Advertisement

Differences in observed and self-reported compliance with ‘Five Moments for Hand Hygiene’ as a function of the empathy of healthcare workers

Open AccessPublished:July 11, 2022DOI:https://doi.org/10.1016/j.jhin.2022.07.008

      Summary

      Background

      Hand hygiene at critical time-points (as established by the World Health Organization's model ‘Five Moments for Hand Hygiene’) remains the leading measure for minimizing the risk of healthcare-associated infections. While many interventions have been tested to improve hand hygiene compliance (HHC) of healthcare workers (HCWs), little is known about the relationship between HHC and empathy of HCWs.

      Aim

      To investigate the relationship between moment-specific HHC rates and empathy of HCWs at both individual and ward levels.

      Methods

      HHC data were collected via observation and self-report, and empathy levels were measured using an established questionnaire. The survey was conducted on 38 wards of three tertiary care hospitals in Germany. Observation data were obtained via in-house observations conducted ≤8 months before or after the survey.

      Findings

      Evidence for the expected correlation between empathy of HCWs and moment-specific HHC was found for both observed HHC (Moment 1: r=0.483, P=0.031; Moment 2: r=588, P=0.006) and self-reported HHC (Moment 1: r=0.093, P=0.092; Moment 2: r=0.145, P=0.008). In analyses of variance, the critical interaction effect between empathy (i.e. lower vs higher empathy) and designated time-point of hand hygiene (i.e. before vs after reference task) was also significant.

      Conclusion

      Empathy of HCWs should be considered as an important factor in explaining differences between moment-specific HHC rates. In consequence, empathy comes into focus not only as a crucial factor for high-quality patient care, but also as an important contributor to improving HHC.

