- Email: [email protected]
A Profile of Indian Health Service Emergency Departments
HEALTH POLICY/BRIEF RESEARCH REPORT
A Profile of Indian Health Service Emergency Departments Kenneth Bernard, MD, MBA*; Kohei Hasegawa, MD, MPH; Ashley Sullivan, MPH, MS; Carlos Camargo, MD, DrPH *Corresponding Author. E-mail: [email protected].
Study objective: The Indian Health Service provides health care to eligible American Indians and Alaskan Natives. No published data exist on emergency services offered by this unique health care system. We seek to determine the characteristics and capabilities of Indian Health Service emergency departments (EDs). Methods: All Indian Health Service EDs were surveyed about demographics and operational characteristics for 2014 with the National Emergency Department Inventory survey (available at http://www.emnet-nedi.org/). Results: Of the forty eligible sites, there were 34 respondents (85% response rate). Respondents reported a total of 637,523 ED encounters, ranging from 521 to 63,200 visits per site. Overall, 85% (95% confidence interval 70% to 94%) had continuous physician coverage. Of all physicians staffing the ED, a median of 13% (interquartile range 0% to 50%) were board certified or board prepared in emergency medicine. Overall, 50% (95% confidence interval 34% to 66%) of respondents reported that their ED was operating over capacity. Conclusion: Indian Health Service EDs varied widely in visit volume, with many operating over capacity. Most were not staffed by board-certified or -prepared emergency physicians. Most lacked access to specialty consultation and telemedicine capabilities. [Ann Emerg Med. 2016;-:1-6.] Please see page XX for the Editor’s Capsule Summary of this article. 0196-0644/$-see front matter Copyright © 2016 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2016.11.031
INTRODUCTION American Indians and Alaskan Natives experience higher mortality, a disproportional burden of chronic and mental illness, poor access to health care services, and substantial unmet medical needs compared with the general US population.1,2 Contributing to the issue are several factors, including increasing disparities in social determinants of health such as poverty, education, and substance abuse.2 The Indian Health Service, a division of the Department of Health and Human Services, is bound by trust to provide health care services to American Indians and Alaskan Natives, eg, primary care, prescription drugs, mental health, and emergency care.1,2 Emergency departments (EDs) are an integral part of any health care system, including the Indian Health Service, which is the only source of care for many American Indians and Alaskan Natives. However, to date, to our knowledge there are no published data on EDs operating within the Indian Health Service. To address this knowledge gap, we surveyed all Indian Health Service EDs in regard to operational characteristics, staffing, use and availability of electronic resources, ED capabilities, and crowding and capacity. Given the historical and persistent health disparities, as well as Volume
-,
no.
-
:
-
2016
well-researched barriers to primary and subspecialty care, these data are vital to understanding the context within which emergency care is delivered to American Indians and Alaskan Natives.1,2 MATERIALS AND METHODS Study Design and Data Collection and Processing We performed a cross-sectional study surveying all EDs operating within the Indian Health Service with the National Emergency Department Inventory survey instrument (Appendix E1, available online at http://www. annemergmed.com). Consistent with previous National Emergency Department Inventory studies, EDs defined as emergency care facilities were open at least 24 hours per day, 7 days per week, and available for use by the general public.3,4 The survey instrument was made available to ED directors or knowledgeable proxies (eg, chief medical officer, nurse manager) by hard copy and e-mail.5 Data were collected during a 6-month period starting in November 2015 in regard to the 2014 calendar year. Contact initiated by e-mail and follow-up e-mails or telephone calls was continued until a target of greater than Annals of Emergency Medicine 1
Indian Health Service Emergency Departments
Bernard et al
Editor’s Capsule Summary
Categorical variables are presented as proportions with confidence intervals (CIs). For ease of analysis and comparison, we categorized sites by annual visit volume into 3 groups: small (<10,000), midsized (10,000 to 19,999), and large (20,000). Next, we compared ED characteristics and capabilities by ED volume status, using c2, Fisher’s exact, or Kruskal-Wallis tests as appropriate. Statistical analyses were performed with Stata (version 12.0; StataCorp, College Station, TX).
What is already known on this topic The Indian Health Service is a national system to provide health care for Native Americans. What question this study addressed The state of emergency services offered within the Indian Health Service is defined from survey responses in 2014. What this study adds to our knowledge The 34 of 40 emergency departments who responded handle between 500 and 63,000 visits per year, with 85% having continuous physician coverage, although a median of only 13% of physicians were board certified in the specialty. How this is relevant to clinical practice This study provides a current assessment of the state of care to help identify gaps in care at a national level.
