Arnetz 2011

The Gerontologist
© The Author 2011. Published by Oxford University Press on behalf of The Gerontological Society of America.
Vol. 51, No. 6, 739–749 All rights reserved. For permissions, please e-mail: [email protected].
doi:10.1093/geront/gnr053 Advance Access publication on June 27, 2011
Organizational Climate Determinants of

Resident Safety Culture in Nursing Homes
Downloaded from https://academic.oup.com/gerontologist/article/51/6/739/818625 by Universidade Federal da Santa Maria user on 20 June 2022
Judith E. Arnetz, PhD, MPH, PT,1,2,* Ludmila S. Zhdanova, PhD,3
Dalia Elsouhag, MD, MS,1 Peter Lichtenberg, PhD,4 Mark R. Luborsky, PhD,4
and Bengt B. Arnetz, MD, PhD, MPH1,2
1
Department of Family Medicine and Public Health Sciences, Wayne State University, Detroit, Michigan.
2
Department of Public Health and Caring Sciences, Uppsala University, Sweden.
3
Department of Psychology, Carleton University, Ottawa, Ontario, Canada.
4
Institute of Gerontology, Wayne State University, Detroit, Michigan.
*Address correspondence to Judith E. Arnetz, PhD, MPH, PT, Department of Family Medicine and Public Health Sciences, Wayne State University
School of Medicine, 3939 Woodward Ave, Detroit, MI 48201. E-mail: [email protected]
Received January 14, 2011; Accepted April 13, 2011

Decision Editor: Rachel Pruchno, PhD
Purpose of the Study: In recent years, there examine whether these results can be replicated with
has been an increasing focus on the role of safety larger samples.
culture in preventing costly adverse events, such as Key Words: Long-term care, Work environment,
medication errors and falls, among nursing home res- Quality of care
idents. However, little is known regarding critical
organizational determinants of a positive safety cul-
ture in nursing homes. The aim of this study was to
identify organizational climate predictors of specific
aspects of the staff-rated resident safety culture Introduction
(RSC) in a sample of nursing homes. Design and Providing quality nursing home care for the fast-
Methods: Staff at 4 Michigan nursing homes growing segment of older people in society is
responded to a self-administered questionnaire becoming a major public health priority. Despite
measuring organizational climate and RSC. Multiple years of regulatory oversight (Scott-Cawiezell et al.,
regression analyses were used to identify organiza- 2005), nursing home care is often characterized
tional climate factors that predicted the safety culture by high workloads and staff turnover (Gruneir &
dimensions nonpunitive response to mistakes, com- Mor, 2008), difficulties in recruiting new and com-
munication about incidents, and compliance with petent staff (R. Stone & Harahan, 2010), and
procedures. Results: The organizational climate suboptimal quality of care (Castle & Ferguson,
factors efficiency and work climate predicted non- 2010; Institute of Medicine, 2001; Maas, Specht,
punitive response to mistakes (p < .001 for both Buckwalter, Gittler, & Bechen, 2008). Taken
scales) and compliance with procedures (p < .05 together, these factors have serious implications
and p < .001 respectively). Work stress was an for the health and safety of nursing home residents.
inverse predictor of compliance with procedures Patient safety culture, previously studied in hospi-
(p < .05). Goal clarity was the only significant tals (Sorra & Nieva, 2004), is an emerging concept
predictor of communication about incidents in nursing homes (Handler, Castle, et al., 2006).
(p < .05). Implications: Efficiency, work climate, However, little is known regarding organizational
work stress, and goal clarity are all malleable orga- determinants of a positive safety culture in nursing
nizational factors that could feasibly be the focus of homes. This study focused on organizational mea-
interventions to improve RSC. Future studies will sures, which are malleable and could potentially
Vol. 51, No. 6, 2011 739

be the object of future interventions aimed at (Gruneir & Mor, 2008). Research on patient safety
improving the safety, care, and health of nursing culture in hospitals has identified several key
home residents. measurement domains, including staffing, commu-
nication openness, nonpunitive response to error,
Challenges to Resident Safety in Nursing Homes feedback and communication about errors, hand-
offs and transitions, management support for
Approximately 1.6 million elderly and disabled patient safety, and organizational learning (Sorra &
persons receive nursing home care in the United Nieva, 2004). In surveys using similar domains,
States each year (Centers for Disease Control, safety culture ratings among nursing home
2009). As the aging population lives longer with staff (Handler, Castle, et al., 2006) and administra-
more medically complex chronic conditions (He, tors’ (Castle & Sonon, 2006) were significantly lower
Sengupta, Velkoff, & DeBarros, 2005), those compared with those of corresponding groups in
receiving care are often frail, with multiple illnesses hospitals. In both hospital and nursing home set-
characterized by physical, cognitive, and sensory tings, a nonpunitive work environment and com-
impairments. Adverse events, such as medication munication about adverse events were identified
errors (Barker, Flynn, Pepper, Bates, & Mikeal, as key aspects of safety culture (Bonner, Castle,
2002; Hansen et al., 2006; Scott-Cawiezell et al., Perera, & Handler, 2008). In adapting the hospital
2007), falls (Kamel, 2005; Rubenstein, Josephson, & safety culture survey (Sorra & Nieva, 2004) to
Robbins, 1994), pressure ulcers (Berlowitz et al., nursing homes (Handler, Castle, et al., 2006), addi-
2003; Saliba et al., 2003), and urinary tract infec- tional dimensions were added, namely training
tions (Kamel, 2005) are recognized as common and skills, teamwork, and compliance with proce-
problems in nursing homes (Maas et al., 2008). Few dures (Sorra, Franklin, & Streagle, 2008). Compli-
studies have compared the incidence of adverse ance with procedures was considered especially
events in nursing homes with other health care facil- important because a strong safety culture had in
ities, such as hospitals. However, it has been sug- previous research been linked to compliance with
gested that the risk of adverse events may be higher safe work practices (Handler, Wright, et al., 2006).
among nursing home residents due to the greater Based on the literature on safety culture in health
number and severity of chronic comorbid condi- care settings, this paper focuses on a nonpunitive
tions (Handler, Wright, Ruby, & Hanlon, 2006). response to mistakes, communication about inci-
Most direct care in nursing homes is provided by dents, and compliance with procedures as three
certified nursing assistants or licensed practical key aspects of resident safety culture (RSC) in
nurses that may lack the level of knowledge neces- nursing homes. Following is a brief description of
sary to care for the elderly residents (J. E. Arnetz & these constructs.
Hasson, 2007; Bonner, Castle, Men, & Handler,
2009; Gruneir & Mor, 2008). Registered nurses are Nonpunitive Response to Mistakes and Communi-
fewer in number, and physicians are most often not cation About Incidents
on site (Johnson, 2010), leaving many nursing
homes with limited access to more advanced medi- The fear of reporting errors or safety issues
cal expertise. Thus, keeping nursing home resi- among nursing home staff has been identified as
dents safe is clearly a major challenge, necessitating one of the major barriers to quality improvement
the creation and maintenance of an effective safety in the nursing home industry today (Gruneir &
culture (Scott-Cawiezell et al., 2006). Mor, 2008; Scott-Cawiezell et al., 2006). In one
study, administrators from a nationally represen-
tative sample of nearly 2,800 nursing homes
Safety Culture
reported mistakes significantly less frequently than
Patient safety culture describes the commitment their hospital counterparts (Castle & Sonon,
to safety that permeates a health care organiza- 2006). In a study of 26 Ohio nursing homes
tion. It encompasses an open atmosphere for (Hughes & Lapane, 2006), 8% of nurses and 9%
discussion of errors, process improvements, and of nursing assistants avoided reporting mistakes,
system issues without fear of reprisal (Dana-Farber whereas 27% of nurses and 22% of nursing assis-
Cancer Institute, 2000). A culture of safety empha- tants felt they were being punished after filing an
sizes open communication regarding adverse events incident report. This reluctance to report may in
and strategies for preventing incident recurrence part be due to the strict and punitive nature of
740 The Gerontologist