      Keywords

      Introduction

      Hand hygiene compliance (HHC) of healthcare workers (HCWs) at five designated moments (World Health Organization's ‘Five Moments for Hand Hygiene’) using alcohol-based hand rub (ABHR) is considered one of the most effective infection control and prevention (IPC) measures [
      KRINKO
      Händehygiene in Einrichtungen des Gesundheitswesens: Empfehlung der Kommission für Krankenhaushygiene und Infektionsprävention (KRINKO) beim Robert Koch-Institut (RKI).
      ,
      World Health Organization
      WHO guidelines on hand hygiene in health care: first global patient safety challenge: clean care is safer care.
      ]. Extensive observation data from routine in-house surveillance within the German national clean hands campaign [‘Aktion Saubere Hände’ (ASH)] present a promising picture indicating a general upwards trend in HHC over the last years [
      • Wetzker W.
      • Bunte-Schönberger K.
      • Walter J.
      • Pilarski G.
      • Gastmeier P.
      • Reichardt Ch
      Compliance with hand hygiene: reference data from the national hand hygiene campaign in Germany.
      ,
      • Kramer T.S.
      • Bunte K.
      • Schröder C.
      • Behnke M.
      • Clausmeyer J.
      • Reichardt C.
      • et al.
      No increase in compliance before aseptic procedures in German hospitals. A longitudinal study with data from the national surveillance system over four years.
      ]. Taking a closer look at moment-specific HHC reveals, however, systematic differences between moments. HHC with ‘before moments’ (i.e. Moments 1 and 2) typically seems to be lower than HHC with ‘after moments’ (i.e. Moment 3–5, see Figure 1). This pattern is evident in both large-scale data, obtained via institutionalized in-house HHC observations within the German ASH framework [
      • Wetzker W.
      • Bunte-Schönberger K.
      • Walter J.
      • Pilarski G.
      • Gastmeier P.
      • Reichardt Ch
      Compliance with hand hygiene: reference data from the national hand hygiene campaign in Germany.
      ,
      • Kramer T.S.
      • Bunte K.
      • Schröder C.
      • Behnke M.
      • Clausmeyer J.
      • Reichardt C.
      • et al.
      No increase in compliance before aseptic procedures in German hospitals. A longitudinal study with data from the national surveillance system over four years.
      ], and research data (see [
      • Erasmus V.
      • Daha T.J.
      • Brug H.
      • Richardus J.H.
      • Behrendt M.D.
      • Vos M.C.
      • et al.
      Systematic review of studies on compliance with hand hygiene guidelines in hospital care.
      ,
      • Kingston L.
      • O’Connell N.H.
      • Dunne C.P.
      Hand hygiene-related clinical trials reported since 2010: a systematic review.
      ,
      • Lydon S.
      • Power M.
      • McSharry J.
      • Byrne M.
      • Madden C.
      • Squires J.E.
      • et al.
      Interventions to improve hand hygiene compliance in the ICU: a systematic review.
      ] for systematic reviews, see [
      • Scheithauer S.
      • Haefner H.
      • Schwanz T.
      • Schulze-Steinen H.
      • Schiefer J.
      • Koch A.
      • et al.
      Compliance with hand hygiene on surgical, medical, and neurologic intensive care units: direct observation versus calculated disinfectant usage.
      ] for a 96-h observation in three intensive care units, and see [
      • Jeanes A.
      • Coen P.G.
      • Wilson A.P.
      • Drey N.S.
      • Gould D.J.
      Collecting the data but missing the point: validity of hand hygiene audit data.
      ] for a comparison of routine in-house data and research data). Considering HHC rates separately for each moment provides not only detailed information for performance feedback, and also enables differentiated analysis of potential predictors, which may differ between the five moments. Considering the empathy of HCWs may help to explain the consistent finding of lower HHC with ‘before moments’ compared with ‘after moments’.
      Figure 1
      Figure 1Overview of the World Health Organization's ‘Five Moments for Hand Hygiene’ in relation to the two dimensions of analysis: designated time-point (before vs after reference task) and type of reference task (reference task related to approaching/leaving the patient vs reference task related to critical sites).
      Psychological research has shown that empathy increases pro-social behaviour (towards specific persons [
      • Batson C.D.
      • Chang J.
      • Orr R.
      • Rowland J.
      Empathy, attitudes, and action: can feeling for a member of a stigmatized group motivate one to help the group?.
      ,
      • Batson C.D.
      • Eklund J.H.
      • Chermok V.L.
      • Hoyt J.L.
      • Ortiz B.G.
      An additional antecedent of empathic concern: valuing the welfare of the person in need.
      ,
      • Sassenrath C.
      • Pfattheicher S.
      • Keller J.
      I might ease your pain, but only if you’re sad: the impact of the empathized emotion in the empathy-helping association.
      ] as well as towards unknown individuals or abstract entities [
      • Nook E.C.
      • Ong D.C.
      • Morelli S.A.
      • Mitchell J.P.
      • Zaki J.
      Prosocial conformity: prosocial norms generalize across behavior and empathy.
      ,
      • Pfattheicher S.
      • Sassenrath C.
      • Schindler S.
      Feelings for the suffering of others and the environment: compassion fosters proenvironmental tendencies.
      ]). Emotional empathy can be defined as an emotional reaction directed towards others' well-being which involves connecting affectively with another person [
      • Galinsky A.D.
      • Maddux W.W.
      • Gilin D.
      • White J.B.
      Why it pays to get inside the head of your opponent: the differential effects of perspective taking and empathy in negotiations.
      ]. As a consequence of this affective connection, valuing the other person's well-being increases, which, ultimately, fosters motivation to act in the other person's best interests [
      • Batson C.D.
      • Chang J.
      • Orr R.
      • Rowland J.
      Empathy, attitudes, and action: can feeling for a member of a stigmatized group motivate one to help the group?.
      ,
      • Batson C.D.
      • Eklund J.H.
      • Chermok V.L.
      • Hoyt J.L.
      • Ortiz B.G.
      An additional antecedent of empathic concern: valuing the welfare of the person in need.
      ]. This study on the role of empathy in behavioural outcomes can be linked to the different beneficiaries of HHC for ‘before moments’ and ‘after moments’. ‘After moments’ focus mainly on the HCWs' own health and safety (in addition to preventing environment contamination), whereas the rationale for ‘before moments’ lies in the health and safety of patients [
      • Sax H.
      • Allegranzi B.
      • Uçkay I.
      • Larson E.
      • Boyce J.
      • Pittet D.
      ‘My five moments for hand hygiene’: a user-centred design approach to understand, train, monitor and report hand hygiene.
      ].
      Empirical evidence for other-oriented motivation in the context of the hand hygiene behaviour of HCWs was provided in a set of two studies that compared signs emphasizing personal (i.e. HCWs') safety with signs emphasizing patient safety [
      • Grant A.M.
      • Hofmann D.A.
      It’s not all about me: motivating hand hygiene among health care professionals by focusing on patients.
      ]. The two experimental before–after trials demonstrated significant increases in the use of ABHR and observed HHC following the patient safety intervention, while no significant changes were documented in the personal safety intervention. Similarly, another study investigated the effect of empathy-activating photographs, and found a significant increase in electronically counted dispenser activations between baseline and the intervention phase on the ward with empathy-activating materials, while a control ward experienced a decrease in dispenser activations [
      • Sassenrath C.
      • Diefenbacher S.
      • Siegel A.
      • Keller J.
      A person-oriented approach to hand hygiene behaviour: emotional empathy fosters hand hygiene practice.
      ].
      However, these studies looked at hand hygiene behaviour overall, irrespective of the moment of hand hygiene. Given the mechanisms by which empathy influences human behaviour, namely increased relevance of the well-being of others and, thus, increased motivation to act accordingly, the empathy of HCWs should not only play a role in hand hygiene behaviour overall, but should be particularly relevant in HHC with ‘before moments’, which are linked more directly with patient outcomes [
      • Sax H.
      • Allegranzi B.
      • Uçkay I.
      • Larson E.
      • Boyce J.
      • Pittet D.
      ‘My five moments for hand hygiene’: a user-centred design approach to understand, train, monitor and report hand hygiene.
      ].

      Methods

      This article addresses the question whether differences in the emotional empathy of HCWs can help to explain differences in moment-specific HHC. As elaborated above, ‘before moments’ and ‘after moments’ differ in their conceptual proximity to patient outcomes, which renders the empathy of HCWs a relevant construct in explaining such differences. Thus, the designated time-point with respect to the reference task (i.e. use of ABHR indicated before or after a reference task) was used as the key dimension in the analyses. To achieve comparability and facilitate a meaningful interpretation of differences in HHC between ‘before moments’ and ‘after moments’, a second dimension of analysis was also used based on the reference task itself. Moment 1 (i.e. before touching a patient) is most comparable with Moment 4 (i.e. after touching a patient), and Moment 2 (i.e. before clean/aseptic procedure) is most comparable with Moment 3 (i.e. after body fluid exposure risk). This second dimension was named ‘type of reference task’, differentiating between reference tasks related to approaching/leaving the patient (zone) and reference tasks related to critical sites (i.e. clean sites and body fluid sites). Figure 1 provides an overview of the World Health Organization's ‘Five Moments for Hand Hygiene’ along with their grouping into the two dimensions of analysis.
      First, replicating previous findings, HHC was expected to be higher with ‘after moments’ compared with ‘before moments’. Second, replicating and extending previous findings, a positive correlation was expected between HHC and empathy (i.e. higher HHC in more empathic HCWs and vice versa). In addition, two meaningful comparisons of ‘before moments’ vs ‘after moments’ were addressed: (1) before touching a patient vs after touching a patient; and (2) before a clean/aseptic procedure vs after body fluid exposure risk. This leads to the third hypothesis: for both comparisons (i.e. independent of type of reference task), the difference in HHC between ‘before moments’ and ‘after moments’ was expected to be less pronounced for more empathic HCWs compared with less empathic HCWs.
      By introducing empathy as a potential predictor of HHC, a behavioural outcome, which is typically assessed at group level (i.e. HHC score for wards or units based on the individual behaviour observed in the sampled HCWs from each ward), is combined with a personality trait, typically assessed via introspectively informed self-reports. Two strategies were used to facilitate analyses within one level (i.e. group or individual level): (1) combining observed ward-level HHC with an approximated ward empathy score; and (2) combining the original empathy score of each HCW with a self-reported score of their typical HHC.

