527356 research-article2014
WJNXXX10.1177/0193945914527356Western Journal of Nursing ResearchPedrazza et al.
Article
Nurses’ Comfort with Touch and Workplace Well-Being
Western Journal of Nursing Research 1–18 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0193945914527356 wjn.sagepub.com
Monica Pedrazza1, Stefania Minuzzo1, Sabrina Berlanda2, and Elena Trifiletti1
Abstract Touch is an essential part of caregiving and has been proved to be useful to reduce pain. Nevertheless, little attention has been paid to nurses’ perceptions of touch. The aim of this article was to examine the relationship between nurses’ feelings of comfort with touch and their well-being at work. A sample of 241 nurses attending a pain management training course completed a questionnaire, including the following measures: Comfort with Touch (CT) scale (task-oriented contact, touch promoting physical comfort, touch providing emotional containment), Maslach Burnout Inventory (MBI; emotional exhaustion, cynicism), and Job Satisfaction. Results of structural equation models showed that touch providing emotional containment was the main predictor of emotional exhaustion. Emotional exhaustion, in turn, was positively related to cynicism and negatively related to job satisfaction. In addition, the direct path from touch providing emotional containment to cynicism was significant. Practical implications of the findings are discussed. Keywords comfort with touch, physical touch, pain, burnout, job satisfaction
1University 2University
of Verona, Italy of Trento, Italy
Corresponding Author: Monica Pedrazza, Associate Professor, Department of Philosophy, Education, and Psychology, University of Verona, via San Francesco 22, Verona, 37129, Italy. Email: [email protected]
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
2
Western Journal of Nursing Research
Physical contact with the patient is an intrinsic part of the nursing profession (Chang, 2001; Routasalo, 1999). To provide appropriate assistance, nurses have recourse to various forms of contact that differ in their nature and degree of intimacy depending on whether their purpose is to carry out tasks (e.g., taking vital signs), provide physical comfort (e.g., massaging), or provide emotional containment (e.g., hugging patients; Pedrazza, Minuzzo, Trifiletti, & Berlanda, 2014). The ability of the professional to reassure the patient through touch is particularly important when the patient is suffering from serious pathologies, is in pain, is worried, afraid or anxious, or is physically and emotionally fragile and vulnerable (Bonacini & Marzi, 2005). This study shows how feeling comfortable in giving support to patients through touch may be associated with greater job satisfaction and reduced burnout (emotional exhaustion and cynicism).
Physical Touch in Caring One form of contact is cutaneous stimulation, for example, superficial massage of the back, hands, feet, or face. Cutaneous stimulation has been shown to be useful in reducing pain because it activates the large-diameter fibers that transmit tactile information, which in turn antagonize the small-diameter fibers that transmit the pain stimulus (Melzack, 1996). Cutaneous stimulation also stimulates the production of endorphins, relaxes the muscles, sedates the patient, and acts as a distracting stimulus. Evidence-based guidelines highlight the effectiveness of cutaneous stimulation in relieving pain associated with muscle tension or spasms (Grade C, Level IV; National Guideline Clearinghouse). Many studies have confirmed that cutaneous stimulation is useful in reducing pain and anxiety and inducing relaxation in oncology patients (Jane et al., 2011). Hand massage, on its own or together with soothing music, has been shown to reduce restless and aggressive behavior in demented subjects (Fu, Moyle, & Cooke, 2013; Remington, 2002), and to reduce pain and lower the heart rate and blood pressure in patients in an emergency department (Kubsch, Neveau, & Vandertie, 2001). Superficial hand and foot massage is also an effective, inexpensive, risk-free, flexible, and easily applied strategy for managing postoperative pain (Han & Lee, 2012; H. L. Wang & Keck, 2004). In general, touch is an essential part of caregiving and an excellent way of communicating attention, sympathy, closeness, reassurance, and presence (see Routasalo, 1999). This approach is particularly recommended for suffering patients, the frail and solitary elderly, the terminally ill, and the dying. It may be a way of taking charge of the patient to reduce psychological suffering, feelings of loneliness, difficulty in communicating, and the fear and
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Pedrazza et al.
