Interaction between overconfidence effects and training formats in nurses’ education in hand hygiene

Hand hygiene is the cornerstone of infection prevention and is widely acknowledged as the most efficient and cost-effective intervention in reducing hospital-acquired infections. Furthermore, hand hygiene represents a substantial component of patient safety [1]. Hence, training programs in hand hygiene are widely integrated into the curricula of physicians, nurses, and other health care professionals (HCPs). Beyond the procedural aspect, the competency to “speak up” is vital in fostering a secure environment for both HCPs and patients, particularly in instances where hand hygiene is omitted or performed incorrectly [2]. Several studies have demonstrated potential for improvement of both hand hygiene practices and speaking up [2‐5].

Since 2018, our workings group and other authors have identified overconfidence effects (OCE) in hand hygiene practices and feedback reception skills among HCPs across different professions and workplace settings within local and national populations [6‐10]. OCE [11] are also incorrectly referred to as “flawed self-assessment” or the Dunning-Kruger-effect in the popular science [12]. OCE are considered inherent to human nature and manifest in various social contexts, such as economy [13] criminollogy [14], and education [15]. While many scientists interpret the Dunning-Kruger-effect as statistical bias [16], only one study has described detectable physiological reactions associated with the Dunning-Kruger-effect [17]. In contrast to the Dunning-Kruger-effect OCE is considered a genuine phenomenon and can be categorized into three sub-effects: the overestimation effect, which is absolute (“I am better than objectively measured”), overplacement or better-than-average, a relative effect (“I am better than average”), and overprecision, the strong conviction of accuracy in self-assessment [11]. Collectively, these effects not only influence subjective self-assessment but also impede self-reflection and metacognition [18]. Given that traditional learning formats often fail to reduce OCE [19] interactive or constructive learning approaches are deemed more suitable for addressing OCE and stimulating metacognition (the “thinking about thinking”). Metacognition in education, initially conceptualized as the capacity to observe, evaluate, and strategize one’s own learning process [20], hinges on self-reflection. It consists of two key components: metacognitive knowledge and metacognitive control [21]. Knowledge includes understanding information processing and learning tasks, while control regulates cognitive processes for effective task completion [22]. Feedback plays a crucial role in promoting metacognition by enhancing self-reflection and awareness of one’s learning process. When individuals receive feedback on their performance, they are usually prompted to evaluate their understanding, identify areas for improvement, and adjust their strategies accordingly. This reflective process enhances metacognitive skills, such as monitoring and regulating cognitive processes, ultimately leading to more effective learning outcomes [23, 24]. To provide a structured approach to understanding various learning formats in medical education the Interactive-Constructive-Active-Passive (ICAP) framework offers a valuable lens. This framework delineates between four distinct formats: passive (e.g. passive absorption of the learning content), active (e.g. presentations with active components like writing, answering questions), constructive (e.g. reflection, written homework, development of mind maps and completing tasks in the classroom), and interactive learning formats (e.g. learning in interacting groups). The model offers a detailed taxonomy of cognitive engagement modes of the learners crafted by Chi and Wylie (2014) [25]. This model enables educators to discern how students interact with instructional tasks, ranging from passive listening to active knowledge co-construction. By elucidating these behaviors, the framework facilitates the design of more effective pedagogical approaches within medical education contexts [25, 26].

In addition to findings of two prior studies on hand hygiene conducted by Lengerke and colleagues [10], our previous research involved a cohort of more than 1000 medical students from various German universities. Our study revealed that these effects are consistent across all universities and manifest from the early stages of medical education. Furthermore, we could show that medical students assessed themselves to have superior hand hygiene practices compared to postgraduate supervisors and nurses [27]. This phenomenon can be attributed to the clinical tribalism, wherein the sense of optimal hand hygiene compliance extends beyond individual behavior to encompass group identification (“we against them”), akin to an “in-group bias” [28].

These findings raise the question of to what extent OCE may impair the quality of hand hygiene training of undergraduate and postgraduate HCPs, such as nurses or specialized medical assistants. They partially share training, fostering interprofessional education (learning together and from each other), rather than solely multi-professional (learning together) education [29].

During the study period in Germany, three main educational courses were available for non-academic HCPs, each spanning three years: nurses (“Gesundheits- und Krankenpfleger: in”, GKP) including general, pediatric, and geriatric nurses; anesthesia technical assistants (“Anästhesietechnische: r Assistent: in”, ATA); and surgical assistants (“Operationstechnischer: r Assistent: in”, OTA). Additionally, qualified nurses had the opportunity to pursue a specialized postgraduate training in anesthesiology and critical care (“Weiterbildung Intensivpflege und Anästhesie”). The nursing and ATA/OTA training curriculum in Germany adheres to government guidelines. Theoretical lessons take place at schools for health professions authorized by the state, while practical training takes place within the different specialized hospital departments. Certain courses currently provide the opportunity to complete the training at universities. The schools develop a curriculum based on the recommendations of the framework curriculum. Hygiene and infection prevention are important aspects of the training content. During learning sessions, topics such as hygienic hand disinfection, infection sources, transmission routes, nosocomial infections, cleaning, disinfection and sterilization processes, and the legal basis for infection protection are comprehensively discussed. However, there is no predetermined number of teaching hours allocated to hygiene education.

When designing lessons, the focus lies on attaining professional competence, and the learning sessions should be designed to be action-oriented and methodically varied. However, it remains challenging to align teaching with the competence goals. The question arises as to whether classic teaching methods outweigh action-oriented teaching, thereby potentially failing to adequately facilitate competence acquisition. With regard to professional competence, the teaching design should be counter-characterized by individual appropriation of the learning content, discursive discussion, independent control and reflection of the learning process, and learning actions within a protected framework [30].

