ABDELHADI N. & DRACH-ZAHAVY A. (2012) Promoting patient care: work engagement as a mediator between ward service climate and patient-centred care. Journal of Advanced Nursing68(6), 1276–1287.
Aims. To test a model that suggests the ward’s climate of service facilitates nurses’ patient-centred care behaviours through its effect on nurses’ work engagement.
Background. Organizational efforts to promote patient-centred care focused on interventions aimed to improve nurses’ communication skills, or to improve patient’s participation in the decision-making process. These interventions have been only partially successful, as they do not take the ward context into account; so caring professionals who attend workshops can rarely apply their newly acquired skills due to the daily pressures of the ward.
Method. A nested cross-sectional research design (nursing staff within wards) was adopted, with three measures of the care behaviour of nurses. Data were collected in 2009, from 158 nurses working in 40 wards of retirement homes in northern Israel. Nurses’ work engagement, ward’s climate for service and control variables were measured via validated questionnaires. Patient-centred care behaviours were assessed by structured observations.
Results. The findings supported our model: service climate proved a link to nurses’ work engagement and patient-centred care behaviours. Nurses’ work engagement mediated the service-climate patient-centred care behaviours.
Conclusion. The research is pioneering in demonstrating a close relation between ward service climate and patient-centred care. In practice, to improve patient-centred care managers should invest in facilitating ward service climate, highlighting the importance of service to the organization through appropriate rewards, guidance and administrative practices.
What is already known about this topic
- • Patient-centred care is recognized as a care strategy that benefits patients, nurses and organizations.
- • Organizational efforts to promote patient-centred care have focused on interventions to improve nurses’ communication skills or patient’s participation in decision-making.
- • These interventions have been only partially successful, as their beneficial effect tends to fade shortly after the intervention ended.
What this paper adds
- • Nurses’ work engagement encourages them to engage in patient-centred care.
- • A service climate can boost nurses’ work engagement; hence motivate them to engage in patient-centred care even in loaded work contexts.
Implications for practice and/or policy
- • Job restructuring efforts should focus on broadening the nursing role beyond physical care, by focusing on nurses’ patient-centred care behaviours in performance appraisals, and as criteria for promotion and reward.
- • The nursing directorate should establish an operational definition of patient-centred care encompassing the essential requirements at the candidate recruitment stages. It may call for a combination of education and experience, aligned with personal skills, for example, in communication and power sharing.
- • Nursing management should remove obstacles to the enactment of patient-centred care, and endeavour to channel resources to nurses’ training and improvement in caring skills.
Patient-centred care (PCC) encourages nurses to understand the disease and the patient when developing a health plan and/or caring for patients. In 2001, the United States Institute of Medicine endorsed PCC as a central component of high-quality health care (The Committee on Quality of Health Care in America, 2001). Additional key organizations worldwide (ACGME 2005; ABMS 2005) identified it as an important competency necessary for healthcare professionals. This is primarily because PCC is well accepted as an important contributor to a host of health outcomes, including patients’ general understanding of their illness, adherence to therapeutic regimens, disease control, quality of life and patient’s satisfaction (Stewart et al. 2000, Mead & Bower 2002, Griffin et al. 2004). Though most studies on PCC outcomes have focused on patient outcomes there is some evidence that providing PCC is also associated with nurses’ satisfaction with their jobs (Suchman et al. 1993, Stewart et al. 2000). Finally, providing PCC was found to be linked to organizational efficiency and reductions in healthcare delivery costs (Redelmeier et al. 1995, Stewart et al. 2000, Epstein et al. 2005).
