Health Policy and Systems Article Critique Paper

HEALTH POLICY AND SYSTEMSNurse Burnout, Nurse-Reported Quality of Care, and Patient
Outcomes in Thai Hospitals
Apiradee Nantsupawat, PhD, RN1 , Raymoul Nantsupawat, PhD, RN2 , Wipada Kunaviktikul, PhD, RN, FAAN3 ,
Sue Turale, DEd, RN, FACN, FACMHN4 , & Lusine Poghosyan, PhD, MPH, RN5
1 Phi Omega, Instructor, Faculty of Nursing, Chiang Mai University, Chiang Mai, Thailand
2 Phi Omega, Associate Professor, Faculty of Nursing, Chiang Mai University, Chiang Mai, Thailand
3 Phi Omega, Professor, Faculty of Nursing, Chiang Mai University, Chiang Mai, Thailand
4 Tau Nu, Professor, Faculty of Nursing, Chiang Mai University, Chiang Mai, Thailand
5 Alpha Zeta, Assistant Professor, Columbia University, School of Nursing, New York, NY, USA
[Correction made after online publication December 9, 2015: Author name has been updated.]
Key words
Nurse burnout, patient outcomes, quality of
care, Thailand
Correspondence
Dr. Apiradee Nantsupawat, Faculty of Nursing,
Chiang Mai University, 110/406 Inthawaroros
Rd. Sriphum, Chiang Mai, 50200, Thailand.
E-mail: apiradee.n@cmu.ac.th
Accepted: October 29, 2015
doi: 10.1111/jnu.12187
Abstract
Purpose: The purpose of this study was to investigate the effect of nurse
burnout on nurse-reported quality of care and patient adverse events and outcomes in Thai hospitals.
Methods: Cross-sectional analysis of data from 2,084 registered nurses working in 94 community hospitals across Thailand. Data were collected through
survey questionnaire, including the Maslach Burnout Inventory (MBI), which
measures of nurse perceived quality of care and patient outcomes. Multiple
logistic regression modeling was performed to explore associations between
nurse burnout on quality of care and patient outcomes.
Findings: Thirty-two percent of nurses reported high emotional exhaustion,
18% high depersonalization, and 35% low personal accomplishment. In addition, 16% of nurses rated quality of care on their work unit as fair or poor,
5% reported patient falls, 11% reported medication errors, and 14% reported
infections. All three subscales of the MBI were associated with increased reporting of fair or poor quality of care, patient falls, medication errors, and infections. Every unit of increasing emotional exhaustion score was associated
with a 2.63 times rise in reporting fair or poor quality of care, a 30% increase
in patient falls, a 47% increase in medication errors, and a 32% increase in
infection.
Conclusions: Findings clearly indicate that nurse burnout is associated with
increased odds of reporting negative patient outcomes. Implementing interventions to reduce nurse burnout is critical to improving patient care in Thai
hospitals.
Clinical Relevance: Hospital administrators, nurse managers, and health
leaders urgently need to create favorable work environments supporting nursing practice in order to reduce burnout and improve quality of care.
Burnout is a prolonged psychological reaction to chronic
emotional and interpersonal stressors on the job and
is defined by three aspects: emotional exhaustion, depersonalization, and reduced personal accomplishment
(Maslach, 1982). Previous work has found a high prevalence of burnout among nurses globally (Poghosyan,
Journal of Nursing Scholarship, 2016; 48:1, 83–90.

C 2015 Sigma Theta Tau International
Clarke, Finlayson, & Aiken, 2010), and various factors,
including poor work environments and low staffing, have
been reported to lead to burnout among nurses. For
example, nurses often work long irregular hours, have
high workload and work demand, and have traumatic
experiences (Oulton, 2006; Sundin, Hochwalder, Bilt, &
83
Nurse Burnout and Patient Outcomes
Lisspers, 2007). Also, poor work environments and low
staffing challenge nurses’ ability to deliver care of the
highest quality, which might further increase stress and
burnout (Sundin et al., 2007). Studies demonstrate that
nurse burnout has negative consequences not only for
patient care and safety (Laschinger & Leiter, 2006; Teng,
Shyu, Chiou, Fan, & Lam, 2010; You, Aiken, Sloane, Liu,
& He, 2013), but can also affect the overall organization
by decreasing the productivity of employees in terms of
increased absenteeism, increased turnover, and reduced
quality of services (Borritz et al., 2006). On the other
hand, reducing nurse burnout has been found to have a
positive impact on patient care, such as reducing patient
infections by 30% (Cimiotti, Aiken, Sloane, & Wu, 2012).
As evidence is clear that nurse burnout has negative
consequences for care, research has examined the
relationships between burnout, quality of care (QoC),
and patient outcomes in many countries; however,
Thailand has been overlooked by the literature despite
their experiencing a nursing shortage, which in other
countries has created challenging work environments
contributing to burnout (Almalki, FitzGerald, & Clark,
2011). Examining nurse burnout in Thailand can help
identify practices that promote nurse work environment,
improve patient safety and care, and improve quality of
services. Furthermore, this research can advance health
outcomes in Thai hospitals as well as the development
of the nursing profession in Thailand. Here we examine
the relationship between nurse burnout and patient
outcomes in Thai community hospitals.
Background
Burnout was first defined by Freudenberger (1974) as
a state of fatigue or frustration resulting from professional
relationships that failed to produce expected rewards.
Maslach (1982) later expressed burnout as a psychological syndrome characterized by emotional exhaustion,
depersonalization, and diminished personal accomplishment. Emotional exhaustion refers to a lack of energy and
a feeling that one’s emotional resources are used up due
to excessive psychological demands. Depersonalization is
regarded as the treatment of others as objects, rather than
people, through cynical, callous, or uncaring attitudes
and behaviors. Diminished personal accomplishment denotes a tendency to evaluate oneself negatively because
of failure to produce results (Maslach, Jackson, & Leiter,
1996). Today, the Maslach Burnout Inventory is the most
widely used tool to measure nurse burnout internationally (Poghosyan et al., 2010), and prior evidence indicates
that nurses in hospitals experience high levels of jobrelated burnout (McHugh, Kutney-Lee, Cimiotti, Sloane,
& Aiken, 2011), which may affect patient outcomes.
84
Nantsupawat et al.
Patient outcomes are considered nurse sensitive if
they improve because of increased levels of nursing care
(American Nurses Association [ANA], 2010). Studies
looking at nurse-sensitive patient outcomes particularly
investigated patient falls, infections, and medication
errors (Ausserhofer et al., 2013; Zhu et al., 2012). These
outcomes are important indicators for care quality and
are directly affected by nursing care. Medication errors
can originate at various stages of patient care, including
at dispensing or administering, and effective nursing
care can prevent them. In addition, patient falls that
are defined as an unplanned descent to the floor with
or without injury to the patient (National Database
of Nursing Quality Indicators, 2013), and hospitalacquired infections, which patients acquire during their
hospital stay (World Health Organization, 2002), can
be directly linked to nursing care. Moreover, patient
falls and infections are included in the nursing care
performance measures adopted by the National Quality
Forum (2004, 2009) and ANA (2002, 2010). Previous
work has investigated nurse burnout and its effect on
patient outcomes. For example, the study of Laschinger
and Leiter (2006) suggested that nurse burnout is related to self-reported adverse events. Other work has
demonstrated an association between nurse burnout
and patient falls, medication errors (Van Bogaert et al.,
2014), and nosocomial infections (Cimiotti et al., 2012).
