Changes in inpatient and outpatient stroke admissions during COVID-19: a difference-in-differences analysis based on claims data from 242 acute care hospitals in Japan

Introduction While COVID-19 has been associated with an elevated stroke risk due to its prothrombotic and inflammatory effects, previous studies have reported a decreased number of stroke hospitalisations during COVID-19. Patients with a stroke who could not be admitted might have been followed up in outpatient settings of these hospitals. Our study aimed to investigate the change in the number of patients who had a stroke admitted to inpatient and outpatient settings during COVID-19. Methods A difference-in-differences design using claims data in 242 acute-care hospitals was adopted to examine stroke admissions in Japan, contrasting the pre-pandemic period (2015–2019) to the pandemic period (2020). The Japanese government’s declaration of emergency in April 2020 was considered an exogenous shock. The outcomes were the numbers of ischaemic stroke (both embolic and non-embolic), intracerebral haemorrhage and subarachnoid haemorrhage admissions, with in-hospital mortality rates, modified Rankin Scale and procedure uses. Results From a total of 27 631 stroke cases (44.3% women, median age 75 years (IQR 66–83)), no statistically significant changes were observed in the overall count of stroke admissions during the pandemic. A decrease in ischaemic stroke hospitalisations was detected (incidence-rate ratios (95% CI), 0.83 (0.73 to 0.95)), which was somewhat offset by an increasing trend for patients treated in the outpatient setting (1.06 (0.91 to 1.22)). No significant differences were found in in-hospital mortality rate, proportion of severe modified Rankin Scale at discharge and thrombolysis treatment rate. Rates of mechanical thrombectomy in ischaemic stroke were reduced (0.60 (0.36 to 0.99)), and length of stay was reduced for all types of stroke except intracerebral haemorrhage. Conclusion When considering both inpatient and outpatient admissions for stroke, we found no statistically significant change in the number of stroke admissions in 2020 in Japan. In addition, our findings suggest that acute stroke inpatient care, at least during 2020, was appropriately managed despite limited resources.