Article Text

Differing perceptions of safety culture across job roles in the ambulatory setting: analysis of the AHRQ Medical Office Survey on Patient Safety Culture
  1. John Hickner1,
  2. Scott A Smith2,
  3. Naomi Yount2,
  4. Joann Sorra2
  1. 1Family Medicine, University of Illinois at Chicago School of Medicine, Chicago, Illinois, USA
  2. 2Westat, Rockville, Maryland, USA
  1. Correspondence to Dr John Hickner, Family Medicine, University of Illinois at Chicago School of Medicine, Chicago, IL 60612, USA; hickner{at}uic.edu

Abstract

Background Experts in patient safety stress the importance of a shared culture of safety. Lack of consensus may be detrimental to patient safety. This study examines differences in patient safety culture perceptions among providers, management and staff in a large national survey of safety culture in ambulatory practices in the USA.

Methods The US Agency for Healthcare Research and Quality Medical Office Survey on Patient Safety Culture (SOPS) assesses perceptions about patient safety issues and event reporting in medical offices (ie, ambulatory practices). Using the 2014 data, we analysed responses from medical offices with at least five respondents. We calculated differences in perceptions of patient safety culture across six job positions (physicians, management, nurse practitioners (NPs)/physician assistants (PAs), nurses, clinical support staff and administrative/clerical staff) for 10 survey composites, the average of the 10 composites and an overall patient safety rating using multivariate hierarchical linear regressions.

Results We analysed data from 828 medical offices with responses from 15 523 providers and staff, with an average 20 completed surveys per medical office (range: 5–367) and an average medical office response rate of 65% (range: 3%–100%). Management had significantly more positive patient safety culture perceptions on nine of 10 composite scores compared with all other job positions, including physicians. The composite that showed the largest difference was Communication Openness; Management (85% positive) was 22% points more positive than other clinical and support staff and administrative/clerical staff. Physicians were significantly more positive than PAs/NPs, nursing staff, other clinical and support staff and administrative/clerical staff on four composites: Communication About Error, Communication Openness, Staff Training and Teamwork, ranging from 3% to 20% points more positive.

Conclusions These findings suggest that managers need to pay attention to the training needs of office staff, since this was an area with one of the greatest gaps in perceptions. In addition, both office managers and physicians need to encourage more open communication. As medical offices innovate to improve value, efficiency and patient-centred care, it is important that they continue to foster shared perceptions about what organisational members need, understanding that those perceptions may differ systematically by job position.

  • Medication safety
  • Patient safety
  • Ambulatory care
  • Safety culture

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Background

From its roots in traditional high reliability organisations (HROs), such as commercial aviation and nuclear power, to its more recent application in healthcare, the concept of safety culture refers to the integral framework that advances institutional policies into real actions and safe outcomes. Studies of safety in HROs repeatedly find that, in spite of any written standards or expectations, new staff in an organisation quickly acquire the practices and behaviours of those working around them, adopting the culture of the organisation.1

An organisation's safety culture is shaped by the shared beliefs and behaviours of all of its members, but particularly the beliefs and behaviours that are encouraged and rewarded by owners and/or managers.2 Reviews of safety culture literature indicate that the safest possible organisation is one in which all staff are united in the belief that safety comes first.3 Communication that moves in at least two directions and emphasises shared values and goals for the organisation is necessary to make real improvements in an organisation's safety culture.4

Experts note the importance of communicating a shared expectation of safety to all employees. Despite the value of agreement among all employees, some studies have found that managers respond more positively to patient safety culture surveys than other job positions. On the Agency for Healthcare Research and Quality (AHRQ) Hospital Survey on Patient Safety Culture, administrators and managers respond more positively on all 12 areas of patient safety culture.5 Another patient safety culture survey in a California healthcare system found clinicians to be more likely to provide negative responses than non-clinicians, and senior managers to be less likely to provide negative responses than non-senior managers.6