U.S. nursing home guidelines (Bonner et al., 2008; noncompliance of safety guidelines (Maas et al.,
Kapp, 2003). Nursing homes are governed by both 2008). In nursing homes that emphasized innova-
federal and state regulations, with strict require- tion and teamwork, staff were more likely to report
ments to report and correct adverse events. These adhering to pressure ulcer clinical guidelines
guidelines may make staff and nursing homes (Berlowitz et al., 2003). Although there is a lack of
reluctant to report incidents and errors for fear studies specifically examining the role of organiza-
that it might lead to unwanted attention from tional factors in nursing home safety compliance,
regulatory authorities (Hughes & Lapane, 2006). the literature cited here suggests that such factors
Conversely, research has shown that health care may affect staff compliance behavior.
staff are more likely to report adverse events in
nonpunitive environments. For example, a study Organizational Climate
on patient violence toward staff in health care
settings, including long-term care, found a higher Organizational climate is defined as “employees’
likelihood of reporting violent incidents in care shared perceptions about the norms, including
facilities that discouraged blame and encouraged decision making and collaboration,” that charac-
patient and employee safety (J. E. Arnetz & Arnetz, terize their workplace (P. W. Stone et al., 2007).
2000). In nursing homes, staff were more likely to Within organizations, individual employees’ per-
report resident falls in facilities with higher staff ceptions of patient safety culture are believed to
ratings of the RSC (Bonner et al., 2009). Young- influence collective safety behavior (Neal, Griffin, &
berg (2008) suggests that health care comprised Hart, 2000). Despite a general shift in the study of
complex work processes and that errors are virtually patient safety from a focus on individual-level
inevitable. However, rather than blaming individ- factors as a cause of poor patient safety to organi-
ual health care providers for errors, root causes zational factors (Neal et al., 2000; Scott-Cawiezell
must be sought in organizational and work char- et al., 2006; Vogelsmeier, Scott-Cawiezell, Miller, &
acteristics, which are closely linked to attitudes Griffith, 2010), there is limited knowledge regard-
regarding blame and incident reporting (Youngberg, ing which specific aspects of the organizational
2008). Thus, there is a need to identify specific climate that may affect specific facets of safety
organizational climate characteristics of a non- culture. A study in a single Australian hospital
punitive response to mistakes that encourages open found that general organizational climate was a
reporting of errors and near-misses. predictor of staff perceptions of work safety
(which the authors termed “safety climate”),
which was linked to safe work behaviors (Neal
Compliance With Procedures et al., 2000). However, both organizational climate
Adverse events in nursing homes are often pre- and safety climate were aggregate measures, and
ventable, and standard safety guidelines have been relationships between specific subscales were not
developed to enhance resident safety (The Joint investigated.
Commission, 2010; van Gaal et al., 2009). How- Thus, although research suggests that there is
ever, research has shown generally low levels of an association between organizational climate and
adherence (Saliba et al., 2003) and significant var- patient safety (Neal et al., 2000; Scott-Cawiezell
iation among nursing homes regarding adherence et al., 2005; P. W. Stone et al., 2007), knowledge
to guidelines, for example, in the case of pressure is lacking regarding specific organizational climate
ulcer care (Berlowitz et al., 2003; Saliba et al., determinants as these studies have used composite
2003). Adherence in these studies was measured measures of organizational climate. There is a need
via staff self-report (Berlowitz et al., 2003) as well to study the relationship between organizational
as medical record review (Berlowitz et al., 2003; climate and patient safety culture in greater detail
Saliba et al., 2003). The barriers to compliance in order to create a fact-based foundation for
with these procedural guidelines are not well improving the quality of nursing home care. Based
understood, although staffing shortages, a top– on previous research (Hughes & Lapane, 2006;
down organizational structure, and the adversarial Neal et al., 2000; Scott-Cawiezell et al., 2005;
regulatory environment have been suggested as P. W. Stone et al., 2005), we theorized that the
possible explanations (Gruneir & Mor, 2008). patient safety culture—a facet of a health care
Time restraints and lack of knowledge among organization’s performance—is in part determined
individual staff members may also be factors in by the organizational climate. The aim of this study
Vol. 51, No. 6, 2011 741