3
anxiety of death (Bonacini & Marzi, 2005). Kübler-Ross (2008) claimed that gentle hand pressure is the most effective form of communication with dying patients. The positive effects of touch in caring are, therefore, well-established. However, little attention has been paid to how physical touch is perceived by patients and nurses (but see Hollinger-Smith & Buschmann, 1993). Some attention has been paid to the influence of gender on perceptions of touch in caring. Research has shown that although male patients perceive touch from female nurses more positively than do female patients, female nurses consider female patients more receptive to touch and feel more comfortable touching female than male patients (Lane, 1989). As to men nurses, the stereotype of men as sexual aggressor may generate a sense of vulnerability and a need for caution in the use of touch (Evans, 2002). More in general, touching the patient’s body may be associated with negative feelings. Picco, Santoro, and Garrino (2010), using data from in-depth interviews with 14 nurses, found an ambivalent attitude toward taking care of the patient’s body. They showed that although the body is regarded as a privileged element of nursing care, it is also perceived as a source of uneasiness and negative feelings. Physical touch requires great involvement on the part of nurses, because it implies physical, cognitive, and emotional proximity to the patient. Moreover, nurses may expect patients to refuse or react negatively to touch, as touching the patient’s body implies entering his or her privacy (Picco et al., 2010).Touching the patient may be therefore associated with different levels of comfort among nurses. Because touch is an integral part of the nurse–patient relationship (Hollinger-Smith & Buschmann, 1993; Routasalo, 1999) and is fundamental in the provision of nursing care (Chang, 2001), it is important to understand how nurses experience touch and how their feelings about touch relate to workplace well-being. If positive feelings of ease during contact are likely to be associated with well-being (e.g., greater job satisfaction, reduced burnout), a lack of comfort in touch may result in negative organizational outcomes and, consequently, in lower quality of care. The literature on physical touch generally distinguishes between two forms of touch: necessary and non-necessary (Routasalo, 1999). Taskoriented or necessary touch is intended for carrying out a task or procedure (e.g., taking vital signs); non-necessary touch, in contrast, is spontaneous and affective (e.g., reassuring a patient by holding his or her hand), and is not necessary for accomplishing a task. These two forms of contact obviously imply different degrees of cognitive and affective proximity to the patient. Based on previous literature on physical touch, Pedrazza et al. (2013) developed and validated the Comfort With Touch (CT) scale. The scale is articulated into five subscales: Task-Oriented Contact, Personal Care, Physical Comfort, Reassurance and Emotional Containment. The concept of
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
4
Western Journal of Nursing Research
emotional containment refers to the ability to receive, process, and respond with understanding to the emotional reactions of another person (Douglas, 2007). Three subscales of the CT scale were selected for the present study— Task-Oriented Contact, Physical Comfort and Emotional Containment— which correspond to low, medium and high levels of cognitive and affective proximity to the patient. This study aims to verify the effects of comfort with touch on two measures of workplace well-being—burnout and job satisfaction—which appear to be associated with a wide range of individual and organizational outcomes. Burnout is a reaction to prolonged exposure to interpersonal and environmental stressors on the job (Maslach, Schaufeli, & Leiter, 2001). Three dimensions have been generally distinguished: exhaustion, cynicism, and reduced personal efficacy. However, Schaufeli and Bakker (2004) argued that personal efficacy can be considered a dimension of work engagement rather than burnout. Moreover, researchers have found consistent results for emotional exhaustion and cynicism but not for personal efficacy (Laschinger, Finegan, & Wilk, 2011). Therefore, recent studies have adopted a two-factor conceptualization of burnout. The present study focuses on emotional exhaustion and cynicism as outcomes of (dis)comfort with physical touch. Emotional exhaustion, generally regarded as the core of burnout, is a state of physical and emotional depletion that manifests itself both in physical fatigue and a feeling of being “burned out” by one’s job. Sustained emotional exhaustion generally results in cynicism (Leiter & Maslach, 2004), namely, detachment from the job. Burnout in nursing has been widely investigated. Negative work conditions, such as nursing shortage, work overload, lack of job autonomy, and leader–member exchange, appear to be major determinants of burnout (Laschinger et al., 2011; Moustaka & Constantinidis, 2010). Although previous research on burnout has primarily focused on environmental factors, individual difference variables are also likely to play a role in the development of burnout (Maslach et al., 2001). In a meta-analysis, Alarcon, Eschleman, and Bowling (2009) found that self-core evaluations, agreeableness, consciousness, extraversion, optimism, hardiness, and proactive personality were negatively related to both emotional exhaustion and cynicism (for the relationship between self-core evaluations and burnout in nursing, see Laschinger et al., 2011). Positive affectivity yielded a negative association with the two burnout dimensions, whereas negative affectivity showed a positive relationship. Given the costs of burnout for both people and organizations, it is important to further explore its possible antecedents. The current study investigates, for the first time, how nurses’ feelings of comfort with physical touch relate to emotional exhaustion and cynicism. Job satisfaction is the affective orientation that an individual has about his or her work (Price, 2001). Satisfaction at work appears to be a key predictor of Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Pedrazza et al.