      Samples

      In March 2017 and March 2018, nurses and physicians from a total of 38 wards from three tertiary care medical centres with in-house observation data were invited to participate in this study. After providing informed consent, 329 eligible HCWs (82% nurses, 18% physicians) from 31 different wards who had worked in their respective professions for an average of 13.2 years [standard deviation (SD) 10.8, range 1–49 years] filled in a questionnaire including self-reported measures of HHC [
      • Diefenbacher S.
      • Pfattheicher S.
      • Keller J.
      On the role of habit in self-reported and observed hand hygiene behavior.
      ] and empathy [
      • Davis M.H.
      Measuring individual differences in empathy: evidence for a multidimensional approach.
      ,
      • Paulus C.
      Der Saarbrücker Persönlichkeitsfragebogen SPF (IRI) zur Messung von Empathie. Psychometrische Evaluation der deutschen Version des Interpersonal Reactivity Index.
      ]. Out of 349 participating HCWs, individuals were excluded if their profession was neither nurse nor physician (N=2), if they had difficulties in understanding the questionnaire (N=5), or if they did not complete all the measures relevant to the analyses in full (N=20). Two individuals withdrew from the study.
      The procedure regarding the ward data (observed HHC and ward empathy scores, see description below) resulted in the inclusion of 20 wards; 12 of these were intensive care units or intermediate care units, and eight were general wards. Regarding specialities, the distribution was as follows: seven medical wards, nine surgical wards, and four neonatological or paediatric wards. Overall, 4328 hand hygiene opportunities (HHOs) were observed on the 20 included wards. Ward averages for empathy were calculated using the data of the 275 HCWs working on the 20 selected wards.

      Observed hand hygiene compliance – ward level

      Direct observation of HHC was performed by local IPC staff as part of routine in-house surveillance according to the guidelines of the national nosocomial infection surveillance system (Krankenhaus-Infektions-Surveillance-System) [
      • Krüger I.
      Anleitung zur Beobachtung der Händedesinfektion (Bestimmung der Compliance).
      ]. Overall HHC rates were calculated by dividing the number of uses of ABHR in response to the five moments by all HHOs. For moment-specific HHC, HHOs and indicated use of ABHR corresponding to the moment in question were used. During work routines, situations may arise that combine more than one of the five moments into one hand hygiene opportunity. For example, an HCW completes their work with one patient, performs hand hygiene adequately (according to Moment 4 or 5), then moves to the next patient to start their work there. In this case, use of ABHR after the first patient also covers Moment 1 (or 2) at the second patient. Therefore, no additional ABHR is required (provided that nothing is touched in between), so observers would record adequate hand hygiene for the ‘after moment’ as well as for the ‘before moment’.
      The resulting HHC rates are expected to represent the current status quo of a ward, and do not provide information about HHC at the individual level. This study included wards that had observed at least 150 opportunities within a period ≤8 months prior to or after the survey. For wards with multiple observations, the one with the shortest period between observation and survey was used.

      Self-reported hand hygiene compliance – HCW level

      Not all HCWs on a specific ward can be expected to have the same HHC; they differ in their individual HHC from the ward score and between each other. Such differences are invisible when using aggregated ward data. An easy way of acquiring data on the individual level is the use of survey methods [
      • Diefenbacher S.
      • Siegel A.
      • Keller J.
      Verfahren zur Erfassung des Händehygieneverhaltens-Eine methodische Betrachtung aus verhaltenswissenschaftlicher Perspektive [Methods for measuring hand hygiene behavior – a methodological examination from a behavioral scientific perspective].
      ]. Thus, HCWs' self-reported HHC was also assessed. As part of a questionnaire, HCWs indicated on a scale of 0–100 how often they disinfected their hands when necessary according to hand hygiene recommendations [
      • Diefenbacher S.
      • Pfattheicher S.
      • Keller J.
      On the role of habit in self-reported and observed hand hygiene behavior.
      ], thus providing their responses within the same theoretical range as HHC observation data (0–100%). HCWs estimated their overall and moment-specific HHC.

      Emotional empathy – individual level

      Emotional empathy was assessed using a German version [
      • Paulus C.
      Der Saarbrücker Persönlichkeitsfragebogen SPF (IRI) zur Messung von Empathie. Psychometrische Evaluation der deutschen Version des Interpersonal Reactivity Index.
      ] of the empathic concern subscale of the Interpersonal Reactivity Index [
      • Davis M.H.
      Measuring individual differences in empathy: evidence for a multidimensional approach.
      ], comprising four items (‘I often have tender, concerned feelings for people less fortunate than me’, ‘When I see someone being taken advantage of, I feel kind of protective towards them’, ‘I am often quite touched by things that I see happen’, ‘I would describe myself as a pretty soft-hearted person’). Items were measured on a seven-point Likert scale. Reliability was good (Cronbach's α=0.749), so the empathy of HCWs was calculated by averaging all items. Individual HCW responses were then used to calculate a ward score (see next paragraph). Two equal-sized groups of HCWs were created via median split: ‘lower empathy’ and ‘higher empathy’.

