JOURNAL OF PALLIATIVE MEDICINE

Volume 16, Number 11, 2013

ª Mary Ann Liebert, Inc.
DOI: 10.1089/jpm.2013.0160

Job Satisfaction, Intent To Stay,

and Recommended Job Improvements:
The Palliative Nursing Assistant Speaks

Barbara A. Head, PhD, CHPN, ACSW, FPCN,1 Karla T. Washington, PhD, LCSW,2
and John Myers, PhD, MSPH1

Abstract
Background and Objective: There are few studies specifically focused on the job satisfaction of hospice and
palliative nursing assistants (NAs). The goal of this study was to ascertain factors contributing to NAs’ job
satisfaction and intent to remain with their current employer, garner suggestions for improving NA jobs, and
inform hospice and palliative care providers concerned about retaining qualified, satisfied NAs.
Methods: A nonexperimental, cross-sectional survey design was used to collect data from a broad sample of NAs
employed in hospice and palliative care settings between December 2010 and May 2011.
Results and Conclusions: This study found hospice and palliative NAs to be satisfied and committed to their
jobs, which most viewed as long-term careers. However, organizations committed to recruitment and retention
of committed and satisfied NAs would do well to ensure high-quality NA supervision, include NAs as valued
team members, encourage positive work relationships, work to reduce problems identified by NAs as interfering
with their work, and continue efforts to increase NA compensation.

Introduction Yet, NAs often subsist on low wages, receive limited

fringe benefits, carry heavy workloads that are physically

A lthough there have been numerous studies asses-

sing the job satisfaction of nursing assistants (NAs) in
nursing homes and one major national survey of home health
and hospice aides, there are few studies specifically focused
on the job satisfaction of hospice and palliative NAs. In 2011,
with funding from the Hospice and Palliative Nurses Foun-
dation, 876 hospice and palliative NAs from across the United
States were surveyed to assess their job satisfaction using the
Better Jobs Better Care Survey of Direct Care Workers.1 The
goal of the study described herein was to ascertain factors
contributing to NAs’ job satisfaction and intent to remain with
their current employer, garner suggestions for improving NA
jobs, and inform hospice and palliative care providers con-
cerned about retaining qualified, satisfied NAs.
The role of the NA is integral to the provision of quality
hospice and palliative care. These team members provide the
majority of ‘‘hands-on’’ patient care and are often the first
providers to observe and identify significant physical, psy-
chosocial, and spiritual issues. Their ongoing relationships
with patients, families, and team members and the provision
of top-quality personal care are often critical components of
care plan success.
challenging, are offered little opportunity for advance-
ment,2 and may feel ‘‘personally and professionally dis-
missed.’’3 A 2011 national study of home health and
hospice aides found that their average hourly wage was
$10.88, and two-thirds of the 160,700 aides surveyed had
annual family incomes of less than $40,000.2 Although most
NAs earn above the United States minimum wage of $7.25
per hour, the earnings of NAs working full time and pro-
viding for several children or other family members could
easily fall below the poverty level according to the Federal
Poverty Guidelines.4 Although extrinsic factors such as
wages, benefits, policies, and workload have proven to be
important to NAs, intrinsic rewards (those matters related
to worker self-actualization and sense of accomplishment)
have also been shown to affect job satisfaction among NAs.5
Not surprisingly, high job satisfaction has been shown to
correlate with lower intention to leave.6 Job dissatisfaction
results in high employee turnover and negatively impacts
organizational effectiveness, morale, productivity, and pa-
tient satisfaction.7 Unhappy employees who stay on the job
may exhibit poor work performance and can undermine the
efforts of other employees.8

