The Effect Of Adding Telemedicine To Comprehensive Care For Children With Medical Complexity: A Randomized Quality Improvement Trial

Research Objective Telemedicine ( TM ) is widely used but has uncertain value. We assessed TM as method to further improve the outcomes and reduce costs of comprehensive care ( CC ) for children with medical complexity ( CMC ). Study Design Randomized quality improvement trial comparing CC with TM and CC alone using Bayesian analyses with neutral prior (assuming no benefit) and stratifying by age (<2 years or ≥ 2 years) and estimated baseline risk (risk level 1 [positive‐pressure ventilation], risk level 2 [> expected median risk but no positive‐pressure ventilation], and risk level 3 [≤ expected median risk]). All study patients received CC that included primary care providers ( PCPs ) and specialists in the same clinic, 24/7 direct phone access to PCPs, low patient‐to‐PCP ratio (≤100:1), hospital consultation from PCPs during hospitalizations, and multiple other features to promote prompt effective care at all hours. The TM group also received audio‐visual communication (via Zoom for Healthcare) with the PCPs. The study outcomes were the Bayesian posterior probability of reducing days of care outside the home (in a clinic, ED, or hospital; primary outcome ), the rate of children developing a serious illness (causing death, pediatric intensive care unit [ PICU ] admission, or hospital stay >7 days), and health system costs. Population Studied High‐risk CMC (with a chronic disease and ≥ 2 hospitalizations or ≥ 1 PICU admission in the year before enrolling in our CC program) treated in the High‐Risk Children's Clinic at The University of Texas Health Science Center at Houston, Texas. Principal Findings Between August 22, 2018 and March 23, 2020, we randomized 422 CMC, 209 to CC with TM and 213 to CC alone before meeting the predefined stopping rule (≥75% probability of reduced care days outside the home). In intent‐to‐treat analyses, the probability of a reduction with CC with TM compared to CC alone was 99% for days of care outside the home (12.94 vs.16.94 per child‐year; Bayesian rate ratio [ RR ], 0.80 [95% Credible Interval, [0.66–0.98]) , 95% for rate of children with a serious illness (0.29 vs. 0.62 per child‐year; RR, 0.68 [0.43, 1.07]), 90% for admissions (1.01 vs. 1.23 per child‐year; RR 0.83 [0.62–1.11), 97% for PICU admissions (0.38 vs. 0.67 per child‐year; RR 0.66 (0.42–1.03), and 91% for mean total health system costs ($33,718 vs. $41,281 per child‐year; Bayesian cost ratio, 0.85 [0.67–1.08]). Conclusions The addition of TM to CC for children with medical complexity likely reduced their total days of care outside the home, serious illnesses, other adverse outcomes, and health system costs. Implications for Policy or Practice Our findings indicate that TM can be safely used with CC for low‐income, mostly Medicaid‐insured high‐risk CMC like ours to provide them with more convenient, readily accessible, and cost‐effective care while minimizing their exposure in medical settings. Reducing unnecessary exposures for CMC is especially important during the current COVID‐19 pandemic or any other outbreaks of contagious and seasonal illnesses. Trial Registration ClinicalTrials.gov Identifier: NCT03590509. Primary Funding Source Texas Medical Center Health Policy Institute Grant, UTHealth Learning Healthcare Scholars' Award,from grant 5KL2TR000370 from the Center for Clinical and Translational Sciences, grant 5 UL1TR00371 from the National Center for Advancing Translational Sciences.