Demand and supply of invasive and noninvasive ventilators at the peak of the COVID‐19 outbreak in Okinawa

Abstract While Okinawa has been facing outbreak of the coronavirus disease 2019 (COVID‐19) pandemic, healthcare collapse should be prevented by sufficient supply of ventilators for caring the rapidly growing number of critically ill patients with COVID‐19. We estimated the number of invasive and noninvasive ventilators that would be required in Okinawa at the peak of the COVID‐19 outbreak based on recent data of COVID‐19 cases in Okinawa and data on the proportion of patients with COVID‐19 in the ICU requiring ventilation. Based on our results using the current supply of all ventilators, demand for ventilators could be prepared for patients with COVID‐19 who would require it and demand for noninvasive ventilators could also be prepared for those with COVID‐19 who would require it. The higher supply over the demand would be achieved by flattening the epidemic curve by implementing public health interventions to delay and suppress the epidemic peak in Okinawa.


| INTRODUC TI ON
Hospitals in Okinawa have been facing outbreak of the coronavirus disease 2019 (COVID-19) because of its pandemic spread. 1 Although the outbreak has still been in the early phase in Okinawa, it would become serious factor related to healthcare collapse to have insufficient supply of intensive care unit beds and ventilators to care the rapidly growing number of critically ill patients with  Okinawa is a group of small remote islands having no industry that could rapidly produce ventilators 3 ; it may require several days of shipping time for importing ventilators. Thus, we estimated the number of ventilators and noninvasive ventilators that would be required in Okinawa at the peak of the COVID-19 outbreak to address the intensive care preparation. Our estimates are based on recent data of COVID-19 cases in Okinawa and data on the proportion of patients with COVID-19 in the ICU requiring ventilation.

| Model setting
We used the well-known SEIR epidemic model. 4 The meaning and value of each parameter are shown in Table 1. We regard t = 0 as of February 14, 2020, which is the day when the first COVID-19 case was confirmed in Okinawa. 13 Thus, Y(0) = 1, and hence, I(0) = Y(0)/ (pN) ≈ 8.62 ×10 −6 . For simplicity, we assume that E(0) = R(0) = 0, and thus, S(0) = 1 -I(0) ≈ 1. The basic reproduction number R 0 14 is calcu- in the ICU requiring ventilation. Based on our results using the current supply of all ventilators, demand for ventilators could be prepared for patients with COVID-19 who would require it and demand for noninvasive ventilators could also be prepared for those with COVID-19 who would require it. The higher supply over the demand would be achieved by flattening the epidemic curve by implementing public health interventions to delay and suppress the epidemic peak in Okinawa.

| Parameter estimation
To estimate the infection rate β, we apply the least-squares-based method 4 to the data of the daily confirmed cases of COVID-19 in Okinawa from February 14 to April 30, 2020. As a result, we obtain the estimated value β = 0.14 (95% CI, 0.12-0.16), and thus,

| Epidemic curves
Using the parameters in Table 1, we obtain the estimation of the epidemic curve of COVID-19 in Okinawa as shown in Figure 2.
As of April 30, 2020, it seems that the epidemic in Okinawa is well controlled and the estimated epidemic curve is sufficiently flattened so that the peak comes late (t = 413, ie, April 2, 2021).
However, if it goes to the worse scenario (β = 0.16 and R 0 = 1.6), then the peak would come within this year (t = 304, ie, October 14, 2020) and the number of confirmed cases at the peak (≈6334) would become about 1.8 times larger than that for the baseline scenario (≈3532).

| Number of required ventilators
We obtain the estimation of the number of required ventilators for COVID-19 in Okinawa as shown in Figure 3. There is a margin of available ventilators even at the peak in the worse scenario (β = 0.16 and R 0 = 1.6).  17 Second, the pandemic might subside from specific climate conditions in Okinawa such as higher temperature and humidity along with ultraviolet radiation exposure in summer. 18 Third, the coinfection of COVID-19 and influenza in winter may lead to higher incidence of severity, which may increase the demand. 19 In conclusion, although it would be disastrous to wait and see until the possible supply shortage of life-saving ventilators, the requirement of ventilators would not exceed its availability during the period of expected peak based on our estimate about the highest demand scenario. Urgent actions conducted in Okinawa would save lives and avoid devastating rationing that would require physicians not to be able to allocate ventilators to some of patients with COVID-19.

ACK N OWLED G EM ENT
None.

CO N FLI C T O F I NTE R E S T
The authors have stated explicitly that there are no conflicts of interest in connection with this article.