In light of the enormously beneficial outcomes of PCC, the question of how to encourage nurses to engage in these behaviours arises. A review of the literature reveals that the previous research focused largely on PCC outcomes, with the study of PCC antecedents lagging far behind (Drach-Zahavy 2009). Accordingly, this study was designed to narrow this gap in the literature. We developed a multi-level model for predicting PCC, suggesting that the ward’s service climate facilitates nurses’ PCC behaviours through its effect on their work engagement (Figure 1). The model suggests that a climate for service, namely nurses’ perception that providing high quality service is what really counts in their ward, may enhance PCC behaviour. Moreover, the model implies that nurses’ work engagement, defined as a positive, fulfilling, work-related state of mind (Schaufeli et al. 2002), serves as a vehicle whereby service climate affects nurses’ PCC behaviours.
Recognizing the importance of approaching the patient as a whole person, healthcare theoreticians and clinicians expanded the ‘Biomedical Model’ to a broader bio-psychosocial orientation (Engel 1977, 1980), emphasizing the need to include patients’ personal needs and wishes in their healthcare plan (McWhinney et al. 1997). From this orientation, the concept of PCC evolved. The PCC acknowledges the network of relationships and social-cultural contexts within which a patient acts, and advocates the pursuit of understanding the patient’s perspective on his or her illness, including his or her concerns, ideas, expectations, needs, feelings and functioning; PCC promotes a patient–provider partnership by encouraging patient’s empowerment and involvement in making decisions that concerns his or her health status and the planned treatment process. It calls for reaching a shared understanding of the patient’s health problem and its treatment, in keeping with the patients’ values. Scholars also agree that for this, health providers need to develop excellent communication skills, self-awareness, reflective listening, and adequate use of empathy (Mead & Bower 2002, Epstein et al. 2005, Stewart et al. 2005).
Means for improving PCC have typically focused on the provider–patient encounter and included ‘patient activation’ interventions to increase patient participation in the care process (Greenfield et al. 1985, Post et al. 2002) and health providers’ interventions to improve the consultation process (Lewin et al. 2001). Patient activation interventions try to educate patients to prepare themselves better for the visit, to ask questions and demand clarifications during the encounter, and to pursue their wishes and expectations (Sharp et al. 1996; Griffin et al. 2001). Research has shown that such interventions led patients to exhibit greater control during the medical visit, and experience a more balanced power relationship in the encounter (Douglas et al. 2002). Other research indicated that patient activation interventions tend to improve health outcomes but may have a neutral or negative effect on the patient–physician relationship (Oliver et al. 2001). At the same time, health providers’ interventions focus on improving the professionals’ communication skills, including better consultation style, use of empathy, and resolving emotional problems that arise during the encounter. Such interventions have been shown to improve the degree of patient-centredness in the consultation, although inconsistent effects were found in some clinical outcomes (Brown et al. 1995, Epstein et al. 2005).
Although these interventions have notable advantages, they do not take into account nurses’ work environment. In healthcare systems driven by technology and beset by productivity pressures and financial concerns that threaten their survival, and conditions of inadequate training and mentoring time, nurses may be limited in their responsiveness to patients’ needs (Epstein et al. 2005). These circumstances suggest that theory and practice could benefit from exploration of the role of the organization (the ward in this study) in encouraging and maintaining PCC, more particularly from conceptualization and measurement of the service climate in the ward.
Service climate and PCC behaviours
Service climate is commonly defined as employees’ shared perceptions of the practices, procedures and behaviours of effective service that are expected, supported and rewarded (e.g. Schneider et al. 1994). This definition highlights several important themes. First, because multiple climates often exist simultaneously in a single organization, climate is best regarded as a specific construct having a referent – in this case, service (Schneider et al. 1994). Hence, a service climate exists in a ward when the nurses’ shared perceptions are integrated into a theme that indicates the importance of service in their ward. In the specific context of healthcare environments, quality service calls for providing PCC (Mead & Bower 2002, Epstein et al. 2005). So although PCC may differ from quality service in several respects (e.g. intensity of emotions involved in the encounter: Zapf 2002), both concepts refer to investing extra effort in meeting the clients’ expectations (Parasuraman et al. 1985).