was to identify malleable aspects of organizational 40-mile radius of the city of Detroit. The facilities
climate that predict specific key factors of safety had between 74 and 108 beds and between 120
culture in nursing homes, namely nonpunitive and 200 employees.
response to mistakes, communication about inci-
dents, and compliance with procedures. Procedures and Data Collection
The questionnaire study was advertised before-
The Current Study
hand at each nursing home via informational posters.
In the present study, we examined relationships Surveys were distributed at each facility by two
between specific aspects of organizational climate members of the research team at designated time
and each of the outcomes of interest, respectively. periods over a two-day period in order to enable as
We operationalized organizational climate by using many staff as possible to participate. At all four
nine distinct dimensions of leadership, participa- facilities, all categories of staff on all work shifts
tion, feedback, competence development, employ- were invited to participate. Data collection ses-
eeship (a measure of worker engagement), efficiency, sions were held during work hours and were pre-
goals, work climate, and also work stress. These ceded by brief information sessions where staff
nine dimensions were selected because they have received both oral and written information about
been previously associated with higher staff per- the study. Those choosing to participate filled out
ceptions of quality of care (Arnetz, 1999) and bet- an anonymous paper questionnaire and were
ter organizational productivity (Anderzen & Arnetz, compensated for their participation with a $10 gift
2005), both of which could feasibly be related to card (gift cards were not permitted at one of the
safety culture. These dimensions have also been four facilities). Completed questionnaires were
previously validated among large groups of health returned directly to the researchers who adminis-
care workers (B. B. Arnetz & Blomkvist, 2007; tered the survey.
B. B. Arnetz, Lucas, & Arnetz, 2011), including
staff in nursing homes and home-based elderly
Measures
care (J. E. Arnetz & Hasson, 2007).
We hypothesized that all nine of the organiza- Resident Safety Culture.—RSC was measured
tional climate dimensions would be associated with the Nursing Home Survey on Patient Safety
with at least one of the three resident safety out- Culture (NHSPSC), a version of the Hospital
comes. Specifically, we expected that leadership, Survey on Patient Safety Culture (HSPSC) that
participation, feedback, competence development, was adapted for nursing homes (Handler, Castle,
and work climate would all be positively associated et al., 2006; Sorra et al., 2008). The adapted ques-
with nonpunitive response to mistakes. We further tionnaire was pilot tested among more than 5,000
hypothesized that leadership, participation, employ- staff in 40 nursing homes across the United States
eeship, and efficiency would be positively associ- and was found to have sound psychometric prop-
ated, and work stress negatively associated, with erties (Sorra et al., 2008). For a detailed comparison
communication about incidents. Finally, we expected of the HSPSC and NHSPSC surveys, see Castle,
feedback, competence development, work climate, Wagner, Perera, Ferguson, and Handler, 2011.
and goal clarity to be positively associated with Psychometric testing of the NHSPSC has shown
compliance with procedures, with work stress acceptable validity and reliability (Castle et al., 2011).
inversely related. The following RSC dimensions were utilized in
this study: nonpunitive response to mistakes,
Methods feedback and communication about incidents, and
compliance with procedures. “Nonpunitive response
Setting and Participants to mistakes” is a four-item factor measuring
A purposeful convenience sample of four nursing nursing home staff’s perceptions of the workplace
homes in Detroit, MI, participated in the question- atmosphere with regard to mistakes. A sample
naire study between January and July of 2010. item from this scale is “Staff are blamed when
Two of the nursing homes were run by the same a resident is harmed.” Internal reliability for the
for-profit organization, whereas the other two scale, measured with Cronbach’s alpha, was .70.
were both nonprofit church-related organizations. “Feedback and communication about incidents”
All four nursing homes were located within a comprised four items that concern the degree of

workplace discussion of adverse or potentially to use and develop their skills at work. A sample
harmful incidents. “Staff tell someone if they see item is, “I feel that I develop at my work.”
something that might hurt a resident,” is a sample “Employeeship” is a four-item scale (Cronbach’s
item from this scale, which had an alpha of .83. alpha .75) that reflects an employee’s attitudes
The third variable, “Compliance with procedures,” toward and responsibility for one’s work. A sam-
is three items that measure the degree to which ple item is, “I take responsibility for keeping myself
staff follow standard procedures at work. One informed.” “Efficiency” is a four-item measure of
item in this scale is, “To make work easier, staff the employee’s perception of how well work pro-
often ignore procedures.” Cronbach’s alpha for cesses function at their workplace. “Resources are
the scale was .54. All items use a 5-point response optimally utilized” is a sample item from this scale,
scale (1 = strongly disagree; 5 = strongly agree) and which had an alpha of .89. The “Goals” scale
include an additional alternative for “Does not (Cronbach’s alpha .81) comprised four items
apply or don’t know.” Scores for each RSC scale that summarize the clarity of the workplace goals.
were calculated for each respondent by summing A sample item is “The goals are well-defined.”
scores for component items and converting that “Work climate” is three items measuring percep-
sum to a percentage of the maximum possible tions of social support and cohesion among one’s
score. Scale scores thus range between 0 and 100%. colleagues. The alpha for work climate was .89.
A detailed description of all items in these RSC “There is a pleasant atmosphere at my workplace”
scales is available in the Supplementary Appendix. is a sample item. Finally, “Work stress” is a four-
item scale (alpha .84) that focuses on staff percep-
Organizational Climate.—The nursing home’s tions of the adequacy of time available to do one’s
organizational climate was measured with the work well. A sample item is “Do you have time to
Quality-Work Competence (QWC) questionnaire plan your work tasks ahead of time?”
(B. B. Arnetz, 1997; Dunn, Arnetz, Christensen, & All QWC items use a 4-point Likert-type
Homer, 2007). The QWC has been previously val- response scale (e.g., 1 = disagree strongly; 4 = agree
idated among large numbers of health care profes- strongly). Scores for each scale were calculated
sionals and has good psychometric properties according to the same methods used in calculating
(B. B. Arnetz, 1997; Anderzen & Arnetz, 2005; RSC scale scores and also range between 0 and
J. E. Arnetz & Hasson, 2007; B. B. Arnetz et al., 100%. Higher values indicate more positive staff
2011). Nine QWC dimensions were measured in ratings for all scales except work stress, where
this study: leadership, participation, performance lower values are more desirable, indicating lower
feedback, competence development, employeeship, levels of stress. A detailed description of all items
efficiency, goal clarity, work climate, and work in these organizational climate scales is available
stress. “Leadership” is a four-item scale measuring in the Supplementary Appendix.
staff perceptions of their immediate supervisor in
terms of communication, consistency, goal achieve-
Statistical Analyses
ment, and workplace development and improve-
ment. A sample item is, “My immediate supervisor Bivariate analysis was used to examine correla-
communicates clearly.” Cronbach’s alpha for the tions for all 12 variables of interest (9 QWC pre-
Leadership scale was .85. The “Participation” dictor and 3 RSC outcome variables) in the study.
scale comprised six items (Cronbach’s alpha .86) Multiple regression analyses were conducted on
measuring staff’s perceived influence over and nonpunitive response to mistakes, feedback and
involvement in their work. A sample item includes, communication about incidents, and compliance
“Are you able to influence decisions made at your with procedures, respectively. In each analysis,
workplace?” “Performance feedback” comprised all nine predictor variables were entered simulta-
three items that measure feedback received from neously. Due to the low alpha value for the organi-
one’s immediate supervisor. A sample item is, zational climate measure for feedback, all three
“Does your manager make clear what is expected regressions were also run excluding that inde-
of you in your work?” Cronbach’s alpha for the pendent variable, yielding identical results. Thus,
feedback scale in this study was .48. “Competence reported results include all hypothesized predic-
Development” (Cronbach’s alpha .86) comprised tors. For all regression analyses, variance inflation
four items measuring the employee’s opportunities factor values were less than 3.0, indicating that
Vol. 51, No. 6, 2011 743