Recent studies examining differences in patient safety culture perceptions in medical offices have been conducted in Switzerland, Scotland, Turkey and Germany and in US Air Force clinics.7–11 Three of the studies found some differences in perceptions by job type, but the findings were not consistent across these studies. The study from Turkey found no significant differences among job positions, and the study of four Air Force clinics found no differences by job position in overall ratings of patient safety or on five of six subscales of the survey used. To further investigate the relationship between safety culture and job positions (ie, we examined differences in patient safety culture perceptions using a large US dataset from the 2014 AHRQ Medical Office SOPS comparative database.12

Methods

Data sources and measures

The AHRQ Medical Office SOPS assesses perceptions about patient safety issues, medical error and event reporting in medical offices. The survey was developed as an extension of the Hospital Survey on Patient Safety Culture. The development team reviewed research pertaining to safety, patient safety, healthcare quality, ambulatory medicine, medical errors, error reporting, safety climate and culture, and organisational climate and culture. In addition, existing medical office surveys were reviewed and more than two dozen experts in medical office practice, patient safety and medical office providers and staff were consulted to identify key topics and issues. Based on these activities, the researchers identified a potential list of composites to include in the survey. The survey was pilot tested, revised, reviewed by the technical experts and then AHRQ released it in 2009. More information on the Medical Office SOPS may be found on the AHRQ web site: http://www.ahrq.gov/professionals/quality-patient-safety/patientsafetyculture/medical-office/index.html

The survey includes 38 items that measure 10 dimensions or ‘composites’ of patient safety culture and 14 additional items about patient safety and quality in office practice. These additional 14 items were included to help respondents understand patient safety issues that may occur in medical offices, and they are not included in the composite measures in the survey or in this study. The survey also asks the respondent to provide an overall patient safety rating for their office and to indicate their job position: (1) physician; (2) physician assistant (PA), nurse practitioner (NP), clinical nurse specialist, nurse midwife, advanced practice nurse, etc; (3) management (defined as the practice or office manager/administrator, nurse or laboratory manager or office administrator); (4) administrative or clerical staff; (5) registered nurse (RN), licensed vocational nurse (LVN), licensed practical nurse (LPN); (6) other clinical staff or clinical support staff and (7) other position. Staff that selected other position or did not respond were excluded from these analyses.

Each of the 10 patient safety culture composites is listed and defined in table 1. The items that make up these composites and the overall patient safety rating item, as well as the internal consistency and reliability of the composites, are provided in the online supplementary appendix.

Table 1

Agency for Healthcare Research and Quality (AHRQ) Medical Office Survey on Patient Safety Culture (SOPS) composites and definitions

AHRQ funds a comparative database to enable medical offices to compare their survey results with other medical offices.12 Data for this study came from the 2014 Medical Office SOPS database, which included data from 27 103 providers and staff in 935 medical offices across the US Medical Offices administer the AHRQ survey and voluntarily submit their survey data for inclusion in the database. Prior to analysis, the data were cleaned as described in the notes of the Medical Office SOPS 2014 User Comparative Database Report.12 Medical offices missing data on the measures used in this study were excluded from this analysis dataset.

The analysis dataset, therefore, consisted of data from 828 medical offices with responses from 15 523 providers and staff. There was an average of 20 completed surveys per medical office (range: 5–367), with an average medical office response rate of 65%. (The response rates by job position could not be calculated since medical offices did not provide denominators broken out by job position, but rather provided just an overall number of people surveyed).

This study examined differences in scores among job positions on the 10 patient safety culture composites, the average score across the 10 composites and the overall patient safety rating (ie, Question G2: ‘Overall, how would you rate the systems and clinical processes your medical office has in place to prevent, catch, and correct problems that have the potential to affect patients?’). The survey's composite items use a 5-point Likert-type scale (eg, strongly agree to strongly disagree and always to never). For positively worded items, per cent positive response was the combined percentage of respondents who answered strongly agree or agree, or always or most of the time. Negatively worded items were reverse coded to calculate a per cent positive response, which was the combined percentage of respondents who answered strongly disagree or disagree, or rarely or never. The overall patient safety rating was calculated as the combined percentage of respondents who answered very good or excellent. We calculated a per cent positive score for each of the composites and the overall patient safety rating for each job position within each medical office.