      Emotional empathy – ward level

      The ward empathy score was calculated by averaging individual responses of all participating HCWs on a given ward. Wards with fewer than five responses were excluded to reduce the risk of bias in these ward empathy approximations. Scores were then matched with observational HHC data. Two equal-sized groups of wards were created via median split: ‘lower empathy’ and ‘higher empathy’.

      Statistical analysis

      Bivariate correlations were calculated via Pearson's r and visualized for the four correlations of interest (i.e. moment-specific HHC rates for Moments 1–4 and empathy) using scatterplots with linear regression lines and 95% confidence intervals (CI). Two separate analyses of variance with repeated measures were run to test whether empathy, designated time-point and type of reference task are good predictors of moment-specific HHC; one with directly observed HHC (i.e. wards as unit of analysis) and one with self-reported HHC (i.e. HCWs as unit of analysis) as the dependent variable. The independent variables of these analyses were empathy (lower vs higher) as between-group factor, and designated time-point (before vs after) and type of reference task (reference task related to approaching/leaving the patient vs related to critical sites) as repeated measures. By using median splits to create a two-group factor of empathy, the variance in empathy was reduced artificially. Therefore, additional analyses were performed with the continuous measure. The results can be found in Table S2 (see online supplementary material). All analyses were performed using SPSS Version 28.0.1.0 (IBM Corp., Armonk, NY, USA).

      Results

      Descriptive results and bivariate correlations of observed and self-reported HHC

      Objectively observed and self-reported HHC rates were similar, except for Moments 2 and 3. For these two moments, HHC self-reported by HCWs was considerably higher than HHC observed on wards (Moment 2: 59.7% observed HHC vs 86.2% self-reported HHC; Moment 3: 80.5% observed HHC vs 93.5% self-reported HHC). Descriptive results of both ward level and HCW level data for all variables are shown in Table S1 (see online supplementary material).
      All moment-specific HHC rates were strongly and significantly correlated with the overall HHC rate. These correlations ranged between 0.617 and 0.902 for directly observed HHC, and between 0.369 and 0.607 for self-reported HHC. Table S1 (see online supplementary material) shows all bivariate correlations.

      HCW empathy and HHC with ‘before moments’ and ‘after moments’

      First, bivariate correlations were examined. Importantly, significant or marginally significant correlations between the empathy of HCWs and HHC were found for both Moment 1 (observed HHC: r=0.483, P=0.031; self-reported HHC: r=0.093, P=0.092) and Moment 2 (observed HHC: r=588, P=0.006; self-reported HHC: r=0.145, P=0.008). Table S1 (see online supplementary material) shows all bivariate correlations. Associations between the four critical moment-specific HHC rates and empathy are shown in Figure 2.
      Figure 2
      Figure 2Scatterplots of observed hand hygiene compliance (HHC) (dark grey) and self-reported HHC (light grey) and empathy by designated time-point (before vs after reference task) and type of reference task (reference task related to approaching/leaving the patient vs related to critical sites), all with linear regression lines (continuous lines) and 95% confidence intervals (dotted lines). M, moment.
      Next, two analyses of variance were run to test the hypotheses. An overview of all test statistics and relevant descriptive statistics is presented in Table I. As expected, HHC with ‘before moments’ was lower compared with HHC with ‘after moments’ for both observed and self-reported HHC. Likewise, the expected difference between lower and higher empathy of HCWs was found, with HHC being higher among HCWs with higher empathy compared with lower empathy. Furthermore, the interaction effect between empathy and designated time-point was significant for both observed and self-reported HHC, qualifying these two main effects. Specifically, HHC was highest with ‘before moments’ on wards and HCWs with higher empathy. For wards and HCWs with lower empathy, ‘before moments’ had the lowest HHC, while ‘after-moment’ HHC in the lower-empathy group was similar to HHC in the higher-empathy group. For observed HHC, this two-way interaction was further qualified by type of reference task. That is, the difference between ‘before-moment’ and ‘after-moment’ HHC varied depending on empathy and type of reference task, with greater differences for tasks related to critical sites (i.e. Moments 2 and 3) than for tasks related to approaching/leaving the patient (i.e. Moments 1 and 4).
      Table IResults of two analyses of variance on moment-specific hand hygiene compliance (HHC) and relevant descriptive results from ward-level data and healthcare worker (HCW)-level data
      Ward-level dataHCW-level data
      F
      Degrees of freedom (df1,df2) of analyses with ward-level data: 1,18.
      Ppart.η2F
      Degrees of freedom (df1,df2) of analyses with HCW-level data: 1327.
      Ppart.η2
      Designated time-point (dTP)29.804<0.0010.623170.446<0.0010.343
      Empathy (Emp)11.5170.0030.3905.3140.0220.016
      Type of reference task (Task)12.9020.0020.418131.450<0.0010.287
      dTP x Emp6.7230.0180.2727.0270.0080.021
      dTP x Task15.9330.0010.47016.539<0.0010.048
      Emp x Task7.9390.0110.306<10.990<0.001
      dTP x Emp x Task6.8100.0180.274<10.696<0.001
      Relevant descriptive results
      95% CI95% CI
      MLBUBMLBUB
      HHC ‘before moments’ (Moments 1 and 2)67.5%58.876.380.5%78.688.7
      HHC ‘after moments’ (Moments 3 and 4)82.1%77.287.190.1%88.791.6
      CI, confidence interval; M, mean; LB, lower boundary; UB, upper boundary.
      Independent variables: empathy (Emp: lower vs higher) as between-subject factor, designated time-point with respect to reference task (dTP: before vs after), and type of reference task (task: reference task related to approaching/leaving the patient vs related to critical sites) as within-subject factors.
      a Degrees of freedom (df1,df2) of analyses with ward-level data: 1,18.
      b Degrees of freedom (df1,df2) of analyses with HCW-level data: 1327.