collinearity was not a concern (Neter, Kutner, Table 1. Characteristics of Questionnaire Respondents
(n = 312)
Nachtsheim, & Wasserman, 1996).
Intracluster correlation coefficients (ICC) were
n %
calculated to measure how much of the total vari-
ance in each outcome variable was accounted for by Job category
Nursing assistant/aide 117 40
group membership, that is, nursing home (Cohen, Support staff 52 18
Cohen, West, & Aiken, 2003). ICCs range from 0 Administrator/manager 40 14
(complete independence) to 1 (complete depen- Direct care staffa
33 11
dence). ICCs of .05 or less do not pose a threat to Licensed nurse 27 9
the analyses, whereas ICCs of greater than .05 Administrative support staff 25 8
Otherb 1 <1
indicate that clustering effects should be taken
Length of employment at current nursing home
into account (Thomas & Heck, 2001). The ICCs Less than 2 months 10 3
for communication about incidents and compli- 2–11 months 53 18
ance with procedures were equal to .03 and .04, 1–2 years 52 17
respectively; thus, interdependence did not pose a 3–5 years 81 27
problem for those regression analyses. The ICC 6–10 years 59 20
11 years or more 45 15
for nonpunitive response was .06, which is slightly Hours worked per week
above the .05 cutoff. For this reason, a .01 alpha 15 or fewer 9 3
level was chosen for the regression analyses with 16–24 27 9
this dependent variable (Thomas & Heck, 2001). 25–40 198 66
For all other analyses, statistical significance was More than 40 68 22
Shift
set at p < .05 (two-tailed). The statistical package
Days 196 66
SPSS (version 18.0) was used for all analyses. Evenings 69 23
Approval for this study was granted by the Nights 34 11
Human Investigation Committee at Wayne State
Notes: aExamples of direct care staff: activities staff
University, Detroit, MI. member, dietician/nutritionist, physical therapist, and social
worker.
Results b
This alternative was marked but left blank.
A total of 312 staff from the four nursing homes
responded to the questionnaire, with an overall Table 3 resents the results of the respective
response rate of 64%. The percentage of missing multiple regressions with nonpunitive response,
values was under 3% for all individual items, with communication about incidents, and compliance
the exception of background questions, where with procedures as the dependent variables. The
missing values ranged from 3.2% to 5.4%. organizational climate measures efficiency and
Characteristics of questionnaire respondents are work climate were the only significant predictors
summarized in Table 1. Nursing aides comprised of nonpunitive response to mistakes, explaining
the largest group of respondents (40%), followed 38% of the total variance. The goal clarity dimen-
by support staff (e.g., maintenance, food service/ sion was the only significant predictor of commu-
dietary, housekeeping, 18%). More than 70% of nication about incidents, explaining 28% of the
respondents (n = 211) worked directly with resi- total variance. Finally, the organizational climate
dents most of the time, and slightly more than dimensions efficiency and work climate emerged as
10% (n = 34) were paid by staffing agencies. The significant positive predictors, and work stress
majority of respondents (62%) had worked 3 years was a significant negative predictor of compliance
or longer at their current nursing home. Among with procedures. Together, these three variables
those employed by staffing agencies, the majority explained 28% of the variance in compliance with
(71%) had worked 1 year or longer. procedures.
Table 2 presents bivariate correlations and
descriptive statistics for the variables included in
the study. All predictors were strongly and moder- Discussion
ately related with all criterion variables, with cor- The aim of this study was to identify organiza-
relation coefficients ranging from 0.15 to 0.53. As tional climate predictors of three key aspects of the
expected, work stress was inversely associated RSC in nursing homes. We hypothesized that all nine
with all other variables. of the organizational climate dimensions would

