Emergency Physician Perceptions of Patient Safety Risks
Article Outline
Study objective
Although national standards to address patient safety exist, their relevance to emergency department (ED) patient safety is unclear. We survey practicing emergency physicians to assess their perceptions of the relative importance of patient safety concerns and how these varied by urban/rural location and practice characteristics.
Methods
We developed and analyzed electronically collected survey data that assessed emergency physician perceptions of patient safety risks. American College of Emergency Physicians (ACEP) members rated 16 patient safety concerns with a 5-point Likert scale.
Results
Of 2,507 emergency physician respondents, 1,114 (44%) practiced in urban, 1,056 (42%) in suburban, and 337 (13%) in rural settings. Crowding from inpatient boarding (mean Likert scale score 4.3), availability of specialty consultation (mean 4.1), and nursing shortages (mean 3.9) were the greatest concerns. Rural respondents ranked consultant availability (mean 4.3), lack of follow-up after ED care (mean 3.8), and nurse shortages (mean 3.8) as top concerns. Crowding was the greatest concern for suburban (mean 4.3) and urban emergency physicians (mean 4.5) but was ranked seventh by rural emergency physicians (mean 3.5). Crowding was perceived as a greater problem as hospital size, ED volume, and the percentage of patients who left without being seen increased, regardless of practice location.
Conclusion
In this sample of practicing emergency physicians, rural emergency physicians' patient safety concerns differ from those of their urban/suburban counterparts. For urban/suburban emergency physicians, crowding is the greatest safety concern; for rural emergency physicians, consultant availability was the greatest concern. Emergency physicians' greatest concerns are not routinely measured and reported as part of national patient safety benchmarking programs.
Introduction
Since the publication of the Institute of Medicine's report To Err Is Human,1 medical error and patient safety have received increased public attention. In response, the medical community has attempted to identify areas of highest risk and design interventions aimed at them.2 The emergency department (ED) has been identified as an area of particular risk.3 Various organizations have published lists with priority areas, including accrediting organizations such as The Joint Commission,4 the Center for Medicare & Medicaid Studies,5 and national independent medical quality organizations (National Quality Forum).6 These lists and quality measures are generally selected through a consensus process using experts in various fields.
What is already known on this topic
National concerns about the safety of health care have led to the promulgation of national safety goals for health care organizations.
What question this study addressed
Emergency physicians' perceptions of the most important safety problems in their practices were compared with published national patient safety goals.
What this study adds to our knowledge
The 2,817 emergency physicians' (14% of those eligible; 38% of those who opened the e-mail) primary safety concerns were crowding, lack of nursing staff, and unavailability of consultants. None of these were included in national safety goals, and most national goal items were not highly ranked.
How this might change clinical practice
The evident disconnect between national safety goals and locally perceived safety concerns affects the credibility and acceptance of the national goals. Steps should be taken to align national goals with emergency physicians' actual concerns.
Little is known, however, about how national patient safety measures apply to the ED or how they may vary by different care settings, such as rural/urban differences. Some measures may apply to all areas of health care but others may not be as important in the ED. Characteristics of the hospital or ED may also affect what issues are deemed most important for patient safety, yet little is known about how these characteristics affect the perceptions of emergency physicians.
We surveyed emergency physician members of the American College of Emergency Physicians (ACEP) and asked them to rate various areas of concern and risk for patient safety in the ED. We expected that perceived safety risks would vary by practice and physician characteristics and that emergency physicians would rate most highly areas that are presently neither measured nor rewarded by governmental and voluntary quality organizations. We also suspected that urban/suburban emergency physicians' perceptions would differ from those of rural practitioners.
Materials and Methods
Study Design
We report descriptive survey data collected electronically through the Internet of ACEP members to assess their perceptions of patient safety risks and how these perceptions may vary by practice and related factors. The study design was reviewed and approved by our institutional review board.
