Comparative Effectiveness Research in Pediatric Infectious Diseases.

Comparative-effectiveness research (CER) is a hot topic. There are several definitions for CER, but the Institute ofMedicine defines CER as “the generation and synthesis of evidence that compares the benefits and harms of alternative methods to prevent, diagnose, treat andmonitor a clinical condition or to improve the delivery of care” [1]. Comparative-effectiveness research focuses more on effectiveness than efficacy; that is, how well something works when implemented in a realworld setting and on comparing alternative treatments, prevention, or diagnostic strategies. Comparativeeffectiveness research also seeks to develop evidence that improves health by focusing studies on outcomes that matter to patients. Comparative-effectiveness research is receiving a lot of attention right now and, frankly, a lot of money. The American Recovery and Reinvestment Act in 2009 devoted over $1 billion to CER. The Patient Protection and Affordable Care Act led to the creation of the Patient-Centered Outcomes Research Institute, which, as of 2014, has awarded over $400 million in CER research. The conduct of CER studies involves a range of methodologies including randomized controlled trials (RCTs), cluster randomized trials, observational studies including those that use large administrative databases, as well as cost-effectiveness analysis and decision analysis [2].Although randomization remains the gold standard for controlling unmeasured confounding in comparative studies, randomized trials have inherent limitations including power and feasibility (especially when studying rare events), high costs, and poor generalizability to real-world care. Advanced methods to control for confounding using observational data, including multivariable modeling, propensity scoring and weights and instrumental variables improve the reliability of observational studies and can greatly improve the ability to do CER when randomized trials are neither available, feasible, nor affordable. These methods of special importance for pediatric infectious disease (ID) research, because children are underrepresented in therapeutic clinical trials [3]. The emerging prominence of CER is in part a response to the fact that healthcare delivery in the United States is highly variable. At the heart of the issue of variability are at least 3 domains: (1) economics: there are geographic differences in reimbursement that create financial incentives to provide more or less care; (2) evidencebase: there is a lack of comparative evidence to support the best and most cost-effective treatment, prevention, or delivery strategies for many common conditions; and (3) implementation: there is failure to translate established evidence into practice, often leading to overuse of unnecessary treatments and services. The field of pediatric IDs is well positioned to embrace CER. All 3 domains mentioned above are relevant for our field. A prominent example is the variability in antibiotic prescribing observed by region [4], across health plans [5], and between hospitals [6]. As pediatric ID physicians, we often seek to understand clinical outcomes for diseases that in some cases are relatively rare and the evidence base is weak. Comparative-effectiveness research therefore represents a natural approach to addressing the types of clinical research questions that we face. Other important examples include comparing shorter versus standard courses of antibiotics for common infections, prolonged intravenous versus early conversion to oral therapy for a variety of serious infections, and preemptive versus prophylactic antiviral medications after transplantation. Regarding implementation, despite clinical guidelines and evidence supporting the use of narrow-spectrum therapies for conditions including pneumonia and pharyngitis, guideline implementation and successful translation to clinical practice has only beenmodestly successful [7,8]. Editorial Commentary