A SUPPORTIVE NURSING CARE UNIT FOR REGISTERED NURSES

1
A SUPPORTIVE NURSING CARE UNIT FOR REGISTERED
NURSES

• Debra S. Hall, RN, PhD candidate
• University of Kentucky, College of Nursing
• Advisor Julie Sebastian, PhD, RN, CS, FAAN

2
PURPOSE

• investigate the relationship between a specific
  type of nursing unit governance structure and
  nurse retention, job satisfaction, job control,
  self-efficacy, collective efficacy, job stress,
  co-worker support, supervisor support, somatic
  complaints, and absenteeism due to illness of
  Registered Nurses in a hospital recognized for
  good nursing care

3
SIGNIFICANCE

• In an ANA website survey, 70.5 of nurses cited
  the acute and chronic effects of stress and
  overwork among their top three health and safety
  concerns

4
SIGNIFICANCE

• With a projected shortage of nearly 20 of RNs by
  2020, and a cost of 42,000 for the turnover of
  each general medical-surgical nurse and 64,000
  for turnover of a specialty nurse (Advisory Board
  Company, 2000) determining strategies to recruit
  and retain nursing personnel is a critical issue
  (Buerhaus, 2000)

5
DESIGN

• cross-sectional
• two phase (quantitative qualitative)
• comparative study of the effects of unit
  governance structure on RNs working in three
  different patient care units within the same
  hospital

6
DESIGN

• Unit 1 traditional nursing unit structure with
  a non-specialized patient population (n 28)
• Unit 2 a shared governance structural framework
  (n 24)
• Unit 3 a specialized, homogeneous inpatient
  population that does not change, in which there
  are few off-service patients
• (n 29)

7
METHODS Phase I

• Phase I quantitative survey comprised of nine
  instruments and single item
• questions for age, marital status, gender,
  ethnic/racial background, level of nursing
  education, length of time working in present
  unit, and amount of experience as a bedside RN

8
INSTRUMENTS

• The Decision Latitude Scale of the Job Content
  Questionnaire (JCQ) (Karasek et al., 1998)
• Maastricht Autonomy Questionnaire (MAQ)
• (de Jonge, 1995)
• Inventory of Socially Supportive Behaviors
  (ISSB) (Barrera, Sandler, Ramsay, 1981)
• Personal Efficacy Beliefs tool (10-items) and
  collective efficacy instrument (Riggs, Warka,
  Babasa, Betancourt, and Hooker, 1994)

9
INSTRUMENTS

• Nurse Work Stress Scenarios (NWSS) - 24-item
  instrument designed by the primary investigator
• Single-item question on job satisfaction (Warr,
  Cook, Wall, 1979)
• The Physical Symptoms Inventory (PSI) (Spector
  Jex, 1998)
• Staff turnover was calculated for each unit

10
SAMPLE Phase I

• Eighty-one nurses (n 69) working at least 24
  hours per week in a typical staff position in one
  of three adult nursing units were approached
• RNs were at least 18 years old and could not be
  on orientation, probation, light duty assignment,
  or other non-typical assignments during the
  study

11
METHODS Phase II

• Phase II focus groups with 5 RNs from each of
  the three units
• Nurses were asked open-ended questions related to
  stressors, coping mechanisms used, types of
  interventions they used to change stressful
  situations, and social support they received

12
SAMPLE Phase II

• Fifteen RNs (n 13) working in the same units
  were purposively chosen based on age for Phase II
  (response rate 87)
• SETTING - a south central teaching hospital
  serving patients from suburban and rural areas

13
PROCEDURE

• Obtained IRB approval
• Contacted Chief Nursing Officer and Unit Managers
• Mailed cover letter, questionnaire, postcard, and
  return SASE to each nurse
• Used Dillmans Total Design Method (Dillman, 2000)

14
PROCEDURE

• Two weeks later, sent reminder postcard
• Four weeks later, sent a follow-up letter and
  replacement questionnaire, postcard, and return
  envelope to non-respondents
• Sent thank you letter to the respondents
• of the survey

15
DATA ANALYSIS

• Descriptive, univariate statistics (M, SD)
• Spearman-S Ranking
• Independent t-test
• Chi-square test of independence
• One-way analysis of variance (ANOVA)
• One-way multivariate analysis of variance
  (MANOVA)
• Multiple regression