Methods of Measurement
We developed and pilot tested a brief survey designed to assess practicing ACEP physician members' concerns about patient safety. Content areas about patient safety were chosen according to their appearance in general and emergency medical literature on patient safety, ACEP areas of patient safety priority, and finally Center for Medicare & Medicaid Studies and The Joint Commission standards. Questions were field tested among experienced emergency physician researchers for clarity before use.
We identified 16 patient safety concerns, which are listed in Table 1. Respondents rated each concern for patient safety importance, using a 5-point Likert scale ranging from 1 (not important) to 5 (very important). The participants were also asked to provide demographic and practice characteristics.
Table 1. Mean rating and rank importance of patient safety factors by emergency physician practice location.
| Rural | Suburban | Urban | Total | |||||
|---|---|---|---|---|---|---|---|---|
| Patient Safety Factor | Mean (SD) | Rank | Mean (SD) | Rank | Mean (SD) | Rank | Mean (SD) | Rank |
| Crowding from inpatient boarding | 3.5 (1.4) | 7 | 4.3(1.1) | 1 | 4.5(0.9) | 1 | 4.3(1.1) | 1 |
| Availability of specialty consultation | 4.3(0.9) | 1 | 4.2(1.0) | 2 | 4.0(1.1) | 2 | 4.1(1.1) | 2 |
| Nurse shortages | 3.8(1.1) | 3 | 3.9(1.1) | 3 | 3.9(1.1) | 3 | 3.9(1.1) | 3 |
| Lack of follow-up for ED care | 3.8(1.1) | 2 | 3.7(1.1) | 5 | 3.8(1.1) | 4 | 3.8(1.1) | 4 |
| Laboratory/radiograph time | 3.7(1.1) | 5 | 3.7(1.1) | 4 | 3.7(1.0) | 5 | 3.7(1.1) | 5 |
| Information systems/data sharing including between institutions | 3.6(1.2) | 6 | 3.6(1.1) | 6 | 3.6(1.1) | 6 | 3.6(1.1) | 6 |
| Aspirin–β-blockers for myocardial infarction | 3.7(1.4) | 4 | 3.6(1.3) | 7 | 3.5(1.3) | 7 | 3.6(1.3) | 7 |
| Hand offs | 3.1(1.2) | 10 | 3.1(1.1) | 10 | 3.3(1.1) | 8 | 3.2(1.1) | 8 |
| Availability of language interpreters | 2.9(1.2) | 12 | 3.1(1.2) | 8 | 3.3(1.2) | 9 | 3.1(1.2) | 9 |
| Handwashing | 3.2(1.2) | 9 | 3.0(1.2) | 11 | 3.1(1.2) | 10 | 3.1(1.2) | 10 |
| Nonurgent use of the ED | 3.3(1.3) | 8 | 3.1(1.3) | 9 | 3.0(1.3) | 12 | 3.1(1.3) | 11 |
| Antibiotics within 4 h for pneumonia | 3.1(1.3) | 11 | 2.9(1.2) | 12 | 2.8(1.3) | 14 | 2.9(1.3) | 12 |
| Medication errors | 2.9(1.1) | 13 | 2.8(1.2) | 14 | 2.9(1.2) | 13 | 2.9(1.1) | 13 |
| Ambulance diversions | 1.9(1.2) | 16 | 2.8(1.3) | 13 | 3.1(1.3) | 11 | 2.8(1.4) | 14 |
| Physician order entry | 2.7(1.2) | 14 | 2.7(1.2) | 15 | 2.8(1.2) | 15 | 2.7(1.2) | 15 |
| Time-outs to verify correct side and site for procedure | 2.1(1.1) | 15 | 2.0(1.1) | 16 | 2.1(1.1) | 16 | 2.1(1.1) | 16 |
Setting and Selection of Participants
In March 2007, all ACEP members with an e-mail address on file with ACEP were sent an e-mail soliciting their input on a study of patient safety concerns observed in their practice. In this solicitation, members were asked to fill out a brief survey through the Internet. Nonrespondents received 2 additional solicitations to participate. Using computer tracking software, we determined that of the 19,236 members with e-mail addresses, 99% (19,092) received the e-mail and 39% (7,384) opened it. Of those who opened the e-mail, 2,817 (38%) returned completed responses. Three respondents did not provide a location of practice and were excluded. Only respondents who identified themselves as practicing emergency physicians were included in this analysis, thus excluding 307 responses from residents, for a final sample of 2,507 for this analysis.
