Discussion

Among 469 patients across 16 hemodialysis centers, we found that a minority of eligible participants tried the PVI intervention, and many of those who tried it were not consistently using it. Participants’ most frequent reason for not using PVI was that they forgot to use it or that staff did not offer it to them. Consistent with two prior studies of short term use among surgical patients,^{Reference Schweizer, Herwaldt and Boyken10,Reference Maslow, Hutzler, Cuff, Rosenberg, Phillips and Bosco11} and other studies of longer term use for rhinosinusitis,^{Reference Frank, Capriotti, Brown and Tessema14,Reference Mullings, Panchmatia and Samoy15} most participants in our study who used intranasal PVI found it neutral or pleasant to use. Despite participants’ reported satisfaction with PVI, adherence to the PVI protocol was low, which could be due to several factors including competing priorities for both patients and staff, especially during the COVID-19 pandemic, and patients’ misconception that their risk of infection is low.^{Reference Dukes, Hockett Sherlock and Racila18}

Hemodialysis patients are at high risk for multiple types of infections due to coexisting illnesses, presence of central venous catheters that can become infected, and frequent interaction with the healthcare system.^{Reference Dalrymple, Mu and Nguyen19–Reference Patel, Kallen and Arduino21} In fact, data from the Centers for Disease Control and Prevention document that the rate of S. aureus bloodstream infections is 100-times higher among adults on hemodialysis compared with adults not on hemodialysis (4,248 vs. 42 S. aureus bloodstream infections per 100,000 person-years).^{Reference Rha, See and Dunham4,22} Yet, in our survey, over half of our study participants reported they considered themselves at average or lower risk of infections than an average person in the US. This result highlights the need to improve patient education regarding their risk of hemodialysis-associated infection, the patient’s role in preventing infections, and thereby preventing the serious adverse outcomes of these infections. Further education might also help motivate patients to participate in infection prevention efforts.^{Reference See, Shugart, Lamb, Kallen, Patel and Sinkowitz-Cochran23}

Hemodialysis patients tend to have multiple co-existing conditions such as diabetes and heart failure.²⁴ Moreover, rates of depression, chronic pain, and difficulties with social support are also high among patients on hemodialysis.^{Reference Weisbord25} Given their already complicated treatment regimens and their underestimation of their infection risk, patients may have placed a low priority on adding and adhering to another routine task, leading to low participation and adherence rates. Other investigators have noted low compliance rates for hemodialysis-related treatments and behavior modifications.^{Reference Patel, Yi and Booth1} Thus, we likely need multifactorial strategies, such as health coaching and peer mentoring, to engage patients on hemodialysis in active infection prevention efforts.^{Reference Hussein, Bennett, Abra, Watson and Schiller26} We may also need to integrate infection prevention interventions into staff workflows at hemodialysis centers to ensure that patients consistently receive interventions such as PVI.^{Reference Dukes, Hockett Sherlock and Racila18} Previously reported interviews with hemodialysis staff during this study found that ensuring access to PVI supplies, and deliberate, repeated messaging about PVI to staff and patients facilitated use of PVI.^{Reference Dukes, Hockett Sherlock and Racila18}

Prior studies have shown that patients on in-center hemodialysis report a low degree of activation.^{Reference Hussein, Bennett, Abra, Watson and Schiller26,Reference Van Bulck, Claes and Dierickx27} In each survey, we assessed perceived activation and we asked about their willingness to exert effort to prevent infections. We hypothesized that involving patients in an infection prevention intervention would increase their perceived activation and willingness to take an active role in preventing infections. Contrary to our hypothesis, significantly fewer participants reported that they were willing to expend “a lot of effort” to prevent an infection at the later time points, which also corresponded with decreased use of intranasal PVI. Yet, the majority of patients still agreed with the perceived activation statements at each time point. The inconsistency in the responses to the activation and willingness questions suggests that patients on hemodialysis believe they are actively involved in their healthcare, but they may be unable or are unwilling to participate in a long-term infection prevention intervention.

Our study had some limitations. A minority of eligible patients participated in the intervention and completed the study surveys. Our surveys did not ask patients why they did not want to use PVI. Thus, we could not identify all motivational barriers to patients’ using PVI or other infection prevention measures. Anecdotally, patients often expressed apprehension about having their noses swabbed with PVI given past experiences with nasopharyngeal COVID-19 test swabs, which patients felt were uncomfortable. In addition, all hemodialysis facilities had masking requirements due to the COVID-19 pandemic, which made it difficult for participants to use intranasal PVI during dialysis sessions. Staffing issues, which were exacerbated during the COVID-19 pandemic, also hampered some sites from fully participating.^{Reference Dukes, Hockett Sherlock and Racila18}

Hemodialysis center staff commonly cite a lack of dedicated time as a barrier to implementing infection prevention interventions such as PVI.^{Reference Dukes, Hockett Sherlock and Racila18,Reference Fitzgerald, Tyner, Drake, Beach, Rupp and Schwedhelm28} Fragmentation of care for patients on hemodialysis leads to nephrologists and dialysis staff becoming responsible for much of the general health maintenance for these complex patients, allowing little time for infection prevention.^{Reference Sloan, Zhong and Mohottige29} This is also exacerbated by the shrinking nephrology workforce.^{Reference Sloan, Zhong and Mohottige29}

In summary, most participants found intranasal PVI to be acceptable and few participants reported side effects. Participants substantially underestimated their infection risk. Complex health conditions, complex treatment requirements, competing priorities, and social support needs may adversely affect patients’ interest in and ability to actively participate in infection prevention activities. Given the barriers that patients encounter, future studies should evaluate multifactorial strategies for educating patients on hemodialysis about their high risk of infection, helping patients maintain long-term engagement in infection prevention measures, providing personalized support for patients with health challenges, and optimally incorporating intranasal nasal decolonization and other infection prevention interventions into hemodialysis workflows. These multifactorial strategies could be informed by principles of adult learning, implementation science, and human factors engineering.