Robust infection prevention and control is critical to the safe care of nursing home (NH) residents. US payment regulations require NHs to have an infection prevention and control (IPC) program that includes “a system for preventing, identifying, reporting, investigating, and controlling infections” and written policies, standards, and program procedures. 1 A designated infection preventionist (IP) responsible for this program is also required. 1 However, meeting these requirements has proven difficult for many NHs, due in part to resource constraints and staff turnover. Reference Loomer, Grabowski, Yu and Gandhi2
To bolster NH IPC programs and reduce healthcare-associated infections, we implemented a 12-month program, Preventing Resistance and Infections by Integrating Systems in Michigan (PRIISM), to enhance relationships between NHs and hospitals. Reference Mody, Washer and Flanders3 PRIISM, which began January 2018, brings NH IPC programs and their affiliated acute-care hospital IPC programs together aligning prevention initiatives and provider training across the healthcare continuum. The overarching goals of PRIISM were to reduce infections and enhance resident safety, with specific focus on reducing catheter-associated and non–catheter-associated urinary tract infections (CAUTIs and UTIs). PRIISM was initially developed, tested, and refined in 13 NHs in southeastern Michigan and their 3 main referring hospitals. In the 3 years since it began, an additional 67 Michigan NHs have implemented PRIISM.
The purpose of this evaluation study, which focused on 56 NHs participating during the first 2 years, was to understand key contextual factors among NHs participating in PRIISM at program outset. This analysis included NH IPC program characteristics, IP and staff training, CAUTI and UTI prevention strategies, and communication surrounding resident transfers with hospitals. This information was used to modify program content, to identify specific needs, and to inform ongoing efforts to support NHs meeting regulatory requirements and ensuring safe resident care.
Methods
Study design
In this mixed-methods study, we integrated data from cross-sectional NH surveys and semistructured interviews conducted with PRIISM participants during the first 2 years of the program, cohort 1 (ie, the pilot year) and cohort 2. This research was deemed exempt from approval by the University of Michigan Medical School Institutional Review Board.
Sampling and recruitment
Participating PRIISM NHs were recruited following a recommendation from a local participating hospital or request to participate after learning about the project at a hospital-led postacute care meeting. There were no specific enrollment criteria beyond the expectation that NHs would remain engaged throughout the 12-month participation period. At the beginning of each project year (2018 and 2019), participating NHs received a study survey for completion by the NH administrator, director of nursing (DON), or IP. During these project years, we recruited clinical leadership and PRIISM NH participants, primarily DONs, assistant directors of nursing (ADONs), and IPs to participate in qualitative semistructured interviews. Potential participants from each NH were emailed twice. For those who agreed, an interview was scheduled at their convenience. No response was considered a passive decline.
Data collection
Quantitative data collection
The survey, distributed between January 2018 and March 2019, comprised 36 questions. Survey domains included facility characteristics; IPC program, including experience and training of the individual responsible for IPC and antibiotic stewardship; surveillance activities; CAUTI/UTI prevention practices; and communication with other healthcare facilities.
Qualitative data collection
Semistructured interviews with NH staff were conducted between April 2018 and November 2019, while the NHs were participating in the PRIISM program. Interview topics included the same survey domains to gather more contextualized information across NHs (Appendix A online). The questions also focused on experiences with the PRIISM program, relationships with local hospitals, resident transfer between hospitals and NHs, and barriers to and facilitators of implementing IPC practices. Prior to the interview, verbal consent was obtained. Interviews were conducted in person, before or after a PRIISM meeting, or by telephone, and interviews were audio-recorded and transcribed verbatim.
Data analysis
Quantitative analysis
Responses for each survey question were summarized using Stata version 15 software (StataCorp, College Station, TX). Categorical responses were summarized as percentages, excluding any missing responses. Numeric responses were summarized as mean (range).
Qualitative analysis
A descriptive matrix analysis approach was used to analyze patterns of responses among participants. Reference Averill4 This approach involves creating a table (ie, rows and columns) to analyze domains of interest across sites and/or participants. Within each cell, participant data are synthesized, but they reflect participant perspectives and often contain direct quotes supporting the interpretation. This approach is valuable in identifying congruence and incongruence across domains and participant experience. For this study, the matrix was constructed with rows representing interview topics and columns representing participant roles and NH affiliation. Transcripts were read by 2 study team members (S.L.K. and M.H.), and the responses were summarized and added to the matrix along with applicable supporting quotes. The full study team reviewed the matrix, and if necessary for clarification, transcripts were revisited for additional information to ensure accuracy of interpretation. Each domain was reviewed across participants and was summarized to develop the overall findings. Findings were confirmed through discussion among the study team.