Secondly, service climate derives from a consensual understanding of how to behave in different settings and with different customer populations (Hui et al. 2007). Socially shared climate perceptions are valuable in situations where it is unclear which performance facet should be prioritized. Such ambiguities often arise from a discrepancy between formally espoused policies and enacted practices (e.g. Zohar & Tenne-Gazit 2008). For example, despite the overall focus on PCC in today’s healthcare organizations, nurses’ service role behaviours tend to be differentially supported or rewarded under changing task conditions (e.g. falling behind schedule), and head nurses occasionally disregard service clashes depending on situational demands (Schneider et al. 2009).
Furthermore, most conceptualizations of service climate pertain to employees’ concomitant cognitive appraisals of management’s concern for their own well-being, and the customers’ (e.g. Johnson 1996, Schneider et al. 1998, Borucki & Burke 1999). Concern for employees is expressed in such management gestures as support for and appreciation of nurses’ service performance, investment in service training, removal of obstacles to providing good service, and distribution of rewards for it. Concern for customers or patients is manifested in management’s focus on seeking feedback from patients and acting on it (e.g. Schneider et al. 1998, Borucki & Burke 1999).
Given the current primacy of providing patient-centred care in today’s healthcare organizations (Epstein et al. 2005), the scarcity of research on the link between service climate and service quality in the healthcare industry is surprising. Nevertheless, in various studies of other service industries, such as banking, insurance, retail stores and restaurants, findings have robustly supported the relationship between service-climate and service quality (e.g. Schneider et al. 1998, 2005; Liao & Chuang 2004; Salanova & Agut 2005; Wall & Berry 2007). For example, emphasizing a climate for service was linked (among other things) to employees’ quality service behaviours such as empathy, courtesy and organizational citizenship behaviours towards customers (Salanova & Agut 2005, Schneider et al. 2005, Dimitriades 2007, Hui et al. 2007). In addition, studies in health care demonstrated positive associations between a general supportive work environment and employees’ service behaviour (Glisson & Hemmelgan 1998; McCusker et al. 2002).
Hypothesis 1: A ward’s service climate will be positively associated with nurses’ PCC: The higher the service climate in the ward the higher the PCC behaviors displayed by the nurses.
How does (a ward’s) service climate generate PCC behaviours among nurses?
The literature reveals two theoretical explanations of how service climate affects nurses’ behaviours. (a) The managerial argument maintains that socially and organizationally supported values will bring about particular types of management practices; these in turn will influence employees’ perceptions of the work environment and the behaviours which are expected, appreciated and rewarded in the ward (Borucki & Burke 1999). For example, we expect PCC behaviours to flourish in wards where they are rewarded and serve as a criterion for hiring and promotion, and where resources are directed to training and improving clinicians’ PCC skills (namely in high as against low service-climate wards). (b) The ‘fit’ argument suggests that people strive to harmonize with their environment (Smith-Crowe et al. 2003). To achieve this they seek information on proper behaviour from their surroundings. Therefore, when the organization-promoted climate is explicit, and organizational policies and practices are directly aligned with it, people will identify and recognize what is important and display more role-fit behaviours (Schneider 1975, Burke et al. 1992). This is especially important, given the primacy of service quality in daily nursing activities, and the fact that it often entails making a greater effort, going at a slower pace or disrupting the flow of activities (Drach-Zahavy 2009, Drach-Zahavy & Somech 2010). Under these circumstances, nurses are likely to make sense of the ward’s climate in their search for social information on the relative priorities of these competing demands in their daily exchanges;
Although sound, these two explanations have not stimulated much empirical research on how service climate generates PCC behaviours among its nurses. Nevertheless, the two explanations point at an underlying motivational mechanism directing nurses to adhere to management’s expectations or to harmonize with their environment. In this study we suggest that nurses’ work engagement might mediate the service climate-PCC link, thereby explaining how a ward’s service climate affects nurses’ PCC behaviours.
