Are emergency departments really a “safety net” for the medically indigent?

Original Contributions

Are Emergency Departments Really a "Safety Net" for the Medically Indigent? KENNETH V. ISERSON, MD, MBA, TAMMY Y. KASTRE, MD This study was designed to quantify the willingness of emergency departments (EDs) and private care practitioners to see medically indigent patients. Three case scenarios were developed to represent severe, moderate, and mild problems that typically confront ED physicians. A female investigator made telephone calls using these scenarios, each time declaring herself to be medically indigent. All EDs received calls about all three scenarios, but only the least severe scenario was used for private practitioners. The timing and order of all calls were randomized. A control survey of the same population was subsequently performed in which the caller related that she had third-party insurance and had the minimal (rash) problem. The participants were all 54 nonmilitary EDs in Arizona and 69 randomly chosen private primary care practitioners in the same locales as the EDs. Calls to EDs were made during all time periods and days of the week; private practitioners were called only during their weekday office hours. The majority of all EDs were willing to see medically indigent patients, recommending that the caller come to the ED immediately 76% of the time. This response did not vary by geography or the facility's size, although ED personnel suggested initial home treatment more commonly at smaller hospitals (P = .02), and suggested coming to the ED more often on weekends (P < .02). Some EDs, however, clearly did not comply with their own telephone advice policies, and some ED personnel failed to give medically appropriate advice, In contrast to the EDs (P < .001), 62% of private practitioners' staffs stated they were not taking new patients or required at least $30 in advance. Private practitioners in the largest communities were significantly more reluctant to see the medically indigent than their peers in smaller communities (P < .05). For an insured caller, 55% of private practitioners would see the caller for <$30 and only 35% were not taking new patients or provided referral. In contrast to most private primary care practitioners, EDs are at least willing to serve as a triage point for the medically indigent and are often the primary-care "safety net" for the medically indigent. (Am J Emerg Med 1996;14:1-5. Copyright © 1996 by W.B. Saunders Company)

As soaring health care costs lead the United States to reexamine its health treatment system and decide what health care services should be provided to Americans, we

From the Arizona Bioethics Program and Section of Emergency Medicine, University of Arizona College of Medicine, Tucson. Manuscript received March 21, 1994, returned April 11, 1994; revision received May 25, 1994; accepted February 7, 1995. Supported by the Arizona Bioethics Program. Address reprint requests to Dr Iserson, Arizona Bioethics Program and Section of Emergency Medicine, University of Arizona College of Medicine, 1501 N. Campbell Ave, Tucson, AZ 85724. Key Words: Emergency medical services, emergency services, hospital, health policy, medical indigency, medically uninsured, public policy. Copyright © 1996 by W.B. Saunders Company 0735-6757/96/1401-0001 $5.00/0

still do not know what access indigent patients have to medical treatment. Up to 15% of Americans lack the resources to pay for any medical care. 1,2 Yet the United States has a traditional moral commitment to provide medical treatment for all citizens. A commitment honored with varying success in different historic eras and geographic locales, support for the medically indigent has ranged from pro bono care that individual practitioners extended to some of their private patients or donated at free clinics, to charity or governmental support of hospitals) Passage of federal Medicaid legislation and subsequent enabling programs briefly raised the hope that medical treatment could be universally available in the United States. Sadly, the medically indigent encounter increasing difficulty obtaining access to medical treatment as the financial pressures on practitioners and institutions increase. Emergency medicine has, in recent years, described itself as the "safety net" of the American health care system. 4,5 The Emergency Care Guidelines of the American College of Emergency Physicians (ACEP) state that emergency department (ED) care is designed not only for those with threats to life or limb, but also "for those individuals whose health needs are less urgent but for whom the ED may be the only entry point into the broader health care system." 6 According to the College's Ethics Manual, "emergency care is a fundamental individual fight and should be available to all who seek it . . . . Denial of emergency care or delay in providing emergency services based on race, religion, gender, ethnic background, social status, type of illness/ injury, or ability to pay is unethical. ''7 The College not only describes an altruistic position, but also recognizes legal mandates from courts and legislators establishing a right to emergency care and a corresponding physician duty to evaluate all patients who present to EDs, to treat emergencies, and not to transfer unstable patients for whom they can provide definitive care. 8 Some overcrowded EDs skirt these minimum requirements for triage of any patient despite warnings against this practice. 9-n ED physicians and their institutions, however, are not immune to increasing economic pressures on medical practice. Patients made more than 92 million visits to US EDs in 1990, nearly a 19% increase since 1983.12 In some large cities, ED visits increased significantly between 1980 and 1990: Seattle (104%), Columbus, OH (69%), Nashville (67%), Cleveland (59%), San Francisco (57%), Memphis (51%), San Antonio (48%), Los Angeles (33%), Dallas (32%), San Diego (30%), and Chicago (29%). 12