16
QUALITATIVE ANALYSIS

• Substantive coding of work-related stressors,
  coping mechanisms, and social support comments
• Recoded
• Table for each topic
• Selective coding
• Table of coping mechanisms to compare groups

17
RESULTS - Sociodemographic

• Male RNs scored significantly higher for overall
  work stress (M 75, SD 35) than female RNs (M
  40, SD 19), t(51) -2.98, p .04
• Caucasian nurses used a larger number of direct
  coping mechanisms in the eight scenarios (M
  50.9, SD 5.8), than did African-American
  nurses, (M 42, SD 6), t(53) 3.09, p .03

18
RESULTS - Sociodemographic

• Nurses who were married or had steady partners
  perceived higher levels of job control, higher
  collective efficacy, job satisfaction, and used
  more indirect coping methods, experienced less
  somatic symptoms, and had less days in which they
  felt ill but came to work than RNs who were
  single or divorced

19
RESULTS - Occupational Stress

• Hypothesis - the level of job stress for RNs in a
  clinical role would vary by type of nurse unit
  governance
• Not supported by findings
• Most commonly occurring stressor organizational
  barriers such as a lack of supplies and equipment
  
• The scenario eliciting the most stress involved a
  lack of skilled labor

20
RESULTS - Occupational Stress

• Supervisor support predicted 7 of the variance
  in frequency of encountering stressful situations
  at work, F (1,65) 5.10, p lt ,05, adjusted R2
  .06
• Supervisor support and type of unit predicted 18
  of the variance in experiencing work stress, F
  (1, 49) 5.39, p lt .05, adjusted R2 .15

21
RESULTS - Job control

• Hypothesis - levels of job control would vary by
  type of nurse unit governance
• Not supported by findings as there were no
  significant differences among units on amount of
  job control overall average scores were low
• MAQ (M 2.6, SD .66) between little control
  and some control
• Decision Latitude Scale had higher scores for
  skill discretion (learning new things, requiring
  high level of skill) than for decision authority
  (say about what happens on job) (M 67.3, SD
  9.2)

22
RESULTS

• Supervisor support predicted 18 of the variance
  in job control as measured by the MAQ, F (1,64)
  13.62, p lt .005, adjusted R2 .16
• Supervisor support predicted 12.3 of the
  variance in job control based on the decision
  latitude scale, F (1,65) 9.09, p lt .005,
  adjusted R2 .11

23
RESULTS Coworker Support

• Hypothesis - level of support from co-workers for
  RNs in a clinical role will vary by type of nurse
  unit governance not supported
• Co-worker support was uniformly high among all
  three units (M 4.0, SD 1.1)

24
RESULTS Coworker Support

• Positive correlation between amount of co-worker
  support and supervisor support, unit efficacy,
  and job satisfaction
• Regression equation combining number of years in
  current unit with supervisor support and number
  of years as an RN predicts 25 of the variance in
  co-worker support, F (3,62) 6.93, p lt .001,
  adjusted R2 .22

25
RESULTS Self-efficacy

• Hypothesis RNs working in shared governance or
  specialty unit will have higher self-efficacy
  levels not supported by findings
• Nurses ranked themselves high on work related
  self-efficacy in all three units with a mean
  score of 4.8 (SD .7)

26
RESULTS Self-efficacy

• Positive correlation between job self-efficacy
  and age, years of experience as an RN, and years
  working in the current unit
• Regression equation with age and type of unit
  predicts 15.6 of the variance in job
  self-efficacy, F (2,65) 6.01, p lt .005,
  adjusted R2 .13

27
RESULTS Collective efficacy

• Hypothesis - level of collective efficacy for RNs
  in a clinical role will vary by type of unit
  governance
• Not supported by findings as the overall rating
  of unit efficacy was similarly high among all
  three units (M 4.4, SD .8)

28
RESULTS Collective efficacy

• Positive correlation between unit efficacy and
  job control, co-worker and supervisor support,
  job satisfaction, and indirect coping mechanisms
• Regression equation with supervisor support
  predicted 10.6 of the variance in unit efficacy,
  F (1, 65) 7.7, p lt .05, adjusted R2 .09

29
RESULTS Coping with Stress

• Hypothesis - RNs working in the clinical role in
  a shared governance unit would have more
  proactive or direct coping methods and use an
  increased number of coping methods than RNs
  working in traditional or specialized units
• Results did not support this hypothesis