5
nurses’ turnover and intentions to leave (Blegen, 1993; Irvine & Evans, 1995; Lu, Barriball, Zhang, & While, 2012; Sourdif, 2004; L. T. H. Wang, Ellenbecker, & Liu, 2012) and is therefore important to understand its sources. Researchers have outlined a number of environmental and personality factors associated with job satisfaction. In a systematic review, Lu et al. (2012) found that organizational environment and working conditions (e.g., structural empowerment and leader–member exchange; Laschinger et al., 2011), role conflict and ambiguity, and organizational and professional commitment were main predictors of nurses’ job satisfaction. Moreover, Laschinger et al. (2011) found that both emotional exhaustion and cynicism were negatively related to nurses’ satisfaction at work. As to personal dispositional factors, Judge, Heller, and Mount (2002) reported significant correlations between job satisfaction and the Big Five traits of extraversion, neuroticism, agreeableness, and consciousness (for similar results, see Zhai, Willis, O’Shea, Zhai, & Yang, 2013). Self-core evaluations (Judge & Bono, 2001; Srivastava, Locke, Judge, & Adams, 2010), positive and negative affectivity, and affective disposition (Connolly & Viswesvaran, 2000) are also related to satisfaction at work. The present study examines for the first time the relationship between job satisfaction and comfort with touch among nurses. The purposes of the present study are (a) to examine whether nurses report similar levels of (dis)comfort in relation to the three dimensions of touch (taskoriented contact, touch promoting physical comfort, and touch aimed at emotional containment); (b) to examine the associations between the three dimensions of touch, emotional exhaustion, cynicism, and job satisfaction; (c) to test the influence of touch on emotional exhaustion, cynicism, and job satisfaction.
Method Design, Sample, and Procedure The study was approved by the ethics committee of the researchers’ institution. A cross-sectional survey design was adopted. A questionnaire was administered to a convenience sample of 400 registered nurses, working in the departments of medicine, surgery, intensive care, oncology, and geriatrics in several hospitals in Northeast Italy. Data were collected during 2010. Participants were contacted during a 2-day pain management training course held by one of the authors. They were given the questionnaire at the end of the first day and asked to return it on the following day. Informed consent was obtained from each participant. The final sample included 241 nurses (241 of 400; 60.25% return rate). The mean age was 41.14 years (SD = 7.65, range = 24-60). Females comprised 85.9% of the sample (207 of 241). The mean length of service was 17.37 years (SD = 8.07, range = 1-35). Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
6
Western Journal of Nursing Research
Instrument The survey contained the following measures. CT scale. Nurses’ feelings of ease with touch were measured with eight items (Pedrazza et al., 2014). Each item describes a specific form of contact with the patient. Nurses are asked to indicate to what extent they feel at ease at performing each contact behavior. Responses are given on a 7-point scale, ranging from 1 (not at all) to 7 (very much). The measure is articulated into three subscales. Two items measure task-oriented contact (e.g., “Touching the patient to take his or her pulse”); three items measure touch aimed at promoting physical comfort (e.g., “Massaging the patient’s hands to reduce pain”); three items measure touch aimed at emotional containment (e.g., “Letting the patient cry in my arms”). Pedrazza et al. (2014) confirmed the construct validity of the three-factor scale using exploratory factor analysis (EFA). Cronbach’s alpha reliabilities in their study ranged between.91 (taskoriented contact) and .96 (emotional containment). For the present study, Cronbach’s alphas were .78 (task-oriented contact), .83 (promotion of physical comfort), and .87 (emotional containment). Burnout. The Emotional Exhaustion and Cynicism scales of the Maslach Burnout Inventory (MBI; Maslach & Jackson, 1981) were used in the Italian version validated by Sirigatti and Stefanile (1993). Six items measured emotional exhaustion (e.g., “I feel burned out from my work”), and five items measured cynicism (e.g., “I feel I treat some recipients as if they were impersonal objects”). Responses were given on a 7-point scale, ranging from 1 (completely disagree) to 7 (completely agree). Cronbach’s alpha coefficients for the Italian adaptation are generally higher than .87 for emotional exhaustion and .67 for cynicism (e.g., Martini & Converso, 2012). In this study, Cronbach’s alphas were .87 (emotional exhaustion) and .72 (cynicism). Job satisfaction. Job satisfaction was measured with four items (e.g., “I am satisfied with my job”), adapted from Dazzi, Voci, Capozza, and Bergamin’s (1998) scale. The 7-point response scale ranged from 1 (completely disagree) to 7 (completely agree). Bobbio, Manganelli Rattazzi, and Muraro (2007) demonstrated the construct validity of the one-factor scale with EFA and reported a Cronbach’s alpha of .86. The Cronbach’s alpha coefficient for this study was .64.
Data Analysis Data analysis was performed using SPSS 21.0 and LISREL 8.7 (Jöreskog & Sörbom, 2004). PRELIS (LISREL 8.7) was used for the imputation of
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Pedrazza et al.