2

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 1 • January 1996

We believe that these additional patients appear to be medically indigent. Typical of inner city EDs, a Torrance, California ED recently reported that nearly 80% of its patients were uninsured and were presumably unable to pay the bill. 4 ED physicians believe that private practitioners refer medically indigent patients to the ED. If EDs are acting as the safety net for America's medically indigent, then this reality must be factored into any revision of the US health treatment system. However, the role of EDs in indigent care has primarily been evaluated on anecdotal evidence. This study compares and quantifies the willingness of EDs and private primary care practitioners to see medically indigent patients and delineates each group's role in providing this treatment. METHODS

We called each of the EDs in Arizona's 54 nonmilitary acute care hospitals to determine their response to an indigent person requesting medical care. The caller told the person giving medical telephone advice that she was indigent and requested care. Three scripted scenarios (Appendix) represented various levels of acuity: minimal (rash), moderate (wrist injury), and serious (worst headache of life). Each ED was called once for each scenario, with random calls on a weekday or on a weekend (Friday through Sunday). A random- number computer program generated the calling sequence, with no site receiving all three calls during the same shift or time of the week. The caller asked to speak to a nurse, and when ED personnel referred health-related questions to a separate nurse-staffed telephone service, that service was contacted, and the caller identified the referring hospital. The caller waited on hold a maximum of 20 minutes and re-called the ED at another time if the waiting time exceeded 20 minutes. The caller collected data about the time of the call ("day" noted as 7 AM to 7 PM and "night" shift as 7 PM to 7 AM), the day of the call (weekend or weekday), who spoke to the caller, their advice, and the timing of any recommendations to seek treatment. Standard sources provided the location and number of annual ED visits for each facility. 13,14 The caller also telephoned the offices of 69 private primary care practitioners (internal medicine, family practice, general practice) in the same locales as the EDs. She related only the minimal (rash) scenario because this was the only one that all practitioners had the experience and ability to evaluate in their offices. An informal survey of primary practitioners showed that many felt uncomfortable treating either a probable wrist fracture or "worst headache of my life" in their office. All calls to physician offices were made during their daytime weekday office hours. The number of physicians contacted was proportionate to the number of nonmilitary acute care hospitals in a particular area. The caller contacted these private practitioners on different occasions as a medically indigent and then as an insured patient. To minimize discrepancies caused by caller variance, the same person placed all of the calls. The female caller is a native English speaker with no discernable accent. We also sent letters requesting all Arizona hospital EDs to send their written policies describing how their personnel should give health-related telephone advice and how they accept patients for treatment. The Institutional Review Board exempted this study from formal review. Data analysis was done by John Roy of the Arizona Biostatistics department. SPSS crosstabs procedure for Windows produced n-way crosstabulation for variables that have a limited number of numeric or string values. SPSS crosstabulation version of ×2 test were performed, and values of P < .05 were considered significant.

