The impact of the state of Ohio stay-at-home order on non-COVID-19 intensive care unit admissions and outcomes

Dear Editor,

Hospitals reported a decline in emergency room (ER) visits, hospitalisations, and elective procedures during the coronavirus disease 2019 (COVID-19) pandemic [1–4]. This raised concerns over delays in seeking care [5].

Cleveland Clinic, the largest healthcare system in northeast Ohio with its 10 hospitals witnessed a significant decline in ER visits and intensive care unit (ICU) admissions since the March 16 Ohio school closure order and the March 23 stay-at-home order by the Governor of Ohio. This study reviews non-COVID-19 patient ICU admissions and outcomes during the above social distancing measures.

We analysed the Cleveland Clinic health care system quality data registry for all non-COVID-19 ICU admissions from March 15 to April 30 2020. The Ohio stay-at-home order expired on May 1. This data was compared to the same period last year (2019). We collected demographics, ICU admission sources, hospital and ICU length of stay (LOS), hospital and ICU mortality, admission acute physiology score (APS), acute physiology and chronic health evaluation (APACHE III) score, and admission principal diagnosis to the ICU. Additionally, we collected Department of Health (DOH) in Ohio mortality data excluding COVID-19 for the months of March and April 2020 and compared those to the mortality counts and ratios for the same two months in 2019 for the seven counties in northeast Ohio where the 10 hospitals serve over 2.7 million population [6]. Two-sample t-test or Wilcoxon rank-sum test were used to compare continuous variables; the c² test was applied to compare categorical variables. The institutional review board at the Cleveland Clinic approved this study and waived patient informed consent.

The number of patients presenting at all 10 hospital ERs from March 15 to April 30 2020 was 39,970, a decrease of 40.5% from 67,217 during the same period last year, with incidence rate ratio (IRR): 0.5946 (95% CI: 0.5873–0.6020). With universal COVID-19 testing for all admissions, ICU admissions for non-COVID-19 cases decreased by 38.1% from 2573 to 1592, IRR: 0.6187 (95% CI: 0.5812–0.6586). During the above same period the total number of ICU admissions of COVID-19-confirmed cases was 274, and the total overall number of hospitalised COVID-19-confirmed patients was 656. Table 1 summarises demographic and characteristic data for 2019 compared to 2020 for non-COVID-19 ICU admissions. Only the principal diagnosis on admission to the ICU was reported. A decrease in patient counts for all admission diagnoses was noted except for cardiogenic shock and the acute respiratory distress syndrome (ARDS). Non COVID-19 ICU admissions had a statistically significant higher APS score and APACHE III score in 2020 compared to 2019 (P < 0.0001 for both). Proportionally more patients were admitted with acute respiratory distress syndrome (P = 0.0041), sepsis (P = 0.0193), cardiogenic shock (P = 0.002), respiratory failure on mechanical ventilation (P < 0.0001), and patients on chronic dialysis (P = 0.0323). On the other hand, there were proportionally fewer chronic obstructive lung disease (COPD) exacerbations (P = 0.0003), chest pain admissions (P = 0.0142), and post-operative surgical patients (P = 0.0004). Despite higher acuity at presentation, there were no statistical differences in ICU or hospital mortality within the Cleveland Clinic healthcare system. Also, no differences in mortality for non-COVID-19 patients at the county level in Northeast Ohio were seen (Table 2).

During the COVID-19 stay-at-home order in the state of Ohio, ER visits declined by 40.5% and ICU admissions by 38.1%. Sicker admissions of non-COVID-19 patients to the ICU with higher APS and APACHE score were noted. This could be due to the significantly higher proportion of higher acuity admission diagnoses in the 2020 cohort and lower proportion of the lower acuity presentations, with patients who are less sick probably being more reluctant to seek care and to present to the ER. However, the decline across all patient principal diagnoses and admission numbers, along with an increase in the APS and APACHE III score and the increase in the number of patients with ARDS and cardiogenic shock in the 2020 cohort are alarming. This confirms a trend where patients most often reached sicker health status, seeking care later, and avoiding hospitals due to fear of the COVID-19 infection. A Similar decline of 42% was reported by the Centres for Disease Control and Prevention (CDC) in emergency room visits as of June 3, 2020 because of the pandemic [7], a decline of 42% in Veterans Affairs hospitals admissions [4] and 33.7% in hospital admissions for eight other acute care hospitals [9], a decline of 38% to 40% in cardiac catheterisation laboratory ST-segment elevation myocardial infarction (STEMI) activations across the United States and Spain [1, 8], and finally a decline in the use of stroke imaging by 39% [10]. All are consistent with the above 38.1% decline seen in our non-COVID-19 ICU admissions during the pandemic compared to the same period last year.

The limitations of the study are mainly the retrospective nature and the single healthcare system data, which might not be generalisable. Another limitation was the use of the principal diagnosis on admission, which could have overlooked other comorbidities and critical care illnesses that were not reported. In a short-term follow-up and when the data was obtained from the Ohio DOH in June 2020, the ICU, hospital, and county mortalities were not statistically different, as mentioned above. However, the long-term impact of such delays in care remains unknown. Patient education and care planning will be needed especially if a second wave with new stay-at-home orders are to be expected.