Reasons Why Patients Choose an Ambulance and Willingness to Consider Alternatives Lalena M. Yarris, MD, Raymond Moreno, MD, Terri A. Schmidt, MD, MS, Annette L. Adams, PhD(c), MPH, Heather S. Brooks, BS
Abstract Objectives: To test a hypothesis that patients would accept alternatives to transport to an emergency department (ED) by ambulance and to evaluate factors related to patient willingness to consider alternatives. Concerns about resource utilization have prompted emergency medical services (EMS) systems to explore alternatives to ambulance transport to an ED, but studies have evaluated the safety of alternatives, not patient preferences. Methods: Trained research assistants surveyed patients transported by ambulance to a university ED. Interfacility transfers, trauma patients, and critically ill patients were excluded. The primary outcome was willingness to accept one of several presented alternatives to ambulance transport to the ED for that visit. Demographic and clinical factors were evaluated for association with willingness to consider alternatives. Relative risks (RR) and 95% confidence intervals (95% CI) were determined by using Mantel-Haenszel stratified methods. Results: Three hundred fifteen subjects completed the survey. Two hundred forty-seven (78.4%) were willing to consider at least one alternative. One hundred ninety-four (61.6%) were willing to consider transportation by car, and 177 (56.2%) were willing to consider transportation by taxi. Factors associated with willingness to consider alternatives included the following: age 18–65 years (RR, 1.25; 95% CI = 1.03 to 1.49), being unemployed (RR, 1.08; 95% CI = 1.08 to 1.33), use of the ED for routine care (RR, 1.25; 95% CI = 1.17 to 1.35), and not being admitted to the hospital (RR, 1.19; 95% CI = 1.04 to 1.40). Race, gender, health insurance status, and EMS interventions en route were not associated with willingness to consider transportation alternatives. Conclusions: Many patients transported by ambulance to an ED would have considered an alternative, if one were offered. ACADEMIC EMERGENCY MEDICINE 2006; 13:401–405 ª 2006 by the Society for Academic Emergency Medicine Keywords: paramedic, emergency medical services, triage, transport
From the Center for Policy and Research in Emergency Medicine, Department of Emergency Medicine, Oregon Health and Science University (LMY, RM, TAS, ALA, HSB), Portland, OR; Providence St. Vincent Medical Center (RM), Portland, OR; Department of Public Health and Preventive Medicine, Oregon Health and Science University (ALA), Portland, OR; American Medical Response, Northwest (TAS), Portland, OR. Received September 9, 2005; revision received November 20, 2005; accepted November 21, 2005. Presented as a poster at the Western Regional Research Forum of the Society for Academic Emergency Medicine, Marina Del Rey, CA, May 2005, and presented at the Annual Meeting of the Society for Academic Emergency Medicine, New York, NY, May 2005. Address for correspondence and reprints: Lalena M. Yarris, MD, Department of Emergency Medicine, Oregon Health and Science University, 3181 SW Sam Jackson Park Road, Mailcode CDW-EM, Portland, OR 97239-3098. Fax: 503-494-8237; e-mail:
[email protected].
ª 2006 by the Society for Academic Emergency Medicine doi: 10.1197/j.aem.2005.11.079
T
he emergency medical services (EMS) system was designed to treat and transport seriously ill and injured patients. Currently, patients access the EMS system for many different reasons. In some EMS systems, a significant number of 911 calls do not ultimately result in ambulance transport to an emergency department (ED). In addition, several studies suggest that many patients who are seen in the ED, or who contact 911, could be treated elsewhere.1–3 Faced with mounting call volumes and increasing costs, many EMS systems are exploring alternatives to ambulance transport to a hospital ED. These alternatives have included treat-and-release programs and transport to primary care physician offices or urgent care clinics. Several studies have evaluated potential triage guidelines, either at the time of the 911 call or in the field.4–12 These studies evaluated the safety of such programs and have reported varying under-triage rates. The need for further study in this area is well recognized, with a recent conference of
ISSN 1069-6563 PII ISSN 1069-6563583
401
402
Yarris et al.