Researchers collect data about patient outcomes using various data sources; however, most nursing studies
demonstrate that nurses are a reliable source of information regarding the QoC they deliver (Aiken, Sloane,
Bruyneel, Van den Heede, & Sermeus, 2012; You et al.,
2013). Nurses perceive QoC as the degree to which patients’ physical, psychosocial, and extra care needs are
met (Williams, 1998), and they can provide valuable
information about the QoC they deliver. Nurse ratings
of QoC aggregated to the hospital level provide related
yet distinct information about patient outcomes when
compared with evidence derived from administrative
databases (Aiken, Clarke, & Sloane, 2002). Nurses’ assessments of overall quality have been used in a number of
studies, and researchers found it to be strongly associated
with patient outcomes (Aiken et al., 2012; Lucero, Lake,
& Aiken, 2010; McHugh & Witkoski Stimpfel, 2012).
Thus, in this study we focus on nurse-reported QoC and
patient outcomes.
Prior studies have also examined the relationship between nurse burnout and QoC. For instance, Poghosyan
et al. (2010) examined the relationship between nurse
burnout and ratings of QoC in 53,846 nurses from
six countries: Canada, Germany, Japan, New Zealand,
the United Kingdom, and the United States. Findings
indicated that across countries, higher levels of burnout
Journal of Nursing Scholarship, 2016; 48:1, 83–90.

C 2015 Sigma Theta Tau International
Nurse Burnout and Patient Outcomes
Nantsupawat et al.
were associated with poor QoC. Other work by Van
Bogaert, Clarke, Roelant, Meulemans, and Van de
Heyning (2010), Van Bogaert, Kowalski, Weeks, Van
Heusden, and Clarke (2013), and Van Bogaert et al.
(2014) corroborated these findings, reporting that QoC
was associated with important dimensions of burnout,
specifically emotional exhaustion and depersonalization.
A body of empirical literature has identified characteristics of practice environments that positively and negatively affect nurse burnout. For example, a number of
studies have indicated that promoting nursing leadership, nursing foundations for QoC, nurse staffing and resources, and collegial nurse–physician relations, as well
as enhancing nursing participation in management and
decision making, may reduce nurse burnout (KanaiPak, Aiken, Sloane, & Poghosyan, 2008; Van Bogaert
et al., 2013). On the other hand, high workloads and
low staffing levels have been identified as antecedents of
nursing burnout (Aiken et al., 2011; Doef, Mbazzi, & Verhoeven, 2012; Lang, Patrician, & Steele, 2012; Liu et al.,
2012; Teng et al., 2010).
In Thailand, concern of burnout among nurses has
grown due to nursing staff shortages, particularly in
community hospitals. Most Thai hospitals are operated
by the Ministry of Public Health and fall into one of
three classifications: community, general, or regional
hospitals (Bureau of Policy and Strategy, Ministry of
Thai Public Health, 2014). Community hospitals, the
settings for data collection in this study, are located at
the district level and typically have a capacity of 10 to
150 beds. They provide general or specialized nonurgent
and short-term care to patients and refer those patients
in need of more advanced care to larger general or
regional hospitals. Three years ago it was predicted that a
further 10,446 replacement nurses would be required for
these community hospitals (Division of Administration,
Ministry of Public Health, 2011). A recent Thai study
described that a nurse in community hospitals cared for
approximately 11 patients and on average worked 55 hr
per week (Nantsupawat, Nantsupawat, Kulnaviktikul,
& McHugh, 2014). Extended nursing hours of work is a
common issue in these hospitals since 80% of registered
nurses (RNs) reported working 20 more hours beyond
the usual 40 hr per week, and 82% of extended work
hours had been assigned to nurses by their administrators on work-shift rotation (Supamanee, Kunaviktikul,
& Keitlertnapha, 2014). Additionally, these extended
hours were negatively correlated with nurses’ work–life
balance and adequacy of rest and sleep, which led to
their physical impairment and burnout.
Although prior researchers have established an association between nurse burnout, QoC, and patient
outcomes, this association has not been shown in
Journal of Nursing Scholarship, 2016; 48:1, 83–90.

C 2015 Sigma Theta Tau International
Thailand. We argue that delivering high-quality care to
all patients in healthcare settings is a priority. To ensure
better patient outcomes and patient safety, it is vital to
investigate factors promoting or hindering the delivery
of QoC to identify measures to change the status quo,
and keep nurses in nursing without feeling burnt out.
Therefore, the aim of this study was to explore the effect
of nurse burnout on nurse-reported QoC and patient
adverse events and outcomes.
The conceptual framework for this study is based on
the work of Donabedian 1988. Donabedian suggested
that outcome is a result of structure and process. In this
study, the structure variables are the characteristics of
patient units and hospitals (e.g., organizational environment), while the process variables are defined as nursing
care, and outcomes are patient outcomes (e.g., patient
falls, infections, medication errors). The possible explanation is that burnout results from the gap between individuals’ expectations to fulfill their professional roles and the
failure of organizational structures (Leiter, 1991, 1992).
When the organizational environment does not support
nurse practice or allow nurses to deliver care according
to nursing philosophy of care, nurses may feel emotionally overextended and exhausted, develop impersonal
responses toward patients, and experience less competence and successful achievement in their work with
patients. These feelings may contribute to adverse
job performance, which may lead to adverse events
such as patient falls, infections, and medication errors
(outcomes).
Methods
Study Design and Sampling
Using a cross-sectional design, we surveyed nurses
working in 92 community hospitals across Thailand. Administrators from every hospital with at least 90 beds
were asked to distribute the survey instrument among
their registered nursing staff. Nurses were eligible to participate in the study if they were RNs providing direct patient care in inpatient units and if they had worked in
their positions for at least 1 year. Written explanation of
the study was given along with the research instrument,
and nurses choosing to participate gave written informed
consent; 2,450 surveys were distributed. The nurses were
surveyed between May and July 2012. Of 2,415 RNs
who returned the survey via mail (the response rate was
98.6%), 2,084 met the inclusion criteria and returned
fully completed forms. This study was approved by the
Faculty of Nursing and the Institutional Review Board at
Chiang Mai University and was endorsed by all participating hospitals. Confidentiality of data was maintained
throughout the study.
85
Nurse Burnout and Patient Outcomes
Nantsupawat et al.