or equal to an 80% response rate was reached. If participation was declined at any time, no further contact was attempted. Completed survey data were entered and managed with REDCap secure electronic data management and capture tools hosted by Partners Healthcare (Boston, MA).6 Approval was obtained from the institutional review boards of the Massachusetts General Hospital (Boston, MA) and the Indian Health Services (Rockville, MD). We surveyed variables divided into 5 categories: ED basic characteristics; ED staffing, including availability of specialists; electronic resources in the ED; ED capabilities, eg, timing of consultations, tests, transfers; and ED crowding and capacity. Basic characteristics included annual ED volume and hospitalization rates. Staffing variables included the number of full-time equivalent physicians and advance practice clinicians (ie, physician assistant and nurse practitioners) and continuous in-house physician coverage. We assessed use of electronic resources and ED capabilities such as telemedicine, availability of clinical laboratory, and access to a dedicated computed tomography (CT) or magnetic resonance imaging (MRI) scanner in the ED. Finally, correlates of ED crowding, such as boarding in the ED greater than 2 hours or patient care in hallways, was also assessed. Primary Data Analysis Continuous variables with a non-normal distribution are presented as medians with interquartile ranges (IQRs). 2 Annals of Emergency Medicine
RESULTS Characteristics of Study Subjects Of the 40 Indian Health Service EDs, most densely clustered in Alaska, the Upper Northwest, and the Southwest United States, there were 34 respondents (response rate 85%) (Figure 1). When compared with respondents, nonrespondents had similar geographic distribution and were contiguous with hospitals. All respondent EDs treated both children and adults. In 2014, there were a total of 637,523 ED encounters, of which 172,214 (27%) were by children (Figure 2). The median annual ED visit volume was 15,234 (IQR 10,347 to 23,618). Approximately one third (32%) of respondents reported annual ED volume greater than or equal to 20,000 patient visits, and these sites accounted for 59% of the total visits. Of the 34 respondents, 29 (85%; 95% CI 70% to 94%) reported continuous physician coverage. Of all physicians staffing Indian Health Service EDs, a median of 13% (IQR 0% to 50%) were board certified or board prepared in emergency medicine. Compared with small (50%; 95% CI 17% to 83%) and midsized (93%; 95% CI 66% to 100%) EDs, large (100%; 95% CI 68% to 100%) EDs were more likely to have continuous physician coverage. A median of 63% (IQR 0% to 46%) of physicians at large EDs were board certified or board prepared in emergency medicine compared with medians 5% (IQR 0% to 46%) and 0% (IQR 0% to 18%) at small and midsized EDs, respectively. Advanced practice clinicians were staffed at 50% (95% CI 17% to 83%), 53% (95% CI 27% to 78%), and 82% (95% CI 48% to 97%), of small, midsized, and large EDs, respectively (Table). Overall, the most widely available specialists were pediatricians (71%), obstetricians (41%), and general surgeons (35%). In contrast, few EDs reported access to cardiologist (6%) or neurologists (3%), and 0% had access to plastic or hand surgery (Table). Overall, large EDs had higher access to specialty consultation. For example, compared with small (0% and 13%) and midsized EDs (13% and 7%), 55% and 64% of large EDs had access to an anesthesiologist and orthopedic surgeon, respectively (Table). Volume
-,
no.
-
:
-
2016
Bernard et al
Indian Health Service Emergency Departments
Figure 1. Location of Indian Health Service EDs. Numbers correspond to site number on Figure 2.
The most widely available electronic resources were laboratory results, previous ED notes, and previous ECGs. Few EDs reported access to patient trackers and real-time clinical data collection. Telemedicine was used at only 4 sites (12%; 95% CI 5% to 27%) (Table).
B
Denotes nonrespondents.
Cardiac monitoring was universally available at all sites (100%; 95% CI 90% to 100%). In contrast, only 2 EDs (6%; 95% CI 2% to 19%) reported continuous radiology coverage. Compared with small and midsized EDs, large EDs were nonsignificantly more likely to have access to
Indian Health Service Annual Visits by Site
60000
Annual Visit Volume
50000
40000
Children Adult
30000
20000
10000
0 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Figure 2. Annual ED visits by Indian Health Service site (2014). Volume
-,
no.
-
:
-
2016
Annals of Emergency Medicine 3
Bernard et al
Indian Health Service Emergency Departments Table. Summary of National Emergency Department Inventory Indian Health Service survey data.* Annual ED Visits, No. (% [95% CI]) NEDI Category ED staffing (n) Continuous physician staffing Board certified or board prepared in emergency medicine (IQR) Staffed by advance practice clinician Specialist availability Anesthesiologist Cardiologist General surgeon Hand surgeon Neurologist Neurosurgeon OB/GYN Orthopedic surgeon Pediatrician Psychiatrist Plastic surgeon Electronic resources Telemedicine Real-time clinical data Patient tracking Critical value notification Computerized error checking Previous discharge summary Previous ED visits Previous ECG Previous radiographic images Outpatient medications Laboratory results Physician order entry ED capabilities Cardiac monitoring Continuous laboratory Continuous radiology Mechanical ventilation Respiratory isolation CT scanner MRI scanner ED boarding and capacity Patients in hallway Boarding >2 h Left before being seen At or over capacity
<10,000, Small (n[8 [24%])
10,000–19,999, Midsize (n[15 [44%])
>20,000, Large (n[11 [32%])
4 (50 [17–83]) 5 (0–46)
14 (93 [66–100]) 0 (0–18)
11 (100 [68–100]) 63 (0–46)
4 (50 [17–83])
8 (53 [27–78])
9 (82 [48–97])
0 [0–40] 0 [0–40] 1 (13 [1–53]) 0 [0–40] 0 [0–40] 0 [0–40] 2 (25 [4–64]) 1 (13 [1–53]) 5 (63 [26–90]) 1 (13 [1–53]) 0 [0–40]
2 (13 [2–41]) 0 [0–25] 4 (27 [9–55]) 0 [0–25] 1 (7 [0–34]) 0 [0–25] 5 (33 [13–61]) 1 (7 [0–34]) 9 (60 [33–83]) 2 (13 [2–41]) 0 [0–25]
6 (55 [25–82]) 2 (18 [3–52]) 7 (64 [32–88]) 0 [0–32] 0 [0–32] 1 (9 [0–43]) 7 (64 [32–88]) 7 (64 [32–88]) 10 (91 [57–99]) 4 (36 [12–68]) 0 [0–32]
1 (13 [1–53]) 3 (38 [10–74]) 5 (63 [26–90]) 6 (75 [36–96]) 5 (63 [26–90]) 6 (75 [36–96]) 6 (75 [36–96]) 6 (75 [36–96]) 8 (100 [60–100]) 6 (75 [36–96]) 7 (88 [47–99]) 6 (75 [36–96])
1 (7 [0–34]) 9 (60 [33–83]) 7 (47 [22–73]) 9 (60 [33–83]) 13 (87 [59–98]) 13 (87 [59–98]) 14 (93 [66–100]) 14 (93 [66–100]) 13 (87 [59–98]) 15 (100 [75–100]) 15 (100 [75–100]) 15 (100 [75–100])
2 (18 [3–52]) 6 (55 [25–82]) 5 (45 [18–75]) 6 (55 [25–82]) 9 (82 [48–97]) 10 (91 [57–99]) 10 (91 [57–99]) 7 (64 [32–88]) 10 (91 [57–99]) 10 (91 [57–99]) 10 (91 [57–99]) 10 (91 [57–99])
8 (100 [60–100]) 8 (100 [60–100]) 7 (88 [47–99]) 7 (88 [47–99]) 4 (50 [17–83]) 5 (63 [26–90]) 0 [0–40]
15 (100 [75–100]) 14 (93 [66–100]) 14 (93 [66–100]) 7 (47 [22–73]) 12 (80 [51–95]) 11 (73 [45–91]) 1 (7 [0–34])
11 (100 [68–100]) 11 (100 [68–100]) 11 (100 [68–100]) 11 (100 [68–100]) 9 (82 [48–97]) 8 (73 [39–93]) 3 (27 [7–61])
0 [0–40] 2 (25 [4–64]) 4 (IQR 2–5) 6 (75 [36–96])
2 (13 [2–41]) 5 (33 [13–61]) 5 (IQR 2–12) 8 (53 [27–78])
5 (45 [18–75]) 9 (82 [48–97]) 6 (IQR 4–13) 11 (100 [68–100])
*Proportions are reported with 95% CIs and medians are reported with IQRs.