      Discussion

      This article explores the role of emotional empathy of HCWs in explaining differences between moment-specific HHC rates, specifically between ‘before moments’ and ‘after moments’. Four of the five moments for hand hygiene were grouped according to two dimensions: designated time-point (i.e. before or after a reference task) and type of reference task (i.e. reference tasks related to approaching/leaving the patient or related to critical sites). In doing so, corresponding ‘before moments’ and ‘after moments’ could be compared. Importantly, ‘before moments’ and ‘after moments’ differ in their conceptual proximity to patient outcomes. HHC of HCWs with ‘before moments’ is especially beneficial for patients by preventing colonization (Moment 1) and infection in patients (Moment 2); HHC with ‘after moments’, in contrast, besides preventing contamination of subsequently touched surfaces, is mainly beneficial for the HCWs themselves [
      • Sax H.
      • Allegranzi B.
      • Uçkay I.
      • Larson E.
      • Boyce J.
      • Pittet D.
      ‘My five moments for hand hygiene’: a user-centred design approach to understand, train, monitor and report hand hygiene.
      ]. This renders the empathy of HCWs a relevant construct in explaining differences in moment-specific HHC rates and, indeed, the present results could link such differences to the empathy of HCWs. First, significant correlations between ward empathy level and observed HHC were found for the two ‘before moments’ (Moments 1 and 2); likewise, these correlations were significant or marginally significant for self-reported HHC but much smaller. Second, in analyses of variance, the critical interaction between empathy and designated time-point (i.e. before vs after reference task) was significant, pointing towards the hypothesized role of the empathy of HCWs in ‘before-moment’ HHC. The difference between ‘before moments’ and ‘after moments’ was less pronounced for more empathic HCWs or wards, and larger for less empathic HCWs or wards, with lower ‘before-moment’ HCC compared with ‘after-moment’ HHC. Previous findings consistently revealed that HHC with ‘before moments’ was lower than HHC with ‘after moments’ [
      • Erasmus V.
      • Daha T.J.
      • Brug H.
      • Richardus J.H.
      • Behrendt M.D.
      • Vos M.C.
      • et al.
      Systematic review of studies on compliance with hand hygiene guidelines in hospital care.
      ,
      • Kingston L.
      • O’Connell N.H.
      • Dunne C.P.
      Hand hygiene-related clinical trials reported since 2010: a systematic review.
      ,
      • Lydon S.
      • Power M.
      • McSharry J.
      • Byrne M.
      • Madden C.
      • Squires J.E.
      • et al.
      Interventions to improve hand hygiene compliance in the ICU: a systematic review.
      ]. The present study is no exception to this pattern, but provides a theoretically sound explanation based on the interpersonal orientation of HCWs. In addition, the observational data highlighted the relevance of the type of reference task for interpreting this expected effect: the empathy of HCWs mattered the most when comparing ‘before moments’ and ‘after moments’ related to critical sites (i.e. Moment 2 vs Moment 3). For Moment 2 – the moment that might be most critical for patient safety – problematically low HHC rates (i.e. <40%) were observed in a substantial number of the sampled wards, and the largest correlation between empathy and HHC was found for Moment 2. This points to the role of empathy-based strategies in HHC interventions in order to increase other-oriented motivations of HCWs. The analysis of self-reported HHC did not, however, replicate the three-way interaction, but also produced the highest correlation between empathy and HHC for Moment 2. Of note, the strength of the empathy–HHC association may be biased; for ward-level data, due to the limited realized range in ward empathy scores in the study sample; and for HCW-level data, due to the clear ceiling effect in the self-reported measure of HHC.
      The present results are in line with previous work [
      • Grant A.M.
      • Hofmann D.A.
      It’s not all about me: motivating hand hygiene among health care professionals by focusing on patients.
      ] providing evidence for other-oriented motivation in the context of the hand hygiene behaviour of HCWs by demonstrating increased hand hygiene performance when highlighting its use for patient safety (compared with the personal safety of HCWs). Importantly, other-oriented motivation and self-oriented motivation to perform hand hygiene should not be viewed as opposing poles but as two complementing motivational processes which can be addressed simultaneously in future interventions. In the before-mentioned study, baseline HHC was already high at approximately 80% on both wards [
      • Grant A.M.
      • Hofmann D.A.
      It’s not all about me: motivating hand hygiene among health care professionals by focusing on patients.
      ]. Emphasizing patient safety may have added other-oriented motivation to a seemingly more basic personal safety motivation. A personal-safety-first perspective would also explain the typical pattern for lower HHC with ‘before moments’ compared with ‘after moments’. From this perspective, fostering emotional empathy can be expected to attenuate this difference [
      • Kramer T.S.
      • Bunte K.
      • Schröder C.
      • Behnke M.
      • Clausmeyer J.
      • Reichardt C.
      • et al.
      No increase in compliance before aseptic procedures in German hospitals. A longitudinal study with data from the national surveillance system over four years.
      ,
      • Sassenrath C.
      • Diefenbacher S.
      • Siegel A.
      • Keller J.
      A person-oriented approach to hand hygiene behaviour: emotional empathy fosters hand hygiene practice.
      ]. Recent research further substantiated this perspective, showing a positive effect of highlighting consequences for others on the importance of ‘before-moment’ HHC and behavioural intentions in the light of the coronavirus disease 2019 pandemic [
      • Sassenrath C.
      • Diefenbacher S.
      • Kolbe V.
      • Niesalla H.
      • Keller J.
      The impact of activating an empathic focus during COVID19 on healthcare workers motivation for hand hygiene compliance in moments serving the protection of others: a randomized controlled trial study.
      ]. Taking into account that empathic reactions and empathic behaviour can depend on the neediness of the target of empathy [
      • Sassenrath C.
      • Pfattheicher S.
      • Keller J.
      I might ease your pain, but only if you’re sad: the impact of the empathized emotion in the empathy-helping association.
      ], the strength of the empathy–HHC association may vary with respect to specific patient profiles (e.g. children or neonates vs adults); working with specific patient profiles may also attract HCWs based on their empathic skills (e.g. geriatric or oncological patients vs general surgical patients). Independent of patient profiles, empathy training has been conducted successfully within the healthcare setting [
      • Winter R.
      • Issa E.
      • Roberts N.
      • Norman R.I.
      • Howick J.
      Assessing the effect of empathy-enhancing interventions in health education and training: a systematic review of randomised controlled trials.
      ]. In work promoting empathy in health care, empathy is frequently conceptualized with direct reference to patients (e.g. involving the ability to understand the patient's situation, perspective and feelings [
      • Winter R.
      • Issa E.
      • Roberts N.
      • Norman R.I.
      • Howick J.
      Assessing the effect of empathy-enhancing interventions in health education and training: a systematic review of randomised controlled trials.
      ]). In this study, emotional empathy was measured more broadly, referring to other individuals in general.
