This issue of Hospital Pediatrics includes a study by Samady et al,1 in which different models of delivering inpatient asthma education were compared. This contribution is important for two reasons. One, it provides useful information to help inform the optimal approach to teaching patients and families about chronic asthma management. Two, the study team used a randomized controlled trial (RCT) design to test their intervention. RCTs are the gold standard for rigorously testing interventions, but they continue to be rare in the field of pediatric hospital medicine (PHM).
This study was motivated by the authors’ observation of important gaps in the literature on asthma education: a lack of studies on asthma education interventions in the inpatient setting and a lack of studies comparing the effectiveness of different asthma education models. They conducted a single-site, investigator-blinded RCT comparing a standard didactic and passive educational approach to an interactive approach (which relied on the learner’s active participation and incorporated the use of a low-literacy asthma action plan composed primarily of pictures).
The trial included 140 participants, and the primary outcome was emergency department (ED) visits after hospitalization (within 6 or 12 months). Although there were no differences in the primary outcome, some benefits were noted in secondary outcomes. At 12 months after hospitalization, children in the interactive education group had improved inhaler technique, improved patient-reported symptoms, and fewer hospitalizations. They hypothesized that the lack of difference in ED visits was driven by parents or caretakers seeking ED care earlier, in the context of milder asthma exacerbations (perhaps in better alignment with asthma action plan instructions). The authors also impressively collected a balancing measure, time spent to deliver both types of education. They found that the interactive education took about 6 minutes longer than the more traditional, didactic approach. The study was limited by single-center design, inclusion of only English-speaking patients and families, and collection of reutilization data only from that single-center’s electronic health records. Nonetheless, this simple educational intervention has strong potential to improve outcomes for children with asthma.
The Importance of PHM RCTs
A huge strength of this study by Samady et al is the RCT design, the gold standard for comparing interventions. Observational studies are limited by the fact that investigators can only adjust for measured patient characteristics, so these studies will always be subject to the possibility of residual bias, including confounding. For instance, if the authors of this study had instead chosen an observational design, and if clinicians tended to provide the longer interactive asthma education to patients with more severe chronic asthma, we may have not discovered the benefits of this intervention. Whereas observational studies are an important research tool, overreliance on observational evidence in lieu of RCTs risks keeping potentially helpful interventions out of use and/or exposing patients to interventions that are ineffective and potentially harmful. Prominent examples of widely accepted practices founded in observational evidence but later contradicted by RCTs include prophylactic surgical ligation of patent ductus arteriosus in preterm infants2,3 and endotracheal suctioning for neonates born through meconium-stained fluid.4,5
Unfortunately, pediatric RCTs are in short supply. Whereas publication of adult RCTs have been rising since 2005, the rate of publication of pediatric RCTs has not changed over the same time period.6 Most recently, despite impassioned calls for pediatric trials in the setting of COVID-19,7,8 pediatric RCT findings to guide clinical decisions for COVID-19 are nonexistent, nearly 2 years after the start of the pandemic. For example, there are no published RCTs and only 2 RCTs (NCT04898231; NCT04826588) registered on clinicaltrials.gov that examine optimal pharmacologic interventions for multisystem inflammatory syndrome in children. Examination of submissions to the 2 recent national PHM conferences suggest that PHM RCTs are rare. A total of 536 research abstracts were submitted to the 2020 and 2021 PHM conferences, of which only 3 abstracts (<1%) involved RCT designs (E.C., unpublished observations).
A myriad of barriers hinder conduct of RCTs in PHM. It takes a robust research infrastructure to execute the processes that go into a successful RCT (eg, planning and registering the trial, screening and recruiting participants, implementing randomization and treatment allocation, measuring outcomes). These processes are fundamental to all RCTs, but pediatric RCTs come with extra challenges. The low prevalence of childhood diseases (compared to adult diseases) means that more intensive screening and longer trial durations are required. Similarly, markers of severe disease, such as mortality, are thankfully uncommon for children, but rare outcomes also translate into larger sample sizes and/or reliance on surrogate outcomes. Lower disease prevalence likely contributes to less federal and industry funding for pediatric research.9,10 Finally, the consent process is more complicated for pediatric trials because children are a vulnerable population and, in most cases, cannot independently consent to participation.11,12
Despite these barriers, we are optimistic that PHM can rise to the challenge and increasingly answer our field’s most pressing research questions with RCTs. PHM only became an accredited specialty within the last 5 years, and the number of hospitalists who are receiving research training through PHM fellowships increases every year. Thus, the number of hospitalists with the skills to contribute to and/or lead RCTs will only increase for the foreseeable future. Additionally, there are 2 national PHM networks that can potentially help support different aspects of PHM RCTs (Pediatric Research in Inpatient Settings Network and Value in Inpatient Pediatrics Network). To increase feasibility of PHM RCTs, the field can leverage emerging trial innovations such as electronic consent and integrated consent (which allows medical providers to obtain informed consent during their routine care of patients).13,14 To promote high impact PHM RCTs, the field will also need to crystallize a research agenda. Several recent publications have provided updated data on the most prevalent and costly conditions among hospitalized children to guide the creation of such a research agenda.15,16
The authors should be commended for successfully conducting a PHM RCT and overcoming the aforementioned barriers. This study is a testament to the increasing capacity of pediatric hospitalists to answer fundamental questions in PHM with RCTs. As PHM continues training new investigators, developing research network infrastructure, and embracing innovation, our facility with RCTs will only grow. This most rigorous approach to clinical research is integral to advancing our field and improving care and outcomes for severely ill children.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest to disclose.
Drs Kaiser, Schroeder, and Coon drafted the initial manuscript, revised the manuscript, and approved the final manuscript as submitted.
Comments