1
School of Medicine, 2Kent School of Social Work, University of Louisville, Kentucky.
Accepted May 28, 2013.

1356
NURSING ASSISTANT JOB SATISFACTION
Attraction and retention of qualified, committed NAs
should be a top priority for all palliative and hospice provider
organizations wishing to provide the highest quality of care to
patients and families and minimize the costs and burden of
turnover. Therefore, research specific to the satisfaction and
intent to stay of hospice and palliative NAs is of value to
employers and colleagues of these team members.
Methods
Prior to the initiation of any research activities, the study
plan and all related processes were reviewed and approved
by the Human Subjects Protection Program of the University
of Louisville. A nonexperimental, cross-sectional survey de-
sign was used to collect data from a broad sample of NAs
employed in hospice and palliative care settings between
December 2010 and May 2011. Potential participants were
invited to participate via multiple modalities—letters of in-
vitation sent to mailing lists of nursing assistant members
of the Hospice and Palliative Nurses Association and the
National Hospice and Palliative Care Organization, e-mail
invitations, organizational newsletters, and letters sent to
employers. Organizational members of the Population-based
Palliative Care Research Network (PoPCRN), a national
consortium of hospices involved in multi-site research pro-
jects, were also asked to invite their employed NAs to com-
plete the survey instruments.
Instrument
Study participants completed the Better Jobs Better Care
Survey of Direct Care Workers. This self-administered survey
was developed as a result of a 4-year, multimillion-dollar
study funded by the Robert Wood Johnson Foundation and
the Atlantic Philanthropies to gain an understanding of direct
care workers’ (NAs, home health aides, and personal care
attendants) perceptions of their work and job environments.
The survey was developed by the Survey Research Center at
Pennsylvania State University.1 Because this instrument had
been tested with NAs and other direct care workers in various
settings, the wording of the questions was not altered. Find-
ings of this study are therefore limited to the queries and
response options of the instrument.
Data collection
Participants were encouraged to complete the survey on-
line but, if preferred, they could request a printed copy of the
survey. Those choosing the online method were directed to a
website that offered a brief explanation of the study followed
by preamble consent and a link to the online survey. Data
collected online were later downloaded into a Statistical
Package for the Social Sciences (SPSS), version 20 (SPSS Inc.,
Chicago, IL) data file. As explained in the preamble materials,
participant consent was indicated by completion of the online
or printed survey.
Those requesting a printed survey were sent the preamble
consent and survey with a return, stamped envelope. A
number of employers requested packets of printed surveys
and returned those in bulk. Once returned, the data from
those surveys were entered into the data file.
A total of 876 hospice and palliative NAs completed the
survey; however, due to a problem in the administration of
1357
the online survey, only 626 indicated the extent to which they
were satisfied with their current job. Data from these 626 cases
were analyzed in the quantitative portion of the study, the
first of two separate analyses conducted. Approximately 14%
(n = 125) of the 876 original cases contained no response to the
item asking, ‘‘What is the single most important thing your em-
ployer could do to improve your job?’’ leaving a total of 751 cases
that were analyzed in the text analysis portion of the study,
separate from the purely quantitative data. It is difficult to
estimate what percentage of the total number of NAs pro-
viding hospice and palliative care are represented in this
sample. The National Board for Certification of Hospice and
Palliative Nurses reports that 3770 NAs have obtained spe-
cialty certification, but this is only a portion of those working
in this specialty.
Quantitative data analysis
Both the hand-written surveys (n = 287, 45.8%) and those
submitted online (n = 339, 54.2%) were entered into SPSS-20
for analysis. Initially, researchers investigated if differences
existed between NAs who completed hand-written surveys
and those who submitted online surveys. No significant dif-
ferences existed; therefore, all 626 surveys were combined into
one complete dataset for analysis. Although it would have
been possible for an NA to complete the survey more than
once, this threat was outweighed by the need to recruit par-
ticipants via various means and to allow both online and
written responses.
Counts and percentages for categorical characteristics
(gender, ethnicity, satisfied with job, likely to leave job within
12 months, view job as short-term/long career, etc.) were
calculated, as well as mean and standard deviations for all
continuous variables (age, hourly wage, time as NA, and
time with employer). Perceived personal characteristics (self-
efficacy, pride in employer, challenging job, etc.) as well as
perceived work-related characteristics (dead-end job, dis-
crimination, can find a higher paying job, etc.) were explored.
Estimated odds ratios (OR) and 95% confidence intervals
(CI) were calculated using multivariable logistic regression9
to predict the odds that an NA was satisfied with his or her
job. Similarly, estimated OR and 95% CI were calculated to
predict the odds an NA viewed his or her job as long term. All
significantly different demographic variables were included
in the multivariable logistic regression model. Significance
was set at an alpha = 0.05. Because of the exploratory nature of
this study, no adjustments were made to the alpha when
performing multiple statistical tests.
Text analysis
Responses to the open-ended question ‘‘What is the single
most important thing your employer could do to improve your job?’’
were analyzed through a template approach to text analysis,10
using a coding template and corresponding definitions es-
tablished by Kemper and colleagues11 in their study of direct
care workers in nursing facilities, home health agencies, as-
sisted living organizations, adult day centers, and other per-
sonal care organizations. Using this a priori template of codes,
two members of the research team independently coded all
responses included in the dataset, applying one of 16 different
codes to individual responses, and subsequently grouping the
coded responses into four broad categories. Although the
1358 HEAD ET AL.

