Contributions to the literature

• Our study found changes in the implementation of several COVID-19 infection control best practices in nursing homes amid the pandemic, especially practices related to visitation policies, communication with families and residents, hygiene, management of deceased residents, and communal restrictions.

• We found that culture stress was an important inner setting factor that may have impacted implementation activities in nursing homes.

• More research is needed to continue evaluating implementation measurement in the nursing home context to support nursing home administrators and staff in closing the gap between infection control guidelines and practice.

Introduction

The COVID-19 pandemic has claimed over 1.1 million lives in the United States [1]. As of April 2023, more than 166,000 nursing home residents and 3,000 staff have died from COVID-19 [2]. Advanced age, underlying frailty, and communal living conditions make nursing home residents especially vulnerable, and their reliance on nursing home staff members put these essential workers at high risk. In addition to the devastating mortality rates experienced by nursing home residents and staff, nursing home administrators (NHAs) and staff have faced constant challenges with heavy workloads, burnout, staff shortages, and changing guidelines regarding infection prevention and control since March 2020 when COVID-19 was declared a pandemic [3]. These challenges can contribute to increased levels of organizational culture stress.

To prevent COVID-19 deaths and rising cases in nursing homes, facilities were required to increase engagement of infection control staff [4]. However, only 3% of designated infection preventionists in nursing homes had basic infection control training prior to the pandemic [4, 5]. With the constant change in information about COVID-19, the need for knowledge and skill sharing for infection control best practices was critical, including information about the correct use of personal protective equipment (PPE), improved hygiene practices, and how to better communicate accurate COVID-19 education with residents and their families [6, 7]. NHAs also needed to learn strategies on how to deal with staff shortages, stress, and burnout [3]. In addition, training that integrates opportunities to improve the inner setting capacity (e.g., learning climate, leadership engagement, organizational culture) of nursing homes to implement COVID-19 infection control guidelines was needed. Although guidelines may appear relatively straightforward, implementation requires organizational capacity, staff engagement, and problem-solving which can strain organizations lacking appropriate training, resources, and support [8].

In response to the training and implementation needs of nursing homes during the pandemic, the Agency for Healthcare Research and Quality (AHRQ) and the University of New Mexico’s ECHO (Extension for Community Health Outcomes) Institute partnered to create the ECHO National Nursing Home COVID-19 Action Network (hereinafter, “the Network”). The Network provided free training and mentorship to nursing homes across the country to increase the implementation of best-practice infection prevention and safety to protect residents and staff. Project ECHO at the Penn State College of Medicine was part of the Network, connecting infectious disease, geriatric, and nursing home experts with NHAs and staff in Northeast and Midwest states to provide needed infection control training [8]. The present study sought to assess the implementation of COVID-19 infection control best practices in participating nursing homes.

Methods

Participants and procedures

This study was part of a larger cluster randomized controlled trial (RCT) enrolling nursing homes that took the Network’s training. The study team recruited facilities from Connecticut, Delaware, Illinois, Indiana, Maryland, New Hampshire, New Jersey, New York, Ohio, Pennsylvania, Vermont, Virginia, and Wisconsin. A random sample of 290 nursing homes were assessed for eligibility from national nursing home lists that comprised 2,000 facilities, with 136 facilities enrolled into the RCT [8]. Eligibility criteria for nursing home facilities to participate included United States-based, Centers for Medicare & Medicaid Services (CMS)-eligible skilled nursing facilities with access to a computer or electronic device for intervention participation. Facilities were ineligible if they had previously participated in a Project ECHO-led COVID-19 series. Briefly, the RCT was conducted in two phases. During phase one, nursing homes in both intervention arms (ECHO and ECHO+) received the Network’s 16-week infectious disease and quality improvement training via real-time, interactive videoconferencing [8]. These 16 sessions were 90 min in length each. During phase two, nursing homes randomized to the ECHO arm were offered optional 60-minute office hours for 9 weeks, and facilities in the ECHO + arm attended 9 weeks of 60-minute sessions on emerging topics and an additional 8-session refresher course on infection control [8]. The RCT aims were to compare the effectiveness of ECHO versus ECHO + on the number of nursing home residents with COVID-19 between baseline and 18 months (primary) and other outcomes, including quality of life, flu-like symptoms, hospitalizations, and deaths (secondary). Additional details about the protocol [8] and findings of this cluster RCT [9] are available elsewhere.