Table 2. Means, Standard Deviations, and Bivariate Associations Between Safety Culture (1–3) and Organizational
Climate (4–12) Measures
M SD 1 2 3 4 5 6 7 8 9 10 11 12
1 Nonpunitive response 57.22 19.74 1
2 Communication about 87.71 15.3 0.41 1
incidents
3 Compliance with 63.02 18.92 0.45 0.37 1
procedures
4 Leadership 75.34 20.66 0.40 0.35 0.34 1
5 Participation 78.21 19.50 0.40 0.35 0.29 0.54 1
6 Feedback 82.32 16.27 0.38 0.35 0.31 0.69 0.48 1
7 Competence Development 84.83 18.16 0.36 0.31 0.28 0.49 0.53 0.48 1
8 Employeeship 82.38 14.46 0.21 0.27 0.15 0.45 0.52 0.35 0.39 1
9 Efficiency 71.28 20.32 0.53 0.38 0.42 0.64 0.53 0.51 0.44 0.40 1
10 Goals 79.44 18.18 0.39 0.45 0.34 0.57 0.55 0.60 0.47 0.37 0.56 1
11 Work climate 74.59 21.55 0.49 0.26 0.43 0.50 0.34 0.39 0.35 0.22 0.53 0.40 1
12 Work stress 26.70 21.66 −0.36 −0.23 −0.34 −0.44 −0.55 −0.37 −0.46 −0.38 −0.43 −0.54 −0.33 1
Note: N = 312. All correlation coefficients are significant at p < .001.
predict at least one of the safety culture outcomes. is a pleasant atmosphere at work; and whether
However, only four of the nine dimensions— colleagues are supportive of one another. Together,
efficiency, work climate, goal clarity, and work these two factors explained 38% of the total vari-
stress—emerged as significant predictors. ance in nonpunitive response, indicating that when
With regard to nonpunitive response to mistakes, work processes operate smoothly and staff work
our hypothesis was only confirmed with regard to well together, the culture of blame is less strong.
the work climate predictor. None of the other These same factors, along with less work stress,
hypothesized dimensions—leadership, participa- explained 28% of the variance in compliance with
tion, feedback, or competence development—were procedures. The work stress subscale measures to
significant predictors of this aspect of RSC. Our what degree one has enough time to plan, carry
prediction that feedback and communication out, reflect over, and consider ways to improve
about incidents would be positively associated work tasks. Thus, based on these results, staff
with leadership, participation, employeeship, and would be more likely to follow standard safety
efficiency and negatively associated with work procedures in nursing homes they judge to be effi-
stress was also not confirmed. Instead, only a single cient, with pleasant and supportive colleagues and
organizational climate factor, goal clarity, was a with adequate time to carry out work tasks. These
significant predictor of communication about inci- results partly support those of Neal and colleagues
dents. Finally, our hypothesis regarding compli- (2000), who found that general organizational
ance with procedures was partially confirmed in climate influenced reports of individual safety
that work stress was a significant inverse predic- behavior, including compliance with safety regula-
tor. No significant association was seen with the tions and procedures, in hospital employees. Previ-
other predicted variables. ous studies have noted the importance of a positive
Three of the four organizational climate work climate for nursing home performance,
variables—efficiency, work climate, and work reporting that higher performing nursing homes
stress—were significant predictors of compliance were characterized by more positive and cohesive
with procedures, and efficiency and work climate relationships (Scott-Cawiezell et al., 2005) and
were also significant determinants of nonpunitive better “connectivity” (Forbes-Thompson, Leiker,
response to mistakes. The efficiency subscale mea- & Bleich, 2007) among staff, compared with low-
sures staff perceptions of the degree to which staff performance facilities. However, neither of these
plan the work at one’s workplace; everyone works studies examined safety culture or safety perfor-
toward a common goal; the decision-making mance, specifically.
process works well; and whether resources are Few previous studies have examined staff per-
optimally utilized. Work climate measures how ceptions of work efficiency in nursing homes. In
well staff get along at the workplace; whether there a qualitative study of 30 nurse’s aides in nursing
Vol. 51, No. 6, 2011 745

Table 3. Multiple Regression Analysis Predicting group ratings of these constructs were not mea-
Nonpunitive Response to Mistakes, Communication About
Incidents, and Compliance With Procedures
sured. In another qualitative study of four nursing
homes, Forbes-Thompson and colleagues (2007)
Variable R2 B SE B b reported that staff found positive and creative
ways of getting their work done and meeting resi-
DV: nonpunitive response to mistakes
Model 0.38*** dent needs in two high-performing nursing homes;
Leadership −0.06 0.09 −0.07 in two low-performing nursing homes, staff tended
Participation 0.08 0.08 0.08 to work in isolation, and staff relationships were
Feedback 0.07 0.11 0.06 fragmented. Again, perceptions of efficiency per se
Competence 0.07 0.08 0.06 were not measured.
development
Employeeship −0.11 0.10 −0.08
Work stress—measured in terms of the ade-
Efficiency 0.31*** 0.08 0.31*** quacy of time to do one’s work—was an inverse
Goals 0.02 0.09 0.01 predictor of compliance with procedures as we
Work climate 0.28*** 0.06 0.30*** had predicted. Even in other industries, employees
Work stress −0.08 0.07 −0.08 often use short cuts to complete tasks when there
DV: communication about incidents
is pressure for production, which often compro-
Model 0.28***
Leadership −0.01 0.07 −0.01 mises safety (Hansez & Chmiel, 2010). Nursing
Participation 0.08 0.07 0.10 homes working understaffed, with large work-
Feedback 0.03 0.08 0.03 loads per care giver, are pressured to perform
Competence 0.12 0.07 0.14 scheduled duties with tight deadlines, which may
development result in caregiver stress and safety violations
Employeeship 0.04 0.08 0.04
Efficiency 0.09 0.07 0.12
(Maas et al., 2008; Pekkarinen, Sinervo, Perala, &
Goals 0.24*** 0.07 0.29** Elovainio, 2004).
Work climate 0.06 0.05 0.08 Goal clarity was the only predictor of commu-
Work stress 0.09 0.06 0.13 nication about incidents. Communication about
DV: compliance with procedures incidents refers to the organization’s actions when
Model 0.28***
staff report possible harm to a resident, discussions
Leadership 0.03 0.09 0.03
Participation −0.04 0.08 −0.04 of ways to avoid reoccurrence of adverse incidents,
Feedback 0.04 0.10 0.03 staff’s reporting of possible harm to residents,
Competence 0.05 0.08 0.04 and the degree of discussion about how to keep
development residents safe. Previous studies have reported
Employeeship −0.06 0.09 −0.05 associations between communication and both
Efficiency 0.21** 0.08 0.22**
Goals −0.02 0.09 −0.02
turnover (Anderson, Corazzini, & McDaniel,
Work climate 0.25*** 0.06 0.27*** 2004) and work satisfaction (Scott-Cawiezell et al.,
Work stress −0.15** 0.07 −0.17* 2004) among nursing home staff. However, few
Notes: B = unstandardized beta; b = standardized beta;
studies have explored the relationship between
DV = dependent variable. communication and organizational goals. Forbes-
*p < .05. **p < .01. ***p < .001. Thompson and colleagues (2007) reported that a
strong, coherent organizational mission was asso-
ciated with both positive staff relationships and
homes, Bowers and Becker (1992) found that aides information flow in two high-performing nursing
often increased their own work efficiency by homes. Staff perceptions of the organization’s
developing individual strategies for taking care goals were not specifically measured, however. In
of residents. These strategies often entailed cutting another qualitative study, facilitators to positive
corners and breaking rules by not following rou- organizational culture change in nursing homes
tine procedures. Although these strategies were a included shared values and goals at the facility
means of decreasing work stress, they usually level (Scalzi, Evans, Barstow, & Hostvedt, 2006).
required neglecting resident needs in some way, Shared values and goals were accomplished through
with negative repercussions for resident safety and open communication between management and
care quality. Thus, work efficiency in that study staff, which was considered key to creative problem
was inversely related to both work stress and com- solving and thus to organizational change. These
pliance with procedures. However, both efficiency findings, in line with our results, suggest that
and stress were defined by the individual, and when the goals of the organization are shared and