7
Figure 1. Hypothesized model of the effects of comfort with touch on emotional exhaustion, cynicism, and job satisfaction.
missing data with the Expectation-maximization (EM) algorithm. Only 1.44% of the total responses (80 of 5,543) were missing scores. Following Schafer and Graham’s (2002) recommendations, maximum likelihood imputation (EM algorithm) was used to estimate values for missing scores. Maximum likelihood procedures provide more accurate estimates of population parameters than list-wise deletion or mean substitution (Schafer & Graham, 2002). The validity of the CT scale was tested by applying CFA (LISREL 8.7) to the eight items. Descriptive statistics were obtained with SPSS 21.0. For each variable, a composite score was computed by averaging the respective items. To test whether nurses reported similar levels of (dis) comfort in relation to task-oriented contact, touch promoting physical comfort, and touch aimed at emotional containment, a one-way repeated-measures ANOVA (SPSS 21.0) was applied, using the Bonferroni correction for multiple comparisons. Pearson correlation (SPSS 21.0) was used to examine the relationship between the three CT subscales and well-being outcomes (emotional exhaustion, cynicism, and job satisfaction). Structural equation modeling (SEM; LISREL 8.7) was applied to test the relationship between the three CT subscales and well-being outcomes. It was also examined whether cynicism and job satisfaction are predicted by emotional exhaustion, as found by previous research (Laschinger et al., 2011). To this aim, a mediation model was tested, in which CT subscales predict emotional exhaustion, which, in turn, predicts cynicism and job satisfaction (see Figure 1). Two parcels were computed for each of the following constructs: physical comfort and emotional containment (CT scale), emotional exhaustion, cynicism, and
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
8
Western Journal of Nursing Research
job satisfaction. Parceling is a measurement approach commonly used in latent-variable analysis techniques (Little, Cunningham, Shahar, & Widaman, 2002). A parcel is an aggregate-level indicator formed by averaging two or more items. The use of parcels limits measurement error and provides a good ratio of participants to variables. Moreover, compared with items, aggregatelevel data have some psychometric merits, such as higher reliability and a lower likelihood of distributional violations. Items also have fewer, larger, and less equal intervals between scale points than do parcels (Little et al., 2002). For task-oriented contact, the two respective items were used as indicators. In testing the model, the regression paths from the three CT subscales to cynicism and job satisfaction (residual direct effects) were estimated. The goodness of fit of the CFA and the SEM model was evaluated using the chisquare test, the comparative fit index (CFI), and the standardized root mean squared residual (SRMR). A model fits the data well when chi-square is nonsignificant, CFI is ≥.95, and SRMR is ≤.08 (Hu & Bentler, 1999). Mediation was tested with the Sobel test. For all analyses, the p value for statistical significance was .05.
Results The fit of the model for the three-factor structure of the CT scale (CFA) was acceptable: χ2(17) = 90.40, p ≅ .00; SRMR = .059; CFI = .96. Although chisquare was significant, the other two indices satisfied the respective criterion. Loadings were all significant and higher than .75 (p < .001). The correlation (phi coefficient) between task-oriented contact and emotional containment was .50 (95% confidence interval [CI] = [.40, .60]; p < .001). The correlation between task-oriented contact and physical comfort was .63 (95% CI = [.53, .73], p < .001). The correlation between physical comfort and emotional containment was 71 (95% CI = [.67, .75], p < .001). None of the 95% CIs included 1 (the perfect correlation), thus indicating that the three components actually reflected three distinct dimensions. Means and standard deviations for the study variables are reported in Table 1. To test whether nurses reported different levels of comfort in relation to the three CT subscales, a one-way repeated-measures ANOVA was applied. Using the Greenhouse–Geisser correction, we found that the three subscales differed significantly in the degree of perceived comfort, F(1.843, 442.393) 2 = 143.35, p < .001, ηp = .37. The highest perception of comfort was reported in relation to task-oriented contact (M = 6.72, SD = 0.72), followed by touch aimed at physical comfort (M = 5.93, SD = 1.37) and touch aimed at emotional containment (M = 5.35, SD = 1.60). Multiple comparisons with the Bonferroni correction were all significant, p < .001.
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
9
1.30 1.04 1.23
2.78 1.95 4.96
Note. CT = Comfort with Touch.
0.72 1.37 1.60
6.72 5.93 5.35
1. Task-oriented contact (CT scale) 2. Physical comfort (CT scale) 3. Emotional containment (CT scale) 4. Emotional exhaustion (burnout) 5. Cynicism (burnout) 6. Job satisfaction
SD
M
Measure — .66, p < .001
2
—
3
4
5
6
— −.13, p = .04 −.28, p < .001 −.29, p < .001 — −.14, p = .03 −.21, p = .001 −.32, p < .001 .54, p < .001 .19, p = .03 −.54, p < .001 −.34, p < .001 — .12, p = .064 .14, p = .03
— .61, p < .001 .47, p < .001
1
Table 1. Descriptive Statistics and Pearson Correlations Between the Study Variables (N = 241).
10
Western Journal of Nursing Research
Table 2. Standardized Measurement Loadings for the Latent Constructs (N = 241).
1. Task-oriented contact (CT scale) 2. Physical comfort (CT scale) 3. Emotional containment (CT scale) 4. Emotional exhaustion (burnout) 5. Cynicism (burnout) 6. Job satisfaction
Item 1a
Item 2
1.00
.72, p
WJNXXX10.1177/0193945914527356Western Journal of Nursing ResearchPedrazza et al.
