Presentations

Abstract

Introduction:

Literature regarding injury comorbidity in schizophrenia patients in the inpatient setting is lacking in recency and breadth. The objective of this cross-sectional study was to investigate the association between schizophrenia inpatients and injury comorbidities in the inpatient setting using nationally representative data.

Methods:

This study was cross-sectional and included data from the 2016-2019 National Inpatient Sample (NIS). The database was analyzed for patients with a diagnosis of schizophrenia or any of its subtypes (ICD-10: F20). All hospitalizations that were not associated with a diagnosis of schizophrenia were categorized as the control group. The presence of injuries was analyzed by ICD-10 codes and included injuries of the head, neck, thorax, abdomen/pelvis/external genitalia, hip/thigh, ankles/feet, and upper or lower extremities. Injuries were also categorized by type, including superficial injury, open wound, fracture, dislocation, nerve injury, muscle injury, crush injury, traumatic amputation, and "other". Chi-square analysis was used to determine associations between schizophrenia and various combinations of injury types and locations. Multivariable logistic regression was performed on injuries that showed statistically significant relationships via chi-square to generate adjusted odds ratios (aOR) and 95% confidence intervals (CI). The presence of schizophrenia was used as independent variable with individual and categorical injuries as the dependent variable. Odds ratios were adjusted using the following covariates: age (continuous), sex (male, female), race/ethnicity (White, Black, Hispanic, Asian, Native American/Other), and type of insurance (Medicare, Medicaid, private, self-pay, no-charge/other).

Results:

Between 2016 and 2019, a total of 121,074,975 hospitalizations were identified in the US, with 1,626,729 (1.30% (1.3 - 1.4 CI, P < .001)) having an associated schizophrenia diagnosis. Demographic trends indicated that schizophrenia patients were more likely to be between the ages of 40-59, male, identify as White, and have Medicare compared to the control group.

After adjustment, multivariable logistic regression demonstrated that 19 out of 89 of the investigated injury types/locations had a significantly positive associated adjusted odds ratio with schizophrenia. Abdominal/pelvic/external genitalia amputation (aOR: 10.819, 95% CI: 7.816 - 14.977, P <.001), "other" wrist/hand injury (aOR: 1.916, 95% CI: 1.740 - 2.110, P <.001), superficial ankle/foot injury (aOR: 1.863, 95% CI: 1.798 - 1.931, P <.001), superficial head injury (aOR: 1.600, 95% CI: 1.571 - 1.630, P <.001), and "other" ankle/foot injury (aOR: 1.482, 95% CI: 1.330 - 1.652, P <.001) demonstrated the highest adjusted odds ratios. Notably, once the odds ratios were adjusted, the association of schizophrenia with any type of injury at any of the categorized sites became positive with an adjusted odds ratio of 1.113 (95% CI: 1.106 - 1.120).

Conclusion:

Schizophrenia was positively associated with comorbid superficial injuries, including superficial injuries of the head, neck, shoulder, elbow, wrist/hand, knee/leg, and ankle/foot. Other positive associations included comorbid open wound injuries at all sites, including the head, neck, elbow/forearm, and wrist/hands. There was a strong positive association with pelvic/external genitalia/abdominal amputations. Surprisingly, schizophrenia was negatively associated with comorbid fractures.