Discussion

In this paper, we report the findings of an in-depth and theory-based analysis of the behavioral determinants of hand hygiene adherence in neonatology. With this qualitative study, which allows a deeper understanding of HCP perceptions and drivers compared to quantitative methods, we identified 16 determinants. Twelve were considered of high priority to increase hand hygiene adherence in our setting. Four high-priority determinants were identified in each of the COM-B domains “psychological Capability,” “environmental and social Opportunity,” and “reflective and automatic Motivation.” While themes from “Capability” and “Opportunity” were both facilitators and barriers, high-priority “motivational” determinants exclusively were facilitators.

Several other study groups have investigated drivers for (non-)adherence in hand hygiene, some in neonatology, ^{Reference Pasricha, Singh, Pasricha, Jimal and Khurshid2,Reference Pessoa-Silva, Pfister, Touveneau, Perneger and Pittet18} but the majority in non-neonatal settings. Similar to the present study, most investigations identified a range of barriers and facilitators rather than just one or a few. ^{Reference Boscart, Fernie, Lee and Jaglal19,Reference Yehouenou, Abedinzadeh and Houngnihin20} Fuller et al., who investigated reasons for hand hygiene non-adherence on ICUs and acute care units by interrogating HCP immediately after hand hygiene observation, identified “memory/attention/decision making” and “knowledge” as the main determinants for hand hygiene noncompliance. ^{Reference Fuller, Besser, Savage, McAteer, Stone and Michie21} Similarly, Pasricha et al. who assessed barriers through questionnaire and thus with a greater distance from the action, identified “forgetfulness,” “lack of awareness,” and “lack of knowledge.” ^{Reference Pasricha, Singh, Pasricha, Jimal and Khurshid2} Such determinants categorized under the COM-B “Capability” domain, are per definition intrinsic to HCPs, and were confirmed by several other authors. ^{Reference Dyson, Jackson and Cheater22–Reference Le, Lehman, Nguyen and Craig25} While our interviewees also mentioned some of these as barriers (i.e., “lack of attention control” or “planning of workflows”), initially they expressed confidence in their knowledge and ability to perform hand hygiene. However, upon further questioning, some acknowledged knowledge gaps. On the contrary, the IPC experts, who were conducting hand hygiene interventions and adherence observations in the past, mentioned “knowledge” as one of the most common barriers in frontline staff. This could be attributed to individuals either not recognizing their own knowledge gaps or feeling hesitant to admit them in an interview situation. While “knowledge,” “attention control,” “planning of workflow” sure all are important for hand hygiene adherence, they all require active investment of HCP. Ideally, hand hygiene is performed automatically as a habitual behavior. As such, hand hygiene habits developed through years of work experience were often cited as facilitators.

The most frequently mentioned determinants in our study were those lying outside of the HCP’s perceived own influence, in the COM-B “Opportunity” domain. The high workload and patient emergency situations were often cited as barriers. This appears to be a common hindering factor in various hospital settings worldwide, ^{Reference Pasricha, Singh, Pasricha, Jimal and Khurshid2,Reference Song, Stockwell, Floyd, Short and Singh3,Reference Dyson, Jackson and Cheater22,Reference Deshommes, Nagel and Tucker26–Reference Lambe, Lydon and Madden28} likely linked to the high cognitive load, distraction, and competing priorities such situations bring along. On the other hand, the “availability of hand disinfectant” was perceived to be sufficiently present in our setting and thus was seen as non-priority determinant. In other hospitals, both in resource-limited countries, where there might be fewer hand disinfectant dispensers, and in higher resource settings, where dispensers were e.g., reported to be empty, non-availability of hand disinfectant was identified as one of the most relevant barriers. ^{Reference Deshommes, Nagel and Tucker26,Reference Bezerra, Valim and Bortolini29–Reference Vaughan-Malloy and Sandora31} In our neonatal setting, social influences, such as “having role models” or “being observed by others,” particularly parents, were important facilitators. This was also shown in an interview study with HCPs from ICUs that describes the sense of “being watched” and “reminders and encouragement from peers” to improve adherence, ^{Reference Lambe, Lydon and Madden28} and a questionnaire study that found that the “opinion of important others” influences hand hygiene behavior. ^{Reference Pessoa-Silva, Pfister, Touveneau, Perneger and Pittet18} These findings encourage the use of peer feedback and the fostering of a speak-up culture.