      This study provided a novel approach for analysing HHC by grouping four of the five moments for hand hygiene based on two dimensions – designated time-point and type of reference task – and using these dimensions as within factors in joint analyses. While the first dimension is rather straightforward, the latter seems more ambiguous. The two types of reference task were labelled according to the spatial distinctions provided by Sax et al. [
      • Sax H.
      • Allegranzi B.
      • Uçkay I.
      • Larson E.
      • Boyce J.
      • Pittet D.
      ‘My five moments for hand hygiene’: a user-centred design approach to understand, train, monitor and report hand hygiene.
      ], where the transition between healthcare zone and patient zone is at the core of Moments 1, 4 and 5 [i.e. reference tasks related to approaching/leaving the patient (zone)] and Moments 2 and 3 occur within the patient zone involving procedures at clean or body fluid sites, respectively (i.e. reference tasks related to critical sites). The meaning of these two types of reference task in the light of empathy and other motivational elements is, however, rather complex. First, the two types seem to differ in the severity of consequences for patients in the case of non-compliance. Moments 1, 4 and 5 mainly serve the purpose of preventing cross-contamination of patients, HCWs and/or the environment; the two moments involving critical sites are aimed directly at preventing infections [
      • Jeanes A.
      • Coen P.G.
      • Wilson A.P.
      • Drey N.S.
      • Gould D.J.
      Collecting the data but missing the point: validity of hand hygiene audit data.
      ]. Thus, Moments 1, 4 and 5 could stand for lower severity of consequences and Moments 2 and 3 for higher severity of consequences. Second, the complexity of the two types of reference task likely differs, with Moments 2 and 3 representing typically higher complexity compared with Moments 1, 4 and 5. Thus, higher severity of consequences may motivate higher HHC, while higher complexity may, at the same time, decrease HHC through both motivational and cognitive processes. Accordingly, disentangling the oppositional influences of severity of consequences and task complexity could be a focus of future research. In addition, evidence from psychological research about habit formation suggests that it might be cognitively easier to achieve automatization if the target behaviour (e.g. flossing) is to be performed after an event (e.g. tooth brushing) than before the event [
      • Judah G.
      • Gardner B.
      • Aunger R.
      Forming a flossing habit: An exploratory study of the psychological determinants of habit formation.
      ]. These results provide a cognitive explanation for differences in HHC with ‘before moments’ and ‘after moments’. This further highlights the need for more research about the psychological processes underlying HHC in order to optimize training and intervention programmes.
      A strength of this study is that HHC was assessed by two complementary methodological approaches: direct observation, resulting in ward-level HHC rates, and self-reports of HCWs, resulting in HCW-level HHC rates. By doing so, the individual behaviours of HCWs come into focus. This approach provides additional insights compared with the standard approach, looking at overall HHC of wards, which is blind to variation between HCWs [
      • Diefenbacher S.
      • Sassenrath C.
      • Tatzel J.
      • Keller J.
      Evaluating healthcare workers’ hand hygiene performance using first-person view video observation in a standardized patient-care scenario.
      ]. The present results reveal substantial variation in self-reported HHC between HCWs (SDs between 16.2 and 24.1). Complementing direct observation (gold standard in the field [
      World Health Organization
      WHO guidelines on hand hygiene in health care: first global patient safety challenge: clean care is safer care.
      ]) with comparable self-reports (i.e. applying the World Health Organization's ‘Five Moments for Hand Hygiene’ and providing a response scale from 0 to 100) is also informative, given that both methods underlie different mechanisms of bias [
      • Diefenbacher S.
      • Siegel A.
      • Keller J.
      Verfahren zur Erfassung des Händehygieneverhaltens-Eine methodische Betrachtung aus verhaltenswissenschaftlicher Perspektive [Methods for measuring hand hygiene behavior – a methodological examination from a behavioral scientific perspective].
      ,
      • Jeanes A.
      • Coen P.G.
      • Gould D.J.
      • Drey N.S.
      Validity of hand hygiene compliance measurement by observation: a systematic review.
      ], which can frequently (but not exclusively) lead to over-reporting in self-reports compared with observations [
      • Contzen N.
      • De Pasquale S.
      • Mosler H.-J.
      Over-reporting in handwashing self-reports: potential explanatory factors and alternative measurements.
      ]. The present results suggest over-reporting for some moments (i.e. Moments 2 and 3) but not for others (Moments 1 and 4). This is in line with the results of Lydon et al. [
      • Lydon S.
      • Greally C.
      • Tujjar O.
      • Reddy K.
      • Lambe K.
      • Madden C.
      • et al.
      Psychometric evaluation of a measure of factors influencing hand hygiene behaviour to inform intervention.
      ], who found some over-reporting for Moment 1 and strong over-reporting for Moments 2 and 3. However, it needs to be acknowledged that the clear ceiling effect in the present self-report measure of HHC reduces the validity of these results to some degree. Reasons for differential over-reporting can be motivational (e.g. due to self-deception and/or impression management [
      • Paulhus D.L.
      Self-deception and impression management in test responses.
      ] in the case of higher expected severity of consequences) or due to cognitive processes (i.e. HCWs have more difficulty remembering and/or recalling more complex situations, see also [
      • Contzen N.
      • De Pasquale S.
      • Mosler H.-J.
      Over-reporting in handwashing self-reports: potential explanatory factors and alternative measurements.
      ]). More research is needed to investigate these mechanisms more systematically.
      A clear limitation of this study is its cross-sectional approach. While the applied design can contribute to establishing the relevance of empathy with regard to HHC, it is not sufficient to establish a clear causal link. Future studies could apply an experimental design, preferably as a cluster-randomized controlled intervention study, to provide further insights. Another limitation of this study was that some wards had low samples of observations for Moment 2 (five wards) and/or Moment 3 (eight wards) with fewer than 20 HHOs. Although this is a widespread problem in HHC observations on general wards, future studies about moment-specific HHC should include larger samples of HHOs for these moments to corroborate the reported findings. Finally, as the observation data were collected as part of the routine in-house observation in the participating hospitals, the only information available about sampled HCWs was their professional group. Thus, no information about the mobility of HCWs among wards can be given. It was also not possible to sample the same HCWs for the questionnaire. In other words, empathy scores and self-reported HHC are not necessarily based on data from the same HCWs as those who provided the data for the observed HHC measure.
      In conclusion, this study provides the first evidence that the empathy of HCWs plays a differential role for HHC with ‘before moments’ and ‘after moments’, reflecting the other-oriented relevance of ‘before moments’ (i.e. patient safety). Hence, by enabling this differential analysis of potential predictors of HHC, the moment-specific perspective proved useful. The results can inform interventions specifically targeting ‘before moments’.