Table 1. Percentages of Nurses Reporting Burnout
MBI-HSS Subscale
High Emotional Exhaustion (EE score > 27)
High Depersonalization (DP score > 10)
Low Personal Accomplishment (PA score < 5) n (%) 671 (32.2) 371 (17.8) 729 (34.5) Note. MBI-HSS = Maslach Burnout Inventory Human Service Survey. Measurement Burnout. The Maslach Burnout Inventory Human Service Survey (MBI-HSS; Maslach & Jackson, 1986) was used to measure nurse burnout. Mind Garden, Inc. translates and adapts the MBI-HSS to be used in research in Thailand. We used the Thai version of the MBI-HSS in this study. The 22-item tool measures burnout using three subscales: Emotional Exhaustion (EE), Depersonalization (DP), and Reduced Personal Accomplishment (PA). Each item asks respondents about the reports of job-related feelings (e.g., “I feel emotionally drained from my work,” “I do not really care what happens to my patients,” “I have accomplished many worthwhile things in this job”), and their feelings or experiences are rated on a 7-point scale ranging from 1 = “never having them,” to 7 = “having them every day.” Levels of burnout are estimated separately for EE (nine items), DP (five items), and PA (eight items) by using numerical cutoffs listed on the scoring key (Maslach et al., 1996). EE scores equal to or higher than 27 indicate high emotional exhaustion, DP scores greater than or equal to 10 suggest high depersonalization, and PA scores less than 5 indicate low personal accomplishment. Mean scores of 19 to 26 on the EE subscale, 6 to 9 on the DP subscale, and 34 to 39 on the PA subscale demonstrate evidence of average burnout. Mean scores equal to or lower than 18 on the EE subscale, equal to or lower than 5 on the DP subscale, and equal to or higher than 40 on the PA subscale demonstrate evidence of low burnout. The MBI-HSS has high internal consistency reliability measured by Cronbach’s alphas ranging from 0.71 to 0.90 (Maslach et al., 1996). Reliability coefficients (Cronbach’s alphas) for the present study were 0.91 for the EE subscale, 0.77 for the DP subscale, and 0.84 for the PA subscale. Quality of care and patient outcomes. Nurseperceived QoC and patient outcomes were measured by question items with four response categories. RNs were asked to assess QoC during their last shift using a 4-point scale, with higher scores indicating poorer levels of quality (1 = “very good,” 2 = “good,” 3 = “fair,” 4 = “poor”). These questions are used widely in international research (Bruyneel, Van den Heede, Diya, Aiken, & Sermeus, 2009). The item measure of QoC has been 86 presented to be strongly related to measures of patient outcomes (Laschinger, 2008; McHugh & Witkoski Stimpfel, 2012). A single question item also measured respondents’ perception of patient outcomes on their unit. In this study, patient outcomes included patient falls, medication errors, and nosocomial infections. Respondents were asked to rate the degree to which these patient outcomes occurred using a 4-point scale (1 = “never,” 2 = “rarely,” 3 = “sometimes,” 4 = “often”). Previous research has demonstrated that nurses’ reports of frequency of adverse events are associated with measures of care quality (Lucero et al., 2010). Demographic characteristics including age, sex, education, and number of years working as an RN were also collected. Data Analysis Descriptive statistics were used to calculate the means and standard deviations for continuous variables and frequencies and percentages for categorical variables. Logistical regression models were used in both bivariate and multivariate analyses to obtain odds ratios for nurse-reported patient outcomes in relation to nurse burnout. First, the bivariate relationship between patient outcomes and each MBI-HSS subscale was examined. Next, multiple logistic regression models were built to study the relationship between each MBI-HSS subscale and the outcome variables. The models were adjusted for age, sex, education, and number of years working as an RN. All models used a generalized estimating equation to account for the clustering of nurses within hospitals. All analyses were completed using STATA 10.1 (STATA Corp., College Station, TX, USA). The statistical level was specified at p < .05. Results The participants in our study were predominantly female (82%), with an average age of 33 years. All nurses had baccalaureate degrees. The average length of work experience as RNs was 9 years, ranging from 1 to 36 years. The proportion of RNs reporting burnout for the three dimensions of MBI-HSS are shown in Table 1. Approximately 32% of nurses had high emotional exhaustion, 18% had high depersonalization, and 35% had low personal accomplishment. Table 2 describes nurses’ reports on QoC and on occurrence of patient falls, medication errors, and infections on their units. Sixteen percent of nurses reported the QoC delivered on their units as poor or fair. Table 3 displays odds ratios and corresponding confidence intervals (confidence level = 95%) from Journal of Nursing Scholarship, 2016; 48:1, 83–90.  C 2015 Sigma Theta Tau International Nurse Burnout and Patient Outcomes Nantsupawat et al. Table 2. Nurses’ Reports of Quality of Care and Patient Outcomes n % 339 1,744 16.3 83.7 106 1,978 5.1 94.9 219 1,865 10.5 89.5 284 1,800 13.6 86.4 Outcomes Quality of care Poor/fair Very good/good Patient falls Sometimes/often Never/rarely Medication errors Sometimes/often Never/rarely Infection Sometimes/often Never/rarely unadjusted and adjusted models predicting nurses’ ratings of patient outcomes from the three dimensions of burnout. After controlling for nurse characteristics (age, sex, education, years as an RN), high emotional exhaustion was associated with increased odds of reporting QoC as fair or poor (2.63 times) and increased odds of reporting medication errors and infections (1.47 and 1.32 times, respectively). High depersonalization was
also associated with increased odds of reporting QoC as
fair or poor (3.21 times) and increased odds of reporting
medical errors and infections (1.83 and 1.74 times,
respectively). In addition, high depersonalization showed
an association with a 2.06 times increase in the odds of
nurses reporting patient falls. Lastly, low personal accomplishment among nurses was associated with a 1.73 times
increase in the odds of reporting QoC as fair or poor, a
1.61 times increase in the odds of reporting patient
falls, and a 1.49 times increase in the odds of reporting
medication errors.
Discussion
To our knowledge, this is the first study investigating
nurse burnout in relation to patient outcomes in the context of community hospitals in Thailand. After controlling
for nurse characteristics, results demonstrated an association between all three dimensions of burnout (emotional
exhaustion, depersonalization, and personal accomplishment) and nurses’ perceptions of adverse patient outcomes on their units. Specifically, high burnout increased
the odds of RNs reporting poor QoC, patient falls, medication errors, and infections. The study results are consistent with previous research showing that high levels
of burnout are associated with reduced QoC (Poghosyan
et al., 2010; Van Bogaert et al., 2010, 2013, 2014), increased medication errors, increased patient falls (Van
Bogaert et al., 2014), and increased infections (Cimiotti
et al., 2012) and demonstrate a similar link between
nurse burnout and patient outcomes in the Thai setting.
The results also show that around one third of nurses
report high emotional exhaustion and low personal accomplishment, and about two fifths experience high depersonalization. A possible explanation for these findings
in this setting is that community hospitals, one type of
Table 3. Logistic Regression Models Testing the Relationship Between Nurse Burnout, Quality of Care, and Patient Outcomes
Model 1
Patient outcomes
Poor/fair quality of care
High emotional exhaustion
High depersonalization
Low personal accomplishment
Patient falls
High emotional exhaustion
High depersonalization
Low personal accomplishment
Medication errors
High emotional exhaustion
High depersonalization
Low personal accomplishment
Infections
High emotional exhaustion
High depersonalization
Low personal accomplishment
Model 2
OR (95% CI)
p
OR (95% CI)
p
2.63 (2.05–3.37)∗∗∗
3.19 (2.46–4.14)∗∗∗
1.72 (1.34–2.21)∗∗∗
.000
.000
.000
2.63 (2.07–3.34)∗∗∗
3.21 (2.46–4.19)∗∗∗
1.73 (1.36–2.19)∗∗∗
.000
.000
.000
1.32 (0.82–2.11)
2.07 (1.34–3.18)∗∗∗
1.61 (1.15–2.26)∗∗
.245
.001
.005
1.31 (0.87–1.98)
2.06 (1.33–3.20)∗∗∗
1.61 (1.08–2.40)∗
.181
.001
.017
1.47 (1.05–2.07)∗∗
1.83 (1.34–2.48)∗∗∗
1.49 (1.13–1.96)∗∗
.025
.000
.004
1.47 (1.10–1.97)∗∗
1.83 (1.31–2.55)∗∗∗
1.49 (1.12–1.99)∗∗
.009
.000
.006
1.33 (1.00–1.75)∗
1.75 (1.28–2.39)∗∗∗
1.22 (0.93–1.61)
.044
.000
.142
1.32 (1.02–1.72)∗
1.74 (1.29–2.34)∗∗∗
1.22 (0.94–1.58)
.033
.000
.121
Note. Model 1 = not controlling for nurse characteristics; Model 2 = controlling for nurse characteristics (age, sex, education, years as registered nurse).