While our study identified several determinants in the COM-B “reflective and automatic Motivation” domain, such determinants were only rarely reported by other authors. ^{Reference Lambe, Lydon and Madden28,Reference Squires, Linklater and Grimshaw32} We hypothesize that the relevance of the motivational determinants is linked to the neonatal setting where patients are highly vulnerable and dependent, and the frontline staff feel a particularly high obligation to protect them from adverse events. Additionally, in neonatology, pathogen transmission events are frequently detected due to high microbiological screening activity. The determinants we identified in the “motivational” domain included active cognitive processes and reflections that – similar to the behavioral model of the “theory of planned behavior” ^{Reference Ajzen33} – all ultimately influenced the “intention or goal to adhere” to hand hygiene. Most motivational determinants were positive influencers, leading to an increase of the “intention to do hand hygiene.” However, how much the “intention to perform hand hygiene” predicts actual hand hygiene adherence is unclear, and was questioned by O’Boyle et al., who postulated that hand hygiene behavior may be more sensitive to external factors than to internal motivational factors. ^{Reference O’Boyle, Henly and Larson34}

As several other study groups, we perceived the use of the TDF to guide our analysis as useful. ^{Reference Yehouenou, Abedinzadeh and Houngnihin20,Reference Fuller, Besser, Savage, McAteer, Stone and Michie21} Assigning the determinant themes also to the COM-B model will later facilitate the tailoring of implementation strategies using the Behavior Change Wheel, a comprehensive framework that guides the design of behavior change interventions. In the process of introducing behavior change, prioritization of determinants – i.e., identifying important determinants – is one of the crucial steps to be carried out. In our setting, four determinants were indeed found but their addressing was perceived not necessary, either because the concepts or circumstances were well understood, obvious, or had already been addressed in the past. While “importance” is one prioritization criterion, “changeability” – the ease or difficulty of changing a factor – could be another. ^{Reference Green, Gielen, Peterson, Kreuter and Ottoson35} Highly changeable determinants, often termed as low-hanging fruits, could be initially targeted in behavior change interventions, especially when they are also considered important. Notably, not only the determinants themselves but also their classification based on importance and changeability can vary across settings, based upon specific contexts and priorities.

While this study has several strengths such as the theory-based approach, that facilitated comparison of our results with other papers, and the cumulative 16 hours of interviews with purposefully selected HCP, it also has some limitations. First, as in all interview studies, a social desirability bias cannot be excluded. Interviewees might have answered the questions in a way they believed to be more acceptable or favorable rather than expressing their true thoughts and behaviors. To mitigate this potential bias, we selected an interviewer who was not part of the IPC or the neonatology team, and who made efforts to create an atmosphere of openness and trust. We also included IPC experts as interview partners, who, based on their experience and close contact with frontline HCP, were able to provide a likely more objective perspective on the frontline staff’s hand hygiene behavior. Second, there is a risk of selection bias, as the interviews were conducted on a voluntary basis and employees with a positive attitude towards hand hygiene might have been more likely to participate. Third, the interviews were conducted in German, which might have led to employees with German as a second language being less likely to participate, potentially leading to reduced diversity of interviewees. Lastly, this was a single-center study conducted in a high-resource setting, and the findings may thus not be transferable to all other settings.

In conclusion, this qualitative interview study identified several determinants influencing hand hygiene adherence in a neonatology setting. Some are well-known, lying inside an individual HCP such as knowledge about indications or automatisms, or outside of HCP such as high workload and being watched by others. Others are new and likely particularly specific for the neonatal setting, such as the motivational determinants of self-reflection on hand hygiene behavior or the very strong intention to adhere to hand hygiene. Our findings can now inform hand hygiene interventions that are tailored to the needs of HCPs in the neonatology wards, optimizing the use of time and personnel resources. A study comparing the determinants of hand hygiene adherence in both adult ICUs and NICUs is needed to confirm and further elucidate the hypothesized differences between these settings.