Impact of firearm injury in children and adolescents on health care costs and use within a family

Abstract

In 2020, firearm injury became the leading cause of death in U.S. children and adolescents. This study examines sequelae of firearm injury among children and adolescents in terms of health care costs and use within a family over time using an event study design. Using data from a large U.S. commercial insurance company from 2013 to 2019, we identified 532 children and adolescents aged 1-19 years who experienced any firearm-related acute hospitalization or emergency department (ED) encounter and 1667 of their family members (833 parents and 834 siblings). Outcomes included total health care costs, any acute hospitalization and ED visits (yes/no), and number of outpatient management visits, each determined on a quarterly basis 2 years before and 3 years after the firearm injury. Among injured children and adolescents, during the first quarter after the firearm injury, quarterly total health care costs were $24,018 higher than pre-injury; probability of acute hospitalization and ED visits were 27.9% and 90.4% higher, respectively; and number of outpatient visits was 1.8 higher (p < .001 for all). Quarterly total costs continued to be elevated during the second quarter post-injury ($1878 higher than pre-injury, p < .01) and number of outpatient visits remained elevated throughout the first year post-injury (0.6, 0.4, and 0.3 higher in the second through fourth quarter, respectively; p < .05 for all). Parents' number of outpatient visits increased during the second and third years after the firearm injury (0.3 and 0.5 higher per quarter than pre-injury; p < .05). Youth firearm injury has long-lasting impact on health care within a family.

Keywords: Children and adolescents; Dynamic impact; Firearm injury; Gun violence; Health care costs and use; United States.

Figure 1.. Impact of firearm injury in children and adolescents on health care costs and use among family members from 2013 to 2019

Notes. The points in each figure represent the estimated coefficients of event time (−8 to 12 quarters; reference: quarter −2) with 95% confidence intervals. Linear regression was used for total costs; linear probability regressions were used for any acute hospitalization and emergency department visits; and negative binomial regression was used for number of outpatient visits and marginal effects were reported. All models adjusted for calendar year to account for secular trends in health care costs and use. Standard errors were clustered at the individual level. The vertical red dashed line indicates the second quarter prior to youth firearm injury; its effect is set as 0 (red horizontal line) and serves as the reference point to test that the models satisfied the assumptions of event study design: (1) no anticipation (i.e., no significant change of health care costs and use in the quarter prior to the firearm injury [quarter −1]) and (2) no pre-trends of outcomes before the firearm injury. For example, total health care costs among injured youth was similar in the quarter prior to the firearm injury compared to the second quarter prior to the injury (-$151, p=.62), confirming the assumption of no anticipation for that outcome. Total health care costs included payments for any inpatient and outpatient care and prescription medication, and were adjusted to 2020 dollars using the consumer price index.

Figure 1.. Impact of firearm injury in children and adolescents on health care costs and use among family members from 2013 to 2019

Notes. The points in each figure represent the estimated coefficients of event time (−8 to 12 quarters; reference: quarter −2) with 95% confidence intervals. Linear regression was used for total costs; linear probability regressions were used for any acute hospitalization and emergency department visits; and negative binomial regression was used for number of outpatient visits and marginal effects were reported. All models adjusted for calendar year to account for secular trends in health care costs and use. Standard errors were clustered at the individual level. The vertical red dashed line indicates the second quarter prior to youth firearm injury; its effect is set as 0 (red horizontal line) and serves as the reference point to test that the models satisfied the assumptions of event study design: (1) no anticipation (i.e., no significant change of health care costs and use in the quarter prior to the firearm injury [quarter −1]) and (2) no pre-trends of outcomes before the firearm injury. For example, total health care costs among injured youth was similar in the quarter prior to the firearm injury compared to the second quarter prior to the injury (-$151, p=.62), confirming the assumption of no anticipation for that outcome. Total health care costs included payments for any inpatient and outpatient care and prescription medication, and were adjusted to 2020 dollars using the consumer price index.