The comparative database also includes information on medical office characteristics: (1) type of majority ownership (eg, hospital or health system owned, provider owned), (2) whether the office was single or multispeciality and (3) the number of providers who worked in their medical office during a typical week (a proxy for medical office size). We used these three variables as covariates in the analyses to control for the relationships they have with composites.

Analysis

Multivariate hierarchical linear regressions were used to estimate the per cent positive means of the 10 Medical Office SOPS composites, an average score across composites and the overall patient safety rating for each job position, controlling for the medical office characteristics of majority ownership, single speciality versus multispeciality and medical office size. We included these three characteristics since they have been found to be related to patient safety culture scores.12 ,13 We used two-level hierarchical modelling due to the nested nature of the data. That is, job position averages are nested within medical offices, and this clustering can lead to biased SEs and subsequent significance tests. Intraclass correlations (ICC) and design effects were calculated to determine whether the nested nature of the data would bias the results and indicate that multilevel analyses would be necessary.14 ICC values >0.05 indicate that the multilevel structure of the data needs to be taken into consideration, while ICC values <0.05 signify that the consequences of not using multilevel analyses are minimal.15 In addition, unstandardised estimates are provided to quantify the effect of the medical office characteristics on patient safety culture measures. Estimated means and SEs were calculated accounting for the clustering of respondents within medical offices. To take into account multiple comparisons of least square estimated means, we used Tukey-corrected p values <0.05 to identify significant findings. All analyses were completed using SAS V.9.3.

Results

Table 2 presents descriptive information about the types of medical offices included in the analysis dataset. The majority of medical offices (65.1%) were single speciality. Most practices (44.2%) had between four and nine providers working during a typical week. Three-quarters (74.8%) of medical offices were owned either by a hospital or health system.

Table 2

Characteristics of 828 medical offices in the 2014 Medical Office Survey on Patient Safety Culture database

Ignoring job positions, the 12 measures (10 patient safety culture composites, the average score across composites and the overall patient safety rating) all exhibited normal distributions (skewness max: 2.03, kurtosis max: 4.65). The ICCs for the 12 measures were greater than the 0.05 criterion; with an average of 0.25, ranging from 0.17 to 0.42. Therefore, between 17% and 42% of the variance may be attributed to medical office membership, and this established the need for multilevel analyses.

Each of the 12 measures were analysed independently using multivariate hierarchical linear regressions, controlling for medical office size, single speciality versus multispeciality and type of ownership. Medical office size was a significant covariate predicting 11 of the 12 measures such that smaller offices tended to have higher patient safety culture scores. Type of ownership was also a significant covariate for 6 of the 12 measures, such that physician-owned offices had higher patient safety culture scores than university-owned offices on four composites and lower than system-owned on two composites. However, whether a medical office was single speciality or multispeciality was not a significant covariate in any of the measures.

Table 3 presents per cent positive scores by job position for each of the measures from the multivariate hierarchical linear regressions. Teamwork was the patient safety culture composite with the highest scores across all job positions, whereas Work Pressure and Pace was the composite with the lowest scores across all job positions. Table 4 presents only the differences from table 3 that are statistically significant across the job positions. As shown in table 4, management consistently had more positive patient safety culture perceptions on almost all measures compared with all other job positions. Management was significantly more positive than physicians (on 11 of 12 measures); PAs/NPs (on all 12 measures); nursing staff (on all 12 measures); other clinical and support staff (on 11 measures) and administrative and clerical staff (on 11 measures).

Table 3

Per cent positive scores by six job positions on the 2014 Medical Office Survey on Patient Safety Culture (SOPS)

Table 4

Significant differences among six job positions on the 2014 Medical Office Survey on Patient Safety Culture (SOPS)

The perceptions of management and other non-physician job positions differed most on Communication Openness and Staff Training. Management was 20%–22% points more positive on Communication Openness, and 20%–21% points more positive on Staff Training compared with nursing, other clinical and support staff and administrative/clerical staff. Figures 1 and 2 show per cent positive scores by job position for Communication Openness and Staff Training.