mechanical ventilation, respiratory isolation, and advanced imaging capabilities (Table). When asked about a typical day in the ED, 0% (95% CI 0% to 40%) of small EDs reported patient care taking place in a hallway compared with midsized and large EDs, which reported 13% (95% CI 2% to 14%) and 45% (95% CI 18% to 75%), respectively. Similarly, 82% (95% CI 48% to 97%) of large EDs reported patients boarding greater than 2 hours compared with 25% (95% CI 4% to 64%) of small EDs and 33% (95% CI 13% to 61%) of midsized EDs. All large EDs reported operating at or over capacity (Table). 4 Annals of Emergency Medicine
LIMITATIONS This descriptive, self-reported study using the National Emergency Department Inventory survey instrument to survey Indian Health Service ED directors or knowledgeable proxies was subject to bias and error. To mitigate bias and improve validity of results, ED sites were assigned a random site number and respondents were not required to provide their name or title. We acknowledge that respondents had various knowledge of and involvement in ED operations. However, this allowed some level of anonymity because responses were not tied to a Volume
-,
no.
-
:
-
2016
Bernard et al
Indian Health Service Emergency Departments
particular individual at each site. The National Emergency Department Inventory survey instrument has never been applied in the Indian Health Service setting, to our knowledge, and its performance is untested in this health care system. However, the National Emergency Department Inventory survey has been reviewed and refined by multiple Emergency Medicine Network researchers and American College of Emergency Physicians state chapter boards to remove ambiguous language and focused questions covering topics germane to emergency medicine. Furthermore, it has been administered in several other published studies in the United States and abroad, lending to its performance in a diverse context.4 Finally, although we cannot prove nonrespondents did not vary in any appreciable way from respondents, given an 85% response rate in this national survey, the potential for significant bias is likely very limited. DISCUSSION To our knowledge, this cross-sectional survey is the first to examine the basic characteristics, staffing, and capabilities of Indian Health Service EDs, which are critical health care access points for more than 2 million American Indians and Alaskan Natives. We found, as expected, that most Indian Health Service EDs are located in rural areas on or near reservations.2,3 With regard to staffing, only 85% of respondents had continuous physician coverage. Of EDs staffed by physicians, few physicians were board certified or board prepared in emergency medicine. Small and midsized EDs were more likely to be staffed with nonboard-certified or nonboard-prepared emergency physicians compared with larger-volume centers. This finding is consistent with previous work showing that the rural ED workforce is largely made up of later-career, nonboard-certified and nonboard-prepared emergency physicians.7 This has broad implications in regard to health care equity and quality because it is likely that any further emergency physician shortage will disproportionately affect these rural populations. Moreover, for the foreseeable future, nonboard-certified or board-prepared emergency physicians will continue to provide the bulk of ED care in these rural, low-volume EDs within the Indian Health Service. Thus, it is incumbent on the Indian Health Service, local hospital leadership, and emergency medicine governing bodies to ensure that these providers are capable of delivering high-quality evidence-based ED care. One proposed strategy to cope with emergency physician shortages could be to expand the role and scope of advance practice clinicians, especially at smaller-volume sites.8 Currently, 62% of Indian Health Service EDs use Volume
-,
no.
-
:
-
2016
advance practice clinicians, and, although the data were not statistically significant, large EDs were more likely to staff advance practice clinicians compared with small and midsized EDs. Our study may have identified a potential role for advance practice clinicians at lower-volume sites, which disproportionately experience physician shortages, and may function more like urgent care centers. In this setting, with appropriate supervision and support, solo coverage by an experienced advance practice clinician may be sufficient and, given our findings here, unavoidable at some Indian Health Service sites. However, the case-mix acuity of patients seeking care at lower-volume Indian Health Service EDs is unknown. Without a clear understanding of the complexity of disease and the acuity of patients using these small Indian Health Service EDs, one cannot make conclusions about the intensity of resources needed to meet emergency medical needs of those communities. Further investigation may allow Indian Health Service EDs to find the ideal workforce proportion and cost-effectiveness of advance practice clinicians at these sites. Previous studies examining care delivered in the Indian Health Service ambulatory setting have shown that patients and physicians have difficulty obtaining high-quality subspecialty care, hospital admission, diagnostic studies, and mental health services.1,2,9 Our study highlights that this lack of subspecialty care extends to ED settings because most sites lacked access to cardiologists, neurologists, and specialty surgeons. This dearth of subspecialists appears to be more prevalent at smaller sites compared with midsized and large EDs. Therefore, there may be a role for innovative health information technologies to bridge the gap in subspecialty care available to patients using Indian Health Service EDs.9,10 However, in general Indian Health Service EDs had low access to and use of electronic resources, in particular, telemedicine. In our study, only 4 sites had telemedicine capabilities, which consisted of telestroke or teleradiology services. Moreover, these services were available at midsized and large EDs rather than small ones, which tend to have lower access to subspecialists and are more likely to have a nonboard-certified or nonboard-prepared physician or no physician coverage in the ED. In essence, the EDs and communities that could benefit most from health information technologies and telemedicine did not have access to these resources. Expansion of telemedicine initiatives, especially at lower-volume sites, seems to be a logical and feasible solution, but the low penetration of such programs needs to be addressed.10 Finally, the majority of EDs surveyed reported operating at or over capacity, which may contribute to the higher Annals of Emergency Medicine 5
Bernard et al
Indian Health Service Emergency Departments