RESULTS

No ED refused to see the "medically indigent" caller under any scenario; 72% recommended that the caller come to the ED to be seen (Figure 1). ED personnel recommended that the caller immediately come in for evaluation 80% of the time for the most serious scenario (worst headache), 76% of the time for the moderate scenario (injured wrist), and 58% of the time for the minimal scenario (rash). When the ED personnel did not recommend coming to the ED, they usually suggested that the caller seek care at an urgent care center, often not associated with the hospital: 30% for the minimal scenario, 15% for the moderate scenario and 6% for the serious scenario. Usually the ED personnel explicitly stated that the alternative source of care was less expensive than the ED. Home treatment was seldom recommended: 12% for the rash, 2% for the injured wrist, and 10% for the serious headache. In all instances, the caller was told that if home treatment did not work she could come to the ED to be seen. The ED's patient load significantly influenced the advice given for all of the scenarios (P = .02), with the smallest EDs (0 to 15,000 annual visits) recommending home treatment more often than others (11% of all calls) (Figure 2). Sixty-two percent of private physicians in the same locale as these EDs, when the caller was uninsured claimed that they either were not taking new patients or would not see the caller (minimal scenario, rash) unless she could pay at least $30 in advance (Figure 3). In striking contrast, 55% of these same practitioners told the insured caller they would be willing to see her for less than $30, and only 40% were not taking new patients. They recommended home treatment only 1% of the time. When called about the same scenario, EDs had significantly different responses than the private practitioners (P < .001), recommending that the caller be seen in the ED 58% of the time, giving a specific referral source for treatment in 28% of cases, and describing home treatment methods in 12% of calls. For all three scenarios, whether the caller claimed to be medically indigent or insured, only 4% of private practitioners referred patients to an ED; one private hospital's ED referred callers to the county hospital ED. Five percent of EDs referred patients to private physicians when the caller complained of a wrist injury. The timing of calls to an ED for advice significantly (P < .02) affected the recommendation whether to come to the ED

Go to ED Treat at Home l

Go to Private Doctor

Minimal

]

Go to Other Hospital No Recommendation

i

i

Refused Treatment i

i

.......

10 20

~ ................

T.........................1..................... T

30 40 50

60

70

80 90 100

Percent Respons~

FIGURE 1. Telephone recommendations by ED personnel.

ISERSON AND KASTRE • ARE EDs A MEDICAL SAFETY NET?

3

l

Week, Treat at Home

ll >30,~

l

t 15,00%30,000

1

I I

ED Annual Vis!ts: 0"-~5~0~ I

e Treat Now ] Treat PRN

Week

Go to Other Hospital ! Weekend,

Recommendation1 Refused Treatment] 0

10

20

30 40

50

60

70

80

90 100

Weekend

Percent Response

0

10

20

30

FIGURE 2. Responses versus annual patient visits (P = .02). (Figure 4). It was more likely for the caller to be told to come to the ED for treatment on weekends for all scenarios. Getting through to medically trained personnel was often difficult. The caller waited over 5 minutes for the telephone to be answered or to be taken off hold in 6% of her calls to EDs. The caller was denied access to a medically trained person during 17 of 162 calls (10%) to EDs. Of these 17 calls, 5 then referred the caller to Ask-A-Nurse ® (this is a telephone question-answer line staffed 24 hours a day by registered nurses). In the other 12 cases, the respondent refused to connect the caller with a medical person. Whether the caller was referred for information or told the staff could not give out information depended on the time of day or part of the week in which the call was placed. Yet even those EDs with a written policy forbidding medical advice over the telephone or requiring referral to Ask-A-Nurse occasionally gave telephone advice. Of calls to private physician offices, 60 of 69 (87%) were taken by nonmedical personnel who would not refer to the caller to a medically trained person. Phoenix, Tucson, and northern Arizona contained 48 of the state's 54 nonmilitary EDs. The five categories of responses from these EDs showed no statistical difference by region (P = .31) (Figure 5). There was, however, a significant regional difference (P = .041) in the recommendations from physicians' office staffs (Figure 6). Private physicians in the two smaller communities, Tucson and northern Arizona, were more willing to see patients for less than $30 or make arrangements for payment, rather than demanding payment in advance. Still, an average of 62% of all private practitioners surveyed, in any of the three regions,