PATIENT PREFERENCES REGARDING ALTERNATIVES TO AMBULANCE TRANSPORT
EMS experts agreeing upon a standard set of triage criteria and outcome measures.13 These triage systems assume that the public would be willing to accept alternatives to ambulance transport. Although research is ongoing to evaluate the safety and cost-effectiveness of alternatives to ambulance transport, to our knowledge, there are no published studies that investigate patient preferences with respect to these alternatives. For these field triage systems to be effective, patients must be willing to consider and accept alternatives to the traditional model of ambulance transport to a hospital ED. The purposes of this study were to determine the reasons why patients chose to call an ambulance for transportation to the ED and to evaluate their willingness to consider alternatives to ambulance transport. We also sought to determine whether certain factors were associated with willingness to consider alternatives. We hypothesized that a significant proportion of people who are transported by ambulance to an ED would be interested in alternatives to ambulance transport. METHODS Study Design This was a cross-sectional study that used a convenience sample of patients who were transported in an ambulance to one university-based urban ED. This study qualified for expedited institutional review board (IRB) review status and received approval from the Oregon Health and Science University IRB. Written informed consent was obtained from all subjects. Study Setting and Population Oregon Health and Science University is a Level 1 trauma center with an ED volume of approximately 45,000 visits per year. It receives ambulances from a three-county catchment area, consisting of urban, suburban, and rural zones. The EMS system is a dual-response system with fire-based first response and private ambulance transport. All ambulance transport is advanced life support (ALS). In our EMS system, paramedics cannot refuse to transport patients. The vast majority of patients transported to our institution by ambulance are transported directly to a room, and the average time from registration to triage to room for ambulatory patients typically averages less than 30 minutes. The study population consisted of consecutive patients who arrived in the ED by ambulance during the hours that trained research assistants were available (7:00 AM to 11:00 PM, 7 days per week). Patients were excluded if they were deemed medically unstable by treating physician or were not able to be interviewed within 2 hours of arrival, were a trauma system entry, were a victim of a sexual assault, were on a psychiatric hold, were in police custody, were transported from another medical facility, or were not English speaking. The two-hour time limit for enrollment was selected because it was believed that patients who had been in the ED longer than this would likely have received therapies or diagnoses that might bias their responses. On the basis of a sample-size calculation of 200 patients, we enrolled patients from May 24, 2004 to September 1, 2004.
Survey Content and Administration After medical stabilization, research assistants administered a face-to-face survey to patients or the parents of patients within 2 hours of arrival at the ED. The survey was designed to assess patients’ reasons for using an ambulance and whether they would consider alternatives, previous uses of an ambulance, access to medical care, and opinion of EMS services. The survey instrument was pilot tested on approximately 10 patients before being finalized. The piloting process identified questions that were unclear or difficult to answer, wording that could be interpreted as inflammatory, and formatting improvements. This feedback was used to revise the survey to improve the ease and consistency of administration and ability of the survey to accurately represent the patient perceptions. The survey is available as an online Data Supplement at http://www.aemj.org/ cgi/content/full/j.aem.2005.11.079/DC1. Data Analysis The sample size was calculated by using a desired significance level of 0.05, power of 0.80, and a detectable difference of 10% between the null and alternative hypotheses, for which we anticipated that approximately 60% of our subjects would accept alternative transportation and we wanted this proportion to be statistically significantly greater than 50%. Data collected as continuous variables, such as age, were summarized as means, and comparisons between groups were made by using Student’s t-test with adjustments made for unequal variances. Questions related to alternatives to ambulance transport were collected as binary variables for which respondents could select any of several offered possible alternatives that they would be willing to consider. Respondents were asked to respond to each alternative separately in a yes–no fashion. Respondents who indicated a willingness to consider at least one of the proposed alternatives were classified as willing to consider alternatives. Other survey questions inquiring as to respondent opinions were collected by using Likert scales. These categorical variables were summarized as proportions. Comparisons between groups were made by stratified analysis by using Mantel-Haenszel relative risk (RR) estimates14 and 95% confidence intervals (CI). Age, gender, and race were assessed as potential confounding variables. All results are presented as age- and gender-adjusted. RESULTS Demographics A total of 315 subjects completed the survey. An additional 144 subjects were approached about participating in the study but declined to do so, for an overall response rate of 68.6%. A total of 1,393 patients were excluded from the study because they met one of the defined exclusion criteria (Table 1). The mean age of participants was 44.4 years (range 0 to 94 years), and 140 (44.4%) were female. Table 2 shows the distribution of respondents with respect to age, gender, and race. Most respondents (83.8 %) reported having health insurance. There was no significant association between insurance status and willingness to consider alternatives.
