Abstract
U.S. health care facilities have been encountering a recurrence of medical supply shortage since COVID-19 exploded in March 2020. There is an urgent need for important Personal Protective Equipment (PPE) such as N95 and surgical masks. This project examined the factors that were associated with nursing homes’ N95 and surgical mask supply. We analyzed data from the Nursing Home COVID-19 Public File and conducted a multivariate logistic regression estimating the association between nursing home characteristics and county-level demographic parameters with mask supply. We found that a high number of resident COVID-19 cases contributed to the supply of N95, but not surgical masks, whereas a high number of staff cases did not lead to an adequate supply of either N95 or surgical masks. Compared with not-for-profit (NFP) facilities, for-profit (FP) nursing homes were less likely to get enough masks. A better supply distribution plan is needed to prepare for future possible PPE shortage.
Introduction
The outbreak of COVID-19 in the United States has exposed critical flaws in the Personal Protective Equipment (PPE) supply since March 2020 (Landi, 2020; Ranney et al., 2020). Among all types of PPE, N95, and surgical masks were commonly used by health care workers (Chughtai et al., 2020). During the pandemic, such great dependence on masks brought a surge demand, which could be 17 times than usual according to a survey (Pannic, 2020). Thus, hospitals ranked the supply of N95 and surgical masks as their top concern (Pannic, 2020). According to Landry et al. (2020) and a recent Department of Health and Human Services Office of Inspector General report (Strickler et al., 2020), many health systems struggled with shortages and often competed with each other for necessary supplies. Doctors and nurses were reusing N95 masks for days and even weeks (Wan, 2020). The situation was even worse for nursing homes—the pandemic’s epicenter. Nursing homes either had missing items (fewer than the government promised or nursing home ordered) or received flimsy surgical masks or nonmedical-use masks (Ran, 2020).
Several studies examined factors associated with the nursing home COVID-19 outbreak. State variation was observed in nursing homes’ participation in COVID-19 reporting (Li et al., 2020) and the probability of having COVID-19 cases (Abrams et al., 2020). One recent study showed that nursing homes with COVID-19 cases among residents and staff had a higher probability to report shortages (McGarry et al., 2020).
Given the importance and the possible long-lasting nationwide shortage of face masks, it is urgent to evaluate the mask usage across nursing homes to help coordinate resources. In this study, researchers investigated factors that were associated with N95 and surgical mask supply from two levels: facility level and county level. Using lagged data, we checked the responsiveness of nursing facilities facing supply risk.
Study Data and Method
The Nursing Home COVID-19 Public File included data reported by nursing homes to the Centers for Disease Control and Prevention’s (CDC) National Healthcare Safety Network (NHSN) system COVID-19 Long Term Care Facility Module (Centers for Medicare & Medicaid Services, 2020). We obtained information on COVID-19 resident and staff cases together with the mask supply from 14,062 nursing homes from the public data. This study was exempt from the Institutional Review Board (IRB) review because existing public data sets were used and the unit of the analysis was the nursing facility. A retrospective observational study was performed. Our dependent variables were two important medical supplies of COVID-19: N95 and surgical masks in the week ending June 21, 2020. We lagged the dependent variables by 2 weeks to examine the impact of key independent variables on mask supply. We adopted the nursing home resident and staff cases in the week ending June 7, 2020, as our primary independent variable. Except for the primary independent variable, we also included factors of two levels: facility level (ownership type, years of operation, chain affiliation, hospital affiliation, percentage of patients above 65 years, percentage of the White population, percentage of Medicare residents, percentage of Medicaid residents, and patient daily census) and county level (number of nursing facilities, per capita income, urbanity, and COVID-19 cases per 1,000 citizens). Univariate analyses of the independent variables were based on Student’s t-test for continuous variables and chi-square test for categorical variables.
Next, we fit a multivariate logistic regression model with whether or not there were enough supplies of N95 or surgical masks as the binary dependent variable. Our key independent variables were the number of residents and staff COVID-19 cases. Given the univariate analysis results, we also adjusted for facility ownership (not-for-profit [NFP], for-profit [FP], and government-owned), number of nursing facilities in the county, and other facility-level and county-level factors. To incorporate the state variations, we added state as a fixed effect. To see the responsiveness of the N95 and surgical mask supply, we conducted a trend analysis to see how nursing homes of different ownership types acted to decrease the supply shortage. Analyses were performed using R, version 4.0.1.
Results
The study used data from a total of 14,062 nursing homes, which included both skilled nursing facilities and nursing homes. These facilities reported COVID-19 staff and resident cases in numbers and the adequacy of medical supply in “yes” and “no.” Descriptive analyses showed that there were 681 nursing homes lacking supplies of both N95 and surgical face masks, and 1,266 with a shortage of either one of the two masks. Further analysis on resident and staff cases (Table 1) showed that facilities that received adequate N95 supply had a higher mean (0.52 for staff cases and 0.54 for resident cases) than those that did not get adequate N95 supply (0.41 for staff and 0.39 for residents). Similar results were observed for surgical masks (with mean 0.51 vs. 0.34 for staff cases and 0.53 vs. 0.38 for resident cases). In contrast to NFP and government-owned nursing homes, FP nursing homes reported a higher percentage of mask shortage (76.97% vs. 69% for N95 and 78.51% vs. 69.5% for surgical masks).