Data integration
Using the matrix as an analytic tool, quantitative survey data findings were combined with the qualitative data in the form of a joint display. Reference Guetterman, Fetters and Creswell5 This type of integration facilitated development of further insights as qualitative results helped explain and provided additional context to the quantitative findings. Our analysis focused on 3 specific areas to improve understanding of (1) IPC infrastructure; (2) UTI and CAUTI outcomes, process measures, and prevention strategies; and (3) communication with hospitals.
Results
Surveys were completed by >90% of participating NHs in cohorts 1 and 2. Of 51 NHs, 45 (88.2%) identified as for-profit facilities. The mean bed size was 130 (median, 127; range, 24–286), and 11 (22%) of 50 NHs were enrolled in the National Healthcare Safety Network (NHSN) at the start of the study. Semistructured interviews (average length, 53 minutes) were conducted with 13 participants representing 7 different NH providers: cohort 1 comprised 4 NHs and cohort 2 comprised 3 NHs. Most NHs in the PRIISM program were individual facilities, but some were NH providers with multiple participating facilities. Interviewees included facility administrators, DONs, ADONs, IPs, regional managers, and clinical operations staff (Table 1).
Nursing Home Participation
Note. NH, nursing home; IP, infection preventionist; Director of nursing (DON); vice president (VP).
Infection prevention and control infrastructure
Surveyed NHs reported a mean of 1.11 full-time employees (range, 0–3) dedicated to the IPC program (Table 2). A minority (26.0%) of those in the IP role had been in that facility’s role > 5 years, and most (74.0.%) had <5 years of IP experience. Staff in the IP role frequently held other positions within the facility, often staff education and development (60.8%), employee health (25.5%), and serving as DON (27.5%). On average, IPs performed 1.7 activities (range, 1–6) in addition to their IP responsibilities. In 76% facilities, the IP provided IPC-related training to facility staff.
Infection Prevention and Control Infrastructure
Among facility IPs, 36.7% had no specific IPC training. Nearly all participating facilities (96.1%) had a committee to review healthcare-acquired infections (HAIs). These committees predominantly included unit managers (89.8%), medical directors (85.7%), and nursing administrators (85.7%).
Our qualitative findings supported the survey results (Table 2). Frequent IP turnover, limited IPC experience, and holding multiple roles were commonly described by interviewees. Educating staff about IPC was generally the responsibility of the IP, but educators, DONs and ADONs were often involved. In addition to limited experience, IPs usually had limited training, although some facilities described actively focusing on IP training. Finally, member composition varied among those having an HAI review committee. Although most committees included nurse managers, physicians, and administrators, some NHs included a broader range of staff, such as pharmacy and environmental services.
CAUTI- and UTI-specific processes and prevention practices
Most respondents were aware of their facility CAUTI and UTI rates (85.7% and 92.0%, respectively) (Table 3). A majority indicated using an electronic health record system to collect CAUTI and UTI data (88.0%), yet most respondents (90.0%) also reported reviewing provider notes to determine resident UTI diagnosis. All respondents shared infection data with NH leadership, 72.0% shared infection data with bedside nursing staff, but only 32.0% shared infection data with residents and families.
CAUTI/UTI: Outcomes, Process Measures and Prevention Strategies
UTI and CAUTI prevention strategies included hydration practices (85.7%), nurse-initiated indwelling urinary catheter discontinuation (65.3%), stop orders (49.0%), indwelling device rounds (36.7%), and electronic alerts or reminders to assess indwelling catheter need (26.5%). Nearly two-thirds (63.3%) of NHs reported using cranberry juice and/or tablets as a UTI prevention strategy. Two facilities (4.1%) reported no prevention strategies.
Consistent with quantitative findings, our qualitative results revealed various strategies to collect CAUTI and UTI data, but in general most facilities used a manual data extraction process. This process included reviewing provider notes, laboratory data, antibiotic prescription information and managing the information in spreadsheets. Most interviewees used Revised McGeer Criteria to identify resident infections. Sharing data with floor staff varied across NHs. Data were shared most often in the context of education and practice change. One interviewee stated that although data was shared with floor staff, they felt there was not enough time to discuss what it meant and how to improve.
To prevent CAUTIs, most NHs indicated providers were amenable to discontinuing indwelling urinary catheters on or soon after admission. This was attributed to previous education around prompt catheter discontinuation and catheter assessments at admission. In some NHs, nurse managers assessed catheter appropriateness. To prevent CAUTI and UTI, all NHs mentioned hydration protocols, and at least 2 described using “cranberry protocols,” which included giving high UTI-risk residents UTI-Stat, a cranberry concentrate.