Data Collection and Processing and Primary Data Analysis
Respondents who agreed to participate were directed to an electronic link for a Web-based survey. Respondents answered questions electronically and then confirmed their responses. Although responses were anonymous, we were able to identify nonrespondents for repeated electronic solicitation to participate.
Demographic and practice characteristics were divided into 3 levels: ED volume (fewer than 20,000 visits, 20,000 to 40,000 visits, greater than 40,000 visits), hospital size (fewer than 200 beds, 200 to 400 beds, greater than 400 beds), percentage of patients who leave without being seen (less than 2%, 2% to 6%, greater than 6%), and years of practice (fewer than 10 years, 10 to 20 years, more than 20 years).
Mean survey responses were tabulated, summarized, and presented in tabular form. Responses were then further segmented by practice and demographic characteristics.
Results
Table 1 summarizes mean responses for the 16 patient safety statements, ordered from most important to least important. Overall, crowding from inpatient boarding (mean 4.3), availability of specialty consultation (mean 4.1), and nursing shortages (mean 3.9) were ranked as the greatest concerns. Rural respondents ranked availability of consultants (mean 4.3), lack of follow-up after ED care (mean 3.8), and nurse shortages (mean 3.8) as their top 3 concerns. Crowding was the greatest concern for both suburban (mean 4.3) and urban emergency physicians (mean 4.5) but was ranked as the seventh most important concern by rural emergency physicians (mean 3.5). The least important issue for rural respondents was ambulance diversion (mean 1.9), which they rated lower than suburban (mean 2.8) and urban respondents (mean 3.1) did.
Table 2 characterizes emergency physician responses for crowding from inpatient boarding by practice location and other characteristics. Crowding from inpatient boarding was generally perceived as a more serious problem in the larger hospital sizes and with greater ED volume, regardless of location. Similar trends were noted as the percentage of patients who left without being seen increased from less than 2% to greater than 6%. There were no trends for years of practice by location, except among rural respondents: less experienced rural respondents expressed greater concern for crowding compared with more experienced rural respondents.
Table 2. Rating of crowding, by location and ED/hospital characteristics.
| Rural | Suburban | Urban | Total | |||||
|---|---|---|---|---|---|---|---|---|
| Characteristic | No. (%) | Mean (SD) | No. (%) | Mean (SD) | No. (%) | Mean (SD) | No. (%) | Mean (SD) |
| Overall | 337(13) | 3.5(1.4) | 1,056(42) | 4.3(1.1) | 1,114(44) | 4.5(0.9) | 2,507 | 4.3(1.1) |
| ED volume (visits/y) | ||||||||
| <20,000 | 156(46) | 2.8(1.4) | 63(6) | 3.3(1.4) | 27(2) | 4.0(1.3) | 246(10) | 3.1(1.4) |
| 20,000–40,000 | 128(38) | 3.9(1.3) | 429(41) | 4.1(1.2) | 227(20) | 4.3(1.1) | 784(31) | 4.1(1.2) |
| >40,000 | 53(16) | 4.5(1.0) | 560(53) | 4.5(0.9) | 853(77) | 4.6(0.8) | 1,466(58) | 4.5(0.9) |
| Missing | 0(0) | 4(0) | 7(1) | 11(0) | ||||
| Hospital size, beds | ||||||||
| <200 | 273(81) | 3.3(1.5) | 420(40) | 3.9(1.2) | 146(13) | 4.2(1.1) | 839(33) | 3.8(1.3) |
| 200–400 | 48(14) | 4.4(1.0) | 445(42) | 4.4(0.9) | 461(41) | 4.5(0.9) | 954(38) | 4.5(0.9) |
| >400 | 10(3) | 4.1(1.7) | 155(15) | 4.6(0.8) | 468(42) | 4.6(0.8) | 633(25) | 4.6(0.8) |
| Missing | 6(2) | 36(3) | 39(4) | 81(3) | ||||
| Left without being seen, % | ||||||||
| <2 | 283(74) | 3.4(1.5) | 512(48) | 4.0(1.2) | 333(30) | 4.2(1.1) | 1,128(44) | 3.9(1.3) |
| 2–6 | 84(22) | 3.7(1.4) | 430(41) | 4.5(0.9) | 520(47) | 4.6(0.8) | 1,034(41) | 4.5(0.9) |
