Abstract
IMPORTANCE: Multiple organ dysfunction (MOD) is a leading cause of in-hospital child mortality. For survivors, posthospitalization health care resource use and costs are unknown. OBJECTIVE: To evaluate longitudinal health care resource use and costs after hospitalization with MOD in infants (aged <1 year) and children (aged 1-18 years). DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study used nationwide data from 2004 to 2019 from Optum's deidentified Clinformatics Data Mart Database, an insurance claims database. Infants and children from birth to age 18 years with an index hospitalization between January 1, 2005, and December 31, 2018, were included. Infants (age <1 year) and children (age 1-18 years) with MOD (MOD cohort) or without MOD (non-MOD cohort) were separately identified, and cohorts were propensity score weighted to balance demographics and pre-index hospitalization characteristics, including health care use and comorbidities. The data were analyzed between January 7, 2022, and September 8, 2023. MAIN OUTCOMES AND MEASURES: Weighted generalized estimating equations were used to evaluate differences between cohorts in rehospitalizations, emergency department visits, and health care costs up to 5 years after the index hospitalization. RESULTS: During the study period, 9671 children in the MOD cohort were compared with 1 691 793 children in the non-MOD cohort in the weighted sample. Infants comprised 67.4% of the MOD cohort (mean [SD] age at index hospitalization, 0.1 [0.8] years; 51.2% male) and 87% of the non-MOD cohort (mean [SD] age at index hospitalization, 0.1 [0.8] years; 50.8% male), and children comprised 32.5% of the MOD cohort (mean [SD] age at index hospitalization, 11.6 [5.7] years; 50.7% female) and 13.0% of the non-MOD cohort (mean [SD] age at index hospitalization, 11.5 [5.5] years; 51.3% female). The infant MOD cohort had more emergency department visits, with an adjusted incidence rate ratio of 1.76 (95% CI, 1.56-1.97) at 30 days; this difference persisted for years 1 through 5. Children had a similar pattern except at 30 days among those who acquired new organ-supportive technology during the index hospitalization. Among infants, the MOD cohort had more rehospitalizations, with an adjusted incidence rate ratio of 12.45 (95% CI, 11.40-13.59) at 30 days; this difference persisted for years 1 through 5. A similar pattern was observed among children. Annual health care costs were higher for the MOD cohort in year 1 (infants: mean [SD], $80 133 [$6543] vs $5183 [$19] [P < .001]; children: mean [SD], $54 113 [$17 544] vs $10 935 [$95] [P < .001]) and in all years through year 5. CONCLUSIONS AND RELEVANCE: In this cohort study of nearly 1.7 million children, survivors of MOD accrued substantial ongoing health care resource use and cost burden after the index hospitalization. These findings suggest that follow-up care of survivors of MOD should include economic well-being alongside measures of clinical health.