Another area of focus involved residents admitted with or placed on antibiotics for suspected UTI, without symptoms or urine-culture results. Several interviewees stated educating staff nurses about symptoms and urine testing appropriateness was an ongoing task. To address this issue, one NH implemented testing with a urine dipstick before calling the physician for a urine culture order. Another NH implemented a UTI-focused assessment process or Situation-Background–Assessment–Recommendation (SBAR). As the IP explained,
“It’s just specific for UTIs. It goes over in the first section, the basic, is there a fever, does he have any flank pain, any acute pain, any shaking or chills, any mental status changes even though that’s not a clinical sign of a UTI, but if we have these checked for the patient, then we’ll go on. Is there urgency, is there frequency, is there suprapubic pain, hematuria? These are just stuff that the nurse’s looking at so they could have a clear picture when they call the doctor.” [Infection preventionist, NH 3]
Infection-related communication with hospitals
NHs shared information with other healthcare facilities about infections and MDRO history during resident transfers through a variety of mediums. When transferring residents in or out of the facility, infections were communicated using transfer sheets (83.7%), discharge orders (65.3%), phone calls (38.8%), and uniform assessment instruments such as SBAR (12.2%) (Table 4). MDRO histories of residents were communicated via transfer sheets (85.4%), discharge orders (56.3%), phone calls (39.6%), and uniform assessment instruments (10.4%).
Infection-Related Communication with Hospitals
Although the survey results suggested what appeared to be a relatively standard process for communicating infection-related information, our qualitative findings provided a different picture, highlighting several communication challenges. For example, NHs received either too little or too much information, making it difficult to get a timely and complete picture of the resident’s status. In addition, most interviewees described challenges related to those involved in the communication process, with admission information being sent to an admissions person who may or may not have a clinical background and little to no direct clinician-to-clinician communication between the hospital and NH. Finally, from the NH viewpoint, there was a perception that residents admitted to the hospital were assumed to have an infection, ultimately ending up on antibiotics.
Discussion
Developing and maintaining an effective IPC program is a crucial part of ensuring safe care for NH residents. Our mixed-methods study identified not only several challenges but also positive use of prevention practices and opportunities related to IPC among NHs participating in the PRIISM program. First, IP turnover, filling multiple roles, and having limited IP experience were common challenges. Second, tracking, awareness, and reporting of infection rates (notably CAUTI and UTI) were nearly universal. Most NHs reported using recommended practices to reduce indwelling urinary catheter use and to prevent UTIs, along with some nonbeneficial practices. Finally, although most NHs reported strategies for sharing infection-related information during resident transfers, opportunities for enhancing IPC-related communication between NHs and hospitals still exist.
To receive US federal funding, NHs are required to have a designated IP with specialized IPC training. 1 Our work highlights the specific challenges of maintaining consistent, appropriately trained IP staff, an issue that is both more important and perhaps more difficult due to the COVID-19 pandemic. At most PRIISM NHs, the IP was in the position <5 years, and more than one-third of NHs reported the IP lacked specific IPC training. Often, IPs were responsible for other activities, such as staff education, employee health, or served as the DON. These findings suggest a focused need for identifying and implementing strategies that help NHs recruit and retain dedicated IP staff, and providing these IPs with the specialized education and knowledge they need. In addition to ensuring IP staff have time and resources needed to attend credible training courses, such as those offered through the CDC and Association for Professionals in Infection Control and Epidemiology (APIC), a coaching and mentoring program might be a useful strategy for providing ongoing education and support.