| >6 | 4(1) | 4.0(2.0) | 45(4) | 4.6(0.9) | 154(14) | 4.7(0.7) | 203(8) | 4.7(0.8) |
| Missing | 11(3) | 69(7) | 107(10) | 187(7) | ||||
| Years in Practice | ||||||||
| <10 | 92(27) | 3.9(1.3) | 360(34) | 4.3(1.0) | 382(34) | 4.6(0.8) | 834(33) | 4.4(1.0) |
| 10–20 | 115(34) | 3.2(1.4) | 351(33) | 4.2(1.1) | 363(33) | 4.5(1.0) | 829(33) | 4.2(1.2) |
| >20 | 130(39) | 3.5(1.5) | 343(32) | 4.2(1.2) | 366(33) | 4.5(1.0) | 839(33) | 4.2(1.2) |
| Missing | 0(0) | 2(0) | 3(0) | 5(0) | ||||
Limitations
We used a sample of respondents to an anonymous survey of ACEP emergency physicians. Approximately 15% of members who were sent an e-mail responded to the survey, but 38% of those who opened the survey responded. We believe that the latter percentage better represents the response rate. This is akin to telephone surveys in which “no answers” are not counted as refusals. “Not opening” the e-mail is similar to telephone “no answers.” Our respondents represent a slightly greater proportion of rural physicians and fewer urban physicians compared with the ACEP member profile (8% rural, 39% suburban, and 54% urban). Although response bias may be a problem, we obtained the greatest number of responses ever for an ACEP survey, indicating that these issues resonate strongly with emergency physicians.
Not all emergency physicians are members of ACEP, and rural emergency physicians tend to have a lower membership in ACEP. It is estimated that ACEP, with a membership of 26,000, represents the majority of practicing emergency physicians. Respondents may have felt differently than nonrespondents about patient safety. The survey had a limited number of questions. The l6-item patient safety concerns list, although drawn from the medical literature, may have missed other important concerns. It is possible that areas of greater concern were not actually reflected by the survey questions.
Self-identifications and practice descriptors were self-reported and could not be verified.
Discussion
Emergency physicians identified crowding, availability of specialty consultation, nursing shortages, and challenges in arranging follow-up for their patients as the most important patient safety concerns. Many nationally identified concerns (eg, handwashing, medication errors, time-outs to verify side and patient) did not rank highly. The results identify a potential mismatch between emergency physician provider concerns and existing strategies to improve patient safety.
Rural emergency physicians differ from urban and suburban emergency physicians about patient safety concerns. Our survey suggests that emergency physicians who work in urban and suburban hospitals perceive crowding as a greater concern than rural emergency physicians. Crowding is also a leading area of dissatisfaction by ED patients, who often leave without being seen because of the long waits worsened by crowding.7 Crowding varies across the nation. Because crowding has been linked to boarding of admitted patients, some of this variation may be linked to the number of hospital beds per thousand population, which ranges from 1.7 to 6.2.8
Rural emergency physicians noted availability of consultants as their greatest concern for patient safety, which is understandable because of the isolated nature of many rural hospitals, where life-threatening emergencies may present and require immediate intervention, without the opportunity for specialists to provide assistance. Specialists are less available, are farther away, and often face busy clinic schedules after a night of ED call. In many communities, specialists have limited or refused their ED call responsibility. Rural emergency physicians may not have the capacity to address certain emergencies without specialist consultation.