Concurrent with the RCT delivery, we surveyed NHAs at baseline (December 2020-February 2021) and 6-month follow-up (June-August 2021) inquiring about infection control practices their nursing home implemented throughout the RCT. Participants were provided with the summary of research and voluntary implied consent was obtained from all participants through completion of the baseline survey. Surveys were distributed by e-mail and participant responses were collected through REDCap, a HIPAA-compliant and secure web application for online surveys [10]. Our study was guided by the RE-AIM framework to critically evaluate both the effectiveness and implementation outcomes of the cluster RCT. This report focuses exclusively on the survey data on the implementation aspects of infection control best practices. Approval for this study has been obtained from the Institutional Review Board at the Pennsylvania State University (STUDY00015883).

Measures

The surveys distributed to NHAs assessed the implementation of COVID-19 infection control best practices in nursing homes and selected inner setting factors that may have influenced implementation. First, using the Centers for Disease Control and Prevention’s (CDC) COVID-19 Preparedness Checklist [11], our baseline and follow-up surveys asked 80 items inquiring about best-practice activities organized in 15 major areas: (1) preparedness (5 items), (2) visitation policies (8 items), (3) communication with family members of residents (2 items), (4) internal and external COVID-19 notifications (5 items), (5) providing COVID-19 information and education to staff (5 items), (6) infection control practices among staff (8 items), (7) providing COVID-19 information and education to residents (4 items), (8) infection control practices among residents (8 items), (9) potential resident cases of COVID-19 (6 items), (10) potential staffing crises (3 items), 11) deceased residents (3 items), 12) PPE and critical materials (7 items), 13) use of PPE in staff (5 items), 14) hygiene practices within the nursing homes (6 items), and 15) hygiene practices among staff (5 items). NHAs answered all items using yes/no responses. The complete survey is shown in Appendix 1.

The surveys also asked validated scales about the inner setting of nursing homes, identified as an important set of constructs that can influence the implementation of interventions into practice [12]. Our list of inner setting measures included 38 items to evaluate seven constructs and sub-constructs aligned with the Consolidated Framework for Implementation Research (CFIR) inner setting domain. Culture, defined as the norms, values, and basic assumptions of a given organization, was measured with 9 items. Culture stress, defined as the perceived strain, stress, and role overload, was measured with 4 items. Culture effort, how hard people in organizations work toward achieving goals, was measured with 5 items. Implementation climate, defined as the shared receptivity of involved individuals to an intervention and the extent to which use of that intervention will be rewarded, supported, and expected within their organization, was measured with 4 items. Learning climate, in which team members feel that they are essential, valued, and knowledgeable partners in the change and there is time and space for reflective thinking and evaluation, was measured with 4 items. Readiness for implementation was assessed via leadership engagement and available resources. Leadership engagement is the commitment, involvement, and accountability of leaders and managers, and was measured with 4 items. Available resources, defined as the level of resources dedicated for implementation and ongoing operations including money, training, education, physical space, and time, was measured with 7 items. All items were asked using a 5-point agreement scale with response options ranging from strongly disagree to strongly agree. The selected scales have shown strong psychometric properties (e.g., factorial validity, internal consistency, discriminant validity) in prior research [13].

We also collected information about the characteristics of nursing home facilities, including the location (urban or rural), facility size (< 60 or ≥ 60 beds), payment model (for-profit or not-for-profit), facility type (independent or networked), and memory care status (yes/no).

Data analysis

Descriptive statistics were presented to summarize the sample characteristics. To evaluate the implementation of infection control best practices (all item responses were binary, yes/no), we used a McNemar’s test for paired proportions to determine if there is any difference in the proportion of the outcome between baseline and 6 months. For inner setting factors, we averaged the individual Likert scale items to create the corresponding seven scales. We used a paired t-test for paired means to determine if there is any difference in the mean of the outcome between baseline and 6 months. Statistical tests were two-tailed with a critical alpha of 0.05. All analyses were conducted using SAS.

