Prevalence and impact of frailty among inpatients with schizophrenia: Evidence from the US Nationwide Inpatient Sample 2005–2020

OBJECTIVE: Frailty is more prevalent among individuals with schizophrenia, and could negatively affect their hospitalization outcomes. This study aimed to investigate if co-existing frailty is associated with worse in-hospital outcomes in patients with schizophrenia. METHODS: Data of patients ⩾20 years admitted for schizophrenia from 2005 to 2020 were extracted from the from the US Nationwide Inpatient Sample database. Patients were classified by frailty status using the Hospital Frailty Risk Score into three categories: low, intermediate and high risk. Logistic regressions were used to determine the associations. RESULTS: A total of 956,541 patients with schizophrenia were included, and 8%, 4% and 67% patients were in high, intermediate, and low frailty risk, respectively. Patients with intermediate and high risk significantly increased risk of in-hospital mortality (adjusted odds ratio = 1.89, 95% confidence interval: 1.39-2.57; adjusted odds ratio = 3.56, 95% confidence interval: 2.50-5.07), unfavorable discharge (odds ratio = 1.40, 95% confidence interval: 1.35-1.44; adjusted odds ratio = 1.78, 95% confidence interval: 1.68-1.88), prolonged length of stay (adjusted odds ratio = 1.25, 95% confidence interval: 1.22-1.28; adjusted odds ratio = 1.52, 95% confidence interval: 1.45-1.59) and major complications (adjusted odds ratio = 2.43, 95% confidence interval: 2.37-2.50; adjusted odds ratio = 2.69, 95% confidence interval: 2.60-2.79). Besides, intermediate and high risk patients had greater hospital costs (aBeta = 2.69, 95% confidence interval: 2.37-3.01; aBeta = 5.43, 95% confidence interval: 4.81-6.06, per 1000 dollars). Stratified analysis showed consistent associations between frailty and adverse outcomes across age groups. CONCLUSION: In hospitalized patients with schizophrenia, frailty assessed by the Hospital Frailty Risk Score is associated with worse in-hospital outcomes, independent of other demographic and clinical factors.