The objective of this study was to clarify whether OCEs in hand hygiene, speaking up, and feedback reception skill manifest among German GKP, ATA, and OTA, similar to findings observed in preceding studies involving postgraduate professionals and medical students.

We hypothesized that:

Prior to the COVID-19 pandemic, we conducted a cross-sectional online questionnaire at 25 nursing schools in Germany that offer the GKP, ATA and OTA programs from July to September 2019. The study took place following the investigation of medical students [27]. The study protocol was approved by the ethics committee of the physician association Baden Wurttemberg (Landesaerztekammer Baden-Wuerttemberg). We used the questionnaire platform “umfrageonline.com” by Enuvo GmbH, Zurich, Switzerland. The English version of the questionnaire is provided in Supplement A.

In total, approximately 3000 students (1800 GKP, 750 ATA and 750 OTA) across 25 medical schools were contacted, resulting in 196 responses. The inclusion criteria necessitated the current registration as a GKP, ATA, or OTA student. In Germany, a contractual agreement between students and the school is obligatory due to the practical nature of the work involved. Students have the option to commence nursing school and formalize their commitment after completing the 10th grade at the age of 16. Alternatively, if they leave school after the 9th grade at the age of 15, they must undergo training as a nursing assistant, albeit without registration as GKP, ATA, or OTA students and after that with 16 begin their nursing training. Consequently, participants had to be at least 16 years old to meet the criteria to be registered as nursing students and to take part in this study [31]. There were no other exclusion criteria. Participants voluntarily took part in the study and received information through an invitation letter from their respective faculties along with a pre-briefing provided within the questionnaire.

General data included the educational progress in years, gender, age, previous vocational education, year of exposure to the first training in hand hygiene and skills lab experience. These variables were followed by the SATIS-4D questionnaire [27] using a 5-point Likert scale (five (completely agree), four (moderately agree), three (partially agree), two (agree somewhat), one (completely disagree)).

The bias for convenience sampling was addressed according to Ball [32], as we mainly expected selection bias.

We aimed to reach at least 100 nursing students. This objective could be achieved by repetitive communication and letters to faculties of nursing schools and nationwide organizations and associations.

Self-reported hand hygiene compliance was assessed in 12 questions. The participants were asked to specify how often they perform hand disinfection in nine indicated moments typically encountered in daily routine (one World Health Organisation (WHO) indication 3, two WHO 1, 2, 4, and 5 respectively) in accordance with the WHO Hand Hygiene guidelines [1, 33]: “In the following situations I conduct hygienic hand disinfection”:

Additionally, we added three distractor situations not explicitly indicated by the 5 WHO Hand Hygiene guidelines (e.g., “after preparing sterile IV medication”). Hence, these items tested factual and situational knowledge and the self-assessment. Higher ratings pointed to higher occurrence: (1) never (< 1%), (2) seldom (1–25%), (3) sometimes (25–75%), (4) commonly (75–99%), and (5) always (> 99%). In addition, we computed separate index scores for indicated situations (α = 0.81, ω = 0.81) and non-indicated situations (α = 0.71, ω = 0.76).

Patient safety was assessed using two separate items to measure the self-estimated degree of patient safety. The substantial harm to patients was estimated as: “The credible maximum effect of omitted hand hygiene is insignificant – (1) without consequence, (2) minor – without any long-lasting effect, (3) severe – with longer hospital stay), (4) critical – with long-lasting effect, and (5) lethal “. Furthermore, we tested the estimation of the occurrence of these events: “How often is a patient harmed in your educational environment? (1) Uncommon (once > 3 years), (2) seldom (once every 3 years), (3) moderate (once a year), (4) often (once every 3 months), (5) very often (once a month)". The items are based on the ISO 31000 scales in risk management.

Data analysis was conducted by IBM SPSS Software Version 29 (IBM SPSS Statistics for Windows, Armonk, IBM, New York, NY, USA). After inspecting reliability indices for ICAP Subscales and for OCE dimensions, we tested the first set of hypotheses using one-way ANOVA with bootstrapping (Biased-corrected and accelerated (BCa), based on 1,000 bootstrap samples). Using two-step cluster analysis, we tested the second set of hypotheses. We included the self-assessment of competencies and the instrument measuring the self-reported hand hygiene protocol adherence, i.e., the frequency of hand disinfection in indicated situations following the WHO guidelines. The discrepancies among the identified clusters between self-assessment, self-reported frequency of indicated as well not indicated situations were compared. The clusters with high self-assessment and poor self-reported compliance in indicated situations were considered as overestimated. The third hypothesis was tested using the macro “Process” by Hayes, 2017 (model 1) [34]. The items belonging to one of the ICAP dimensions served as a focal predictor and overplacement (difference between own competencies and the competencies of fellow students) served as the criterion variable. Feedback and metacognition were each considered as moderator variables in the separate regression equation. For testing moderation effects, we applied the BCa Method based on 1,000 bootstrap samples for our calculation of moderation effects via multiple linear regression. Additionally, to obtain data points for visualizing the effects, we employed the previously mentioned Model 1 of the “Process” macro, which employs bias-corrected bootstrapping based on 5,000 bootstrap samples. Furthermore, we mean-centered the variables to avoid multicollinearity, especially between the predictor, moderator, and interaction term, which is a common problem in the moderation analysis.