Article
Nurses’ Comfort with Touch and Workplace Well-Being
Western Journal of Nursing Research 1–18 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0193945914527356 wjn.sagepub.com
Monica Pedrazza1, Stefania Minuzzo1, Sabrina Berlanda2, and Elena Trifiletti1
Abstract Touch is an essential part of caregiving and has been proved to be useful to reduce pain. Nevertheless, little attention has been paid to nurses’ perceptions of touch. The aim of this article was to examine the relationship between nurses’ feelings of comfort with touch and their well-being at work. A sample of 241 nurses attending a pain management training course completed a questionnaire, including the following measures: Comfort with Touch (CT) scale (task-oriented contact, touch promoting physical comfort, touch providing emotional containment), Maslach Burnout Inventory (MBI; emotional exhaustion, cynicism), and Job Satisfaction. Results of structural equation models showed that touch providing emotional containment was the main predictor of emotional exhaustion. Emotional exhaustion, in turn, was positively related to cynicism and negatively related to job satisfaction. In addition, the direct path from touch providing emotional containment to cynicism was significant. Practical implications of the findings are discussed. Keywords comfort with touch, physical touch, pain, burnout, job satisfaction
1University 2University
of Verona, Italy of Trento, Italy
Corresponding Author: Monica Pedrazza, Associate Professor, Department of Philosophy, Education, and Psychology, University of Verona, via San Francesco 22, Verona, 37129, Italy. Email: [email protected]
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
2
Western Journal of Nursing Research
Physical contact with the patient is an intrinsic part of the nursing profession (Chang, 2001; Routasalo, 1999). To provide appropriate assistance, nurses have recourse to various forms of contact that differ in their nature and degree of intimacy depending on whether their purpose is to carry out tasks (e.g., taking vital signs), provide physical comfort (e.g., massaging), or provide emotional containment (e.g., hugging patients; Pedrazza, Minuzzo, Trifiletti, & Berlanda, 2014). The ability of the professional to reassure the patient through touch is particularly important when the patient is suffering from serious pathologies, is in pain, is worried, afraid or anxious, or is physically and emotionally fragile and vulnerable (Bonacini & Marzi, 2005). This study shows how feeling comfortable in giving support to patients through touch may be associated with greater job satisfaction and reduced burnout (emotional exhaustion and cynicism).
Physical Touch in Caring One form of contact is cutaneous stimulation, for example, superficial massage of the back, hands, feet, or face. Cutaneous stimulation has been shown to be useful in reducing pain because it activates the large-diameter fibers that transmit tactile information, which in turn antagonize the small-diameter fibers that transmit the pain stimulus (Melzack, 1996). Cutaneous stimulation also stimulates the production of endorphins, relaxes the muscles, sedates the patient, and acts as a distracting stimulus. Evidence-based guidelines highlight the effectiveness of cutaneous stimulation in relieving pain associated with muscle tension or spasms (Grade C, Level IV; National Guideline Clearinghouse). Many studies have confirmed that cutaneous stimulation is useful in reducing pain and anxiety and inducing relaxation in oncology patients (Jane et al., 2011). Hand massage, on its own or together with soothing music, has been shown to reduce restless and aggressive behavior in demented subjects (Fu, Moyle, & Cooke, 2013; Remington, 2002), and to reduce pain and lower the heart rate and blood pressure in patients in an emergency department (Kubsch, Neveau, & Vandertie, 2001). Superficial hand and foot massage is also an effective, inexpensive, risk-free, flexible, and easily applied strategy for managing postoperative pain (Han & Lee, 2012; H. L. Wang & Keck, 2004). In general, touch is an essential part of caregiving and an excellent way of communicating attention, sympathy, closeness, reassurance, and presence (see Routasalo, 1999). This approach is particularly recommended for suffering patients, the frail and solitary elderly, the terminally ill, and the dying. It may be a way of taking charge of the patient to reduce psychological suffering, feelings of loneliness, difficulty in communicating, and the fear and
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Pedrazza et al.