      Acknowledgements

      The study was investigator initiated. BODE Chemie GmbH, a company of the HARTMANN GROUP, partially funded the study, and additionally covered the costs of making this publication available in open access format. C. Plotzki and H. Niesalla are employees of BODE Chemie GmbH.

      Author contributions

      Conceptualization of the overall research project: JK, WSB and HN.
      Conceptualization of the manuscript: SD.
      Provision of observation data: WSB, AA and TSK.
      Coordination of survey data collection: SD., supported by WSB, AA and TSK at the respective partner hospitals.
      Data preparation and analysis of survey and observation data: SD.
      Interpretation of results: SD.
      Discussion of analyses and results: CP, WSB, AA, HN, CS, JK and S.G.
      First draft of the manuscript: SD and CP.
      Re-editing of the manuscript: SD.
      Critical feedback: CP, JK, CS, HN, PG, TSK, AA and SG.

      Conflict of interest statement

      The study was investigator initiated. BODE Chemie GmbH, a company of the HARTMANN GROUP, partially funded the study. C. Plotzki and H. Niesalla are employees of BODE Chemie GmbH.

      Funding sources

      BODE Chemie GmbH, a company of the Hartmann Group partially funded the study as part of a larger research agenda, and additionally covered the costs of making this publication available in open access format.