Figure 1

Communication Openness average per cent positive scores by job position. (1) PA, NP, etc, are comprised of physician assistant, nurse practitioner, clinical nurse specialist, nurse midwife and advanced practice nurse. (2) Nursing is comprised of nurse, licensed vocational nurse and licensed practical nurse. (3) Per cent positive scores were calculated controlling for speciality, number of providers and ownership.

Figure 2

Staff Training average per cent positive scores by job position. (1) PA, NP, etc, are comprised of physician assistant, nurse practitioner, clinical nurse specialist, nurse midwife and advanced practice nurse. (2) Nursing is comprised of nurse, licensed vocational nurse and licensed practical nurse. (3) Per cent positive scores were calculated controlling for speciality, number of providers and ownership.

Physicians were more positive than PAs/NPs, nursing, other clinical and support staff and administrative/clerical staff on Communication About Error, Communication Openness, Staff Training and Teamwork. The one patient safety culture composite, where management and physicians were in agreement (no significant difference between them) and more positive than all other job positions, was Communication Openness.

The patient safety culture composites with the greatest agreement in perceptions (ie, had the fewest significant differences between job positions) were Organisational Learning and Overall Perceptions of Patient Safety and Quality; although management was still significantly more positive than all other staff positions on both of these composites.

The job positions most in agreement with one another were other clinical and support staff and administrative/clerical staff (no significant differences); nursing and other clinical and support staff (one significant difference) and PAs/NPs and nursing (one significant difference).

Discussion

This is the largest study to date that has explored job position differences in perceptions of patient safety culture in medical offices. Prior studies in hospital settings have found management positions to have more positive perceptions than other job positions.5 ,6 We found this to be true in medical offices as well. We did not, however, expect to see such large and consistent differences between managers and all other staff, including physicians. This was especially true for the key categories of Communication Openness and Staff Training, where managers’ perceptions were about 20 percentage points more positive. Similarly, managers’ Overall Ratings on Patient Safety, a global measure of patient safety culture, were 18–19 percentage points higher than any other job position. It seems clear from this data that managers have a tendency to believe their patient safety culture is stronger that it really is. Physicians, too, have a much more positive view of many areas of patient safety culture compared with other staff.

As noted previously, experts in safety culture consistently stress the importance of a shared culture of safety.3 ,4 Lack of consensus, therefore, may be detrimental to this common goal. As medical offices strive to improve their performance in quality and safety, promoting a strong safety culture is likely to be an ongoing challenge. The results of this analysis suggest that office or practice managers/administrators in particular need to pay attention to the training needs of office staff, since this was an area with one of the greatest gaps in perceptions. In addition, both office managers and physicians need to be more open to staff ideas about how to improve office processes, encourage staff to express alternative viewpoints and ask questions. Our findings reinforce the importance of Communication Openness as one of the essential elements to ensuring a positive patient safety culture within healthcare.

Finally, we note that the patient safety culture area scoring lowest, regardless of job position, was Work Pressure and Pace. Managers, physicians, NPs/PAs, nursing and all other staff in medical offices feel rushed when taking care of the volume of patients that need to be seen and do not feel that there are enough staff to handle their patient loads and work effectively. There is great need to institute processes that improve efficiency and throughput while reducing the frenetic work pace that is common in many offices, which leads to management, physician and staff burnout.

As medical offices innovate to improve value, efficiency and patient-centred care through patient-centred medical homes and new healthcare delivery models, it is important that they not lose sight of the aspects of their organisational culture that are fundamental to quality and safety and that they continue to foster shared perceptions about what organisational members need, understanding that those perceptions may differ.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors All authors contributed to drafting the manuscript and interpreting the results, and all have approved the final manuscript for submission.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Data used in this analysis were from the Agency for Healthcare Research and Quality (AHRQ) Medical Office Survey on Patient Safety Culture Comparative Database. The database is funded by AHRQ and managed by Westat under contract number HHSA 290200710024C.