3
anxiety of death (Bonacini & Marzi, 2005). Kübler-Ross (2008) claimed that gentle hand pressure is the most effective form of communication with dying patients. The positive effects of touch in caring are, therefore, well-established. However, little attention has been paid to how physical touch is perceived by patients and nurses (but see Hollinger-Smith & Buschmann, 1993). Some attention has been paid to the influence of gender on perceptions of touch in caring. Research has shown that although male patients perceive touch from female nurses more positively than do female patients, female nurses consider female patients more receptive to touch and feel more comfortable touching female than male patients (Lane, 1989). As to men nurses, the stereotype of men as sexual aggressor may generate a sense of vulnerability and a need for caution in the use of touch (Evans, 2002). More in general, touching the patient’s body may be associated with negative feelings. Picco, Santoro, and Garrino (2010), using data from in-depth interviews with 14 nurses, found an ambivalent attitude toward taking care of the patient’s body. They showed that although the body is regarded as a privileged element of nursing care, it is also perceived as a source of uneasiness and negative feelings. Physical touch requires great involvement on the part of nurses, because it implies physical, cognitive, and emotional proximity to the patient. Moreover, nurses may expect patients to refuse or react negatively to touch, as touching the patient’s body implies entering his or her privacy (Picco et al., 2010).Touching the patient may be therefore associated with different levels of comfort among nurses. Because touch is an integral part of the nurse–patient relationship (Hollinger-Smith & Buschmann, 1993; Routasalo, 1999) and is fundamental in the provision of nursing care (Chang, 2001), it is important to understand how nurses experience touch and how their feelings about touch relate to workplace well-being. If positive feelings of ease during contact are likely to be associated with well-being (e.g., greater job satisfaction, reduced burnout), a lack of comfort in touch may result in negative organizational outcomes and, consequently, in lower quality of care. The literature on physical touch generally distinguishes between two forms of touch: necessary and non-necessary (Routasalo, 1999). Taskoriented or necessary touch is intended for carrying out a task or procedure (e.g., taking vital signs); non-necessary touch, in contrast, is spontaneous and affective (e.g., reassuring a patient by holding his or her hand), and is not necessary for accomplishing a task. These two forms of contact obviously imply different degrees of cognitive and affective proximity to the patient. Based on previous literature on physical touch, Pedrazza et al. (2013) developed and validated the Comfort With Touch (CT) scale. The scale is articulated into five subscales: Task-Oriented Contact, Personal Care, Physical Comfort, Reassurance and Emotional Containment. The concept of
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
4
Western Journal of Nursing Research
emotional containment refers to the ability to receive, process, and respond with understanding to the emotional reactions of another person (Douglas, 2007). Three subscales of the CT scale were selected for the present study— Task-Oriented Contact, Physical Comfort and Emotional Containment— which correspond to low, medium and high levels of cognitive and affective proximity to the patient. This study aims to verify the effects of comfort with touch on two measures of workplace well-being—burnout and job satisfaction—which appear to be associated with a wide range of individual and organizational outcomes. Burnout is a reaction to prolonged exposure to interpersonal and environmental stressors on the job (Maslach, Schaufeli, & Leiter, 2001). Three dimensions have been generally distinguished: exhaustion, cynicism, and reduced personal efficacy. However, Schaufeli and Bakker (2004) argued that personal efficacy can be considered a dimension of work engagement rather than burnout. Moreover, researchers have found consistent results for emotional exhaustion and cynicism but not for personal efficacy (Laschinger, Finegan, & Wilk, 2011). Therefore, recent studies have adopted a two-factor conceptualization of burnout. The present study focuses on emotional exhaustion and cynicism as outcomes of (dis)comfort with physical touch. Emotional exhaustion, generally regarded as the core of burnout, is a state of physical and emotional depletion that manifests itself both in physical fatigue and a feeling of being “burned out” by one’s job. Sustained emotional exhaustion generally results in cynicism (Leiter & Maslach, 2004), namely, detachment from the job. Burnout in nursing has been widely investigated. Negative work conditions, such as nursing shortage, work overload, lack of job autonomy, and leader–member exchange, appear to be major determinants of burnout (Laschinger et al., 2011; Moustaka & Constantinidis, 2010). Although previous research on burnout has primarily focused on environmental factors, individual difference variables are also likely to play a role in the development of burnout (Maslach et al., 2001). In a meta-analysis, Alarcon, Eschleman, and Bowling (2009) found that self-core evaluations, agreeableness, consciousness, extraversion, optimism, hardiness, and proactive personality were negatively related to both emotional exhaustion and cynicism (for the relationship between self-core evaluations and burnout in nursing, see Laschinger et al., 2011). Positive affectivity yielded a negative association with the two burnout dimensions, whereas negative affectivity showed a positive relationship. Given the costs of burnout for both people and organizations, it is important to further explore its possible antecedents. The current study investigates, for the first time, how nurses’ feelings of comfort with physical touch relate to emotional exhaustion and cynicism. Job satisfaction is the affective orientation that an individual has about his or her work (Price, 2001). Satisfaction at work appears to be a key predictor of Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Pedrazza et al.