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      References

        • KRINKO
        Händehygiene in Einrichtungen des Gesundheitswesens: Empfehlung der Kommission für Krankenhaushygiene und Infektionsprävention (KRINKO) beim Robert Koch-Institut (RKI).
        Bundesgesundheitsblatt. 2016; 59 ([in German]): 1189-1220
        • World Health Organization
        WHO guidelines on hand hygiene in health care: first global patient safety challenge: clean care is safer care.
        WHO, Geneva2009
        • Wetzker W.
        • Bunte-Schönberger K.
        • Walter J.
        • Pilarski G.
        • Gastmeier P.
        • Reichardt Ch
        Compliance with hand hygiene: reference data from the national hand hygiene campaign in Germany.
        J Hosp Infect. 2016; 92: 328-331
        • Kramer T.S.
        • Bunte K.
        • Schröder C.
        • Behnke M.
        • Clausmeyer J.
        • Reichardt C.
        • et al.
        No increase in compliance before aseptic procedures in German hospitals. A longitudinal study with data from the national surveillance system over four years.
        J Hosp Infect. 2020; 106: 71-75
        • Erasmus V.
        • Daha T.J.
        • Brug H.
        • Richardus J.H.
        • Behrendt M.D.
        • Vos M.C.
        • et al.
        Systematic review of studies on compliance with hand hygiene guidelines in hospital care.
        Infect Control Hosp Epidemiol. 2010; 31: 283-294
        • Kingston L.
        • O’Connell N.H.
        • Dunne C.P.
        Hand hygiene-related clinical trials reported since 2010: a systematic review.
        J Hosp Infect. 2016; 92: 309-320
        • Lydon S.
        • Power M.
        • McSharry J.
        • Byrne M.
        • Madden C.
        • Squires J.E.
        • et al.
        Interventions to improve hand hygiene compliance in the ICU: a systematic review.
        Crit Care Med. 2017; 45: e1165-e1172
        • Scheithauer S.
        • Haefner H.
        • Schwanz T.
        • Schulze-Steinen H.
        • Schiefer J.
        • Koch A.
        • et al.
        Compliance with hand hygiene on surgical, medical, and neurologic intensive care units: direct observation versus calculated disinfectant usage.
        Am J Infect Control. 2009; 37: 835-841
        • Jeanes A.
        • Coen P.G.
        • Wilson A.P.
        • Drey N.S.
        • Gould D.J.
        Collecting the data but missing the point: validity of hand hygiene audit data.
        J Hosp Infect. 2015; 90: 156-162
        • Batson C.D.
        • Chang J.
        • Orr R.
        • Rowland J.
        Empathy, attitudes, and action: can feeling for a member of a stigmatized group motivate one to help the group?.
        Pers Soc Psychol Bull. 2002; 28: 1656-1666
        • Batson C.D.
        • Eklund J.H.
        • Chermok V.L.
        • Hoyt J.L.
        • Ortiz B.G.
        An additional antecedent of empathic concern: valuing the welfare of the person in need.
        J Pers Soc Psychol. 2007; 93: 65-74
        • Sassenrath C.
        • Pfattheicher S.
        • Keller J.
        I might ease your pain, but only if you’re sad: the impact of the empathized emotion in the empathy-helping association.
        Motiv Emot. 2017; 41: 96-106
        • Nook E.C.
        • Ong D.C.
        • Morelli S.A.
        • Mitchell J.P.
        • Zaki J.
        Prosocial conformity: prosocial norms generalize across behavior and empathy.
        Pers Soc Psychol Bull. 2016; 42: 1045-1062
        • Pfattheicher S.
        • Sassenrath C.
        • Schindler S.
        Feelings for the suffering of others and the environment: compassion fosters proenvironmental tendencies.
        Environ Behav. 2016; 48: 929-945
        • Galinsky A.D.
        • Maddux W.W.
        • Gilin D.
        • White J.B.
        Why it pays to get inside the head of your opponent: the differential effects of perspective taking and empathy in negotiations.
        Psychol Sci. 2008; 19: 378-384
        • Sax H.
        • Allegranzi B.
        • Uçkay I.
        • Larson E.
        • Boyce J.
        • Pittet D.
        ‘My five moments for hand hygiene’: a user-centred design approach to understand, train, monitor and report hand hygiene.
        J Hosp Infect. 2007; 67: 9-21
        • Grant A.M.
        • Hofmann D.A.
        It’s not all about me: motivating hand hygiene among health care professionals by focusing on patients.
        Psychol Sci. 2011; 22: 1494-1499
        • Sassenrath C.
        • Diefenbacher S.
        • Siegel A.
        • Keller J.
        A person-oriented approach to hand hygiene behaviour: emotional empathy fosters hand hygiene practice.
        Psychol Health. 2016; 31: 205-227
        • Diefenbacher S.
        • Pfattheicher S.
        • Keller J.
        On the role of habit in self-reported and observed hand hygiene behavior.
        Appl Psychol Health Well-Being. 2020; 12: 125-143
        • Davis M.H.
        Measuring individual differences in empathy: evidence for a multidimensional approach.
        J Pers Soc Psychol. 1983; 44: 113-126
        • Paulus C.
        Der Saarbrücker Persönlichkeitsfragebogen SPF (IRI) zur Messung von Empathie. Psychometrische Evaluation der deutschen Version des Interpersonal Reactivity Index.
        (Available at:) ([in German])
        • Krüger I.
        Anleitung zur Beobachtung der Händedesinfektion (Bestimmung der Compliance).
        (Available at:) ([in German])
        • Diefenbacher S.
        • Siegel A.
        • Keller J.
        Verfahren zur Erfassung des Händehygieneverhaltens-Eine methodische Betrachtung aus verhaltenswissenschaftlicher Perspektive [Methods for measuring hand hygiene behavior – a methodological examination from a behavioral scientific perspective].
        Hyg Med. 2016; 41 ([ in German]): D105-D119
        • Sassenrath C.
        • Diefenbacher S.
        • Kolbe V.
        • Niesalla H.
        • Keller J.
        The impact of activating an empathic focus during COVID19 on healthcare workers motivation for hand hygiene compliance in moments serving the protection of others: a randomized controlled trial study.
        J Public Health. 2022; https://doi.org/10.1007/s10389-022-01725-z
        • Winter R.
        • Issa E.
        • Roberts N.
        • Norman R.I.
        • Howick J.
        Assessing the effect of empathy-enhancing interventions in health education and training: a systematic review of randomised controlled trials.
        BMJ Open. 2020; 10e036471
        • Judah G.
        • Gardner B.
        • Aunger R.
        Forming a flossing habit: An exploratory study of the psychological determinants of habit formation.
        Br J Health Psychol. 2013; 18: 338-353
        • Diefenbacher S.
        • Sassenrath C.
        • Tatzel J.
        • Keller J.
        Evaluating healthcare workers’ hand hygiene performance using first-person view video observation in a standardized patient-care scenario.
        Am J Infect Control. 2020; 48: 496-502
        • Jeanes A.
        • Coen P.G.
        • Gould D.J.
        • Drey N.S.
        Validity of hand hygiene compliance measurement by observation: a systematic review.
        Am J Infect Control. 2019; 47: 313-322
        • Contzen N.
        • De Pasquale S.
        • Mosler H.-J.
        Over-reporting in handwashing self-reports: potential explanatory factors and alternative measurements.
        PLoS One. 2015; 10e0136445
        • Lydon S.
        • Greally C.
        • Tujjar O.
        • Reddy K.
        • Lambe K.
        • Madden C.
        • et al.
        Psychometric evaluation of a measure of factors influencing hand hygiene behaviour to inform intervention.
        J Hosp Infect. 2019; 102: 407-412
        • Paulhus D.L.
        Self-deception and impression management in test responses.
        in: Angleitner A. Wiggins J.S. Personality assessment via questionnaires. Springer, Berlin1986: 143-165