Work engagement is defined by Schaufeli et al. (2002) as a positive, fulfilling, work-related state of mind characterized by vigour, dedication and absorption. It refers to a persistent and pervasive affective-cognitive state, which is not focused on any particular object, event, individual or behaviour (Simpson 2009).Vigour is characterized by high levels of energy and mental resilience while working, the willingness to invest effort in one’s work and persistence even in the face of difficulties; dedication means being strongly involved in one’s work and experiencing a sense of significance, enthusiasm, inspiration, pride and challenge; and absorption is one’s being fully concentrated and happily engrossed in work, whereby time passes quickly and one has difficulties detaching oneself from it (Gonzalez-Roma et al. 2000, Schaufeli et al. 2002).
Accordingly, we suggest that a more engaged nurse, characterized by higher vigour, dedication and absorption, will exhibit higher PCC behaviours than a less engaged one, a notion supported by several studies (Salanova & Agut 2005, Laschinger & Leiter 2006, Liorens et al. 2006, Richardson et al. 2006). For example, Laschinger and Leiter (2006) similarly found that nurses’ work engagement predicted safer patient care.
Hypothesis 2: Nurses work engagement will be positively related with PCC.
Service climate and work engagement
The research model illustrated in Figure 1 further suggests that service climate also promotes nurses’ work engagement. This premise is theoretically embedded in the Job Demands–Resources model (Bakker et al. 2002), positing that job resources can boost work engagement. Resources are defined as the objects (e.g. technical aids that make work easier), personal characteristics (e.g. traits and skills), conditions (e.g. organizational support, facilitative climate) or energies (e.g. time, mental and physical energy, knowledge) that are valued in their own right or because they act as conduits to the achievement or protection of valued resources (Hobfoll 1989, 2001). Accordingly, service climate might serve as a contextual resource at work (Drach-Zahavy 2009). In fact, Hobfoll (1989) specifically listed ‘necessary tools for work’, ‘job training’ and ‘help in tasks at work’– three characteristics of service climate – as part of a list of resources (Hobfoll 1989, p. 342). In this vein, empirical research has shown that the general facilitative resource arising in organizations high on service climate includes efforts to remove obstacles to work (Schneider 1988, Burke et al. 1996).
Hypothesis 3: Service climate will be positively linked to nurses work engagement.
Finally, our mediating model proposes complete mediation of nurses’ work engagement in the relation of a ward’s service climate to nurses’ PCC behaviours. This argument is consistent with previous team effectiveness models (e.g. Campion et al. 1996, Kirkman & Rosen 1999). These input-process-output models separate job characteristics from internal responses to these characteristics and subsequent performance. All these models involve a three-stage process: (a) input: leaders take various actions – in our case instituting a service climate; (b) these actions affect members’ processes or motivational and affective states – in our case nurses work engagement; (c) output: important outcomes result from workers’ positive states or processes – in our case nurses’ PCC behaviours. We suggest that the anticipated incremental effect is inherent in work engagement as a motivational force (Gonzalez-Roma et al. 2000, Schaufeli et al. 2002).This is especially important given the attractiveness of non-quality service behaviour (low PCC) stemming from melioration bias (Drach-zahavy & Somech 2010) which often outweighs the expected utility of providing PCC in the long run.
Hypothesis 4: Nurses’ work engagement mediates the relationship between a ward’s service climate and nurses’ PCC.
This article reports a study of the relation of the ward’s service climate to patient-centred care, and the mediating role of nurses’ work engagement in this relation.
The study has a cross-sectional multi-level (nurses within wards) design.
Forty nursing wards were randomly selected from a list of publicly funded retirement homes in the north of Israel. Most of the patients in those wards (62%) were highly dependent older people, whereas the rest required only minimal supervision/assistance. In each ward, all nurses were invited to participate in the study (n = 180). Of the 180 nurses approached, 158 returned a completed questionnaire, making a response rate of 87·7%. Sample size was calculated by the PEPI for Windows software; research strength was 80%, alpha = 0·05.
Data were collected in 2009 by a multi-method strategy consisting of observations, surveys and administrative data.