Results

Thirty-eight NHAs, representing 28% of enrolled facilities in the RCT, responded to both surveys. The majority of respondents worked in nursing home facilities located in rural areas (61%) and were considered large (> 60 beds, 68%, Table 1). Over half of facilities were not-for-profit (53%) and were independent (58%). Only 18% of NHAs in our sample worked for a facility offering memory care services. We observed statistically significant changes (p < 0.05) in eleven best-practice activities, described below, between baseline and 6-month follow-up (Table 2). A summary of statistical analyses for all items is shown in Appendix 2. There were no statistically significant differences between the ECHO and ECHO + participants for the infectious control practices assessed in the present study.

Table 1 Baseline characteristics of nursing homes among implementation survey participants (n = 38)
Full size table
Table 2 Assessment of COVID-19 infection control best-practice activities that nursing homes implemented at baseline and 6 months (n = 38)
Full size table

Visitation policies

NHAs reported that more nursing homes restricted non-essential personnel including volunteers and non-medical providers from entering the building at baseline when compared to 6 months (95% vs. 63%, p = 0.002). Similarly, all visitations, except those for compassionate care, were restricted at baseline versus 6 months (92% vs. 45%, p < 0.001). At baseline, facilities had also sent more communication to families advising them of the change in visitation policy that no visitors would be allowed in the facility except for certain situations when compared to follow-up (97% vs. 68%, p < 0.001). There was also a decrease in signage posted at facility entrance advising no visitors may enter the facility at the 6-month period (95% vs. 61%, p < 0.001).

Communication with family members of residents

NHAs reported an increase in reading-level appropriate informational materials on COVID-19 developed or identified for residents and families between baseline and 6-month follow-up (84% vs. 100%, p = 0.031).

Infection control practices among residents

At baseline, a majority of NHAs reported that facilities stopped group activities inside the facility and events outside of the facility, but a change in this practice was noted at 6 months (87% vs. 61%, p = 0.013). Similarly, many NHAs reported a decrease in the number of facilities that stopped communal dining between baseline and 6 months (87% vs. 50%, p < 0.001). NHAs also reported that more facilities encouraged their residents to stay in their rooms at baseline when compared to 6 months (92% vs. 53%, p < 0.001).

Deceased residents

Between surveys, NHAs reported an increase in the development of contingency plans for postmortem care and disposition of deceased residents (53% vs. 82%, p = 0.013) and the identification of facility areas to be used as a temporary morgue (34% vs. 58%, p = 0.020).

Hygiene practices within the nursing home

NHAs also noted an increased preference for alcohol-based hand sanitizer over soap and water in their facilities from baseline to 6 months (76% vs. 97%, p = 0.008).

Inner context factors

NHAs reported a higher mean score for the culture stress scale at 6-month follow-up when compared to their initial score at baseline (3.58 vs. 3.14, p = 0.004). No statistically significant differences were observed for the other six scales (Table 3).

Table 3 Mean scores of inner setting factor scales at baseline and 6-month follow-up (n = 38)
Full size table

Discussion

COVID-19 has had a disproportionate impact on nursing home residents and staff [14]. This is among the first studies to report changes in the implementation of infection control best-practice activities in nursing homes amid the COVID-19 pandemic. In our study, NHAs reported a decrease in restrictions regarding four types of visitation policies and three communal restriction practices in nursing homes as assessed by the CDC’s COVID-19 Preparedness Checklist. This finding is consistent with trends observed as the pandemic evolved, whereby the CMS directed nursing homes to significantly restrict visitors and nonessential personnel, as well as restrict communal activities inside nursing homes in 2020 [15, 16]. The roll-out of COVID-19 vaccinations in late 2020 and early 2021 provided needed protection to nursing home residents and staff against COVID-19, reducing weekly COVID-19 infections and deaths [17]. Protection against COVID-19 thereby allowed nursing homes to loosen COVID-19 restrictions [18]. Citing a national COVID-19 vaccination rate of 86% among residents and low transmission rates in nursing homes, CMS dropped most COVID-19 restrictions in late 2021, allowing visits to take place indoors for all residents, regardless of vaccination status [19]. Importantly, CMS advised that visits should be conducted following the best practices of COVID-19 infection prevention and control without increasing risk to residents. Loosening these restrictions was important for the well-being of residents as studies show that pandemic restrictions affected nearly every part of their lives, especially their mental health [20]. Since CMS provisions were released during the study period and compliance with such provisions were tied to payment, it can be reasonably assumed that those provisions could have been a major driver for implementation decisions in nursing home facilities.

