Relationship between hospital safety climate and outcomes

1
Relationship between hospital safety climate and
outcomes

• Sara Singer, Alyson Falwell, Shoutzu Lin, Toby
  Rathgeb, Laurence Baker
• AcademyHealth Annual Meeting
• June 26, 2006
• Financial support for this study has been
  provided by the Agency for Healthcare Research
  and Quality RO1 HSO13920 and by the VA HSRD

2
Patient safety and safety culture

• Significant patient safety problems plague US
  hospitals
• A culture of safety is increasingly recognized as
  a key to reducing adverse events in hospitals
• Yet we lack conclusive evidence of the
  relationship between safety culture and patient
  safety outcomes

3
Research questions

• How do individuals working in hospitals perceive
  the culture of safety in their facilities?
• Does perceived safety culture relate to hospital
  safety performance?
• Do both strength and uniformity of safety climate
  matter?
• Whose perception of safety culture corresponds
  most to safety outcomes?

4
Hospital safety culture and its measurement

• Shared values, beliefs, and norms of behavior,
  articulated by senior management and translated
  consistently into effective work practices
• Measured using the Patient Safety Climate in
  Healthcare Organizations (PSCHO) survey
• PSCHO measures safety climate, i.e., perceptions
  of safety culture at a point in time
• We examine rates of problematic responses,
  i.e., those indicating a lack of safety culture

5
Survey content and sample

• Survey includes 38 questions and 9 sub-scales
  specific to individual aspects of safety culture
• Surveyed 18,361 individuals from a stratified
  random sample of 92 US hospitals, representing
  four regions of the US and three size categories
  between March 2004 and May 2005
• 100 of active, hospital-based physicians,
• 100 of senior managers (dept heads or above)
• 10 random sample of all other personnel
• 52 response rate

6
Comparison of safety climate to outcomes

• Two individual-level factors, most proximate
  determinants of safety behaviors
• Willingness to seek help (alpha 0.58)
• Asking for help is a sign of incompetence
• Telling others about my mistakes is embarrassing
• If I make a mistake that has significant
  consequences and nobody notices, I do not tell
  anyone about it
• Fear of blame punishment (alpha 0.61)
• If people find out that I made a mistake, I will
  be disciplined
• Clinicians who make serious mistakes are usually
  punished

7
Examination of strength and uniformity of safety
climate to outcomes

• Strength of safety climate (mean problematic
  response)
• Uniformity of safety climate (variance in
  problematic response)
• Interaction between them

Safety culture uniformity
Safety culture strength
Safety outcomes
8
Consideration of alternative perceptions of
safety culture

• Nurses v. doctors
• Front line workers v. senior managers

9
Outcomes measures and analysis

• Measured clinical outcomes using AHRQ patient
  safety indicators (PSIs), computed using 2000
  HCUP data
• Selected 14 out of 20 PSIs that we hypothesized
  would be related to safety climate. Excluded
  indicators
• Whose outcome is driven by a single unit
  (obstetrics)
• That have been criticized as highly unreliable
  (failure to rescue)
• For which no events occurred in study hospitals
  (transfusion reaction)
• Included in analysis a stratified random sample
  of 47 hospitals from 15 states for which PSCHO
  and PSI data are available
• Analyzed relationship of safety climate to all
  PSIs or groups of PSIs

10
Safety climate in 92 hospitals
My unit recognizes individual safety achievement
through rewards and incentives
If I make a mistake that has significant
consequences and nobody notices, I do not
tell anyone about it
11
Safety climate in 92 hospitals
Telling others about my mistakes is embarrassing
If I make a mistake that has significant
consequences and nobody notices, I do not
tell anyone about it
Asking for help is a sign of incompetence
12
Safety climate in 92 hospitals
If people find out that I made a mistake, I will
be disciplined
Clinicians who make serious mistakes are usually
punished
13
Relationship of safety climate dimensions to PSIs

• Safety climate related more strongly to PSIs than
  did other hospital characteristics
  nurse-to-patient hour ratios, early technology
  adoption, and incident reporting activity

14
Relationship of safety climate dimensions to PSIs

• Safety climate related more strongly to PSIs than
  did other hospital characteristics
  nurse-to-patient hour ratios, early technology
  adoption, and incident reporting activity

15
Relationship of safety climate dimensions to PSIs

• Safety climate related more strongly to PSIs than
  did other hospital characteristics
  nurse-to-patient hour ratios, early technology
  adoption, and incident reporting activity

16
Relationship of safety climate dimensions to PSIs

• Safety climate related more strongly to PSIs than
  did other hospital characteristics
  nurse-to-patient hour ratios, early technology
  adoption, and incident reporting activity

17
Relationship of strength and uniformity of
willingness to seek help to PSIs
18
PSI-safety climate relationship differences by
type of personnel
19
PSI-safety climate relationship differences by
type of personnel
20
PSI-safety climate relationship differences by
type of personnel
21
PSI-safety climate relationship differences by
type of personnel
22
Discussion

• Two dimensions of safety climate were
  statistically significantly associated with PSI
  performance
• Lack of willingness to seek help (plt.05) and
• Fear of blame punishment (plt.01)
• Uniformity moderated the relationship between
  willingness to seek help and safety performance
• Nurse and front line workers perceptions
  correlated more strongly with adverse safety
  events than did physician and senior managers
  perceptions respectively

23
Strengths and limitations

• Strengths
• One of the first studies to link safety climate
  directly to clinical safety outcomes
• Results representative of US hospitals
• Limitations
• Need to explore other dimensions of safety
  climate
• Need verification of adverse event rates
• Need longitudinal study to establish causality

24
Conclusion

• Findings support the claim that culture and
  outcomes are strongly related
• Both strength and uniformity of safety climate
  matter
• Senior managers may misperceive important aspects
  of safety climate
• Presence of blame and unwillingness to seek help
  suggest interventions that address
  deeply-ingrained beliefs are needed to improve
  hospital safety culture