The PCC was assessed by structured observations. Observations have been described as one of the better methods for assessing PCC because studies relying on nurses’ or patients’ retrospective self-report suffer from estimation biases (Epstein et al. 2005). In addition, the ‘good impression bias’ stemming from the presence of the observer in patient–provider encounters is deemed minimal, as healthcare providers typically become quickly accustomed to the observer’s presence and tend to exhibit their natural behaviour (Roter 1989). Observers recorded nurse’s PCC in the course of three separate guidance encounters: patient admission, provision of treatment and guidance, each averaging 20 minutes. These encounters were chosen because they afford nurse–patient interaction. A nine-item structured observation sheet was used (Schirmer et al. 2005, Drach-Zahavy 2009). The measure consisted of nine evaluation criteria on a four-point Likert-type scale (from 0 = not attempted, to 3 = well done, and a ‘not applicable’ option), and a space for comments after the evaluation criteria. PCC was averaged across the nine evaluation criteria and across the three observation instances. Pearson correlations among the three observation instances were high, namely 0·59–0·60, indicating support for the reliability of the measure. Cronbach’s alpha was 0·85–0·88.
Two graduate students, nurses by profession, participated as observers. The ward’s nurses perceived their presence as natural, which might prevent bias, yet their objectivity could be relied on and they were familiar with the best practices of patient-centred care. To ensure inter-rater reliability and the observations’ validity the observers received 10 hours of extensive training. This covered (a) observation techniques in research, (b) a thorough study of PCC and caring, (c) participation in periodical meetings during the observation period when categorization dilemmas were discussed and resolved by the group consensus technique (Kappa = 0·85–0·93).
Data on service climate and work engagement were obtained through questionnaires, distributed to nurses on site by a research assistant.
Service climate was assessed by the eight-item global service-climate scale developed by Schneider et al. (1998). The items refer to a collection of behavioural features of the wards, all focusing explicitly on service quality. A sample item (rated on a 5-point scale from 1 = poor to 5 = excellent) is ‘How would you rate the job knowledge and skills of department employees to deliver superior quality service?’ (Cronbach’s alpha = 0·93). In line with previous research (e.g. Schneider et al. 1998, 2005, Salanova & Agut 2005) to capture service climate as a ward-level construct, we aggregated the individual service climate scores across nurses in the same ward assuring homogeneity and within-group inter-rater agreement among nurses (Rousseau 1985). Tests of rwg (James et al. 1993) indicated agreement of responses at the unit level: the median value was: 0·83. A value of 0·70 or above is suggested as a ‘good’ amount of within-group inter-rater agreement (James et al. 1993); ICC assesses ward homogeneity relative to total variance. ICC for service climate was 0·18 (P < 0·05), indicating that 18% of the variance in individual-level responses can be explained by the ward-level properties of the data (Bliese 2000). These findings give support for the aggregation of service climate individual scores to a ward-level indicator.
Work engagement was assessed by a three-scale tool adapted from Salanova and Agut (2005). Vigour, the first scale, consisted of six items. An example item is ‘At work, I feel full of energy’ (Cronbach’s alpha = 0·84). Dedication, the second scale, consisted of five items. An example item is ‘My job inspires me’ (Cronbach’s alpha = 0·89). Absorption, the third scale, consisted of five items. An example item is ‘Time flies when I’m working’ (Cronbach’s alpha = 0·88). All items were scored on a 5-point scale from 1 = never, to 5 = always. High scores on vigour, dedication and absorption were indicative of work engagement. The correlations between the three scales were high and statistically significant, 0·65–0·72 (P < 0·01). Hence, for the purpose of this study, and in accordance with previous research (e.g. Salanova & Agut 2005), we averaged the three subscales to yield an indicator of work engagement (Cronbach’s alpha = 0·88).
Nurses’ gender and tenure in nursing were controlled for at the individual level; ward load, calculated as the mean nurse/patient ratio across the observational session days, were controlled for at the ward level; other controls accounted for potential differences among wards.