survey instructions explained that respondents should pro- hourly wage ($13.10/hour), and had both been an NA and
vide only one recommendation to improve their jobs, some been with their current employer more than 5 years. Re-
NAs included additional responses. In those cases, only the spondents were from 32 different states and 192 different
first response was coded, consistent with procedures used in organizations; 98% of these were hospice programs. When
previous research in this area.11 asked where they provided patient care, 65% reported seeing
After completing independent coding, the researchers re- patients in their homes, 61% in long-term care or nursing
convened to calculate inter-coder reliability. A comparison of homes, 28 % in hospitals, and 27% in hospice or palliative care
the two separately coded datasets revealed that the research- facilities.
ers’ coding decisions were in agreement 94.6% of the time. The Most NAs were satisfied with their job (n = 603, 96.4%), were
researchers discussed the 5.4% of responses they had coded not at all likely to leave their job in the next 12 months (n = 521,
differently, arriving at full consensus on coding decisions 83.9%), were not thinking of quitting (n = 618, 98.7%), viewed
through discussion and minor modifications or clarifications their job as long term (n = 497, 80.0%), would recommend
of code definitions. Finally, frequencies of responses were their place of employment to family for care (n = 614, 98.1%),
calculated, indicating which changes hospice and palliative would recommend their place for employment (n = 590,
NAs believed would be most likely to improve their jobs. 91.1%), received health insurance (n = 454, 73.3%), were fe-
male (n = 595, 95.8%), had at least a high school diploma
Results (n = 614, 98.1%), and were white (n = 399, 66.7%). Descriptive
demographic data were structured by the survey instrument
As seen in Table 1, on average, NAs in the quantitative
and did not permit collection of other information such as
analysis were middle-aged (43.7 years old), made a low
immigration status or previous employment history. Few NAs
viewed their job as short term (n = 124, 20.0%), had employers
who exhibited extraordinary efforts toward promoting their
Table 1. Demographic Characteristics satisfaction (n = 250, 40.5%), and had been discriminated
Variable N = 626 (%) or Mean (SD) against due to ethnicity (n = 51, 8.1%).
As seen in Table 2, NAs in this sample had predominantly
Age 43.7 (11.4) positive perceptions of their work environment. Higher
Hourly wage 13.1 (2.4) average scores were evident in positive categories such as
Time as NA 6.4 years (2.8 years) self-efficacy, helping others, job challenge, job rewards, and
Time with employer 5.5 years (2.6 years) having decision authority, whereas lower scores were related
Satisfied with job to negative job factors such as feeling job was a dead end,
Satisfied 603 (96.4%) exposure to hazards, experiencing job problems, work over-
Dissatisfied 23 (3.6%) load, and discrimination at work.
Likely to leave job in 12 months As seen in Table 3, the only significant predictors of job
Very likely 22 (3.5%) satisfaction were team spirit-rewards (OR = 2.630, 95% CI
Somewhat likely 78 (12.6%) 1.169–5.919, p = 0.019), quality of supervision (OR = 13.390,
Not at all likely 521 (83.9%) 95% CI 1.309–136.971, p = 0.029), and job problems
Think about quitting
Yes 8 (1.3%)
No 618 (98.7%) Table 2. Perceived Personal and Work-Related
View job as short-term job 124 (20.0%) Characteristics
View job as long-term career 497 (80.0%)
Extraordinary employer efforts 250 (40.5%) Variable N = 626 Mean (SD)
to improve job
Self-efficacy 3.78 (0.70)
Recommend place to family for care Intent to leave 2.65 (0.40)
Yes 614 (98.1%) Pride in employer 3.62 (0.66)
No 12 (1.9%) Team spirit-inclusion 3.41 (0.56)
Recommend place for employment Hazard exposure 1.62 (0.63)
Yes 590 (91.1%) Dead-end job 1.65 (0.66)
No 36 (8.9%) Overload 1.52 (0.46)
Receive health insurance Discrimination 1.30 (0.70)
Yes 454 (73.3%) Challenge 3.57 (0.45)
Offered – not enrolled 130 (21.0%) Acknowledgment 3.27 (0.53)
Not offered 35 (5.7%) Helping others 3.77 (0.35)
Decision authority 3.45 (0.61)
Female 595 (95.8%)
Team spirit-rewards 3.37 (0.58)
High school diploma 614 (98.1%)
Quality of supervision 2.88 (0.52)
Ethnicity Intent to leave 2.62 (0.42)
White 399 (66.7%) Job rewards 3.47 (0.42)
Black 154 (25.7%) Job problems 1.54 (0.51)
Hispanic 25 (4.2%) Can find more pay 2.91 (1.06)
Other 21 (3.6%)
Discriminated due to ethnicity 51 (8.1%) Note: Variables were rated on a scale of 1 to 4. Higher scores
indicate increases in the attribute.
NA, nursing assistant; SD, standard deviation. SD, standard deviation.
NURSING ASSISTANT JOB SATISFACTION 1359