CI = confidence interval; OR = odds ratio. ∗ p ࣘ .05; ∗∗ p ࣘ .01; ∗∗∗ p ࣘ .001.
Journal of Nursing Scholarship, 2016; 48:1, 83–90.

C 2015 Sigma Theta Tau International
87
Nurse Burnout and Patient Outcomes
public hospital, function as the front-line public hospital
providing basic health care for the Thai population. Researchers in Thailand have indicated that high workload
and working extended hours are common in community
hospitals (Nantsupawat et al., 2014; Supamanee et al.,
2014). This may explain nurses experiencing feelings of
being overextended and depleted of their emotional and
physical resources, having uncaring attitudes towards the
recipients of one’s service, and/or feelings of incompetence and a lack of achievement and productivity at work.
Consequently, such feelings may affect nurses by decreasing the effective and efficient performance of work since
their physical and mental wealth may be diminished. And
this suboptimal performance may affect QoC and patient
safety.
The results from this study support previous studies
demonstrating that when nurses experience inadequate
resources from practice environment or staffing. They
may feel emotional exhaustion, depersonalization, and
diminished personal accomplishment, and resulting negative attitudes and emotions towards their job, thereby
reducing job performance and probably threatening
patient outcomes.
One limitation of this study is its cross-sectional design,
which does not confirm a causal link between variables.
Future research should investigate this linkage, and
longitudinal studies would probably generate more solid
evidence. Another limitation is that the adverse patient
outcomes were assessed from nurse-reported measures,
which may increase the possibility of response bias. Thus,
evaluating clinical outcomes from objective measurements should be considered, perhaps in a triangulated
study design. Finally, the sample may not be generally
representative of all Thai nurses because only inpatient
unit nurses were participants.
Implications of the Study
Our study results show that nurse burnout was
prevalent among nurses in Thai community hospitals,
and that high levels of burnout were associated with
negative outcomes for patients. The findings of this study
provide insights for hospital managers, policy makers,
and nursing administrators to take actions to reduce
nurse burnout, and consequently promote patient safety.
Thus, they need to take actions to improve working
conditions for nurses who provide care at the bedside,
such as promoting nurses’ perceptions of being supported
in their work settings, having a sense of accomplishment,
and being satisfied with their work, empowering them to
manage their own work, collaborate effectively in teams,
and deliver high-quality care.
88
Nantsupawat et al.
Conclusions
This study conducted a cross-sectional survey of
nurses practicing in Thai community hospitals. The study
found that burnout is high among these nurses and
has a significant negative impact on quality of patient
care, leading to higher patient falls and infections, all
of which may threaten patient safety. Our findings add
to global research that studies nurses around the world
who are dealing with burnout. Results of this study also
support previously published literature proposing the
relationship between burnout and patient outcomes.
Administrators and policy makers should take actions
to improve the work environment of nurses in these
hospitals, which subsequently will reduce nurse burnout
and promote patient safety. Future research is needed to
better understand how to promote patient safety behind
the phenomena in this study.
Acknowledgments
This study was funded by Chiang Mai University,
Thailand.
Clinical Resources
r Maslach
Burnout
Inventory:
http://www.
mindgarden.com/search?controller=search&
orderby=position&orderway=desc&search_query=
burnout
r Ministry of Public Health Thailand: http://eng.
moph.go.th/
References
Aiken, L. H., Clarke, S. P., & Sloane, D. M. (2002). Hospital
staffing, organization, and quality of care: Cross-national
findings. International Journal for Quality in Health Care,
14(1), 5–13.
Aiken, L. H., Sloane, D. M., Bruyneel, L., Van den Heede, K.,
& Sermeus, W. (2012). Nurses’ reports of working
conditions and hospitals quality of care in 12 countries
in Europe. International Journal of Nursing Studies, 50,
143–153.
Aiken, L. H., Sloane, D. M., Clarke, S., Poghosyan, L., Cho, E.,
You, L., . . . Aungsuroch, Y. (2011). Importance of work
environment on hospital outcome in nine countries.
International Journal for Quality in Health Care, 23(4),
357–364.
Almalki, M., FitzGerald, G., & Clark, M. (2011). The nursing
profession in Saudi Arabia: An overview. International
Nursing Review, 58(3), 304–311.
Journal of Nursing Scholarship, 2016; 48:1, 83–90.

C 2015 Sigma Theta Tau International
Nantsupawat et al.
American Nurses Association. (2002). Nursing-sensitive indica
tors. Retrieved from http://www.nursingworld.org/
MainMenuCategories/ThePracticeofProfessionalNursing/
PatientSafetyQuality/Research-Measurement/The-NationalDatabase/Nursing-Sensitive-Indicators_1
American Nurses Association. (2010). The national data base of
nursing-sensitive indicators. Retrieved from http://www.
nursingworld.org/MainMenuCategories/ANAMarketplace/
ANAPeriodicals/OJIN/TableofContents/Volume122007/
No3Sept07/NursingQualityIndicators.html
Ausserhofer, D., Schubert, M., Desmedt, M., Blegen, M. A.,
De Geest, S., & Schwendimann, R. (2013). The association
of patient safety climate and nurse-related organizational
factors with selected patient outcomes: A cross-sectional
survey. International Journal of Nursing Studies, 50, 240–252.
Borritz, M., Rugulies, R., Bjorner, J. B., Villadsen, E.,
Mikkelsen, O. A., & Kristensen, T. S. (2006). Burnout
among employees in human service work: Design and
baseline findings of the PUMA study. Scandinavian Journal
of Public Health, 34, 49–58.
Bruyneel, L., Van den Heede, K., Diya, L., Aiken, L., &
Sermeus, W. (2009). Predictive validity of the international
hospital outcomes study questionnaire: An RN4CAST pilot
study. Journal of Nursing Scholarship, 41(2), 202–210.
Bureau of Policy and Strategy, Ministry of Thai Public Health.
(2014). General information of health services. Retrieved from
http://bps.ops.moph.go.th/Healthinformation/
index.htm
Cimiotti, J., Aiken, L. H., Sloane, D. M., & Wu, E. S. (2012).
Nurse staffing, burnout, and health care-associated
infection. American Journal of Infection Control, 40, 486–490.
Division of Administration, Ministry of Thai Public Health.
(2011). Summary of health care workforce. Nonthaburi,
Thailand: Ministry of Thai Public Health.
Doef, M. V. D., Mbazzi, F. B., & Verhoeven, C. (2012). Job
conditions, job satisfaction, somatic complaints and
burnout among East African nurses. Journal of Clinical
Nursing, 21, 1763–1775.
Donabedian, A. (1988). The quality of care. Journal of the
American Medical Association, 260(12), 1743–1748.
Freudenberger, H. J. (1974). Staff burnout. Journal of Social
Issues, 30, 159–165.
Kanai-Pak, M., Aiken, L. H., Sloane, D. M., & Poghosyan, L.
(2008). Poor work environments and nurse inexperience
are associated with burnout, job dissatisfaction and quality
deficits in Japanese hospitals. Journal of Clinical Nursing, 17,
3324–3329.
Lang, G. M., Patrician, P., & Steele, N. (2012). Comparison of
nurse burnout across army hospital practice. Journal of
Nursing Scholarship, 44(3), 274–283.