clearly communicated, staff will feel responsible Despite this study’s limitations, the methodo-
for achieving them. Thus, clarity of organizational logical shortcomings are not sufficient to explain
goals may be an important factor in nursing home the fact that our hypotheses were only partially
staff’s willingness to communicate about possible confirmed. This study, among the first to examine
harmful situations as well as to devise strategies organizational climate predictors of specific safety
for how they can be avoided. Open reporting of culture outcomes in nursing homes, has pinpointed
adverse incidents and medical errors in a nonpuni- four distinct factors as significant determinants.
tive environment is a prerequisite for increasing The next step will be to see whether these factors
staff knowledge and awareness of factors and cir- can be confirmed as predictors, and the other fac-
cumstances that can jeopardize patient safety tors can be confirmed as nonpredictors, in future
(Kapp, 2003; Scott-Cawiezell et al., 2006). studies.
Limitations Conclusions
The results of this study were based on ques- Although previous studies have suggested a
tionnaire responses from staff at four nursing relationship between organizational climate, per-
homes in one geographic area and are not neces- ceptions of work safety, and safety outcomes (Neal
sarily generalizable to all nursing homes. In addi- et al., 2000; Scott-Cawiezell et al., 2005; P. W.
tion, our results are based on a convenience sample Stone et al., 2007), this study is, to the best of our
of nursing homes and may not be representative of knowledge, the first to identify specific organiza-
others within this geographic region. The overall tional climate determinants of specific safety cul-
response rate was 64%, and it is not known ture outcomes in nursing homes. Although previous
whether respondents were truly representative of research has shown that safety culture may be
all employees at the respective nursing homes. affected by individual behavior, especially in terms
Moreover, the questionnaire did not ask about of compliance with procedures (Neal et al., 2000),
level of education or language ability, and it is pos- our results suggest that specific organizational
sible that our response population was a selection largely malleable characteristics are stronger pre-
of those with capability of responding to a fairly dictors. These results suggest that changes toward
lengthy questionnaire offered only in English. How- a culture of safety in nursing homes need to be on
ever, the percentage of missing values was quite an organizational level, which has been previously
low in this study, indicating that most respondents suggested (Bowers & Becker, 1992; Buerhaus,
understood and were willing and able to answer 2004; Gruneir & Mor, 2008). Clear organiza-
most of the survey questions. Nevertheless, future tional goals, a focus on efficient work processes,
studies in larger more diverse samples are war- a positive work climate, and manageable levels
ranted. Furthermore, the study was cross-sectional of stress may be key ingredients in nursing home
in design and cause and effect cannot be deter- safety culture.
mined. Results for both organizational climate and
Supplementary Material
RSC dimensions were based on self-reports, with a
Supplementary material can be found at: http://gerontologist.oxford
risk for common method variance (Doty & Glick, journals.org.
1998). Future studies should be prospective and
employ mixed methods, combining questionnaire Funding
surveys with interviews, observations, and other This work was supported by Blue Cross Blue Shield of Michigan Foun-
qualitative methods that may validate quantitative dation, grant number 1547.II, Judith E. Arnetz, Principal Investigator.
findings. Such studies would enable the assessment
of actual responses to reported mistakes, commu- Acknowledgments
nication practices about incidents, and compliance The authors would like to extend their sincere thanks to the staff and
management at all participating nursing homes.The QWC survey instru-
with procedures and would complement the sub- ment is owned and marketed by the Swedish company Springlife AB.
jective ratings reported here. Prospective studies Dr. B. B. Arnetz is the cofounder and co-owner of this company.
would also make it possible to study the ability of
staff-rated scores of organizational climate and References
RSC to predict clinical resident outcomes, mea- Anderson, R. A., Corazzini, K. N., & McDaniel, R. R., Jr. (2004).
Complexity science and the dynamics of climate and communication:
sured by the prevalence of adverse events, such as Reducing nursing home turnover. The Gerontologist, 44, 378–388.
medication errors, falls, and pressure ulcers. doi:10.1093/geront/44.3.378
Vol. 51, No. 6, 2011 747