On the other hand, NHAs reported an increase in the implementation of several best-practice activities, including the availability of informational materials for residents and families, the use of alcohol-based hand sanitizer over soap, the development of contingency plans for increased postmortem care, and the identification of facility areas to be used as a temporary morgue. The preference for alcohol-based sanitizers likely reflects the recommendation made by the CDC that healthcare workers should use alcohol-based hand sanitizer as the primary method for hand hygiene in work settings. Also, the increase in activities for managing postmortem care and disposition of deceased residents at the 6-month follow-up was expected as nursing homes faced many challenges dealing with an excess of deaths during the peak of the pandemic in 2020 [21].

When evaluating the inner setting factors of participating nursing homes, we found that NHAs reported increased levels of culture stress (perceived strain, stress, and work overload) over time, resulting in too much pressure to do their job effectively. This finding is consistent with data showing that 43% of healthcare workers in the US suffered from work overload and 49% had burnout amid COVID-19 [22]. The pandemic added significant strain to an already vulnerable nursing home workforce, which has historically experienced high levels of turnover and burnout [3]. Staffing shortages and heavy workloads lead to increased job pressure and dissatisfaction among nursing home staff, which may generate high levels of collective stress in nursing home facilities [23]. Although not assessed in our study, it is possible that receiving too many regulatory demands from local, state, and federal agencies, often confusing or contradictory, contributed to the reported scores pertaining to culture stress among participants. Several studies suggest that receiving unclear guidance and the evolving nature of COVID-19 information hindered the implementation of infection control activities in nursing homes [24, 25]. No other inner setting factor showed a statistically significant difference between baseline and 6-month follow-up, including learning climate and implementation climate. This may suggest that more education and training is necessary to meet sustainable implementation changes in the nursing home setting [26]. For instance, studies have highlighted the effectiveness of continuing education and workforce development programs to meet required competencies among nursing home staff [26, 27]. Future studies should consider the evaluation of implementation determinants beyond inner setting factors.

Several limitations of the study may be noted. First, data from only 38 NHAs were included in the present analyses as the remainder of NHAs who took the baseline survey did not complete the 6-month follow-up survey. This low response rate was due to several factors, including competing priorities, staff turnover, or disinterest in study activities. Also, the six-month time period of this study further limits our ability to assess long-term maintenance of infection control best practices. Additionally, we were unable to limit participation in other COVID-19-related training, and assessing whether participation in other training may have impacted our findings. Another limitation was that some items (e.g., preferences for alcohol-based hand sanitizers) asked about respondents’ perceptions of staff preferences, but their responses could be biased by mandated practice changes or awareness limited to closest employees and not the entire workforce. Finally, our study findings should be interpreted with caution because respondents came from a convenience sample of NHAs exposed to the ECHO interventions with no comparator group. In terms of strengths, our surveys asked validated items from the CDC’s COVID-19 Preparedness Checklist and the CFIR. Although we reported on a small sample, participants were from varying nursing home facilities, including rural/urban location, facility size, payment model, facility type (independent or networked), and memory care status.

Conclusion

Our study found statistically significant changes in the implementation of several COVID-19 infection control best practices in nursing homes amid the pandemic, especially activities related to visitation policies, communication with families and residents, hygiene practices, management of deceased residents, and communal restrictions. We also found that culture stress was an important inner setting factor that may have impacted implementation activities. More research is needed to continue evaluating implementation measurement in the nursing home context. Project ECHO seems an acceptable training strategy to support nursing homes in closing the gap between infection control guidelines and implementation.