OBJECTIVE

Little is known about factors associated with discharge against medical advice (DAMA) in adolescent acute care hospitalization for suicidal ideation (SI) and suicide attempt (SA). Our study seeks to determine whether certain socioeconomic factors or hospital characteristics are associated with DAMA in this population.

METHOD

This retrospective cross-sectional study used data from the National Inpatient Sample from the 2015 fourth quarter to 2019. We included children 10 to 19 years of age hospitalized with a primary or secondary International Classification of Diseases, Tenth Revision, Clinical Modification diagnosis of SI or SA. Exposures were patient socio-demographics and hospital characteristics. The outcome was DAMA. Logistic regression generated odds ratios (ORs) with 95% CIs to measure the association between each patient and hospital characteristics and DAMA.

RESULTS

Of 476,755 hospitalizations meeting inclusion criteria, 3,825 (0.8%) were DAMA. After adjusting for socio-demographics and hospital characteristics, predictive factors for DAMA were age 16 to 19 years (OR = 1.41; CI = 1.08-1.82), self-pay status (OR = 1.43; CI = 1.12-1.83), hospital region South and West (OR = 1.55; CI = 1.10-2.20 and OR = 1.79; CI = 1.26-2.54, respectively), and urban non-teaching status of the hospital (OR = 1.90; CI = 1.42-2.55). Hispanic patients were less likely to be DAMA (OR = 0.66; CI = 0.51-0.85).

PLAIN LANGUAGE SUMMARY

This study utilized retrospective data from the National Inpatient Sample (2015 to 2019) to determine predictive factors for discharge against medical advice (DAMA) for youth, aged 10 to 19 years old, hospitalized for suicidal ideation or suicide attempt. Of 476,755 hospitalizations for suicidality, 3,825 (0.8%) were discharged against medical advice. Correlates of discharge against medical advice included age 16 to 19 years, self-pay status, admission to an urban non-teaching hospital, or a hospital located in the South or West Regions of the United States. Hispanic patients were less likely to be discharged against medical advice with no other racial/ ethnic differences observed.

CONCLUSION

Variations in DAMA probabilities by age, insurance status, hospital teaching status, and hospital regions suggest a need for a better understanding of this uncommon outcome.