Laschinger, H. K., & Leiter, M. P. (2006). The impact of
nursing work environments on patient safety outcomes:
The mediating role of burnout/engagement. Journal of
Nursing Administration, 36(5), 259–267.
Journal of Nursing Scholarship, 2016; 48:1, 83–90.

C 2015 Sigma Theta Tau International
Nurse Burnout and Patient Outcomes
Laschinger, H. K. S. (2008). Effect of empowerment on
professional practice environments, work satisfaction, and
patient care quality: Further testing of the Nursing Worklife
Model. Journal of Nursing Care Quality, 23(4), 322–330.
Leiter, M. P. (1991). Coping patterns as predictors of burnout:
The function of control and escapist coping patterns.
Journal of Organizational Behavior, 12, 123–144.
Leiter, M. P. (1992). Burnout as a crisis in professional role
structures: Measurement and conceptual issues. Anxiety,
Stress, and Coping, 5(1), 79–93.
Liu, K., You, L. M., Chen, S. X., Hao, Y. T., Zhu, X. W., Zhang,
L. F., & Aiken, L. H. (2012). The relationship between
hospital work environment and nurse outcomes in
Guangdong, China: A nurse questionnaire survey. Journal
of Clinical Nursing, 21, 1476–1485.
Lucero, R. J., Lake, E. T., & Aiken, L. H. (2010). Nursing care
quality and adverse events in U.S. hospitals. Journal of
Clinical Nursing, 19(15–16), 2185–2195.
Maslach, C. (1982). Burnout: The cost of caring. Englewood
Cliffs, NJ: Prentice-Hall.
Maslach, C., & Jackson, S. E. (1986). Maslach Burnout Inventory
(2nd ed.). Palo Alto, CA: Consulting Psychologists Press.
Maslach, C., Jackson, S., & Leiter, M. (1996). Maslach Burnout
Inventory (3rd ed). Mountain View, CA: Consulting
Psychological Press.
McHugh, M., & Witkoski Stimpfel, A. (2012). Nurse reports of
quality of care: A measure of hospital quality. Research in
Nursing and Health, 35(6), 561–679.
McHugh, M. D., Kutney-Lee, A., Cimiotti, J. P., Sloane, D. M.,
& Aiken, L. H. (2011). Nurses’ widespread job
dissatisfaction, burnout, and frustration with health
benefits signal problems for patient care. Health Affairs
(Millwood), 30, 202–210.
Nantsupawat, A., Nantsupawat, R., Kulnaviktikul, W., &
McHugh, M. D. (2014). Relationship between nurse
staffing levels and nurse outcomes in community
hospitals, Thailand. Nursing & Health Sciences, 17,
112–118.
National Database of Nursing Quality Indicators. (2013).
Changes to NDNQI fall indicator coming for 2Q2013.
NDNQI Nursing Quality News, 14(1), 2. Retrieved from
http://www.nursingworld.org/MainMenuCategories/
ANAMarketplace/ANAPeriodicals/OJIN/TableofContents/
Volume122007/No3Sept07/NursingQualityIndicators.html
National Quality Forum. (2004). National consensus standards
for nursing-sensitive care: An initial performance measure set.
Washington, DC: Author.
National Quality Forum. (2009). Nursing-sensitive care: Measure
maintenance. Retrieved from http://www.qualityforum.org/
Projects/n-r/Nursing-Sensitive_Care_Measure_Maintencance/
Nursing_Sensitive_Care_Measure_Maintenance.aspx
Oulton, J. A. (2006). The workplace is the issue—Isn’t it
obvious? International Nursing Review, 53(2), 91.
doi: 10.1111/j.1466-7657.2006.00499.x
89
Nurse Burnout and Patient Outcomes
Poghosyan, L., Clarke, S. P., Finlayson, M., & Aiken, L. H.
(2010). Nurse burnout and quality of care: Cross-national
investigation in six countries. Research in Nursing & Health,
33, 288–298.
Sundin, L., Hochwalder, J., Bilt, C., & Lisspers, J. (2007). The
relationship between different work-related sources of
social support and burnout among registered and assistant
nurses in Sweden: A questionnaire survey. International
Journal of Nursing Studies, 44, 758–769.
Supamanee, T., Kunaviktikul, W., & Keitlertnapha, P. (2014).
Nurses’ extended work hours and nurse outcomes in
community hospitals. Nursing Journal, 40, 48–58.
Teng, C. I., Shyu, Y. L., Chiou, W. K., Fan, H. C., & Lam, S. M.
(2010). Interactive effects of nurse-experienced time
pressure and burnout on patient safety: A cross-sectional
survey. International Journal of Nursing Studies, 47,
1442–1450.
Van Bogaert, P., Clarke, S., Roelant, E., Meulemans, H., &
Van de Heyning, P. (2010). Impacts of unit-level nurse
practice environment and burnout on nurse-reported
outcomes: A multilevel modeling approach. Journal of
Clinical Nursing, 19, 1664–1674.
Van Bogaert, P., Kowalski, C., Weeks, S. M., Van Heusden,
D., & Clarke, S. P. (2013). The relationship between nurse
practice environment, nurse work characteristics, burnout
and job outcome and quality of nursing care: A
90
Nantsupawat et al.
cross-sectional survey. International Journal of Nursing
Studies, 50, 1667–1677.
Van Bogaert, P., Timmermans, O., Weeks, S. M., Van
Heusden, D., Wouters, K., & Franck, E. (2014). Nursing
unit teams matter: Impact of unit-levels nurse practice
environment, nurse work characteristics, and burnout on
nurse reported job outcomes, and quality of care, and
patient adverse events—A cross-sectional survey.
International Journal of Nursing Studies, 51(8), 1123–1134.
Williams, A. M. (1998). The delivery of quality nursing care:
A grounded theory study of the nurses’ perspective. Journal
of Advanced Nursing, 27, 808–816.
World Health Organization. (2002). Prevention of hospitalacquired infections: A practical guide. Retrieved from http://
www.who.int/csr/resources/publications/drugresist/en/
whocdscsreph200212.pdf?ua=1
You, L. M., Aiken, L. H., Sloane, D. M., Liu, K., & He, G. P.
(2013). Hospital nursing, care quality, and patient
satisfaction: Cross-sectional surveys of nurses and patients
in hospitals in China and Europe. International Journal of
Nursing Studies, 50, 154–161.
Zhu, X. W., You, L. M., Zheng, J., Liu, K., Fang, J. B., Hou, S.
X., . . . Zhang, L. F. (2012). Nurse staffing levels make
difference on patient outcomes: A multisite study in
Chinese hospitals. Journal of Nursing Scholarship, 44(3),
266–273.
Journal of Nursing Scholarship, 2016; 48:1, 83–90.

C 2015 Sigma Theta Tau International
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.
