ABSTRACT
Background Current therapies for pediatric and adult eosinophilic esophagitis (EoE) include dietary elimination, proton pump inhibitors, swallowed corticosteroids, and biologics. Our aim is to systematically assess the efficacy and safety of published randomized controlled trials (RCTs) of medical therapies for EoE that compare active treatments to placebo or to an active comparator. We consider RCTs that target the induction and maintenance phases of therapy, separately. Methods A search was designed by a Cochrane information specialist and included Cochrane Gut Register, CENTRAL, MEDLINE, Embase, and clinicaltrials.gov databases, from inception to May 2022. Studies that met our search criteria were imported into Covidence for title and review. All authors participated in study screening, and each study was independently evaluated by two authors. Reports of RCTs that met the inclusion criteria were selected for full text review. Multiple reports of the same RCT were collapsed into the parent study. Data from these studies was then extracted to RevMan Web to assess study characteristics, including study design, EoE definition, inclusion / exclusion criteria, age range, interventions, number of patients randomized, number of patients who completed the study, primary and secondary outcomes and conflicts of interest. Studies were also assessed for potential sources of bias including baseline imbalance, selection, performance, detection, attrition, and reporting biases. We used Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) to assess the overall certainty of evidence supporting the primary outcome. Results As of May 2022, we identified 2,638 reports that met our search criteria of which 14 were duplicates, giving a total of 2,624 reports that were imported into Covidence for further review. Following title and screening, 259 reports were selected for full text review, which were collapsed into 46 distinct RCTs that met the inclusion criteria. The primary outcomes for our systematic review were histological improvement, endoscopic improvement, and clinical symptom improvement, all as defined by the study at study endpoint, and withdrawals due to adverse events. The secondary outcomes of our analysis were serious adverse events, endocrine complications, growth concerns, infections, and health-related quality of life. Study outcomes were analyzed on an intention-to-treat basis. Risk ratios (RRs) and corresponding 95% confidence intervals (95% CI) are reported for dichotomous outcomes, and mean difference and standard deviation are reported for continuous outcomes. The data will be presented in full. Conclusions Results of this analysis inform clinicians about the efficacy and potential side effects of different medical therapies for EoE in both pediatric and adult populations. Deficiencies in our current knowledge will be highlighted and will provide direction for design of future RCTs in the field. Future research should continue to explore factors that influence initial and subsequent medical therapy selection for people with EoE.
ABSTRACT
Background: The COVID-19 public health emergency (PHE) caused pediatric gastroenterologists to rapidly adapt and greatly expand telehealth (TH) clinical care. Objectives: This study assessed pediatric gastroenterology TH access to care, socioeconomic disparities, and geo-mapping. Methods: A large retrospective cohort study involving distinct patient encounters (DPE) of pediatric gastroenterology patients within Nemours Children's Health Care System (NCHCS): Delaware Valley (DV) and Central Florida (CF). We examined two socioeconomically and geographically unique populations in 2019-20 comparing TH visits to in-person (IP) visits with the shared AmWell TH platform and Epic Electronic Medical Record. We used U.S. Census block group data (600 to 3,000 people per unit) in conjunction with the American Community Survey socioeconomic status indicators. Geo-mapping from individual patients' street addresses was performed. Confounding variables of sex, age, race and ethnicity, income, insurance, and interpreter service were used to calculate adjusted odds ratios (aOR). Results: In DV, there were 16 TH visits in 2019 compared to 2,320 TH visits in 2020, comprising 27.1% of the total 8,561 distinct 2020 visits. In CF, there were 65 TH visits in 2019 compared to 2,414 TH visits in 2020, comprising 33.7% of the total 7170 distinct 2020 visits. In DV, Asians (aOR[95% CI], 4.25 [1.73,10.41], p<0.05 ) and whites (aOR[95% CI], 1.5 [1.12,2.01], p<0.05) were more likely to utilize TH than IP-visits. In contrast, blacks (aOR[95% CI], 1.61 [1.11,2.33], p<0.05) and Hispanics (aOR[95% CI], 2.27 [1.27,4.0], p<0.001) were more likely to utilize IP-visits than TH. In CF, Asians (aOR[95% CI], 3.22 [0.59,17.7], p=0.178), whites (aOR[95% CI], 0.99 [0.75,1.31], p=0.96), blacks (aOR[95% CI], 0.78 [0.50,1.21], p=0.267), and Hispanics (aOR[95% CI], 1.14 [0.81,1.59], p=0.46) had no difference in using either TH or IP-visits. In DV, patients insured by Medicaid (aOR[95% CI], 1.18 [1.04,1.33], p<0.01) and patient families with interpreter service needed (aOR[95% CI], 2.44 [1.69,3.57], p<0.001) were more likely to utilize IP-visits than TH. In CF, patients insured by Medicaid (aOR[95% CI], 1.02 [0.90,1.15], p=0.76) had no difference in using either TH or IP-visits;however, patient families with interpreter service needed (aOR[95% CI], 1.37 [1.11,1.69], p<0.005) were more likely to have IP-visit than TH. In DV, census block group data from patients' residing in households with a single female led household (aOR[95% CI], 1.92 [1.10, 3.33] p<0.05) and high school diploma (aOR[95% CI], 3.23 [1.96,5.55], p<0.001) were more likely to use IP-visits than TH. In contrast in DV, patients' residing in households with a Bachelor's degree or higher level of education (aOR[95% CI], 3.99 [2.33,6.83], p<0.001) or an income in past 12 months at or greater than $100,000 dollars (aOR[95% CI], 1.75 [1.30,2.37], p<0.01) were more likely to use TH than IP-visits. In CF, census block group data from patients' residing in households with a single female led household (aOR[95% CI], 0.78 [042,1.47], p=0.44), high school diploma (aOR[95% CI], 0.92 [0.47,1.81], p=0.82) patients' residing in households with a Bachelor's degree or higher level of education (aOR[95% CI], 1.32 [0.88,1.98], p=0.18) or an income in past 12 months at or greater than $100,000 dollars (aOR[95% CI],1.29 [0.85,1.96], p=0.23) had no difference in using either TH or IP-visits. Geographic analysis of DV (Figure 1) and CF showed a similar patient distribution of IP 2019 to TH 2020. Conclusion: Significant racial, ethnic, and socioeconomic disparities were present in our DV cohort but were only significant for language barriers in our CF cohort after correcting for confounding variables. As TH continues to evolve, further investigations should help to improve access to care, address socioeconomic disparities, and assess clinical outcomes of TH.