Table 3. Odds Ratio Estimates and Confidence Table 5. Recommendations Provided by Hospice
Intervals from Multivariable Logistic Regression and Palliative Nursing Assistants Regarding One
Predicting Odds of Being Satisfied with Job Thing Employer Could Do To Improve Their Job

Predictor OR 95% CI of OR P value Type of recommendation Number of responses (%)

Team spirit-rewards 2.630 1.169–5.919 0.019 Satisfied/No improvement needed 120 (16%)
Quality of supervision 13.390 1.309–136.971 0.029 Increase compensation 195 (26%)
Fewer/less intense 0.012 0.001–0.182 0.001 Pay 165 (22%)
job problems Benefits 15 (2%)
Hours 15 (2%)
CI, confidence interval; OR, odds ratio.
Improve work relationships 225 (30%)
Communication 45 (6%)
(OR = 0.012, 95% CI 0.001–0.182, p = 0.001). Within the job Teamwork 22 (3%)
problems subscale, factors cited as the most major problems Supervision 30 (4%)
were no support from supervisor, little chance for promo- Listening 22 (3%)
tions, not having help when needed, and getting hurt. Appreciation 44 (6%)
As seen in Table 4, the only significant predictors of view- Respect 53 (7%)
ing their job as long term were pride (OR = 2.513, 95% CI Motivation and encouragement 9 (1%)
1.021–6.173, p = 0.045), quality of supervision (OR = 9.042, 95% Increase staffing 68 (9%)
CI 2.217–36.874, p = 0.002), age (OR = 1.141, 95% CI 1.103– Improve management systems 143 (19%)
1.180, p < 0.001), and intent to leave (OR = 0.008, 95% CI 0.003– Equipment and supplies 15 (2%)
0.023, p < 0.001). Training 90 (12%)
A summary of recommendations provided by NAs when Scheduling 22 (3%)
they were asked which single change would most improve Misc. work systems 16 (2%)
their jobs is provided in Table 5. Sixteen percent of NAs in-
dicated they were completely satisfied with their jobs in their
staffing did not logically fall into any of the other established
current state and recommended no changes. The majority
categories. Therefore, it was promoted to a broad category of
(84%) of respondents provided specific suggestions for en-
its own. ‘‘Hire more help’’ and ‘‘schedule enough qualified, de-
hancing their job satisfaction. Each suggestion fell into one of
pendable personnel to cover each shift’’ are examples of responses
four broad categories: increase compensation, improve work
that were grouped into this category.
relationships, increase staffing, and improve management
Finally, almost one fifth (19%) of all respondents suggested
systems.
that improvements in management systems would have the
More than one fourth (26%) of all respondents indicated
greatest positive impact on their jobs. The suggested im-
that the most powerful change their employer could make
provements in this category included ensuring that equipment
would be to increase compensation in some fashion. This
and supplies (n = 15) were sufficient, that NAs were adequately
category included responses initially coded with the labels
trained (n = 90), and that scheduling (n = 22) was logical and
increase pay (n = 165), increase benefits (n = 15), and increase
consistent. Recommendations about miscellaneous work sys-
number of hours worked (n = 15). Examples of responses labeled
tems (n = 16) (e.g., ‘‘cut down on paperwork,’’ ‘‘do things to make
with these codes included ‘‘a decent raise,’’ ‘‘better health in-
our jobs interesting’’) were also included here.
surance,’’ and ‘‘give me full-time [hours] back.’’
Slightly more respondents (30%) recommended that their
Discussion
supervisors create changes to improve work relationships.
Specific areas cited as needing improvement were communi- Overall job satisfaction and intent to stay in this sample of
cation (n = 45), teamwork (n = 22), supervision (n = 30), listening NAs were higher than previous studies of NAs working in
(n = 22), appreciation (n = 44), respect (n = 54) and motivation/en- various settings. In a study of home health aides working in
couragement (n = 9). Responses in this category included such hospice and home care, Bercovitz and colleagues reported an
items as ‘‘keep us aware of what is happening in the agency,’’ 87% satisfaction rate.2 A recent study of 2146 NAs working in
