts and decision makers. While the term “CER” is most often used in the US, other countries and regions use related, though not necessarily synonymous terms. For example, in Europe, there has been growing interest in “relative efficacy” and “relative effectiveness” trials for new drugs. This interest derives in part from differing evidence requirements that sometimes arise between regulatory agencies, particularly the European Medicines Agency (EMA), and payment authorities. Approaches under consideration involve identifying circumstances in which pre-licensing (pre-marketing) efficacy trials should use active comparators, and ways to close gaps between efficacy and effectiveness, including whether to conduct pre-licensing or post-licensing practical clinical trials (with more heterogeneous patient populations in real-world settings), or to better understand how such extrinsic factors as physician prescribing or patient adherence affect variability in drug response (Eichler 2009; Glaeske 2012). These proposals may affect the relative roles and relationships of regulators, payers, and HTA agencies.

CER is generating more of certain types of evidence that have been of increasing relevance to HTA. Among the broad set of attributes or impacts assessed in HTA, those that CER emphasizes are effectiveness and safety in real-world patients and health care settings, patient outcomes, and direct comparisons to standards of care. Further, CER is contributing to tools and methods that will broaden the scope and strengthen the quality of available evidence, including the development of better health outcome and quality of life measures, observational data sources such as registries and insurance claims, and alternative designs for clinical trials. As such, when HTA develops evidence questions, conducts systematic literature searches, conducts meta-analyses and other evidence syntheses, and develops findings and recommendations, it will draw on an evidence base enriched by CER findings.