The study was approved by the Institutional Review Board of the University of Haifa. All participants agreed voluntarily to participate in the study after receiving a brief explanation of the general research aims, and being assured that the observational and survey data would be used only for research purposes.
To assess our hypotheses a multi-level analysis was necessary since nurses were nested in 40 different wards. The mixed linear models procedure was conducted, this being appropriate for assessing hypotheses at the individual level when individuals are nested in naturally occurring hierarchies. This procedure makes it possible to calculate the fixed effects (factors deemed to contain all levels of interest for the study) and the random effects (factors that reflect a set of levels representing a sample out of a population of levels) (Singer 1998). In the data analysis the ward was treated as a random effect, and the independent ward-level service climate and the mediating individual-level variable of work engagement were treated as fixed effects (Singer 1998).
Further, our model suggested that nurses’ work engagement mediates the relationship between the wards’ service climate and nurses’ PCC behaviour. Therefore, we followed the four-step procedure to analyse mediation, suggested by Baron and Kenny (1986). According to these authors, full mediation can be supported only when (a) an association exists between the independent variable (service climate) and the dependent variable (PCC behaviour); (b) an association exists between the independent variable (service climate) and the mediating variable (work engagement); (c) an association exists between the mediating variable and the dependent variable (PCC behaviour); and (d) when the effect of the mediating variable (work engagement) is controlled, the association of the independent variable of service for climate and the dependent variable of PCC behaviour is no longer important. These recommendations were implemented in our analyses by means of corresponding mixed models analyses.
Sixty-seven per cent of the nurses were women, aged 33·3 years on average (sd = 7·6). Most were married (75%), with 1·4 (sd = 1·3) children on average. Average ward tenure was 5·4 years (sd = 5·07), and average job tenure was 10·4 years (sd = 7·7). As for their education level, the majority (70·2%) was Registered Nurses, of whom 41·7% had a college degree, 26% a Bachelor’s degree and 2·5% had a Master’s degree. The others were practical nurses.
Table 1 exhibits the means, standard deviations and inter-correlation matrix for all study variables. Close inspection of the correlations in Table 1 revealed that the ward’s service climate was significantly and positively associated with nurses’ PCC behaviour and with nurses’ work engagement (r = 0·26 and 0·54, P < 0·01, respectively). Work engagement was significantly and positively related to PCC behaviour (r = 0·38, P < 0·01). These correlations provided initial support for our hypotheses.
|(2) Nurses’ tenure||10·23||7·70||−0·11||1·00|
|(3) Ward’s load||0·18||0·06||−0·14||−0·08||1·00|
|(4) PCC||2·00||0·49||0·04||0·06||0·26 (**)||1·00|
|(5) Work engagement||3·76||0·59||−0·03||0·17 (*)||0·54 (**)||0·38 (**)||1·00|
|(6) Service climate||3·58||0·65||0·10||0·10||−0·20 (*)||0·26 (**)||0·54 (**)||1·00|
To test the first hypothesis, we conducted a mixed linear models analysis for predicting nurses’ PCC behaviours from service climate. To control for the effects of gender, seniority and work load, these variables were entered into the equation at the first stage, followed by the independent variable of service climate at the second stage. The findings are presented in Table 2. As for the control variables, nurses’ job tenure was positively and significantly associated with nurses’ PCC behaviours (estimate = −0·01, P < 0·05), the ward’s load was negatively and significantly associated with nurses’ PCC behaviours (estimate = −2·18, P < 0·05), and nurses’ gender was not associated with it (P > 0·05). In addition, service climate was positively and significantly associated with nurses’ PCC behaviours (estimate = 0·27, P < 0·05). Thus, hypothesis 1 was supported.