ACAD EMERG MED
April 2006, Vol. 13, No. 4
www.aemj.org
Table 1 Reasons for Exclusion and Corresponding Number of Patients Excluded for Each Reason Reason for Exclusion
n
Trauma system entry Too sick Approached >2 h after arrival Interhospital transfer Nursing home or care facility Missed Code 3 ambulance Treating physician believed patient was inappropriate Already participated Minor without adult present Left without being seen before being enrolled Sexual assault evaluation
376 324 225 145 99 41 37 30 20 13 10 1
Willingness to Consider Alternatives Two hundred forty-seven of 315 respondents (78.4%, 95% CI = 73.9% to 82.9%) indicated willingness to consider at least one of several offered alternatives to ambulance transport to the ED for that visit (Table 3). People aged R 65 years were 20% less likely to be willing to consider alternative modes of transportation than were those aged 18–64 years (RR, 0.80; 95% CI = 0.67 to 0.97). Differences in gender or race were not associated with willingness to consider alternatives. A minority of subjects represented racial minorities, which reflects the racial distribution of the greater Portland area. Thirty-six percent of respondents were admitted to the hospital, and subjects in that group were 16% less likely to be willing to consider alternatives than were those who were not admitted (RR, 0.84; 95% CI = 0.73 to 0.96). Reasons for Taking an Ambulance Subjects were asked if any of a list of reasons for taking an ambulance to the ED applied to them for this visit (Table 4). When considering the association between the reason that subjects took the ambulance and their willingness to consider alternatives, patients who were more likely to consider alternatives were those who called the ambulance because they were unsure whether
Table 2 Demographic Characteristics of the Study Sample and the Association between These Characteristics and Willingness to Consider Alternative Transportation Options to an ED Willing to Consider Alternative Transportation Options Characteristic
Yes (n = 247)
Age, mean ( SD) 43.8 (20.6) Male, n (%) 138 (55.9) Race, n (%) White 195 (81.3) African American 24 (10.0) Native American 17 (7.1) Asian 4 (1.7) Pacific Islander 2 (0.8) Hispanic 10 (4.2) RR = risk ratio.
No (n = 68)
Age and Gender-adjusted RR (95% CI)
46.6 (26.1) 37 (54.4)
— 1.03 (0.91, 1.15)
58 4 3 4 1 2
(87.9) (6.1) (4.6) (6.1) (1.5) (3.0)
0.91 1.12 1.05 0.65 0.80 1.09
(0.80, (0.94, (0.86, (0.33, (0.36, (0.86,
1.04) 1.33) 1.27) 1.29) 1.78) 1.39)
403
Table 3 Alternatives to Ambulance Transportation Offered and Number of Respondents Willing to Consider Each Alternative
Alternatives Offered Coming to ED in a car Coming to ED in a taxi Having the ambulance take patient to a doctor’s office or clinic Taking themselves to a doctor’s office or clinic Being treated by paramedics and not transported Willing to accept any alternative
Respondents Willing to Consider Offered Alternative,* n (%) 194 (61.6) 177 (56.2) 117 (37.1) 81 (25.7) 128 (40.6) 247 (78.4)
* Respondents could have endorsed more than one transportation alternative, so the proportions will not sum to 100%. The denominator for each proportion is 315 subjects.
they needed to come to the ED (RR, 1.18; 95% CI = 1.06 to 1.31), didn’t have health insurance (RR, 1.16; 95% CI = 1.01 to 1.33), didn’t have a doctor (RR, 1.27; 95% CI = 1.17 to 1.38), or had no other way of getting to the ED (RR, 1.14; 95% CI = 1.02 to 1.28). Patients who took the ambulance because they thought treatment would be started sooner were less likely to consider alternatives to ambulance transport to the ED (RR, 0.83; 95% CI = 0.73 to 0.94). Patients who identified hospital EDs or urgent care centers as their usual care providers were more likely to consider alternatives than were those who identified a physician’s office as their site of usual medical care (RR, 1.25; 95% CI = 1.17 to 1.35 and RR, 1.23; 95% CI = 1.11 to 1.36, respectively).