‘‘recognize NAs as part of the team,’’ and ‘‘treat all NAs equally.’’ nursing homes found 83% to be satisfied.5 The Better Jobs
Staffing shortages were the top area of concern for 9% of Better Care demonstration project, which surveyed direct care
respondents. Although originally an individual code, improve workers employed in a variety of settings, found that 57% of
respondents reported they were not likely to leave in the next
year12 compared with 84% in the present sample. Because the
Table 4. Odds Ratio Estimates and Confidence
Intervals from Multivariable Logistic Regression average length of employment was more than 5 years in this
Predicting Odds of Viewing Job as Long Term sample, it is likely that study participants were those NAs
who found meaning and purpose in their work as opposed to
Predictor OR 95% CI of OR P value those who stay in such jobs only a short time.
Because the sample in this study was overwhelmingly
Pride 2.513 1.021–6.173 0.045 satisfied and reported high positive scores on most variables,
Quality of supervision 9.042 2.217–36.874 0.002
the only factors significantly predictive of satisfaction were
Higher age 1.141 1.103–1.180 < 0.001
Intent to stay 0.008 0.003–0.023 < 0.001 team spirit-rewards, quality of supervision, and job problems
(those with a lower rating of job problems being more satis-
CI, confidence interval; OR, odds ration. fied). Pride and quality of supervision predicted viewing the
1360
job as long term as did increased age and intent to stay
(however, it is obvious that older age and length of time on the
job would be predictive of a long-term view). When asked
what would most improve their job, almost one third of our
respondents indicated improvements in work relationships,
which included supervisors and fellow team members.
Multiple studies of NAs in other settings have found the
quality of supervision to be a strong component of job satis-
faction.7,13,14 In the 2004 National Nursing Assistant Survey,
positive assessment of the supervisor’s behavior had the
strongest association with intrinsic satisfaction of all vari-
ables.5 Supervisors who do not offer positive feedback, do not
attend to suggestions and opinions, don’t respect skills and
abilities, and do not treat NAs as equal team members dis-
courage and devalue NAs.15 The Better Jobs Better Care
demonstration project found that direct care workers with
positive perceptions of their supervisors were less likely
to quit,12 and productive relationships with supervisors
and coworkers has been found to be an important factor in
job tenure.16
Knowing that quality supervision is paramount to the job
satisfaction and retention of NAs, it behooves organizations to
evaluate the skills and practices of those supervising NAs.
Supervisors need adequate training in this role as well as the
ability to maintain a supportive work environment that is
responsive to the opinions and feedback of NAs and re-
spectful of their skills, abilities, and contributions to the or-
ganization. Supervisors’ treatment can be viewed by the NA
as representative of how the organization views the NAs as
workers and people.3
As with this study, team spirit and a sense of belonging on a
team has proven to be an important factor in job satisfaction in
multiple NA studies.7,14,17–19 Inclusion of NAs in team meet-
ings, decision-making/policy committees, staff meetings, and
social activities would be practices conducive to job satisfac-
tion and tenure of NAs. When included in attendance, it goes
without saying that NAs should be encouraged to offer
feedback and opinions and their contributions should be re-
spected and supported. Interdisciplinary team care is a core
tenet of palliative care and every organization providing
hospice and palliative care should be a venue where inclu-
siveness and respect for all team members are practiced.
Organizations that invest in teamwork and include NAs as
essential team members will reap the rewards of having sat-
isfied, committed NAs.
Pride in one’s work was found to significantly influence the