|Model 1||Model 2|
|Control variables||Direct influence|
|Global service climate||0·27*||0·12|
|Restricted Log Likelihood−2Δ||2·65*|
|Variance of the department level||0·12**|
To test the second hypothesis, we conducted a mixed linear models analysis for predicting nurses’ PCC behaviours from nurses’ work engagement. To control for the effects of gender, seniority and work load, these variables were entered into the equation at the first stage, followed by nurses’ work engagement at the second stage. The findings are presented in Table 3. As shown in Table 3, nurses’ work engagement was significantly and positively associated with nurses’ PCC behaviours (estimate = 0·27, P < 0·05). Thus, hypothesis 2 was supported.
|Model 1||Model 2|
|Control variables||Direct influence|
|Mean of work engagement||0·27**||0·06|
|Restricted Log Likelihood−2Δ||17·083*|
|Variance of the department level||0·106**|
To test the third hypothesis, we conducted a mixed linear models analysis for predicting nurses’ work engagement from service climate. To control for the effects of gender, seniority and work load, these variables were entered into the equation at the first stage, followed by nurses’ service climate at the second stage. The findings are presented in Table 4. As shown in Table 4, the ward’s service climate was significantly and positively associated with nurses’ work engagement (estimate = 0·55, P < 0·01). Thus, hypothesis 3 was supported.
|Model 1||Model 2|
|Control variables||Direct influence|
|Global service climate||0·55**||0·11|
|Restricted Log Likelihood−2Δ||15·723|
|Variance of the department level||0·09**||0·03|
Finally, our fourth hypothesis suggested that nurses’ work engagement mediated the association between the ward’s service climate and PCC. According to the recommendation of Baron and Kenny (1986), we conducted a mixed linear model analysis to examine whether or not when we controlled for the effect of nurses’ work engagement the association between the ward’s service climate and nurses’ PCC behaviour was eliminated. Thus, the control variables of gender, seniority and work load were entered into the equation at the first stage. Next, to control for the effect of nurses’ work engagement, this variable was entered at the second stage, followed by nurses’ service climate at the third stage (see Table 5). As can be seen from Table 5, the mediator variable of nurses’ work engagement was still important (estimate = 0·25, P < 0·01), but the effect of the independent variable of ward’s service climate was not (estimate = 0·13, P > 0·05). Thus, hypothesis 4 was supported.
|Global service climate||0·13||0·12|
|Mean of work engagement||0·25**||0·06|
|Restricted Log Likelihood−2Δ||13·344|
|Variance of the department level||0·1**|
As providing healthcare in an increasingly financially restricted and competitive market becomes an essential feature of the healthcare system, providing PCC emerges as a major component of organizational strategy (Drach-Zahavy 2009). With this, the question of the aetiology of PCC behaviours arises. The present study addressed this issue, by developing and testing a multi-level model suggesting that the context in which nurses operate, namely the ward’s service climate, triggers nurses’ work engagement and consequently facilitates PCC behaviour. This model contributes to the nursing literature in several aspects.
First, in line with the repeated calls in PCC research for greater attention to nurses’ work context (Epstein et al. 2005; Simpson 2009), our findings are pioneering in indicating the important role of the ward’s service climate in facilitating PCC behaviour. Most previous research on the antecedents of PCC focused on educating staff and patients on how to facilitate PCC (Epstein et al. 2005), while neglecting the context in which patient–nurse encounters occur. Our findings indicate that as in organizations in other settings (Schneider et al. 2009), service climate has an important role in facilitating a specific aspect of service behaviour, namely nurses’ PCC behaviour. Note that the facilitating role of service climate emerged even after we controlled for the ward’s workload. This finding indicates that service climate, in its emphasis on clarifying expectations of high quality service, providing support and training for it, and removing obstacles to providing it, is indeed an important resource for nurses. Moreover, working in a ward characterized by a high service climate can compensate for such inhibiting context factors such as workload, and motivate nurses to exhibit higher PCC behaviours (Salanova & Agut 2005).