40

50

60

70

80

90

100

Percent Response

FIGURE 4. Time ED called versus timing of when told to come to ED (P < .02). either demanded more than $30 in advance or said they would not treat a rash for a self-defined indigent patient. The type of primary care provider (family practice, internal medicine, general practice) did not affect the response (P > .05). Thirty-eight of the 54 EDs (70%) responded to a survey concerning triage and telephone information policies. Of these, all stated that they had written policies describing the triage of patients, but only 74% of these 38 EDs had written policies about telephone advice. Only 64% of these 38 EDs sent a copy of their policy, even after a second request.

DISCUSSION EDs developed from "accident rooms" where acutely injured and ill patients, especially those without a private practitioner, could be seen. Eventually EDs became the site where private practitioners saw many of their patients with emergencies, and relatively recently, the site of full-time practice by emergency medicine specialists. This change in the nature of EDs has paralleled their use by medically indigent patients as their portal of access to the medical treatment system. A recent study of 1,000 consecutive patients noted that payment by "public aid" or "self pay" was significantly associated with routine use of EDs. Income below $10,000 and delay in seeking health care were very

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

....

• Doctors'"] ~J EDS I

.......

Treat at Home

l

i J Northern Arizona Phoenix

Go to

.oTreatment Ir

i

Go to No Recommendation ~

No Recommendation Refused Treatment

Tucson

-0

r

10

20

~

30

40

50

r

60

70

80

............. :_o_J 1

Refused T r e a t ~ u :

i ......

90 100

0

Percent Response

FIGURE 3. Private physician versus ED response to minimal (rash) scenario (P < .001).

:~

10 20

....................... ~

i

30 40

, --==..........r

50

60

70

i

....................... r................. "

80

90 i00

Percent Response

FIGURE 5. .31).

ED responses for all scenarios versus location (P =

4

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 1 • January 1996

Treml Will'

0

10

20

30

40

50

60

70

80

90 100

Percent Response

FIGURE 6. Physician responses versus location (P = .041). significant indicators for patients that were most likely to seek ED care.aS Arizona is a valid sample of the US population and health treatment system. This state ranks 24th in population, and 15.3% of the population is covered by the state's Medicaid equivalent for acute illness (AHCCCS) (personal communication with the US Bureau of the Census and AHCCCS hotline, May 1992). Arizona was the only state never to implement the standard Medicaid system; it now uses a prepaid capitation-based system to fund indigent health care under a federal Medicaid waiver. Arizona also ranks 18th in the nation for the number of physicians per person (191 per 100,000), 24th for the portion of the population receiving Social Security (16%), and 37th for the population percentage receiving food stamps, supplemental income, or aid to dependent children (4.1%). Twenty-six percent of Arizona's population belongs to minority groups (Hispanic, Native American, Asian, or Black). 13 Out study represented a request for care by a medically indigent women who was a native English speaker, the largest subset of the US medically indigent population. 2 The courts have developed the fight for anyone presenting to an ED to get at least an evaluation of whether a medical emergency exists. Some key cases explaining the corresponding ED duty came from Arizona. 16,17This duty was further expanded by the federal government under the Consolidated Omnibus Budget Reconciliation Act of 1986 which requires a medical evaluation of all ED patients regardless of their ability to pay. EDs and their hospitals must also provide all medical care necessary for stabilization or suffer legal and financial penalties. This is one of the few rights to health care in US law. s No equivalent requirements apply to private health care practitioners. The specialty of emergency medicine has also defined the rights of the medically indigent to care in the ED. The American College of Emergency Physicians stated in 1982 that an emergency visit is appropriate when "an unforeseen condition of a pathophysiological or psychological nature develops which a prudent layperson, possessing an average knowledge of health and medicine, would judge to require urgent and unscheduled medical attention..." 18In 1990, the College stated that "Hospital EDs function as society's healthCare safety net 24 hours a day, seven days a w e e k . . . (we) are committed to the principle that emergency care, whenever needed, should be readily available and of highest possible quality for all members of our society." 5