Anderzen, I., & Arnetz, B. B. (2005). The impact of a prospective survey- Forbes-Thompson, S., Leiker, T., & Bleich, M. R. (2007). High-performing
based workplace intervention program on employee health, biologic and low-performing nursing homes: A view from complexity science.
stress markers, and organizational productivity. Journal of Occupa- Health Care Management Review, 32, 341–351. doi:10.1097/01.
tional and Environmental Medicine, 47, 671–682. doi:10.1097/01. HMR.0000296789.39128.f6
jom.0000167259.03247.1e Gruneir, A., & Mor, V. (2008). Nursing home safety: Current issues and
Arnetz, B. B. (1997). Physicians’ view of their work environment and barriers to improvement. Annual Review of Public Health, 29, 369–
organization. Psychotherapy and Psychosomatics, 66, 155–162. 382. doi:10.1146/annurev.publhealth.29.020907.090912
doi:10.1159/000289127 Handler, S. M., Castle, N. G., Studenski, S. A., Perera, S., Fridsma, D. B.,
Arnetz, B. B. (1999). Staff perception of the impact of health care transfor- Nace, D. A., et al. (2006). Patient safety culture assessment in the nursing
mation on quality of care. International Journal for Quality in Health home. Quality and Safety in Health Care, 15, 400–404. doi:10.1136/
Care, 11, 345–351. doi:10.1093/intqhc/11.4.345 qshc.2006.018408
Arnetz, B. B.Blomkvist, V. (2007). Leadership, mental health, and organi- Handler, S. M., Wright, R. M., Ruby, C. M., & Hanlon, J. T. (2006).
zational efficacy in health care organizations. Psychosocial predictors of Epidemiology of medication-related adverse events in nursing homes.
healthy organizational development based on prospective data from American Journal of Geriatric Pharmacology, 4, 264–272. doi:l0. 1016/
four different organizations. Psychotherapy and Psychosomatics, 76, j.amjopharm.2006.09.01 I
242–248. doi:10.1159/000101503 Hansen, R. A., Greene, S. B., Williams, C. E., Blalock, S. J., Crook, K. D.,
Arnetz, B. B., Lucas, T., & Arnetz, J. E. (2011). Organizational climate, Akers, R., et al. (2006). Types of medication errors in North Carolina
occupational stress, and employee mental health: Mediating effects of nursing homes: A target for quality improvement. American Journal of
organizational efficiency. Journal of Occupational & Environmental Geriatric Pharmacotherapy, 4, 52–61. doi:10.1016/j.amjopharm.
Medicine, 53, 34–42. doi:10.1097/JOM.0b013e3181ffo5b 2006.03.007
Arnetz, J. E., & Arnetz, B. B. (2000). Implementation and evaluation of a Hansez, I., & Chmiel, N. (2010). Safety behavior: Job demands, job
practical intervention programme for dealing with violence towards resources, and perceived management commitment to safety. Journal
health care workers. Journal of Advanced Nursing, 31, 668–680. of Occupational Health Psychology, 15, 267–278. doi:10.1037/
doi:10.1046/j.1365-2648.2000.01322.x a0019528
Arnetz, J. E., & Hasson, H. (2007). Evaluation of an educational “tool- He, W., Sengupta, M., Velkoff, V. A., & DeBarros, K. (2005). U.S. Census
box” for improving nursing staff competence and psychosocial work Bureau, current population reports, P23-209, 65+ in the United States:
environment in elderly care: Results of a prospective, non-randomized 2005. Washington, DC: U.S. Government Printing Office.
controlled intervention. International Journal of Nursing Studies, 44, Hughes, C. M., & Lapane, K. L. (2006). Nurses’ and nursing assistants’
723–735. doi:10.1016/j.ijnurstu.2006.01.012 perceptions of patient safety culture in nursing homes. International
Barker, K. N., Flynn, E. A., Pepper, G. A., Bates, D. W., & Mikeal, R. L. Journal of Quality Healthcare, 18, 281–286. doi:10.1093/intqhc/
(2002). Medication errors observed in 336 health care facilities. mzl020
Archives of Internal Medicine, 162, 1897–1903. Institute of Medicine. (2001). Improving the quality of long-term care,
Berlowitz, D. R., Young, G. J., Hickey, E. C., Saliba, D., Mittman, B. S., executive summary. Washington DC: National Academy Press.
Czarnowski, E., et al. (2003). Quality improvement implementation in Johnson, M. A. (2010). Changing the culture of nursing homes. The phy-
the nursing home. Health Services Research, 38, 65–83. doi:10.1111/ sician’s role. Archives of Internal Medicine, 107, 407–409.
1475-6773.00105 Kamel, H. (2005). Managing urinary tract infections: Guide for nursing
Bonner, A. F., Castle, N. G., Men, A., & Handler, S. M. (2009). Certified home practitioners. Annals of Long Term Care, 13, 21–28.
nursing assistants’ perceptions of nursing home patient safety culture: Kapp, M. B. (2003). Resident safety and medical errors in nursing homes.
Is there a relationship to clinical outcomes? Journal of the American Journal of Legal Medicine, 24, 51–76. doi:10.1080/713832123
Medical Directors Association, 10, 11–20. doi:10.1136/qshc.2006. Maas, M. L., Specht, J. P., Buckwalter, K. C., Gittler, J., & Bechen, K.
018408 (2008). Nursing home staffing and training recommendations for
Bonner, A. F., Castle, N. G., Perera, S., & Handler, S. M. (2008). Patient promoting older adults’ quality of care and life: Part 1. Deficits in
safety culture: A review of the nursing home literature and recommen- the quality of care due to understaffing and undertraining. Research
dations for practice. Annals of Long Term Care, 16, 18–22. in Gerontological Nursing, 1, 123–133. doi:10.3928/19404921-
Bowers, B., & Becker, M. (1992). Nurse’s aides in nursing homes: The 20080401-03
relationship between organization and quality. The Gerontologist, 32, Neal, A., Griffin, M. A., & Hart, P. C. (2000). The impact of organizational
360–366. doi:10.1093/geront/32.3.360 climate on safety climate and individual behavior. Safety Science, 34,
Buerhaus, P. I. (2004). Lucian Leape on patient safety in U.S. hospitals. 99–109. doi:10.1016/S0925-7535(00)00008-4
Journal of Nursing Scholarship, 36, 366–370. doi:10.1111/j.1547- Neter, J., Kutner, M. H., Nachtsheim, C. J., & Wasserman, W. (1996).
5069.2004.04065.x Applied linear statistical models, WCB. Boston, MA: McGraw-Hill.
Castle, N., & Ferguson, J. C. (2010). What is nursing home quality and Pekkarinen, L., Sinervo, T., Perälä, M. L., & Elovainio, M. (2004). Work
how is it measured? The Gerontologist, 50, 426–442. doi:10.1093/ stressors and the quality of life in long-term care units. The Gerontolo-
geront/gnq052 gist, 44, 633–643. doi:10.1093/geront/44.5.633
Castle, N., & Sonon, K. E. (2006). A culture of patient safety in nursing Rubenstein, L. Z., Josephson, K. R., & Robbins, A. S. (1994). Falls in the
homes. Quality & Safety in Health Care, 15, 405–408. doi:10.1136/ nursing home. Annals of Internal Medicine, 121, 442–451.
qshc.2006.018424 Saliba, D., Rubenstein, L. V., Simon, B., Hickey, E., Ferrel, B., Czarnowski,
Castle, N., Wagner, L. M., Parera, S., Ferguson, J. C., & Handler, S. H. E., et al. (2003). Adherence to pressure ulcer prevention guidelines:
(2011). Comparing the safety culture of nursing homes and hospi- Implications for nursing home quality. Journal of the American Geri-
tals. Journal of Applied Gerontology, 30, 22–43. doi:10.1177/ atrics Society, 51, 56–62. doi:10.1034/j.1601-5215.2002.51010
0733464809353603 Scalzi, C. C., Evans, L. K., Barstow, A., & Hostvedt, K. (2006). Barriers
Centers for Disease Control. (2009). The National Nursing Home Survey: and enablers to changing organizational culture in nursing homes.
2004 overview. Retrieved from http://www.cdc.gov/nchs/data/series Nursing Administration Quarterly, 30, 368–372.
/sr_13/sr13_167.pdf. Scott-Cawiezell, J., Main, D. S., Vojir, C. P., Jones, K., Moore, L.,
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple Nutting, P. A., et al. (2005). Linking nursing home working conditions
regression/correlation analysis for the behavioral sciences (3rd ed.). to organizational performance. Health Care Management Review, 30,
Mahwah, NJ: Lawrence Erlbaum Associates. 372–380.
Dana-Farber Cancer Institute. (2000). Principles of a fair and just culture. Scott-Cawiezell, J., Pepper, G. A., Madsen, R. W., Petroski, G., Vogelsmeier,
Retrieved from http://www.dana-farber.org/pat/patient-safety/patient- A., & Zellmer, D. (2007). Nursing home error and level of staff
safety-resources/patient-rounding-toolkit.html. credentials. Clinical Nursing Research, 16, 72–78. doi:10.1177/
Doty, D. H., & Glick, W. H. (1998). Common methods bias: Does com- 1054773806295241
mon methods variance really bias results? Organizational Research Scott-Cawiezell, J., Schenkman, M., Moore, L., Vojir, C., Connoly, R. P.,
Methods, 1, 374–406. doi:10.1177/109442819814002 Pratt, M., et al. (2004). Exploring nursing home staff’s perceptions of
Dunn, P. M., Arnetz, B. B., Christensen, J. F., & Homer, L. (2007). Meet- communication and leadership to facilitate quality improvement. Jour-
ing the imperative to improve physician well-being: Assessment of an nal of Nursing Care Quality, 19, 242–252.
innovative program. Journal of General Internal Medicine, 22, 1544– Scott-Cawiezell, J., Vogelsmeier, A., McKenney, C., Rantz, M., Hicks, L., &
1552. doi:10.1007/s11606-007-0363-5 Zellmer, D. (2006). Moving from a culture of blame to a culture of