5
nurses’ turnover and intentions to leave (Blegen, 1993; Irvine & Evans, 1995; Lu, Barriball, Zhang, & While, 2012; Sourdif, 2004; L. T. H. Wang, Ellenbecker, & Liu, 2012) and is therefore important to understand its sources. Researchers have outlined a number of environmental and personality factors associated with job satisfaction. In a systematic review, Lu et al. (2012) found that organizational environment and working conditions (e.g., structural empowerment and leader–member exchange; Laschinger et al., 2011), role conflict and ambiguity, and organizational and professional commitment were main predictors of nurses’ job satisfaction. Moreover, Laschinger et al. (2011) found that both emotional exhaustion and cynicism were negatively related to nurses’ satisfaction at work. As to personal dispositional factors, Judge, Heller, and Mount (2002) reported significant correlations between job satisfaction and the Big Five traits of extraversion, neuroticism, agreeableness, and consciousness (for similar results, see Zhai, Willis, O’Shea, Zhai, & Yang, 2013). Self-core evaluations (Judge & Bono, 2001; Srivastava, Locke, Judge, & Adams, 2010), positive and negative affectivity, and affective disposition (Connolly & Viswesvaran, 2000) are also related to satisfaction at work. The present study examines for the first time the relationship between job satisfaction and comfort with touch among nurses. The purposes of the present study are (a) to examine whether nurses report similar levels of (dis)comfort in relation to the three dimensions of touch (taskoriented contact, touch promoting physical comfort, and touch aimed at emotional containment); (b) to examine the associations between the three dimensions of touch, emotional exhaustion, cynicism, and job satisfaction; (c) to test the influence of touch on emotional exhaustion, cynicism, and job satisfaction.
Method Design, Sample, and Procedure The study was approved by the ethics committee of the researchers’ institution. A cross-sectional survey design was adopted. A questionnaire was administered to a convenience sample of 400 registered nurses, working in the departments of medicine, surgery, intensive care, oncology, and geriatrics in several hospitals in Northeast Italy. Data were collected during 2010. Participants were contacted during a 2-day pain management training course held by one of the authors. They were given the questionnaire at the end of the first day and asked to return it on the following day. Informed consent was obtained from each participant. The final sample included 241 nurses (241 of 400; 60.25% return rate). The mean age was 41.14 years (SD = 7.65, range = 24-60). Females comprised 85.9% of the sample (207 of 241). The mean length of service was 17.37 years (SD = 8.07, range = 1-35). Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
6
Western Journal of Nursing Research
Instrument The survey contained the following measures. CT scale. Nurses’ feelings of ease with touch were measured with eight items (Pedrazza et al., 2014). Each item describes a specific form of contact with the patient. Nurses are asked to indicate to what extent they feel at ease at performing each contact behavior. Responses are given on a 7-point scale, ranging from 1 (not at all) to 7 (very much). The measure is articulated into three subscales. Two items measure task-oriented contact (e.g., “Touching the patient to take his or her pulse”); three items measure touch aimed at promoting physical comfort (e.g., “Massaging the patient’s hands to reduce pain”); three items measure touch aimed at emotional containment (e.g., “Letting the patient cry in my arms”). Pedrazza et al. (2014) confirmed the construct validity of the three-factor scale using exploratory factor analysis (EFA). Cronbach’s alpha reliabilities in their study ranged between.91 (taskoriented contact) and .96 (emotional containment). For the present study, Cronbach’s alphas were .78 (task-oriented contact), .83 (promotion of physical comfort), and .87 (emotional containment). Burnout. The Emotional Exhaustion and Cynicism scales of the Maslach Burnout Inventory (MBI; Maslach & Jackson, 1981) were used in the Italian version validated by Sirigatti and Stefanile (1993). Six items measured emotional exhaustion (e.g., “I feel burned out from my work”), and five items measured cynicism (e.g., “I feel I treat some recipients as if they were impersonal objects”). Responses were given on a 7-point scale, ranging from 1 (completely disagree) to 7 (completely agree). Cronbach’s alpha coefficients for the Italian adaptation are generally higher than .87 for emotional exhaustion and .67 for cynicism (e.g., Martini & Converso, 2012). In this study, Cronbach’s alphas were .87 (emotional exhaustion) and .72 (cynicism). Job satisfaction. Job satisfaction was measured with four items (e.g., “I am satisfied with my job”), adapted from Dazzi, Voci, Capozza, and Bergamin’s (1998) scale. The 7-point response scale ranged from 1 (completely disagree) to 7 (completely agree). Bobbio, Manganelli Rattazzi, and Muraro (2007) demonstrated the construct validity of the one-factor scale with EFA and reported a Cronbach’s alpha of .86. The Cronbach’s alpha coefficient for this study was .64.
Data Analysis Data analysis was performed using SPSS 21.0 and LISREL 8.7 (Jöreskog & Sörbom, 2004). PRELIS (LISREL 8.7) was used for the imputation of
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Pedrazza et al.