NAs’ intention to stay. In a study of job satisfaction among
hospice workers, DeLoach and Monroe found that 64% of the
respondents defined job satisfaction as pride in the contribu-
tions made to the organization,20 dying patients, and their
families. As with other palliative care employees, these NAs’
intention to remain in their job was influenced by the feeling
that they are contributing to an important work or cause.
In the text analysis component of the present study, the
most frequent response to the question regarding changes that
would improve NAs’ jobs was improved work relationships,
including supervision. This further supports the finding that
quality of supervision predicts both overall job satisfaction
and viewing the job as long term. NAs in this sample rated
being heard, appreciated, respected, and encouraged over
increased compensation as the change most needed to im-
prove their job.
HEAD ET AL.
Yet, increased compensation was listed by 26% as the most
needed job improvement. As with all wage earners, NAs view
their compensation as a reflection of the value of their
work and believe that low wages minimize their skills,
knowledge, and experience.3 In a study of NAs in nursing
homes, Wiener and coworkers found that a one dollar in-
crease in hourly wages was associated with 2.1 additional
months of tenure.16 Squillace and colleagues found that,
among certified NAs likely to leave their jobs in the next year,
one in three cited pay as a reason to leave.21 The National
Nursing Home Study concluded that higher wages have more
impact on recruitment than on retention; once employed,
workers want to feel valued in other ways.2 This finding
might explain why higher wages was the second rather than
first most often offered way the NAs felt their job could be
improved. Regardless, hospice and palliative employers of
NAs should work to increase the wages of NAs, many of
whom are living at poverty levels. NAs are often the working
poor and many are providing for single parent families.6
There are multiple study limitations. Although every effort
was made to reach NAs lacking Internet savvy or access and
to allow for written responses, there is no doubt that barriers
to participation existed for those NAs uncomfortable with
locating and completing online instruments or those for
whom English is a second language. This sample consisted
almost exclusively of NAs employed by a hospice/palliative
care provider, which limits generalizability of the findings to
those employed in settings where such care is the focus.
Employing organizations were asked to invite their NAs to
participate and, in a number of cases, the employer circulated
and returned these surveys; therefore, employer coercion
could be a threat to study validity (although most employers
submitting surveys reported efforts to maintain respondent
confidentiality in the process).
Findings related to factors affecting job satisfaction are
limited to the options of the structured survey instrument;
there may indeed be factors specific to hospice and palliative
work that contribute to job satisfaction not captured. For in-
stance, the instrument did not address specific work factors
related to daily work with seriously ill and/or dying persons
and their families.
Conclusions
This study found hospice and palliative NAs to be satisfied
and committed to their jobs, which most viewed as long-term
careers. However, organizations committed to recruitment
and retention of committed and satisfied NAs would do well
to ensure high quality NA supervision, include NAs as valued
team members, encourage positive work relationships, work
to reduce problems identified by NAs as interfering with their
work, and continue efforts to increase NA compensation.
Acknowledgments
This study was made possible by a grant from the Hospice
and Palliative Nurses Foundation. The authors would like to
thank the Foundation and all the NAs and organizations that
participated in this research.
Author Disclosure Statement
No competing financial interests exist.
NURSING ASSISTANT JOB SATISFACTION 1361