Previous studies of rural EDs have focused on workforce and training issues. Rural hospital EDs are much less likely to be staffed by emergency medicine residency–trained physicians than urban and suburban EDs. This difference in staffing makes the difficulty in accessing specialty consultants even more serious because the physicians staffing the ED may not have training appropriate for the problem encountered.9
Our survey suggests that to address safety issues in rural EDs, initiatives to augment the availability of specialty care should be prioritized. Possible options include improved access to specialists through telephone, telemedicine, or actual transfer of patients to centers that offer the needed specialty care. The ACEP has recently released recommendations concerning best practices for reducing boarding and crowding, and investment in these recommendations may address concerns about safety raised in this survey.10
Measures commonly used by national organizations addressing patient safety issues such as “time-out” procedures, handwashing, medication errors, and antibiotics within 4 hours (now changed to 6 hours) for pneumonia patients were ranked substantially lower in importance than most measures. These measures may need review and modification to match the patient safety priorities identified by emergency physicians, which is important because the ED has been recognized as an area of high risk for medical error.3 Efforts to address patient safety should concentrate on those areas in which the yield for such efforts would be the highest. In addition, the mismatch of perceived patient safety problems and nationally sponsored responses may result in cynicism and lack of engagement in the quality improvement process by emergency physicians. The low ranking of these national issues was similar between urban, suburban, and rural hospitals. One area that did differ in low ranking was ambulance diversion, which does not appear to be a safety issue of significance in rural hospitals but is of moderate importance in urban and suburban hospital EDs.
ED safety concerns differ, depending on the characteristics of the department. Furthermore, emergency medicine safety concerns are not being addressed by national bodies charged with addressing safety issues. This disconnect is all the more surprising because the ED has been recognized as an area of high risk for patient care. Future areas of prioritization for patient safety should incorporate concerns of practicing emergency physicians.
References
- . To Err Is Human: Building a Safer Health System. In: Kohn LT, Carrigan JM, Donaldson MS editor. Washington, DC: National Academy Press; 2000;
- . What practices will most improve safety? (evidence-based medicine meets patient safety). JAMA. 2002;288:501–507
- Incidence of adverse events and negligence in hospitalized patients: results of the Harvard Medical Practice Study I. N Engl J Med. 1991;324:370–376
- . 2008 JCAHO National Patient Safety Goals. http://www.jointcommission.org/?Accessed July 12, 2008
- . Hospital Quality Alliance. http://www.cms.hhs.gov/HospitalQualityInits/33_HospitalQualityAlliance.aspAccessed October 1, 2009
- . Safe Practices for Better Healthcare 2006 Update. Irving, TX: American College of Emergency Physicians; 2008;
- . The effect of crowding on access and quality in an academic ED. Am J Emerg Med. 2006;26:787–794
- Stathealthfacts.org Hospital beds per 1,000 population, 2006. http://www.statehealthfacts.org/comparemaptable.jsp?ind=384&cat=8Accessed January 15, 2009
- Assessment of emergency physician workforce needs in the United States, 2005. Acad Emerg Med. 2008;15:1317–1320
- ACEP Task Force Report on Boarding. Emergency Department Crowding: High-Impact Solutions. 2008;Dallas, TX
Supervising editor: Robert L. Wears, MD, MS
Author contributions: All authors conceived the study and developed the design. DPS, CSC, and RC worked with ACEP to develop the electronic data collection form. RC oversaw data collection. CSC conducted the statistical analysis. DPS, CSC, and TZ contributed to writing the article. All authors take responsibility for the paper as a whole.
Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article that might create any potential conflict of interest. The authors have stated that no such relationships exist. See the Manuscript Submission Agreement in this issue for examples of specific conflicts covered by this statement.
Please see page 337 for the Editor's Capsule Summary of this article.
Reprints not available from the authors.
PII: S0196-0644(09)01447-4
doi:10.1016/j.annemergmed.2009.08.020
© 2009 American College of Emergency Physicians. Published by Elsevier Inc. All rights reserved.