30
RESULTS Coping with Stress

• Positive correlation between number of proactive
  coping mechanisms used and number of somatic
  complaints and number of days the RN felt ill but
  came to work

31
RESULTS Coping with Stress

• Positive correlation between the number of
  indirect coping mechanisms used and job control
  (using the MAQ), unit efficacy and job
  satisfaction
• Negative correlation between with the frequency
  of encountering stress and overall stress and
  number of indirect coping mechanisms used

32
RESULTS Days Ill

• Hypothesis - RNs working in shared governance
  will have less absenteeism related to illness
  than RNs working in a unit with traditional
  governance
• Hypothesis supported one-way ANOVA,
  between-groups design revealed a significant
  effect for type of unit governance, F(2, 64)
  3.37 p lt .04

33
RESULTS Days Ill

• Tukeys HSD test showed that nurses in the shared
  governance unit (M .7, SD .9) used
  significantly less sick days than nurses in the
  traditional unit (M 3.4, SD 5.4) (p lt .05)

34
RESULTS Days Ill

• Negative correlation between amount of supervisor
  support and number of days the RN felt ill but
  came to work
• Positive correlation between the number of days
  absent due to sickness, number of symptoms
  experienced and number of direct coping
  mechanisms used

35
RESULTS Days Ill

• Number of years in current unit with type of unit
  predicted 15.7 of the variance in days absent
  from work due to illness, F (2,63) 5.93, p lt
  .005, adjusted R2 .13
• Age and supervisor support predicted 13 of the
  variance in number of sick days used with each
  illness,
• F (5, 52) 5.9, p lt .05, adjusted R2 .13
• Supervisor support predicted 7.4 of the variance
  in number of days the nurse felt ill, F (1,65)
  5.2, p lt .05

36
RESULTS Somatic Complaints

• Hypothesis - RNs working in shared governance or
  specialty units will have a lower number of
  somatic complaints than RNs working in a unit
  with traditional governance
• Hypothesis not supported by findings
• Most of the somatic complaints listed had been
  experienced by the nurses (M 26.1, SD 4.5).

37
RESULTS Somatic Complaints

• Significant positive correlation between the
  number of somatic complaints reported and amount
  of job stress, frequency of stressful job
  situations, number of days feeling ill, number of
  days using sick time, and number of proactive
  coping mechanisms used
• Somatic complaints correlated negatively with
  amount of supervisor support, amount of job
  control, level of job satisfaction, and number of
  indirect coping mechanisms

38
RESULTS Somatic Complaints

• Supervisor support predicted 8 of the variance
  in reported somatic complaints, F (1, 65) 5.5,
  p lt .05, adjusted R2 .06
• It accounted for 18 of the variance in the
  amount of upset stomach/nausea experienced by
  nurses, F (1, 65) 14.25, p .001, adjusted R2
  .167 and for 12 of the variance in the amount
  of diarrhea experienced by nurses, F (1, 65)
  8.55, p .005, adjusted R2 .10
• Supervisor support predicted 9 of the variance
  in reported loss of appetite, F (1, 65) 6.6, p
  lt .05, adjusted R2 .08

39
RESULTS Job satisfaction

• Results supported the hypothesis that RNs working
  in shared governance or specialty units will have
  more job satisfaction that RNs working in a unit
  with traditional governance
• Job satisfaction was highest on the specialty
  unit (M 5.08, SD 1.32) and lowest on the
  traditional unit (M 4.48, SD 1.60) however,
  the traditional unit had the highest level of
  supervisor support

40
RESULTS Job satisfaction

• Supervisor support and type of unit accounted for
  27 of the variance in job satisfaction of
  nurses, F (2, 64) 12.1, p lt .0001, adjusted R2
  .25
• Significant positive correlation between job
  satisfaction and job control, co-worker support,
  supervisor support, and number of indirect coping
  mechanisms
• Significant negative correlation between job
  satisfaction and work stress, frequency of
  encountering stressful situations, and number of
  somatic complaints

41
RESULTS RN Turnover

• Hypothesis - specialty units will have less RN
  turnover than traditional governance units
• A one-way ANOVA, between-groups design revealed a
  significant effect for staff turnover, F(2, 66)
  Infinity p lt .0001