correlates of ED crowding observed in this study (eg, patient boarding). Our study found that this burden lies disproportionately with large EDs, which may be due to their higher patient volumes. However, the contribution of other factors that have been associated with ED crowding, such as patient acuity, inpatient bed capacity, and workforce shortages, is unknown. To date, the current approaches to address boarding in Indian Health Service EDs and the effects on patient care are also unknown and merit investigation. Our study found that Indian Health Service EDs vary widely in regard to staffing, as well as electronic, technical, and diagnostic capabilities. Indian Health Service EDs also have physician shortages and lack access to subspecialty services. Furthermore, despite a dearth of subspecialist support we found generally low use of telemedicine. Finally, Indian Health Service EDs, in particular largervolume ones, reported operating over capacity and higher correlates of ED crowding. Our data identified areas for improvement and facilitate further investigation into physician recruitment and retention initiatives; mix of workforce, including advance practice clinicians; broader implementation of telemedicine; and the causes of and solutions for crowding. The present study provides a foundation for survey research in an area that requires further, more rigorous study to understand and improve the ED care delivered to American Indians and Alaskan Natives. The authors acknowledge the Indian Health Service participants, internal review board, and administration, without whose support these data would not be available and this important work could not progress; and Jane Bittner, MPH, at the Emergency Medicine Network Coordinating Center for her assistance. Supervising editor: Daniel A. Handel, MD, MBA Author affiliations: From the Department of Emergency Medicine, Tuba City Regional Health Care Corp, Tuba City, AZ (Bernard); and the Department of Emergency Medicine and Emergency Medicine Network (EMNet), Massachusetts General Hospital, Boston, MA (Camargo, Hasegawa, Sullivan). Author contributions: KB, KH, and CC contributed to article conception, development, and analysis; drafting and revision of the article; and the final draft. All authors contributed to final article approval. All authors certify that this article is a valid work and is not under review for publication elsewhere, nor have these
6 Annals of Emergency Medicine
data been publicly available. KB takes responsibility for the paper as a whole. All authors attest to meeting the four ICMJE.org authorship criteria: (1) Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND (2) Drafting the work or revising it critically for important intellectual content; AND (3) Final approval of the version to be published; AND (4) Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. 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 as per ICMJE conflict of interest guidelines (see www.icmje.org). This study was supported by a grant from the Massachusetts College of Emergency Physicians. Dr. Bernard, MD, MBA, was awarded an Indian Health Service Health Professions Scholarship and is an emergency physician at an Indian Health Service facility. Publication dates: Received for publication August 30, 2016. Revision received November 14, 2016. Accepted for publication November 18, 2016.
REFERENCES 1. Zuckerman S, Haley J, Roubideaux Y, et al. Health service access, use, and insurance coverage among American Indians/Alaska Natives and whites: what role does the Indian Health Service play? Am J Public Health. 2004;94:53-59. 2. Sequist T, Cullen T, Bernard K, et al. Trends in quality of care and barriers to improvement in the Indian Health Service. J Gen Intern Med. 2011;26:480-486. 3. Indian Health Service. Indian Health Service locations. Available at: https://www.ihs.gov/locations/. Accessed July 1, 2016. 4. Emergency Medicine Network. National Emergency Department Inventories (NEDI). Available at: http://www.emnet-usa.org/nedi/ nedi.htm. Accessed July 1, 2016. 5. Emergency Medicine Network (EMNet). The National ED Inventories survey instrument. 2011. Available at: http://www.emnet-nedi.org. Accessed July 1, 2016. 6. Harris P, Taylor R, Thielke R, et al. Research Electronic Data Capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377-381. 7. Macho M, MacKenzie A, Ginde A. The rural emergency physician workforce. J Rural Emerg Med. 2014;1:3-8. 8. Brown D, Sullivan A, Espinola J, et al. Continued rise in the use of midlevel providers in US emergency departments, 1993-2009. Int J Emerg Med. 2012;5:21. 9. Baldwin L, Hollow W, Casey S, et al. Access to specialty health care for rural American Indians in two states. J Rural Health. 2008;24:269-278. 10. Rheuban K. The role of telemedicine in fostering health-care innovations to address problems of access, speciality shortages and changing patient care needs. J Telemed Telecare. 2006;12: 45-50.
Volume
-,
no.
-
:
-
2016
Bernard et al
Indian Health Service Emergency Departments
APPENDIX E1 National Emergency Department Inventory Indian Health Service Survey
2015
Indian Health Service Emergency Department Study Funded by the Massachusetts College of Emergency Physicians Supported by the Emergency Medicine Network (www.emnet-usa.org)
Dear Colleague: We would like your help with a brief survey of Indian Health Service (IHS) emergency departments (EDs). The aim is to gather information from leaders of all IHS EDs. Your response to this survey will inform ongoing efforts to improve the accessibility and quality of emergency care within IHS and beyond. Please answer questions with estimates for the year 2014. If data are available for a fiscal year (e.g., October 2013 – September 2014), and that’s easier for you, please use the fiscal year data. Although the survey asks for factual information, if you are unable to quickly find the precise answer, please provide your best guess. The survey takes 10-15 minutes to complete, and results will be completely confidential. Your name will not appear on the survey and no individual survey responses will be shared with anyone. A summary of the survey results will use aggregate responses only; this report will be shared with all ED directors statewide. Without your response, this study cannot succeed. We hope that you will help us by completing this survey or asking a knowledgeable colleague for his/her help. Thank you in advance for your assistance.
Best wishes,
Ken Bernard, MD,MBA Senior resident BWH/MGH Boston, MA
Carlos Camargo, MD, DrPH Director, EMNet Mass General Hospital Boston, MA
Please return your completed survey to: Ken Bernard, MD, MBA
SITE #
Volume
-,
no.
-
:
-
2016
Annals of Emergency Medicine 6.e1
Bernard et al
Indian Health Service Emergency Departments
6.e2 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2016
Bernard et al
Volume
-,
no.