Recommended practices for preventing CAUTI and UTI in NH residents include general practices (eg, surveillance and hand hygiene) and specific practices (eg, reducing indwelling catheter use) as well as improving diagnosis. Reference Meddings, Saint and Krein6 More than 80% of participating facilities indicated that they conduct surveillance, that they are aware of their CAUTI and UTI rates, and that they share these data with NH leadership. Indeed, this group of NHs reported engaging in these activities more often when compared to a cohort of NHs participating in a national CAUTI and UTI–focused collaborative program beginning in 2014–2015, in which only 58% knew their CAUTI rates and 70% shared data with leadership. Reference Mody, Meddings and Edson7 Based on the survey and qualitative findings, PRIISM NHs demonstrated a clear focus on strategies to reduce indwelling catheter use, including nurse-initiated discontinuation, and assessment and timely removal if not indicated. Both hydration and use of cranberry products were among the CAUTI and UTI prevention practices reported by most PRIISM NHs. Although evidence to support hydration as a UTI prevention practice is limited, Reference Lean, Nawaz, Jawad and Vincent8 a recent study of structured drink rounds to promote hydration found a reduction in the average number of UTIs requiring antibiotics (1.8 at baseline to .75 after the intervention) and those requiring hospital admission. Reference Lean, Nawaz, Jawad and Vincent8 Further research focusing on hydration practices in NHs for UTI prevention is warranted. However, studies of cranberry use continue to demonstrate little benefit among this population and may be a practice suited for de-implementation. Reference Juthani-Mehta, Van Ness and Bianco9–Reference Ashraf, Gaur and Bushen11
Although not included in the survey, antimicrobial use and urine testing for possible infection was a common topic of discussion during the qualitative interviews. In 2017, point-prevalence data indicated that ∼1 in 38 US NH residents received an antibiotic for a UTI on any given day. Reference Thompson, Penna and Eure12 This included antibiotics prescribed for UTI prevention, the likely overuse of fluoroquinolones, and median planned duration beyond 7 days for almost all commonly prescribed antibiotic classes. Reference Rutten, van Buul and Smalbrugge13 Concerns about antibiotic overuse and misuse prompted development of consensus recommendations specifically related to diagnosis, treatment, and prevention of UTIs in postacute and long-term care settings by the Infection Advisory Subcommittee of AMDA, the Society for Post-Acute and Long-Term Care Medicine. Reference Rutten, van Buul and Smalbrugge13 With several studies and emerging interventions focusing on promoting appropriate antibiotic use and diagnosis of UTIs in NHs, Reference Rutten, van Buul and Smalbrugge13,Reference Aliyu, Travers and Heimlich14 our findings suggest a high level of awareness and interest among NHs with a desire for strategies to address these issues.
Improving IPC-related communication between NHs and local hospitals was one of the motivating factors for launching PRIISM. Reference Mody, Washer and Flanders3 Thus, it is not surprising that our findings identified challenges and areas for improvement in this domain. Although most facilities indicated using transfer sheets, interviews revealed issues with the amount and type of information received, and individuals involved in exchanging information. In general, there appeared to be little inter-clinician communication and the perception that hospitals assumed NH residents sent to the hospital had an infection. Some of these issues are reflected in other studies. Reference Rutten, van Buul and Smalbrugge13,Reference Campbell Britton, Petersen-Pickett and Hodshon15,Reference Shah, Burack and Boockvar16 Nonetheless they highlight the need for strategies and tools to improve interfacility infection-related communication practices. Such strategies include encouraging use of communication tools, such as the CDC’s Inter-Facility Infection Control Transfer Form, 17 INTERACT, 18 and consideration of a statewide registry system (eg, the Illinois “XDRO registry” 19 ) with additional research focusing on developing strategies that improve NH and hospital communication and coordination.
Linking quantitative survey and qualitative interview data to provide a comprehensive understanding of NH IPC programs and UTI prevention practices is a strength of this work. However, our study has several limitations. Surveys and interviews were conducted with PRIISM participants, a selected group of NHs from a single geographic region, with potential for response bias and limited generalizability. Interviews were also conducted during program participation, which may have affected some responses. Finally, data for this evaluation were collected prior to onset of the COVID-19 pandemic, and our findings do not reflect the potential impact of COVID-19 on NH IPC resources and practices.
In conclusion, IPC is a core element for the safe delivery of NH resident care. Unfortunately, workforce-related challenges, including IP turnover, limited education, and filling multiple roles, were common among NHs enrolled in the PRIISM project. Despite these challenges, participating NHs reported awareness of their CAUTI and UTI rates and use of recommended prevention practices. However, communication of infection-related information during transfers was a clear area with opportunities for improvement and further research.
Acknowledgements
Financial support
This research was supported by the Agency for Healthcare Research & Quality (grant no. RO1HS25451). Dr. Mody is also supported by the National Institute on Aging Mid-Career Mentorship (grant no. K24 AG050685); Michigan Institute for Clinical and Health Research (grant no. UL1TR002240); and the Claude D. Pepper Older Americans Independence Center (grant no. P30 AG024824). Dr. Mody is also supported by the Geriatrics Research, Education and Clinical Centers, Veterans’ Affairs Ann Arbor Healthcare System. Dr. Krein is supported through a Research Career Scientist Award from the Department of Veterans’ Affairs, Health Services Research and Development Service (grant no. RCS 11-222).
Competing interests
The authors have declared no conflicts of interest.
Appendix A. Selected Question Domains
Open access