CLINICAL SCHOLARSHIP
Surgical Patient Satisfaction as an Outcome of Nurses’ Caring
Behaviors: A Descriptive and Correlational Study in Six European
Countries
Alvisa Palese, MSc, RN1 , Marco Tomietto, PhDc, RN2 , Riitta Suhonen, PhD, RN3 ,
Georgios Efstathiou, PhDc, RN4 , Haritini Tsangari, PhD5 , Anastasios Merkouris, PhD, RN6 ,
Darja Jarosova, PhD, RN7 , Helena Leino-Kilpi, PhD, RN8 , Elisabeth Patiraki, PhD, RN9 ,
Chrysoula Karlou, PhDc, RN10 , Zoltan Balogh, PhD, RN11 , & Evridiki Papastavrou, PhD, RN12
1 Associate Professor, University of Udine, Italy
2 Research Fellow, University of Udine, Italy
3 Professor (acting), University of Turku, Department of Nursing Science, Finland
4 Research Fellow, Cyprus University of Technology, Cyprus
5 Statistician, Associate Professor, University of Nicosia, Cyprus
6 Associate Professor, Cyprus University of Technology, Cyprus
7 Associate Professor, University of Ostrava, Czech Republic
8 Professor and Chair, University of Turku, Finland, Nurse Manager, Hospital District of South West Finland
9 Associate Professor, National and Kapodistrian University of Athens, Greece
10 Research Fellow, National and Kapodistrian University of Athens, Greece
11 Associate Professor, Semmelweis University, Hungary
12 Lecturer, Cyprus University of Technology, Cyprus
Key words
Caring, Caring Behaviours Inventory, patient
satisfaction, Patient Satisfaction Scale,
cross-national, European, stepwise multiple
regression analysis
Correspondence
Dr. Evridiki Papastavrou, School of Health
Sciences Department of Nursing, Cyprus
University of Technology, 13 Ithakis, Limassol
Cyprus, 3107. E-mail: e.papastavrou@cut.ac.cy
Accepted June 8, 2011
doi: 10.1111/j.1547-5069.2011.01413.x
Journal of Nursing Scholarship, 2011; 43:4, 341–350.

C 2011 Sigma Theta Tau International
Abstract
Purpose: Theoretically, patient satisfaction is correlated with nursing care,
but there is not sufficient evidence to support it. The aim of this study was
to address three research questions: (a) What is the correlation between
caring as perceived by patients and patient satisfaction? (b) Are there differences across various countries on the correlation on caring as perceived
by patients and patient satisfaction? (c) Do caring behaviors affect patient
satisfaction?
Design: A multicenter correlational design was adopted involving surgical patients from six European countries: Cyprus, Czech Republic, Greece, Finland,
Hungary, and Italy.
Methods: A convenience sample of 1,565 patients was recruited in autumn
2009. The short version of the Caring Behaviours Inventory (CBI; 24 items)
and Patient Satisfaction Scale (PSS; 11 items) were used. Data analysis included descriptive statistics, as well as correlation analysis and stepwise multiple regression, to examine relations between caring behaviors and patient
satisfaction.
Findings: According to the patients involved, nurses performed caring behaviors between very frequently (score = 5) and always (score = 6). Patient
satisfaction with nursing care was also high, between satisfied (score = 3)
and very satisfied (score = 4). A positive correlation emerged between CBI
and PPS (r = 0.66, p < .01) ranging between countries from 0.27 to 0.85 (Czech Republic r = 0.27, Cyprus r = 0.76, Finland r = 0.71, Greece r = 0.85, Hungary r = 0.63, and Italy r = 0.45 [p < .01]). Among the CBI dimensions, “connectedness” mainly explains patient satisfaction (R2 = 0.404, p < .001), followed by “assurance” (R2 = 0.032, p < .001) and “respectful” (R2 = 0.005, p < .001). 341 Surgical Patient Satisfaction Palese et al. Conclusions: Caring behaviors enacted by nurses determine a consistent proportion of patient satisfaction. This association between them suggests several implications for nursing education, practice, and management. Clinical Relevance: The results may be utilized by policymakers, nurse ward managers, nurse educators, and clinical nurses as a background for taking appropriate measures to improve nursing care provided, thereby enhancing patient satisfaction. Caring is a core concept of nursing. Caring is an interpersonal process based on professional growth, expert competence, and sensitivity (Finfgeld-Connett, 2008) and is one of the ethical bases of nursing (Watson, 1985). A caring relationship generates a caring moment (Watson & Foster, 2003) in which an encounter between patient and nurse produces physically and psychologically positive outcomes both for patients and nurses. Patients have reported growing self-esteem, less anxiety, a sense of existential growth, awareness, and selfefficacy (Finfgeld-Connett, 2008). Nurses have experienced a sense of competence, capability in managing complexity and uncertainty, decision-making effectiveness, and in-depth understanding of patients’ experiences (Brilowski & Wendler, 2005; Finfgeld-Connett, 2008). Although some evidence is already available, there is uncertainty about the state of caring knowledge in nursing (Watson, 2008). Caring consequences are not easily identifiable (Cutcliffe & McKenna, 2005), though patient satisfaction is considered one of the outcomes theoretically linked with caring behaviors enacted by nurses: it describes the subjective evaluation of patients’ cognitive and emotional reactions to the comparison between caring expectations and caring received (Merkouris, Papathanassoglou, & Lemonidou, 2004; Wagner & Bear, 2008). However, there is a lack of empirical data on the effects of caring on patient satisfaction. This, in turn, has effects on caring as an imperative of nursing education, both at the basic and advanced levels (Cook & Cullen, 2003), on the scientific debate aiming to discover and describe what caring is, in the context of nursing (Cutcliffe & McKenna, 2005), and also on creating work environments that support nurses’ caring (Laschinger & Leiter, 2006). Another reason for investigating the area of caring and patient satisfaction from a multicenter perspective is the freedom of movement within European counties and the increasing mobility of both nurses and patients. This situation requires alignment of educational programs, a fact that has forced healthcare policymakers across Europe to focus on the establishment of common bases on nursing education so as to safeguard equal opportunities for quality care for every European citizen in the 342 Union. In 1993, the Bologna Process established the single most important higher education revolution that has taken place in Europe (Davies, 2008). The Bologna Process focuses on a gradual convergence toward a common framework of qualifications from educational programs. It affects approximately 6 million European nurses (European Parliament, 2010), more than 499 million European citizens (European Commission, 2010), and millions of healthcare workers. This revolution, aiming to develop a common European cultural dimension, affects nursing migration, careers, nursing management policies, and research opportunities. Although European unification opened the borders of each country for nurses, few international cross-cultural nursing studies (Suhonen, Saarikoski, & Leino-Kilpi, 2009) are available. Such studies are important for advancing nursing knowledge and practice across Europe. In this study, the investigators aimed to answer the following research questions: (a) What is the correlation between caring as perceived by patients and patient satisfaction? (b) Are there differences across various countries on the correlation on caring as perceived by patients and patient satisfaction? (c) Do caring behaviors affect patient satisfaction? Literature Review Nursing as a variable associated with patient outcomes has been mainly studied in large observational research conducted in North America and only recently in European countries. As stated by Griffiths (2009), the majority of the available studies on nursing outcomes are mainly based on administrative data from 10 years ago rather than focusing on the patients’ actual perceptions, and they have mainly adopted retrospective study designs instead of being actual or prospective based. Within the international framework of measuring the relationship between nursing and nursing outcomes, caring has received less attention, while other nursing dimensions, such us surveillance, are well described (Aiken, Clarke, Sloane, Sochalski, 2002; Kutney-Lee, Lake, & Aiken, 2009). In a review by Clarke and Surgical Patient Satisfaction