Indian Health Service Emergency Departments
-
:
-
2016
Annals of Emergency Medicine 6.e3
Bernard et al
Indian Health Service Emergency Departments
6.e4 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2016
A Profile of Indian Health Service Emergency Departments Kenneth Bernard, MD, MBA*; Kohei Hasegawa, MD, MPH; Ashley Sullivan, MPH, MS; Carlos Camargo, MD, DrPH *Corresponding Author. E-mail: [email protected].
Study objective: The Indian Health Service provides health care to eligible American Indians and Alaskan Natives. No published data exist on emergency services offered by this unique health care system. We seek to determine the characteristics and capabilities of Indian Health Service emergency departments (EDs). Methods: All Indian Health Service EDs were surveyed about demographics and operational characteristics for 2014 with the National Emergency Department Inventory survey (available at http://www.emnet-nedi.org/). Results: Of the forty eligible sites, there were 34 respondents (85% response rate). Respondents reported a total of 637,523 ED encounters, ranging from 521 to 63,200 visits per site. Overall, 85% (95% confidence interval 70% to 94%) had continuous physician coverage. Of all physicians staffing the ED, a median of 13% (interquartile range 0% to 50%) were board certified or board prepared in emergency medicine. Overall, 50% (95% confidence interval 34% to 66%) of respondents reported that their ED was operating over capacity. Conclusion: Indian Health Service EDs varied widely in visit volume, with many operating over capacity. Most were not staffed by board-certified or -prepared emergency physicians. Most lacked access to specialty consultation and telemedicine capabilities. [Ann Emerg Med. 2016;-:1-6.] Please see page XX for the Editor’s Capsule Summary of this article. 0196-0644/$-see front matter Copyright © 2016 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2016.11.031
INTRODUCTION American Indians and Alaskan Natives experience higher mortality, a disproportional burden of chronic and mental illness, poor access to health care services, and substantial unmet medical needs compared with the general US population.1,2 Contributing to the issue are several factors, including increasing disparities in social determinants of health such as poverty, education, and substance abuse.2 The Indian Health Service, a division of the Department of Health and Human Services, is bound by trust to provide health care services to American Indians and Alaskan Natives, eg, primary care, prescription drugs, mental health, and emergency care.1,2 Emergency departments (EDs) are an integral part of any health care system, including the Indian Health Service, which is the only source of care for many American Indians and Alaskan Natives. However, to date, to our knowledge there are no published data on EDs operating within the Indian Health Service. To address this knowledge gap, we surveyed all Indian Health Service EDs in regard to operational characteristics, staffing, use and availability of electronic resources, ED capabilities, and crowding and capacity. Given the historical and persistent health disparities, as well as Volume
-,
no.
-
:
-
2016
well-researched barriers to primary and subspecialty care, these data are vital to understanding the context within which emergency care is delivered to American Indians and Alaskan Natives.1,2 MATERIALS AND METHODS Study Design and Data Collection and Processing We performed a cross-sectional study surveying all EDs operating within the Indian Health Service with the National Emergency Department Inventory survey instrument (Appendix E1, available online at http://www. annemergmed.com). Consistent with previous National Emergency Department Inventory studies, EDs defined as emergency care facilities were open at least 24 hours per day, 7 days per week, and available for use by the general public.3,4 The survey instrument was made available to ED directors or knowledgeable proxies (eg, chief medical officer, nurse manager) by hard copy and e-mail.5 Data were collected during a 6-month period starting in November 2015 in regard to the 2014 calendar year. Contact initiated by e-mail and follow-up e-mails or telephone calls was continued until a target of greater than Annals of Emergency Medicine 1
Indian Health Service Emergency Departments
Bernard et al
Editor’s Capsule Summary
Categorical variables are presented as proportions with confidence intervals (CIs). For ease of analysis and comparison, we categorized sites by annual visit volume into 3 groups: small (<10,000), midsized (10,000 to 19,999), and large (20,000). Next, we compared ED characteristics and capabilities by ED volume status, using c2, Fisher’s exact, or Kruskal-Wallis tests as appropriate. Statistical analyses were performed with Stata (version 12.0; StataCorp, College Station, TX).
What is already known on this topic The Indian Health Service is a national system to provide health care for Native Americans. What question this study addressed The state of emergency services offered within the Indian Health Service is defined from survey responses in 2014. What this study adds to our knowledge The 34 of 40 emergency departments who responded handle between 500 and 63,000 visits per year, with 85% having continuous physician coverage, although a median of only 13% of physicians were board certified in the specialty. How this is relevant to clinical practice This study provides a current assessment of the state of care to help identify gaps in care at a national level.
or equal to an 80% response rate was reached. If participation was declined at any time, no further contact was attempted. Completed survey data were entered and managed with REDCap secure electronic data management and capture tools hosted by Partners Healthcare (Boston, MA).6 Approval was obtained from the institutional review boards of the Massachusetts General Hospital (Boston, MA) and the Indian Health Services (Rockville, MD). We surveyed variables divided into 5 categories: ED basic characteristics; ED staffing, including availability of specialists; electronic resources in the ED; ED capabilities, eg, timing of consultations, tests, transfers; and ED crowding and capacity. Basic characteristics included annual ED volume and hospitalization rates. Staffing variables included the number of full-time equivalent physicians and advance practice clinicians (ie, physician assistant and nurse practitioners) and continuous in-house physician coverage. We assessed use of electronic resources and ED capabilities such as telemedicine, availability of clinical laboratory, and access to a dedicated computed tomography (CT) or magnetic resonance imaging (MRI) scanner in the ED. Finally, correlates of ED crowding, such as boarding in the ED greater than 2 hours or patient care in hallways, was also assessed. Primary Data Analysis Continuous variables with a non-normal distribution are presented as medians with interquartile ranges (IQRs). 2 Annals of Emergency Medicine
RESULTS Characteristics of Study Subjects Of the 40 Indian Health Service EDs, most densely clustered in Alaska, the Upper Northwest, and the Southwest United States, there were 34 respondents (response rate 85%) (Figure 1). When compared with respondents, nonrespondents had similar geographic distribution and were contiguous with hospitals. All respondent EDs treated both children and adults. In 2014, there were a total of 637,523 ED encounters, of which 172,214 (27%) were by children (Figure 2). The median annual ED visit volume was 15,234 (IQR 10,347 to 23,618). Approximately one third (32%) of respondents reported annual ED volume greater than or equal to 20,000 patient visits, and these sites accounted for 59% of the total visits. Of the 34 respondents, 29 (85%; 95% CI 70% to 94%) reported continuous physician coverage. Of all physicians staffing Indian Health Service EDs, a median of 13% (IQR 0% to 50%) were board certified or board prepared in emergency medicine. Compared with small (50%; 95% CI 17% to 83%) and midsized (93%; 95% CI 66% to 100%) EDs, large (100%; 95% CI 68% to 100%) EDs were more likely to have continuous physician coverage. A median of 63% (IQR 0% to 46%) of physicians at large EDs were board certified or board prepared in emergency medicine compared with medians 5% (IQR 0% to 46%) and 0% (IQR 0% to 18%) at small and midsized EDs, respectively. Advanced practice clinicians were staffed at 50% (95% CI 17% to 83%), 53% (95% CI 27% to 78%), and 82% (95% CI 48% to 97%), of small, midsized, and large EDs, respectively (Table). Overall, the most widely available specialists were pediatricians (71%), obstetricians (41%), and general surgeons (35%). In contrast, few EDs reported access to cardiologist (6%) or neurologists (3%), and 0% had access to plastic or hand surgery (Table). Overall, large EDs had higher access to specialty consultation. For example, compared with small (0% and 13%) and midsized EDs (13% and 7%), 55% and 64% of large EDs had access to an anesthesiologist and orthopedic surgeon, respectively (Table). Volume
-,
no.