References 13. Probst JC, Baek J-D, Laditka SB: The relationship between
workplace environment and job satisfaction among nursing
1. Survey Research Center at Penn State: Confidential Survey of
assistants: Findings from a national survey. J Am Med Dir
Long-Term Caregivers. Pittsburgh: Pennsylvania State Uni-
Assoc 2010;11:246–252.
versity, 2008.
14. Chou SC, Boldy DP, Lee AH: Measuring job satisfaction in
2. Bercovitz A, Moss A, Sengupta M, Park-Lee EY, Jones A,
residential aged care. Int J Qual Health Care 2002;14:49–54.
Harris-Kojetin LD: An overview of home health aides:
15. Shur D, Noelker LS, Looman WJ, Whitlatch CJ, Ejaz FA:
United States, 2007. Natl Health Stat Rep 2011:1–31.
Four steps to more committed nursing assistants. Am Coll
3. Bowers BJ, Esmond S, Jacobson N: Turnover reinterpreted
Health Care Admin 1998;2:29–39.
CNAs talk about why they leave. J Gerontol Nurs 2003;29:36–43.
16. Wiener JM, Squillace MR, Anderson WL, Khatutsky G: Why
4. U.S. Department of Health and Human Services: 2013 Pov-
do they stay? Job tenure among certified nursing assistants
erty Guidelines. Washington, D.C.: Office of the Assistant
in nursing homes. Gerontologist 2009;49:198–210.
Secretary for Planning and Evaluation, 2013.
17. Castle NG, Degenholtz H, Rosen J: Determinants of staff job
5. Decker FH, Harris-Kojetin LD, Bercovitz A: Intrinsic job
satisfaction of caregivers in two nursing homes in Pennsyl-
satisfaction, overall satisfaction, and intention to leave the
vania. BMC Health Serv Res 2006;6:60.
job among nursing assistants in nursing homes. Gerontolo-
18. Secrest J, Iorio DH, Martz W: The meaning of work for
gist 2009;49:596–610.
nursing assistants who stay in long-term care. J Clin Nurs
6. Castle NG, Engberg J, Anderson R, Men A: Job satisfaction
2005;14:90–97.
of nurse aides in nursing homes: Intent to leave and turn-
19. Atchison JH: Perceived job satisfaction factors of nursing
over. Gerontologist 2007;47:193–204.
assistants employed in Midwest nursing homes. Geriatr
7. Parsons SK, Simmons WP, Penn K, Furlough M: Determi-
Nurs 1998;19:135–137; quiz 138.
nants of satisfaction and turnover among nursing assistants.
20. DeLoach R, Monroe J: Job satisfaction among hospice
The results of a statewide survey. J Gerontol Nurs 2003;
workers: What managers need to know. Health Care Man-
29:51–58.
age 2004;23:209–219.
8. Nahavandi A, Malekzadeh A: Organizational Behavior: The
21. Squillace M, Berovitz A, Rosenoff E, Remsburg R: An Ex-
Person-Organization Fit. Upper Saddle River, N.J.: Prentice
ploratory Study of Certified Nursing Assistants’ Intent to Leave.
Hall, 1999.
Washington, D.C.: U.S. Department of Helath and Human
9. Hosmer D LS: Applied Logistic Regression. New York: Wiley,
Services, Office of the Assistant Secretary for Planning and
2000.
Evaluation, 2008.
10. Crabtree B, Miller W: A template approach to text analysis:
Developing and using codebooks. In: Crabtree B, Miller W
(eds.): Doing Qualitative Research. Newbury Park, CA: Sage,
1999:163–177. Address correspondence to:
11. Kemper P, Heier B, Barry T, et al.: What do direct care Barbara A. Head, PhD, CHPN, ACSW, FPCN
workers say would improve their jobs? Differences across University of Louisville School of Medicine
settings. Gerontologist 2008;48(Special Issue 1):17–25. MDR Building, Suite 110
12. Brannon D, Barry T, Kemper P, Schreiner A, Vasey J: Job 511 South Floyd Street
perceptions and intent to leave among direct care workers: Louisville, KY 40202
Evidence from the Better Jobs Better Care demonstrations.
Gerontologist 2007;47:820–829. E-mail: barbara.head@louisville.edu