safety in the nursing home setting. Nursing Forum, 41, 133–140. The Joint Commission. (2010). Accreditation program: Long term care
doi:10.1111/j.1744-6198.2006.00049.x national safety goals, effective January 1, 2011. Retrieved from http://
Sorra, J., Franklin, M., & Streagle, S. (2008). Survey user’s guide, nursing www.jointcommission.org/ltc_2011_npsgs/.
home survey on patient safety culture. Rockville, MD: Agency for Thomas, S. L., & Heck, R. H. (2001). Analysis of large-scale secondary
Healthcare Research and Quality. AHRQ Publication No. 08-0060. data in higher education research: Potential perils associated with com-
Sorra, J., & Nieva, V. (2004). Hospital survey on patient safety culture. plex sampling designs. Research in Higher Education, 42, 517–540.
Rockville, MD: Agency for Healthcare Research and Quality. AHRQ doi:10.1023/A:1011098109834
Publication No. 04–0041. van Gaal, B. G., Schoonhoven, L., Hulscher, M. E., Mintjes, J. A., Borm,
Stone, P. W., Harrison, M. I., Feldman, P., Linzer, M., Peng, T., Roblin,
G. F., Koopmans, R. T., et al. (2009). The design of the SAFE or
D., et al. (2005). Organizational climate of staff working conditions
SORRY? study: A cluster randomised trial on the development and
and safety—An integrative model. In K. Henriksen, J. B. Battles, & E. S.
testing of an evidence based inpatient safety program for the preven-
Marks (Eds.), Advances in patient safety: From research to implemen-
tion of adverse events. BMC Health Services Research, 9, 58–66.
tation, (Vol. 2, pp. 467–482). Rockville, MD: Agency for Health-
care Research and Quality. doi:10.1186/1472-6963-9-58
Stone, P. W., Mooney-Kane, C., Larson, E. L., Horan, T., Glance, L. G., Vogelsmeier, A., Scott-Cawiezell, J., Miller, B., & Griffith, S. (2010).
Zwanziger, J., et al. (2007). Nurse working conditions and patient Influencing leadership perceptions of patient safety through just
safety outcomes. Medical Care, 45, 571–578. doi:10.1097/MLR. culture training. Journal of Nursing Care Quality, 25, 288–294.
0b013e3180383667 doi:10.1097/NCQ.0b013e3181d8e0f2
Stone, R., & Harahan, M. F. (2010). Improving the long-term care work- Youngberg, B. J. (2008). Event reporting: The value of a nonpunitive
force serving older adults. Health Affairs, 29, 109–115. doi:10.1377/ approach. Clinical Obstetrics and Gynecology, 51, 647–655.
hlthaff.2009.0554 doi:10.1097/GRF.0b013e3181899a05
Vol. 51, No. 6, 2011 749

Arnetz 2011

Uploaded by

Copyright:

Available Formats

Arnetz 2011

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Arnetz 2011

Uploaded by

Copyright:

Available Formats

The Gerontologist

Organizational Climate Determinants of

Received January 14, 2011; Accepted April 13, 2011

Vol. 51, No. 6, 2011 739

740 The Gerontologist

Vol. 51, No. 6, 2011 741

742 The Gerontologist

Vol. 51, No. 6, 2011 743

744 The Gerontologist

Vol. 51, No. 6, 2011 745

746 The Gerontologist

Vol. 51, No. 6, 2011 747

748 The Gerontologist

Vol. 51, No. 6, 2011 749

You might also like