42
RESULTS -

• Tukeys HSD test showed that nurses in the
  specialty unit had a significantly lower rate of
  turnover (7.2) than nurses in the traditional
  unit (9.5) who had a significantly lower rate of
  turnover than nurses in the shared governance
  unit (18)
• Supervisor support and type of unit governance
  predicted 17 of the variance in staff turnover,
  F (2,64) 6.51, p lt .005, adjusted R2 .14

43
RESULTS Supervisor Support

• A one-way ANOVA, between-groups design revealed a
  significant effect for supervisor support, F(2,
  65) 7.26 p lt .0014
• Tukeys HSD test showed that nurses in the
  traditional governance unit (M 2.7, SD 1.3)
  had significantly more supervisor support (once
  or twice a week) than nurses in the other two
  units

44
RESULTS Supervisor Support

• Nurses in the specialty unit had significantly
  more supervisor support (M 2.0, SD .9) (p lt
  .05) than nurses in the shared governance unit (M
  1.6, SD .4)
• A one-way MANOVA, between-groups design, revealed
  a significant multivariate effect for supervisor
  support, Wilks lambda .19, F(2, 39) 3.69 p
  lt .05 between the shared governance unit and the
  traditional governance unit

45
RESULTS Supervisor Support

• positive correlations between supervisor support
  and job control, co-worker support, unit
  efficacy, and job satisfaction
• Negative correlations between supervisor support
  and work stress, frequency of encountering
  stressful situations, number of somatic
  complaints, and number of days the nurse felt ill
  but was not absent

46
RESULTS Type of Unit Governance

• Results supported the hypothesis that specialty
  units with homogeneous patient populations will
  have less RN turnover than traditional governance
  units however, there was no support for
  decreased RN turnover in a shared governance unit
  as this unit had a higher rate of turnover than
  either of the other two units and a lower amount
  of perceived supervisor support

47
RESULTS Type of Unit Governance

• Results supported the hypothesis that RNs working
  in a shared governance unit will have less
  absenteeism related to illness than RNs working
  in a unit with traditional governance
• Results supported the hypothesis that RNs working
  in shared governance or specialty units will have
  more job satisfaction that RNs working in a
  traditional governance unit

48
RESULTS Type of Unit Governance

• Although most of the hypotheses for the study
  were not supported, the significant difference in
  amount of supervisor support between the units
  had a greater effect on the dependent variables
  than any other independent variable

49
CONCLUSIONS

• Findings support the effect of shared governance
  unit structure on job satisfaction and
  absenteeism however, they do not support
  positive outcomes on physical symptoms and
  turnover rate
• Although structural environment was not
  significantly related to outcomes, social
  environment in the form of supervisor support was
  related to outcomes
• It is the affirmative contact with the first-line
  supervisor, rather than a shared governance
  model, that relates to physical and psychological
  outcomes

50
CONCLUSIONS

• Male nurses may perceive more work stress related
  to their role as a minority in the work
  environment of nurses
• Direct (proactive) mechanisms of coping may cause
  more perceived work stress and physical symptoms
  
• The positive correlation between using more
  proactive coping mechanisms and having more
  somatic complaints and days the RN felt ill may
  demonstrate the strain that direct action in
  response to stress causes, even if the action is
  successful

51
CONCLUSIONS

• Lack of control over organizational factors
  rather than work tasks appear to create a
  perception of low job control among nurses
• The use of indirect coping mechanisms (primarily
  realizing there is nothing the nurse can do about
  the situation) may have the effect of releasing
  the nurse to focus control on other aspects of
  his/her work

52
LIMITATIONS

• Size of study N 69 and setting limited to one
  hospital
• Small number of male nurses
• Small number of minority nurses
• Reliance on self-report

53
IMPLICATIONS FOR FUTURE RESEARCH

• Supervisor actions perceived by nurses as
  supportive and their effect
• Differences in perceived stress levels between
  male and female nurses
• The use of indirect and direct coping mechanisms
  to further understand the relationship between
  use of specific types of coping mechanisms and
  outcomes

54
ACKNOWLEDGMENTS

• "This research was supported (in part) by a pilot
  project research training grant from the
  University of Cincinnati. The University of
  Cincinnati, an Education and Research Center, is
  supported by Training Grant No. T42/CCT510420
  from the Centers for Disease Control and
  Prevention/National Institute for Occupational
  Safety and Health. The contents are solely the
  responsibility of the author and do not
  necessarily represent the official views of the
  National Institute for Occupational Safety and
  Health."

55
THANK YOU FOR YOUR SUPPORT