Descriptive Analyses.
Note. FP = for-profit; NFP = not-for-profit; NF = nursing facilities.
Finally, the shortage of masks was different across states. We plotted the map showing mask shortage in facilities with staff COVID-19 cases across states (Figure 1). From Figure 1, for example, we can see that California, Alabama, and Indiana had the largest number of facilities lacking both N95 and surgical masks, whereas Montana, Wyoming, and North Dakota had no shortages.
Nursing facilities with a shortage of masks and new weekly staff COVID-19 cases at the week ending June 21, 2020 (Li et al., 2020).
Furthermore, we checked the relationship between supply with different factors, for example, COVID-19 confirmed staff cases and resident cases, ownership, state variations, and so on. To examine the effect of these factors, we used a multivariate logistic regression model and summarized the results in Table 2.
Multivariate Logistic Regression Results (States’ Output Omitted).
Note. OR = odds ratio; CI = confidence interval; FP = for-profit; NF = nursing facilities.
First, the number of resident cases was positively associated with the N95 supply with an odds ratio (OR) of 1.05 (p < .011). A positive relationship between N95 supply and resident cases was also observed with surgical masks but was not significant (OR = 1.04; p = .169). The staff cases were not significant for either N95 or surgical masks. Second, our model showed that ownership and the number of nursing facilities in the same county had significant impacts on the supply of N95 and surgical masks. Specifically, compared with NFP facilities, FP nursing homes were less likely to have N95 (OR = 0.60, p < .001) and surgical masks (OR = 0.55, p = .002). Moreover, there was a negative relationship between the number of nursing facilities and surgical masks (OR = 0.905, p < .001). In addition, we found that state variations existed in N95 and surgical mask supply. For example, compared with California, Texas was more likely to have an adequate supply of N95 (OR = 3.13, p < .001) and surgical mask (OR = 4.18, p =.006), whereas Iowa was less likely to have an adequate supply of N95 (OR = 0.32, p < .001) and surgical masks (OR = 0.19, p < .001).
We also checked how the supply for different ownership facilities changed over time. Compared with May 24, the shortage of N95 and surgical masks was mitigated regardless of the ownership type (Figure 2).
The percentage of N95 and surgical mask shortage by nursing home ownership types over time (Li et al., 2020).
Discussion
Our study explored the factors associated with nursing facilities’ access to adequate N95 and surgical masks. We found that state variations existed and ownership played a significant role. However, the number of COVID-19 cases did not necessarily lead to sufficient mask supplies.
First, a positive relationship was observed between the number of resident cases and the N95 supply. A previous study assumed nursing homes with more resident cases had a higher rate of consuming PPE (McGarry et al., 2020). It is possible that increased resident cases put nursing homes under urgent pressure to actively seek more N95 to prepare for rapid use. Also, nursing homes located in hot spots were on the government’s priority list in getting N95 masks (Sinha, 2020). However, the staff cases did not lead to an adequate N95 and surgical mask supply. Such unclear linkage between staff cases and a sufficient mask supply put the control of COVID-19 at risk, given that staff health played a vital role (Chen et al., 2020).
Second, our study indicated that the number of facilities in the same county had a significant negative effect on the supply of surgical masks. Nursing facilities should contact their local or state public health emergency management agencies to get more supply on critical PPE (Stein et al., 2020; Strickler, 2020). Given this situation, when the number of facilities was large, it was inevitable to cause competition and create a shortage for nursing homes.
Third, compared with NFP nursing homes, FP facilities were more likely to have a shortage of supply, consistent with a recent study (McGarry et al., 2020). It is possible that during the pandemic, PPE were expensive (Diaz et al., 2020), so FP nursing homes might not actively seek supply considering the high cost and devoted fewer resources to direct patient care compared with NFP facilities (Schroeder, 2018). Previous studies indicated that FP nursing homes tended to have more COVID-19 cases, so the shortage of mask supply intensified the situation (e.g., Abrams et al., 2020; He et al., 2020).
Last but not least, state variations of mask shortage existed. For example, for N95 masks, we can see that Texas was more likely to have an adequate supply of N95 and surgical masks compared with California, whereas this was the opposite for Iowa. This might be attributed to many factors of each state, for example, distribution policy and supply budget, and showed effects in getting enough supply (Federal Emergency Management Agency, 2020; Sinha, 2020).
There were several limitations to our study. One was that the nursing home COVID-19 data did not report the actual amount of mask supply, but only provided binary indicators on whether the shortage existed in each facility. This impeded our analysis on further development of a proper distribution plan. Another was that most states did not list how they distributed supplies to different facility types. We could find such statistical results for Massachusetts only; however, these data for all states would be key factors in evaluating states’ performance.
Conclusion
In summary, findings included evidence of nursing homes’ facility-level and county-level factors associated with N95 and surgical mask supply. Larger COVID-19 resident and staff cases did not necessarily lead to an adequate supply. This urges the central planner to develop a more efficient and effective distribution plan. In addition, data should be clear and complete to improve the supply chain transparency for better decision-making. Finally, for potential coming waves of the coronavirus, it is important to guide nursing homes in preparing PPE for future demand surge and supply risk.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