Yet some EDs, severely stressed by financial and logistical constraints, have developed systems to triage patients with relatively minor complaints away from their facility.9,1° Some hospitals have refused to open trauma care units, which frequently operate at a loss, and many trauma centers closed after suffering sizable economic loss. However, most people believe that EDs are a major provider of health treatment to their community's medically indigent. This study quantifies and confirms this belief. Virtually all EDs were willing to see medically indigent patients; their willingness to see them immediately was proportional to the seriousness of the complaint. This contrasts with private primary care practitioners in the same locale who were mostly unwilling to treat even that level of illness (skin rash) that they could easily manage within an office practice, unless the caller was insured. Although private practitioners may believe they are being unfairly chastised because of the actions primarily of their nonmedical staff, it is reasonable to assume that the advice provided by the nonmedical staff who answered most queries in this study reflects what the practitioner has instructed them to say. Some danger signals are evident, however. Some of the telephone advice was given by medically unsophisticated personnel. The fact that the caller was told to try home remedies 10% of the time for "the worst headache I've ever had" suggests that more care is needed in determining who gives advice and what advice is given. Bad advice may be more dangerous to patients than no advice at all. Indeed, ACEP recommends that "telephone advice should be given only by qualified medical professionals who know the limitations and ramifications of providing this service. The quality of telephone advice should be assured through the use of policies, protocols, documentation, and quality assurance programs to monitor outcomes... When any doubt exists, a recommendation should be made for the caller to come to the emergency department. ''18 The Ask-A-Nurse service, to which mainly private hospitals referred callers, either said to go to the ED if home treatment was ineffective, or referred the patient directly to the county hospital ED because the caller was medically indigent. The increased number of weekend recommendations by ED personnel to come to the ED for treatment for all scenarios probably reflects a lack of alternative treatment sites during these periods. It certainly does not reflect a desire for an increased patient volume, because this is the busiest time in virtually all EDs. Although this study showed ED staffs' willingness to evaluate patients, it is silent about what happens when patients arrive at the hospital or practitioner's office. It would be virtually impossible to know the care providers' actions if such a patient arrived, compared with the verbalized policy as expressed by the staff on the telephone. In addition, this study surveyed one state having a unique HMO-based independent health care system with a geographically representative but numerically limited set of primary care providers.

CONCLUSION This study shows that most EDs are at least~willing to serve as a triage point for the medically indigent, and private

ISERSON AND KASTRE • ARE EDs A MEDICAL SAFETY NET?

primary care practitioners are generally u n w i l l i n g to see these patients without s o m e guarantee o f significant advance payment. Alternative public m e d i c a l facilities are generally available to the m e d i c a l l y indigent, but not outside o f w e e k d a y hours and often not on an urgent basis. EDs, however, m u s t i m p r o v e the quality o f their telephone advice, so patients can get appropriate information about their significant m e d i c a l concerns. Physicians within E D s therefore clearly serve as the primary care providers and safety net for m a n y m e d i c a l l y indigent patients. Unfortunately, this appears to be but another s y m p t o m of our failing health care system. 19