Palese et al. Donaldson (2008), the authors stated that more robust measurement instruments and new theoretical models are needed to capture variations in outcomes. Furthermore, they suggested that future research must tackle the “black box” (Clarke & Donaldson, 2008, p. 19) of nursing practice by acknowledging the complexity of nursing assessment and also the anomalies in the evidence base, and perhaps most importantly, the need for taking some intellectual and political risks. Additionally, Mick and Mark (2005) suggested there is a necessity to develop more theory to support the relationship between nursing and nursing outcomes. Patient Satisfaction as a Nursing Outcome Patient satisfaction with nursing care has been defined as the patient’s opinion on the received care from nursing personnel (Merkouris et al., 2004; Wagner & Bear, 2008). Key elements of patient satisfaction are the patients, nurses, and organizational environment (Wagner & Bear, 2008). The main influences on patient satisfaction that have been reported are the patients’ expectations, patients’ demographics, patients’ previous experience as care receivers, length of stay, and cultural and social aspects of personal life (Wagner & Bear, 2008). Furthermore, nurse caring behaviors have also been considered as important for influencing satisfaction (Larrabee et al., 2004; Wagner & Bear, 2008). In a study involving 362 medical, surgical, and intensive care patients, Larrabee and colleagues (2004) identified through causal modeling, using the Caring Behaviors Inventory (CBI; Wolf, Giardino, Osborne, & Ambrose, 1994), that patient-perceived nurse caring is the major predictor of patient satisfaction. Han, Connolly, and Canham (2003) studied 477 surgical and medical patients, documenting the relationship between patient satisfaction and nursing care within a primary nurse working unit in a large Taiwanese teaching hospital. Previously, Wolf, Colahan, and Costello (1998) documented similar results with 335 patients who responded to a mailed questionnaire on their experience of hospitalization within the last year for medical or surgical care. Unfortunately, there is no evidence on the relationship between caring and patient satisfaction within European countries. As recently documented by Griffiths, Jones, Maben, and Murrells (2008), patient satisfaction in association with nursing care is an important nursing outcome. A nursing outcome is a condition, behavior, attitude, or measurable perception of patients or their families, conceptualized as a variable and largely influenced by or “sensitive” to nursing care (Moorhead, Johnson, & Maas, 2003). Patient satisfaction is a critical outcome for several reasons: It influences further health service utilization decision making (e.g., dissatisfied patients may refuse follow-up care; Laschinger, Hall, Pedersen, & Almost, 2005), and it influences the level of patients’ adherence or compliance to prescribed treatments, regimens, and recommendations (e.g., dissatisfied patients following anticoagulant therapy may refuse to follow a nurse’s recommendations; Wagner & Bear, 2008). Appropriate utilization of healthcare services and following prescribed treatments and recommendations might influence the patient’s health status and the clinical severity of his or her disease (e.g., decreasing mortality as documented by Nolte & McKee, 2008). Methods Study Design and Setting A multicenter correlational design was adopted in six countries (Cyprus, Czech Republic, Greece, Finland, Hungary, and Italy). The participation of the specific countries was based on a preexisting established cooperation among them; therefore, for convenience purposes these counties participated in this study. Participating hospitals and wards were selected according to availability, proximity, and convenience by each partner. A total of 88 wards from 34 general hospitals were included: Cyprus (15 wards from 5 hospitals), Czech Republic (19 wards from 5 hospitals), Finland (14 wards from 7 hospitals), Greece (16 wards from 5 hospitals), Hungary (10 wards from 5 hospitals), and Italy (14 wards from 7 hospitals). The selection of hospitals and wards was based on convenience (based on each partner’s access to premises). Instrumentation Necessary data were collected with the use of the Caring Behaviours Inventory 24-item version (CBI-24) and Patient Satisfaction Scale (PSS). A separate questionnaire including demographic background was also distributed (which also included a question on the self-perceived level of health condition on behalf of the patient). The CBI-24 (Wu, Larrabee, & Putnam, 2006) is based on a conceptual definition reporting nurse caring as an interactive and intersubjective process that occurs during moments of shared vulnerability between nurses and patients (Watson, 2008). It was developed by Zane Wolf and is based on Watson’s Transpersonal Caring Theory (Watson, 1985). This instrument was selected because of its conceptual-theoretical basis clarity, its 343 Surgical Patient Satisfaction consistent language, and its comprehensible instructions. It has also been reported that it requires the shortest length of time to complete (Watson, 2008). The instrument includes four factors: “assurance of human presence” (eight items), which deals with patients’ needs and security; “knowledge and skill” (five items), related to nurses as skillful and educated persons; “respectful deference to the other” (six items), dealing with how nurses show interest for the patients; and “positive connectedness” (five items), corresponding to the need for nurses to be ready to help the patients (Wolf et al., 1994). To each item, patients are requested to answer using a 6-point Likert scale (1 = never, 6 = always). The PSS (Kim, 1991) examines patients’ satisfaction with nursing care and is based on 11 items evaluated on a 4-point Likert scale (1 = very dissatisfied, 4 = very satisfied). The PSS measured patients’ satisfaction with nursing care through three factors: “technical-scientific care needs” (three items), “information care needs” (five items), and “interaction-support care needs” (three items; Suhonen, Leino-Kilpi, Välimäki, & Kim, 2007). Preliminary authorizations for using the CBI-24 and PSS were requested and obtained from the authors (Wolf & Kim, personal communication, 2008). Agreements were also obtained for the copyright of each translated version, and the authors also consented to any modifications that the research group deemed necessary. Both instruments (CBI-24, PSS) were translated into the participating countries’ languages by respective partner, using the American-English versions of both. The translation process was not performed in the Finnish language for the PSS because it had already been previously validated (Suhonen et al., 2007). Translation followed a forward and back translation processes (MAPI, 2009). Translated versions were first discussed within a national panel of experts in order to assess their content validity. A discussion within the international group followed to finalize the translation process, ensuring the content, concept, and semantic equivalence of the instruments. Further advice was obtained by the authors of the instruments on the administration process and other queries that arose during the meeting. Questionnaire reliability was assessed using Cronbach’s α coefficient. For the CBI, Cronbach’s α of pooled sample was 0.96 (ranging from 0.87 to 0.97 for each country sample); for the PSS, Cronbach’s α pooled sample was 0.95 (ranging from 0.94 to 0.96 for each country sample). Sampling and Sample Power analysis was performed using the NQuery Advisor Statistical program in order to determine the appropriate sample size of patients, with power 95% and 344 Palese et al. α = 0.01. It was assumed that a difference of ± 0.5 between the means in the items of the CBI-24 was clinically important (SD 0.9), which gave an effect size of 0.0358 and a sample size of 122 patients, and a difference of ± 0.25 between the means in the items of the PSS was clinically important (SD 0.6), which gave an effect size of 0.0194 and a sample size of 223 patients. Therefore, the maximum sample size between the two was selected, indicating that a total of 223 completed questionnaires were needed from each country. A total of 1,971 questionnaires were distributed to a convenience sample of surgical patients admitted