Secondly, our findings show that service climate also facilitates nurses’ work engagement. This finding is in line with previous studies that used the Job Demands–Resources Model (JD–R) as a framework for studying the antecedents of work engagement (Demerouti et al. 2001; Bakker et al. 2007, Simpson 2009). These studies have repeatedly identified job resources (as against job demands) as important predictors of work engagement (Schaufeli & Bakker 2004; Hakanen et al. 2006, Liorens et al. 2006, Mauno et al. 2007). For example, Koyuncu et al. (2006) showed that higher levels of resources at work such as control at work, rewards and recognition from the management, and value fit were associated with more work engagement. Our finding complement existing knowledge by pointing out the ward’s service climate as yet another external resource that can boost nurses’ work engagement.
Thirdly, our findings also highlight the role of nurses’ work engagement in motivating nurses to engage with PCC: nurses who experienced high levels of work engagement provided more PCC than those who experienced less. This finding is in line with Simpson’s (2009) comprehensive review of the work engagement literature, concluding that it exerts a powerful performance-based impact, and more specifically with findings demonstrating an association between work engagement and service behaviours (Hallberg & Schaufeli 2006). Our findings contribute in locating the work engagement–service behaviour link in the specific context of nurses and PCC also.
Fourthly, our findings support our model in depicting nurses’ work engagement as a mediator in the relationship between ward’s service climate and nurses’ PCC behaviours. This model highlights the importance of resources, both internal such as work engagement and external such as the ward’s service climate, as facilitating PCC. Moreover, most previous research on the relationship between service climate and service behaviour, in service organizations in general, has not explored the underlying mechanism by which service climate triggers high quality service behaviours such as PCC (Kuenzi & Schminke 2009). Our study is pioneering in demonstrating service climate as a motivating factor, boosting nurses’ internal resources in the form of work engagement and consequently motivating excellent patient-centred care.
Limitations of the study and suggestions for further research
Several limitations of the study should be acknowledged. The research was cross-sectional, so causal inferences could not be drawn. A longitudinal design would have made it possible to collect data on subsequent PCC behaviours, and provided stronger evidence of a directional relationship between the variables. Secondly, the data were collected in specific nursing setting – retirement homes, which raises the question of the generalizability of the findings. Studies in different organizational settings would help ascertain whether the findings are context-specific or can be generalized across organizational settings. Thirdly, although a multi-method strategy for data collection was used: survey-observation and administrative-data techniques, the observational measure of PCC might be subject to bias. Nevertheless, Roter (1989) concluded that the ‘good impression bias’ that stems from the presence of the observer in patient–provider encounters is minimal, as healthcare providers typically become quickly accustomed to the presence of the observer and tend to exhibit their natural behaviour.
Nurses’ overload has typically been raised as a factor constraining their ability to provide PCC (Epstein et al. 2005). Still, our findings show that a service climate can boost nurses’ work engagement and consequently motivate them to engage with PCC even in loaded work contexts. The question of how to elicit high work engagement in nurses then emerges. Previous research demonstrated the superiority of organizational over personal factors, thereby pointing at all level nursing administration responsibility in fostering nurses’ work engagement (Simpson 2009). To achieve enduring effects, job restructuring efforts should focus broadening the scope of nursing beyond physical care, by stressing nurses’ PCC behaviours in performance appraisals, and as criteria for promotion and reward. The nursing directorate should establish an operative definition of PCC covering the essential requirements at the candidate recruitment stages. This would possibly require a combination of education and experience, along with personal abilities such as communication skills and power sharing. Moreover, nursing management should act to remove obstacles to the enactment of PCC, and ensure that resources are directed to training and improving nurses’ caring skills. Together, these efforts will help in maintaining an organizational climate that is focused on service.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
No conflict of interest has been declared by the authors.
AD and NA were responsible for the study conception and design of the article. They performed the data collection and data analysis and were responsible for the drafting of the manuscript. They also made critical revisions to the article for important intellectual content besides providing statistical expertise. AD provided administrative, technical or material support and supervised the study.