LAUSR

Under the Best Pharmaceuticals for Children Act, the National Institute of Child Health and Human Development (NICHD) is tasked with updating annually a priority list of drugs requiring further study in children.1 The 2015 Priority List comprises 17 therapeutic areas (such as infectious disease or cardiovascular disease) containing 78 specific drugs or items.1 The current prioritization process is primarily based on expertise: NICHD first requests solicitations for candidate drugs from stakeholders, including both the US Food and Drug Administration (FDA) and experts in pediatric medicine, and then NICHD working groups prioritize drug nominations for further study based on criteria including the affected target populations, the level of available evidence, and the feasibility of subsequent studies.1 An empirical population-level perspective is required to balance priorities between common drugs with wide-reaching effects (such as asthma medications) and rarer drugs with smaller target populations or more serious consequences (such as cardiovascular drugs). Between 2004 and 2009, NICHD awarded a series of one-time contracts to gather population-level information about pediatric conditions and drug use and guide prioritization. We propose that the annual selection process should routinely and systematically use populationlevel data to address 5 key considerations: the prevalence of disease, patterns of drug use, adverse drug events (ADEs), drug-drug interactions (DDIs), and drugs with potential for misuse (Box). First, the prevalence of disease has long guided drug prioritization, for example, by the World Health Organization for the study of drugs useful in treating worldwide disease. Systematically examining diseases with high morbidity and mortality, and the availability of drugs to treat them, can identify “pharmaceutical gaps” for research. In 2005, NICHD used the Kids’ Inpatient Database to examine the annual frequency of inpatient pediatric conditions. As patterns of morbidity and mortality change, especially among children with medical complexity (CMC), new therapeutic needs are developing. The Kids’ Inpatient Database data, updated every 3 years, should be supplemented using broader, more immediate data from multiple settings of care to address care across the health continuum and emerging disease management needs. These additional data sources include the Pediatric Health Information System data from freestanding children’s hospitals, improved outpatient sources of all-payer prescription data including Medicaid and private claims databases, and developing networks of clinical information systems. Basic frequency data are enhanced by studies of specific diseases’ currently available therapies, clinical outcomes, and costs. The European Priority Medicines Project examined the number of new chemical entities available for major conditions, finding that diseases with the most significant morbidities (including the neuropsychiatric, cardiovascular, respiratory, and digestive diseases common among CMC), compared with diseases with less morbidity, had a paucity of available drug therapies. Second, population-level studies of patterns of drug use illuminate current clinical practices and trends that should be used to prioritize drugs, especially drugs lacking pediatric FDA approvals. In 2004, NICHD commissioned studies of inpatient and outpatient drug use, but since then prescription data sources have improved significantly, as have large data and longitudinal analytic methods. In 2011, a national study of more than a half million children in a large sample of general hospitals and children’s hospitals in the United States determined the frequency use of all inpatient medications and found that a large proportion of hospitalized children are exposed to significant polypharmacy.2 Population-level studies of drug use in narrower populations can help prioritize drugs within therapeutic areas. While the 2015 Priority List does not contain any specific renal drugs for study, a national study of pediatric patients with acute renal failure found that among the 50 most commonly used renal drugs, only 18% had clear pediatric dosing regimens.3 Alternatively, new trends in drug use, such as emerging off-label uses, can be identified by examining the use of drugs without clear pediatric indications. The Intensive Care Priorities therapeutic area only contains 3 neurological medicines for study, but a national Pediatric Health Information System study demonstrated widespread use of 8 neurological and cardiovascular drugs without FDA approval in children in pediatric intensive care units.4 Third, population-level data about ADEs, including both frequent ADEs and rarer but more severe ADEs, should inform drug research prioritization. While the FDA maintains a voluntary Adverse Event Reporting System to spot potential problematic drugs, these data cannot yield population estimates of actual clinical exposure to risky drugs or occurrence of ADEs. A study using the Nationwide Emergency Department Sample estimated the prevalence of pediatric ADEs to be associated with 0.5% Box. Key Population-Level Considerations for Informing Drug Prioritization