-
:
-
2016
Bernard et al
Indian Health Service Emergency Departments
Figure 1. Location of Indian Health Service EDs. Numbers correspond to site number on Figure 2.
The most widely available electronic resources were laboratory results, previous ED notes, and previous ECGs. Few EDs reported access to patient trackers and real-time clinical data collection. Telemedicine was used at only 4 sites (12%; 95% CI 5% to 27%) (Table).
B
Denotes nonrespondents.
Cardiac monitoring was universally available at all sites (100%; 95% CI 90% to 100%). In contrast, only 2 EDs (6%; 95% CI 2% to 19%) reported continuous radiology coverage. Compared with small and midsized EDs, large EDs were nonsignificantly more likely to have access to
Indian Health Service Annual Visits by Site
60000
Annual Visit Volume
50000
40000
Children Adult
30000
20000
10000
0 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Figure 2. Annual ED visits by Indian Health Service site (2014). Volume
-,
no.
-
:
-
2016
Annals of Emergency Medicine 3
Bernard et al
Indian Health Service Emergency Departments Table. Summary of National Emergency Department Inventory Indian Health Service survey data.* Annual ED Visits, No. (% [95% CI]) NEDI Category ED staffing (n) Continuous physician staffing Board certified or board prepared in emergency medicine (IQR) Staffed by advance practice clinician Specialist availability Anesthesiologist Cardiologist General surgeon Hand surgeon Neurologist Neurosurgeon OB/GYN Orthopedic surgeon Pediatrician Psychiatrist Plastic surgeon Electronic resources Telemedicine Real-time clinical data Patient tracking Critical value notification Computerized error checking Previous discharge summary Previous ED visits Previous ECG Previous radiographic images Outpatient medications Laboratory results Physician order entry ED capabilities Cardiac monitoring Continuous laboratory Continuous radiology Mechanical ventilation Respiratory isolation CT scanner MRI scanner ED boarding and capacity Patients in hallway Boarding >2 h Left before being seen At or over capacity
<10,000, Small (n[8 [24%])
10,000–19,999, Midsize (n[15 [44%])
>20,000, Large (n[11 [32%])
4 (50 [17–83]) 5 (0–46)
14 (93 [66–100]) 0 (0–18)
11 (100 [68–100]) 63 (0–46)
4 (50 [17–83])
8 (53 [27–78])
9 (82 [48–97])
0 [0–40] 0 [0–40] 1 (13 [1–53]) 0 [0–40] 0 [0–40] 0 [0–40] 2 (25 [4–64]) 1 (13 [1–53]) 5 (63 [26–90]) 1 (13 [1–53]) 0 [0–40]
2 (13 [2–41]) 0 [0–25] 4 (27 [9–55]) 0 [0–25] 1 (7 [0–34]) 0 [0–25] 5 (33 [13–61]) 1 (7 [0–34]) 9 (60 [33–83]) 2 (13 [2–41]) 0 [0–25]
6 (55 [25–82]) 2 (18 [3–52]) 7 (64 [32–88]) 0 [0–32] 0 [0–32] 1 (9 [0–43]) 7 (64 [32–88]) 7 (64 [32–88]) 10 (91 [57–99]) 4 (36 [12–68]) 0 [0–32]
1 (13 [1–53]) 3 (38 [10–74]) 5 (63 [26–90]) 6 (75 [36–96]) 5 (63 [26–90]) 6 (75 [36–96]) 6 (75 [36–96]) 6 (75 [36–96]) 8 (100 [60–100]) 6 (75 [36–96]) 7 (88 [47–99]) 6 (75 [36–96])
1 (7 [0–34]) 9 (60 [33–83]) 7 (47 [22–73]) 9 (60 [33–83]) 13 (87 [59–98]) 13 (87 [59–98]) 14 (93 [66–100]) 14 (93 [66–100]) 13 (87 [59–98]) 15 (100 [75–100]) 15 (100 [75–100]) 15 (100 [75–100])
2 (18 [3–52]) 6 (55 [25–82]) 5 (45 [18–75]) 6 (55 [25–82]) 9 (82 [48–97]) 10 (91 [57–99]) 10 (91 [57–99]) 7 (64 [32–88]) 10 (91 [57–99]) 10 (91 [57–99]) 10 (91 [57–99]) 10 (91 [57–99])
8 (100 [60–100]) 8 (100 [60–100]) 7 (88 [47–99]) 7 (88 [47–99]) 4 (50 [17–83]) 5 (63 [26–90]) 0 [0–40]
15 (100 [75–100]) 14 (93 [66–100]) 14 (93 [66–100]) 7 (47 [22–73]) 12 (80 [51–95]) 11 (73 [45–91]) 1 (7 [0–34])
11 (100 [68–100]) 11 (100 [68–100]) 11 (100 [68–100]) 11 (100 [68–100]) 9 (82 [48–97]) 8 (73 [39–93]) 3 (27 [7–61])
0 [0–40] 2 (25 [4–64]) 4 (IQR 2–5) 6 (75 [36–96])
2 (13 [2–41]) 5 (33 [13–61]) 5 (IQR 2–12) 8 (53 [27–78])
5 (45 [18–75]) 9 (82 [48–97]) 6 (IQR 4–13) 11 (100 [68–100])
*Proportions are reported with 95% CIs and medians are reported with IQRs.