REFERENCES 1. Overviewof Entitlement Programs: 1990 Green Book. Washington, DC, US House of Representatives, Committee on Ways and Means, 1990 2. Health Insurance and the Uninsured: Background Data and Analysis. Washington, DC, Congressional Research Service, Library of Congress, 1988 3. Stevens R: In Sickness and in Wealth: American Hospitals in the Twentieth Century. New York, NY, Basic Books, 1989, pp 17-51 4. Baker DW, Stevens CD, Brook RH: Patients who leave a public hospital emergency department without being seen by a physician. JAMA 1991;266:1085-1090 5. American College of Emergency Physicians: Measures to deal with emergency department overcrowding. Ann Emerg Med 1990;19: 944-945 6. American College of Emergency Physicians: Emergency care guidelines. Ann Emerg Med 1991;20:1389-1395 7. SandersAB, DerseA, Knopp R, et al: ACEP ethics manual. Ann Emerg Med 1991;20:1153-1162 8. Knopp RK, Goldfrank LR, Derse AR, et al: An ethical foundation for health care: An emergency medicine perspective. Ann Emerg Med 1992;21:1381-1387 9. Derlet RW, Nishio DA: Refusing care to patients who present to an emergency department. Ann Emerg Med 1990;19:262-267 10. Derlet RW, Nishio DA, Cole LM, et al: Triage of patients out of the emergency department: Three-year experience. Am J Emerg Med 1992;10:105-199 11. Iserson KV: Assessing values: Rationing emergency care. Am J Emerg Med 1992;10:263-264 12. American Hospital Association: Hospital Statistics. Chicago, AHA, 1992 13. US Department of Commerce, Economics and Statistics Administration, Bureau of the Census: Statistical Abstract of the United States, 11th ed. The National Data Book, Washington, DC, Government Printing Office, 1991 14. Office of Health Economics and Facilities Review, Arizona Department of Health Services, Arizona Hospital Statistics, Phoenix, AZ, DHS, 1990 15. Pane GA, Farner MC, Salness KA: Health care access problems of medically indigent emergency department walk-in patients. Ann Emerg Med 1991;20:730-733 16. Thompson vSun City Community Hospital, Inc., 141 Ariz. 597, 688 P.2d 605 (1984) 17. Guerrero vCopper Queen Hospital, 112 Ariz. 104, 537 P.2d 1329 (1975) 18. American College of Emergency Physicians: ACEP Policy Summaries, 1992 Ed, Dallas, TX, ACEP 19. Iserson KV: Limits of health care resources. Am J Emerg Med 1992; 10:588-592

5

APPENDIX: SCENARIOS Standard Patient is a 25-year old right-handed women with no personal physician, no medications, no allergies, no medical history, no money for payment, no health insurance, and with a part-time job that makes patient ineligible for medically indigent state health insurance (AHCCCS). Patient recently moved to local area, has no personal physician, and has never been seen at the particular hospital. Minimal Scenario: "Rash With Itching" Caller: Hello, can I talk with someone about a medical problem? (caller answers) Are you a nurse? I wanted to know if I can be seen for this itching and rash all over my body that is getting worse today. I have no insurance or money. Potential response: How long has this been a problem? Caller: Over the past 2 days. Potential response: Any other problems? Caller: No other problems. For all three scenarios, the responders' possible answers to the question, "Can I be seen for this... ?" are categorized by course of action suggested: 1. Try over-the-counter home treatment. 2. Come to our emergency department. 3. Go to the nearest public clinic. 4. Come to our urgent care clinic. 5. Come to another (specific) clinic at our hospital. 6 Go to other hospital. 7. No recommendation. The responders' answers were also categorized by when they advised the caller to be seen: 1. Should be seen now. 2. Should be seen tomorrow. 3. No need to be seen. 4. Should be seen as needed.

Moderate Scenario: Injured Arm Caller: Can I talk with someone about a medical problem? Are you a nurse? I hurt my wrist today and now it is swollen and really hurts when I move it, can I be seen now? I have no insurance or money. Potential response: Are your fingers numb or tingling? Caller: No. Potential response: How did you hurt it? Caller: I fell off my bike while going to the store. Potential response: Any history of prior injury? Caller: No.

Serious Scenario: Worst Headache Caller: Can I talk with someone about a medical problem? Are you a nurse? I have the worst headache I've ever had. It's all over my head, but worse in the back. It came on all of a sudden. Can I be seen? I have no insurance, and no money. Potential responses: How long have you had this headache? Do you have blurry vision? Have you had any head trauma? Do you have a history of headaches? Do you have any numbness or tingling? Caller: It started all of a sudden. I have no vision problems, no injury, no prior medical history, no numbness or tingling. Potential response: Do you have any neck pain? Caller: A little, when I move my head.