7
Figure 1. Hypothesized model of the effects of comfort with touch on emotional exhaustion, cynicism, and job satisfaction.
missing data with the Expectation-maximization (EM) algorithm. Only 1.44% of the total responses (80 of 5,543) were missing scores. Following Schafer and Graham’s (2002) recommendations, maximum likelihood imputation (EM algorithm) was used to estimate values for missing scores. Maximum likelihood procedures provide more accurate estimates of population parameters than list-wise deletion or mean substitution (Schafer & Graham, 2002). The validity of the CT scale was tested by applying CFA (LISREL 8.7) to the eight items. Descriptive statistics were obtained with SPSS 21.0. For each variable, a composite score was computed by averaging the respective items. To test whether nurses reported similar levels of (dis) comfort in relation to task-oriented contact, touch promoting physical comfort, and touch aimed at emotional containment, a one-way repeated-measures ANOVA (SPSS 21.0) was applied, using the Bonferroni correction for multiple comparisons. Pearson correlation (SPSS 21.0) was used to examine the relationship between the three CT subscales and well-being outcomes (emotional exhaustion, cynicism, and job satisfaction). Structural equation modeling (SEM; LISREL 8.7) was applied to test the relationship between the three CT subscales and well-being outcomes. It was also examined whether cynicism and job satisfaction are predicted by emotional exhaustion, as found by previous research (Laschinger et al., 2011). To this aim, a mediation model was tested, in which CT subscales predict emotional exhaustion, which, in turn, predicts cynicism and job satisfaction (see Figure 1). Two parcels were computed for each of the following constructs: physical comfort and emotional containment (CT scale), emotional exhaustion, cynicism, and
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
8
Western Journal of Nursing Research
job satisfaction. Parceling is a measurement approach commonly used in latent-variable analysis techniques (Little, Cunningham, Shahar, & Widaman, 2002). A parcel is an aggregate-level indicator formed by averaging two or more items. The use of parcels limits measurement error and provides a good ratio of participants to variables. Moreover, compared with items, aggregatelevel data have some psychometric merits, such as higher reliability and a lower likelihood of distributional violations. Items also have fewer, larger, and less equal intervals between scale points than do parcels (Little et al., 2002). For task-oriented contact, the two respective items were used as indicators. In testing the model, the regression paths from the three CT subscales to cynicism and job satisfaction (residual direct effects) were estimated. The goodness of fit of the CFA and the SEM model was evaluated using the chisquare test, the comparative fit index (CFI), and the standardized root mean squared residual (SRMR). A model fits the data well when chi-square is nonsignificant, CFI is ≥.95, and SRMR is ≤.08 (Hu & Bentler, 1999). Mediation was tested with the Sobel test. For all analyses, the p value for statistical significance was .05.
Results The fit of the model for the three-factor structure of the CT scale (CFA) was acceptable: χ2(17) = 90.40, p ≅ .00; SRMR = .059; CFI = .96. Although chisquare was significant, the other two indices satisfied the respective criterion. Loadings were all significant and higher than .75 (p < .001). The correlation (phi coefficient) between task-oriented contact and emotional containment was .50 (95% confidence interval [CI] = [.40, .60]; p < .001). The correlation between task-oriented contact and physical comfort was .63 (95% CI = [.53, .73], p < .001). The correlation between physical comfort and emotional containment was 71 (95% CI = [.67, .75], p < .001). None of the 95% CIs included 1 (the perfect correlation), thus indicating that the three components actually reflected three distinct dimensions. Means and standard deviations for the study variables are reported in Table 1. To test whether nurses reported different levels of comfort in relation to the three CT subscales, a one-way repeated-measures ANOVA was applied. Using the Greenhouse–Geisser correction, we found that the three subscales differed significantly in the degree of perceived comfort, F(1.843, 442.393) 2 = 143.35, p < .001, ηp = .37. The highest perception of comfort was reported in relation to task-oriented contact (M = 6.72, SD = 0.72), followed by touch aimed at physical comfort (M = 5.93, SD = 1.37) and touch aimed at emotional containment (M = 5.35, SD = 1.60). Multiple comparisons with the Bonferroni correction were all significant, p < .001.
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
Downloaded from wjn.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on June 12, 2015
9
1.30 1.04 1.23
2.78 1.95 4.96
Note. CT = Comfort with Touch.
0.72 1.37 1.60
6.72 5.93 5.35
1. Task-oriented contact (CT scale) 2. Physical comfort (CT scale) 3. Emotional containment (CT scale) 4. Emotional exhaustion (burnout) 5. Cynicism (burnout) 6. Job satisfaction
SD
M
Measure — .66, p < .001
2
—
3
4
5
6
— −.13, p = .04 −.28, p < .001 −.29, p < .001 — −.14, p = .03 −.21, p = .001 −.32, p < .001 .54, p < .001 .19, p = .03 −.54, p < .001 −.34, p < .001 — .12, p = .064 .14, p = .03
— .61, p < .001 .47, p < .001
1
Table 1. Descriptive Statistics and Pearson Correlations Between the Study Variables (N = 241).
10
Western Journal of Nursing Research
Table 2. Standardized Measurement Loadings for the Latent Constructs (N = 241).
1. Task-oriented contact (CT scale) 2. Physical comfort (CT scale) 3. Emotional containment (CT scale) 4. Emotional exhaustion (burnout) 5. Cynicism (burnout) 6. Job satisfaction
Item 1a
Item 2
1.00
.72, p