mechanical ventilation, respiratory isolation, and advanced imaging capabilities (Table). When asked about a typical day in the ED, 0% (95% CI 0% to 40%) of small EDs reported patient care taking place in a hallway compared with midsized and large EDs, which reported 13% (95% CI 2% to 14%) and 45% (95% CI 18% to 75%), respectively. Similarly, 82% (95% CI 48% to 97%) of large EDs reported patients boarding greater than 2 hours compared with 25% (95% CI 4% to 64%) of small EDs and 33% (95% CI 13% to 61%) of midsized EDs. All large EDs reported operating at or over capacity (Table). 4 Annals of Emergency Medicine
LIMITATIONS This descriptive, self-reported study using the National Emergency Department Inventory survey instrument to survey Indian Health Service ED directors or knowledgeable proxies was subject to bias and error. To mitigate bias and improve validity of results, ED sites were assigned a random site number and respondents were not required to provide their name or title. We acknowledge that respondents had various knowledge of and involvement in ED operations. However, this allowed some level of anonymity because responses were not tied to a Volume
-,
no.
-
:
-
2016
Bernard et al
Indian Health Service Emergency Departments
particular individual at each site. The National Emergency Department Inventory survey instrument has never been applied in the Indian Health Service setting, to our knowledge, and its performance is untested in this health care system. However, the National Emergency Department Inventory survey has been reviewed and refined by multiple Emergency Medicine Network researchers and American College of Emergency Physicians state chapter boards to remove ambiguous language and focused questions covering topics germane to emergency medicine. Furthermore, it has been administered in several other published studies in the United States and abroad, lending to its performance in a diverse context.4 Finally, although we cannot prove nonrespondents did not vary in any appreciable way from respondents, given an 85% response rate in this national survey, the potential for significant bias is likely very limited. DISCUSSION To our knowledge, this cross-sectional survey is the first to examine the basic characteristics, staffing, and capabilities of Indian Health Service EDs, which are critical health care access points for more than 2 million American Indians and Alaskan Natives. We found, as expected, that most Indian Health Service EDs are located in rural areas on or near reservations.2,3 With regard to staffing, only 85% of respondents had continuous physician coverage. Of EDs staffed by physicians, few physicians were board certified or board prepared in emergency medicine. Small and midsized EDs were more likely to be staffed with nonboard-certified or nonboard-prepared emergency physicians compared with larger-volume centers. This finding is consistent with previous work showing that the rural ED workforce is largely made up of later-career, nonboard-certified and nonboard-prepared emergency physicians.7 This has broad implications in regard to health care equity and quality because it is likely that any further emergency physician shortage will disproportionately affect these rural populations. Moreover, for the foreseeable future, nonboard-certified or board-prepared emergency physicians will continue to provide the bulk of ED care in these rural, low-volume EDs within the Indian Health Service. Thus, it is incumbent on the Indian Health Service, local hospital leadership, and emergency medicine governing bodies to ensure that these providers are capable of delivering high-quality evidence-based ED care. One proposed strategy to cope with emergency physician shortages could be to expand the role and scope of advance practice clinicians, especially at smaller-volume sites.8 Currently, 62% of Indian Health Service EDs use Volume
-,
no.
-
:
-
2016
advance practice clinicians, and, although the data were not statistically significant, large EDs were more likely to staff advance practice clinicians compared with small and midsized EDs. Our study may have identified a potential role for advance practice clinicians at lower-volume sites, which disproportionately experience physician shortages, and may function more like urgent care centers. In this setting, with appropriate supervision and support, solo coverage by an experienced advance practice clinician may be sufficient and, given our findings here, unavoidable at some Indian Health Service sites. However, the case-mix acuity of patients seeking care at lower-volume Indian Health Service EDs is unknown. Without a clear understanding of the complexity of disease and the acuity of patients using these small Indian Health Service EDs, one cannot make conclusions about the intensity of resources needed to meet emergency medical needs of those communities. Further investigation may allow Indian Health Service EDs to find the ideal workforce proportion and cost-effectiveness of advance practice clinicians at these sites. Previous studies examining care delivered in the Indian Health Service ambulatory setting have shown that patients and physicians have difficulty obtaining high-quality subspecialty care, hospital admission, diagnostic studies, and mental health services.1,2,9 Our study highlights that this lack of subspecialty care extends to ED settings because most sites lacked access to cardiologists, neurologists, and specialty surgeons. This dearth of subspecialists appears to be more prevalent at smaller sites compared with midsized and large EDs. Therefore, there may be a role for innovative health information technologies to bridge the gap in subspecialty care available to patients using Indian Health Service EDs.9,10 However, in general Indian Health Service EDs had low access to and use of electronic resources, in particular, telemedicine. In our study, only 4 sites had telemedicine capabilities, which consisted of telestroke or teleradiology services. Moreover, these services were available at midsized and large EDs rather than small ones, which tend to have lower access to subspecialists and are more likely to have a nonboard-certified or nonboard-prepared physician or no physician coverage in the ED. In essence, the EDs and communities that could benefit most from health information technologies and telemedicine did not have access to these resources. Expansion of telemedicine initiatives, especially at lower-volume sites, seems to be a logical and feasible solution, but the low penetration of such programs needs to be addressed.10 Finally, the majority of EDs surveyed reported operating at or over capacity, which may contribute to the higher Annals of Emergency Medicine 5
Bernard et al
Indian Health Service Emergency Departments