in the hospital selected by partners; 1,659 were returned. Of these, 1,565 were eligible for analysis (response rate 84.2% on distributed and returned questionnaires and response rate 78.0% on distributed and eligible for analysis questionnaires). Data Collection Procedures Data collection was performed in autumn 2009. Initially each admitted surgical patient for planned or urgent general surgery (e.g., abdominal, gastroenterical) who was able to communicate and to give his or her consent received a letter after 48 hr of hospital admission. This letter was given to them by contacts appointed by each country partner, explaining the aims of the study and assuring them of the anonymity and confidentiality of the collected data. Hence, patients received the CBI-24 and PSS with a background data sheet. Return of completed questionnaires was considered as consent for participation. To further safeguard the confidentiality of data, the following procedures were adopted: (a) The completed questionnaires were returned in a sealed envelope and then placed in a box clearly identifiable in the ward. This box was regularly emptied by those in charge of collecting the questionnaires (who were not working in the ward). (b) The questionnaires were completed before leaving the ward, on the day of patients’ discharge. (c) Patients were assured they could refuse participation or withdraw from the study, without this affecting by any means the care provided to them. Authorizations and Ethical Issues Each country was responsible for obtaining ethical approval and access to research premises according to local requirements. Completed instruments were sent by each participating country to the project leader country (Cyprus). In each country, data were protected securely (both in electronic and paper form) with restricted access. Surgical Patient Satisfaction Palese et al. Data Analysis Statistical analysis was performed centrally by the project coordinator country (Cyprus). Data were analyzed with SPSS v16 (SPSS Inc., Chicago, IL, USA) using descriptive and inferential statistics. Analysis was performed on the overall sample and at the country level. Descriptive statistics included frequencies, percentages, means, and standard deviations. For each instrument (CBI-24 and PSS), the rank of the average scores obtained by each factor and item was also measured. Inferential statistics examined the relations between the variables of interest, using correlation and regression analysis. Pearson correlation coefficients were calculated between the overall scores obtained using the two tools and within factors. The predictive ability of the factors of the CBI was evaluated by regressing PSS on the four CBI factors and using a stepwise multiple regression analysis. Level of statistical significance accepted was p < .05. mean 5.4, SD 0.9), and “managing equipment skillfully” (factor “knowledge and skill,” mean 5.3, SD 0.9). Overall, the PSS yielded an average score of 3.3 (SD 0.58, minimum 1, maximum 4), with responses following a distribution skewed to the left, tending toward positive answers (skewness −1.14, kurtosis 1.94). The technical-scientific factor of the PSS showed the higher mean score (mean 3.4, SD 0.6), and the informational factor showed the lowest mean score (mean 3.2, SD 0.6; see Table 2). Ranking PSS items by average score, the first 3 items out of the 11 were “general professionalism of the nursing staff” (factor “technicalscientific,” mean 3.5, SD 0.7), “the way the nursing staff approached and dealt with me when I was ill” (factor “technical-scientific,” mean 3.4, SD 0.7), and “standard of care at this hospital” (factor “interactional,” mean 3.4, SD 0.7). Correlation Between Caring and Satisfaction Findings Participants’ Descriptions A total of 1,565 patients participated in the study; approximately half of them were females (51.2%, 801 of 1,565). Their average age was 54.4 years (SD 16.7), and the majority had at least secondary-level education (73.9%, 1,156 of 1,565). Participants were hospitalized on average for 9.7 days (SD 11.9) after having being admitted mainly for planned surgery (67.7%, 1,059 of 1,565). The majority of the patients (75.9%, 1,189 of 1,565) had previous hospital experience and reported their health status as ranging from fair to very good (91.9%, 1,438 of 1,565). Comparisons– using analysis of variance and chi-square tests– showed significant differences between countries on demographics (p < .001; Table 1). Caring and Satisfaction as Perceived by Patients Overall, the CBI-24 index yielded an average score of 4.9 (SD 0.8, minimum 1, maximum 6), with responses following approximately the normal distribution (skewness −0.88, kurtosis 0.57). The CBI dimension “knowledge and skills” showed the highest mean score (mean 5.3, SD 0.8), and “positive connectedness” showed the lowest mean score (mean 4.5, SD 1.1; Table 2). Ranking CBI items by average score, the first 3 items out of the 24 were “knowing how to give yes injections, IVs, etc.” (factor “knowledge and skill,” mean 5.4, SD 0.9), followed by “giving the patient’s treatments and medications on time” (factor “assurance of human presence,” The correlation between the total scores of the CBI24 and PSS was statistically significant and positive (r = 0.66, p < .001), ranging between countries from 0.27 to 0.85 (Czech Republic r = 0.27, Cyprus r = 0.76, Finland r = 0.71, Greece r = 0.85, Hungary r = 0.63, and Italy r = 0.45 [p < .001]). The correlation coefficients between PSS and the factors of the CBI ranged from r = 0.50 (PSS with “assurance of human presence”) to r = 0.62 (PSS with each of the remaining three CBI factors; p < .001). Variance of Patients’ Satisfaction Explained by Perceived Caring Behaviors Using a stepwise multiple regression model on factors affecting the PSS, 44.1% of the PSS variance is explained by all factors of the CBI. The main factor that explains patient satisfaction is “connectedness” (40.4% of satisfaction explained, p < .001), followed by “assurance” (3.2% of satisfaction explained, p < .001), and “respectful” (0.5% of satisfaction explained, p = .001). The factor “knowledge and skills” does not contribute to explaining patient satisfaction. Beta coefficients were all positive, showing that an increase in the value of each CBI factor would cause an increase in the value of the PSS index (Table 3). Discussion As reported before, most of the studies conducted on nurse caring and its relationship with patient satisfaction as an outcome have focused on data deriving from retrospective patients’ data rather than on the actual 345 Surgical Patient Satisfaction Palese et al. Table 1. Participants’ Descriptions Variable All N 1,565 % 100 Age Mean 54.4 SD 16.7 Minimum–maximum 17–94 Gender (%) Male 48.8 Female 51.2 Education (%) None 1.5 Primary 24.7 Secondary 40.6 College 20.6 University 12.7 Days of hospitalization Mean 9.7 SD 11.9 Minimum–maximum 2–120 Previous hospital experience (%) Yes 76.0 No 21.5 Unknown 2.4 Admission (%) Planned 67.7 Emergency 32.3 Perceived health condition (%) Fair to very good 91.9 Bad to very bad 8.2 Cyprus Czech Republic Finland Greece Hungary Italy 220 14.1 280 17.9 291 18.6 250 16.0 274 17.5 248 15.9 47.1 18.2 17–86 51.6 17.1 18–94 59.1 14.4 17–88 53.4 18.4 18–90 56.3 13.5 20–86 57.3 15.8 17–88 54.7 45.3 54.0 46.0 46.8 53.2 52.5 47.5 33.8 66.2 52.2 47.8 2.3 23.8 51.4 11.7 10.7 0.7 16.8 52.0 13.6 16.8 1.1 47.4 24.1 20.4 6.9 3.7 24.0 37.8 15.9 18.7 0 13.7 53.9 21.1 11.3 1.2 20.7 25.2 41.3 11.6 6.3 7.5 2–75 10.6 9.7 2–62 6.0 5.6 2–43 11.0 12.6 2–120 16.7 18.8 2–110 6.7 7.6 2–78 73.8 24.8 1.4 73.4 23.4 3.2 92.0 7.7 0.3 67.3 29.8 2.8 81.3 16.0 2.6 61.1 33.9 5.0 45.1 54.9 62.1 37.9 68.0 32.0 62.2 37.8 83.5 16.5 83.8 16.2 92.6 7.4 87.0 13.0 95.4 4.6 96.3 3.7 83.3 16.7 98.5 1.5 perceptions of patients (Griffiths, 2009). This might have led to the report of results not reflecting the current (at the time that these studies were conducted) situation. This study is a report of contemporary, actual findings, based on recent data. Therefore, the results reflect the current situation, as this appears among patients. Table 3. Stepwise Multiple Regression of the Four Factors of the Caring Behaviours Inventory on the Patient Satisfaction Scale Variable R2 R2 change Beta coefficient p Connectedness Assurance Respectful Knowledge and skills 0.404 0.436 0.441 0.441 0.404 0.032 0.005 0.000 0.126 0.167 0.111 0.036