correlates of ED crowding observed in this study (eg, patient boarding). Our study found that this burden lies disproportionately with large EDs, which may be due to their higher patient volumes. However, the contribution of other factors that have been associated with ED crowding, such as patient acuity, inpatient bed capacity, and workforce shortages, is unknown. To date, the current approaches to address boarding in Indian Health Service EDs and the effects on patient care are also unknown and merit investigation. Our study found that Indian Health Service EDs vary widely in regard to staffing, as well as electronic, technical, and diagnostic capabilities. Indian Health Service EDs also have physician shortages and lack access to subspecialty services. Furthermore, despite a dearth of subspecialist support we found generally low use of telemedicine. Finally, Indian Health Service EDs, in particular largervolume ones, reported operating over capacity and higher correlates of ED crowding. Our data identified areas for improvement and facilitate further investigation into physician recruitment and retention initiatives; mix of workforce, including advance practice clinicians; broader implementation of telemedicine; and the causes of and solutions for crowding. The present study provides a foundation for survey research in an area that requires further, more rigorous study to understand and improve the ED care delivered to American Indians and Alaskan Natives. The authors acknowledge the Indian Health Service participants, internal review board, and administration, without whose support these data would not be available and this important work could not progress; and Jane Bittner, MPH, at the Emergency Medicine Network Coordinating Center for her assistance. Supervising editor: Daniel A. Handel, MD, MBA Author affiliations: From the Department of Emergency Medicine, Tuba City Regional Health Care Corp, Tuba City, AZ (Bernard); and the Department of Emergency Medicine and Emergency Medicine Network (EMNet), Massachusetts General Hospital, Boston, MA (Camargo, Hasegawa, Sullivan). Author contributions: KB, KH, and CC contributed to article conception, development, and analysis; drafting and revision of the article; and the final draft. All authors contributed to final article approval. All authors certify that this article is a valid work and is not under review for publication elsewhere, nor have these
6 Annals of Emergency Medicine
data been publicly available. KB takes responsibility for the paper as a whole. All authors attest to meeting the four ICMJE.org authorship criteria: (1) Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND (2) Drafting the work or revising it critically for important intellectual content; AND (3) Final approval of the version to be published; AND (4) Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. 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 as per ICMJE conflict of interest guidelines (see www.icmje.org). This study was supported by a grant from the Massachusetts College of Emergency Physicians. Dr. Bernard, MD, MBA, was awarded an Indian Health Service Health Professions Scholarship and is an emergency physician at an Indian Health Service facility. Publication dates: Received for publication August 30, 2016. Revision received November 14, 2016. Accepted for publication November 18, 2016.
REFERENCES 1. Zuckerman S, Haley J, Roubideaux Y, et al. Health service access, use, and insurance coverage among American Indians/Alaska Natives and whites: what role does the Indian Health Service play? Am J Public Health. 2004;94:53-59. 2. Sequist T, Cullen T, Bernard K, et al. Trends in quality of care and barriers to improvement in the Indian Health Service. J Gen Intern Med. 2011;26:480-486. 3. Indian Health Service. Indian Health Service locations. Available at: https://www.ihs.gov/locations/. Accessed July 1, 2016. 4. Emergency Medicine Network. National Emergency Department Inventories (NEDI). Available at: http://www.emnet-usa.org/nedi/ nedi.htm. Accessed July 1, 2016. 5. Emergency Medicine Network (EMNet). The National ED Inventories survey instrument. 2011. Available at: http://www.emnet-nedi.org. Accessed July 1, 2016. 6. Harris P, Taylor R, Thielke R, et al. Research Electronic Data Capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377-381. 7. Macho M, MacKenzie A, Ginde A. The rural emergency physician workforce. J Rural Emerg Med. 2014;1:3-8. 8. Brown D, Sullivan A, Espinola J, et al. Continued rise in the use of midlevel providers in US emergency departments, 1993-2009. Int J Emerg Med. 2012;5:21. 9. Baldwin L, Hollow W, Casey S, et al. Access to specialty health care for rural American Indians in two states. J Rural Health. 2008;24:269-278. 10. Rheuban K. The role of telemedicine in fostering health-care innovations to address problems of access, speciality shortages and changing patient care needs. J Telemed Telecare. 2006;12: 45-50.
Volume
-,
no.
-
:
-
2016
Bernard et al
Indian Health Service Emergency Departments
APPENDIX E1 National Emergency Department Inventory Indian Health Service Survey
2015
Indian Health Service Emergency Department Study Funded by the Massachusetts College of Emergency Physicians Supported by the Emergency Medicine Network (www.emnet-usa.org)
Dear Colleague: We would like your help with a brief survey of Indian Health Service (IHS) emergency departments (EDs). The aim is to gather information from leaders of all IHS EDs. Your response to this survey will inform ongoing efforts to improve the accessibility and quality of emergency care within IHS and beyond. Please answer questions with estimates for the year 2014. If data are available for a fiscal year (e.g., October 2013 – September 2014), and that’s easier for you, please use the fiscal year data. Although the survey asks for factual information, if you are unable to quickly find the precise answer, please provide your best guess. The survey takes 10-15 minutes to complete, and results will be completely confidential. Your name will not appear on the survey and no individual survey responses will be shared with anyone. A summary of the survey results will use aggregate responses only; this report will be shared with all ED directors statewide. Without your response, this study cannot succeed. We hope that you will help us by completing this survey or asking a knowledgeable colleague for his/her help. Thank you in advance for your assistance.
Best wishes,
Ken Bernard, MD,MBA Senior resident BWH/MGH Boston, MA
Carlos Camargo, MD, DrPH Director, EMNet Mass General Hospital Boston, MA
Please return your completed survey to: Ken Bernard, MD, MBA
SITE #
Volume
-,
no.
-
:
-
2016
Annals of Emergency Medicine 6.e1
Bernard et al
Indian Health Service Emergency Departments
6.e2 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2016
Bernard et al
Volume
-,
no.
Indian Health Service Emergency Departments
-
:
-
2016
Annals of Emergency Medicine 6.e3
Bernard et al
Indian Health Service Emergency Departments
6.e4 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2016