DISCUSSION Overcrowded EDs and busy EMS systems have motivated previous studies that have attempted to evaluate the safety and cost-efficiency of alternatives to ambulance transport. Those studies have used different criteria for determining the safety of field triage.15 To the best of the authors’ knowledge, they have not considered whether or not patients would be willing to consider alternatives. Even if it could be clearly shown that
Table 4 Reasons Respondents Chose to Take an Ambulance to the ED Reasons Why Ambulance Was Used Someone else called Life-threatening emergency Unsure whether needed to come Paramedic recommendation Thought wait would be shorter Don’t have health insurance Don’t have a doctor No other way of getting to ED Thought treatment would be started sooner * N = 315.
Number of Respondents Who Agreed that Reason Applied to Them, n (%)* 208 131 92 162 48 25 36 144 133
(66.0) (41.6) (29.2) (51.4) (15.2) (7.9) (11.4) (45.7) (42.2)
404
Yarris et al.
PATIENT PREFERENCES REGARDING ALTERNATIVES TO AMBULANCE TRANSPORT
alternatives to ambulance transport are safe and effective, the clinical applicability of this information would be low without patient willingness to accept such alternatives. This study suggests a high level of patient interest in alternatives to ambulance transport. More than three quarters of our respondents reported that they would have been willing to consider at least one alternative, if offered. There is no consensus that alternatives to ambulance transport are either safe or effective, and this study did not attempt to evaluate the safety or appropriateness of using an alternative. Therefore, it cannot add knowledge about the safety of such alternatives. What this study does add to ongoing discussions is the knowledge that if alternatives to ambulance transport were determined to be safe and cost-efficient, the majority of respondents in this study would embrace such alternatives. Thus, although the process of testing and implementing alternative solutions presents many difficult challenges, it is worth continuing research efforts in this direction, at least as far as patients are concerned. In assessing the willingness of respondents to accept alternatives, two categories of options were offered. The first category involved coming to the ED via an alternative means of transport. More than half the respondents would have come to the ED in either a car or taxi if one were provided for them. This correlates with our finding that nearly half of the respondents chose to take an ambulance as a result of lack of access to suitable transportation. The second category addressed whether respondents would be willing to accept an alternative destination of care. A significant percentage of respondents indicated a willingness to consider transportation to a doctor’s office or clinic, and a similar percentage would consider being treated and not transported. This suggests that whereas some patients activate EMS for transportation reasons, a second subset of patients do so as a means of obtaining some form of medical evaluation. Many of our respondents indicate that they do not necessarily believe that this medical evaluation needs to occur in an ED. Furthermore, these two subsets of patients are not mutually exclusive. More than a quarter of patients were willing to accept both an alternative means and alternative destination by indicating they would have been willing to drive themselves to the doctor’s office or clinic. We also explored reasons that patients choose to take an ambulance to the ED. In the majority of cases, patients took the ambulance because someone else called it. This is consistent with an earlier study finding that in most cases, a third party contacted 911 for the patient.4 Whereas a minority of the patients who come by ambulance are critically ill, 42% of respondents in this study came by ambulance because they believed that they were having a life-threatening emergency. This suggests that if they were appropriately educated about the nature of their condition, they might be willing to consider an alternative. A review of patient education in the ED reports that several controlled studies have demonstrated the efficacy of standardized patient education in the ED for various conditions.16 The REACT study found that public education led to increased use of EMS in patients with potential acute coronary disease.17 These studies all
suggest that public education may change behavior and that it may be possible to better inform the public both when to access EMS and when not to do so. Further studies are needed to determine whether patient education interventions in the ED decrease future ambulance utilization. Another potential reason for taking an ambulance is lack of access to a primary care provider. A preliminary report found that patients with a primary care provider were more likely to seek medical care before coming to the ED18 and that access to primary care providers can decrease ED use.19 Another recent study found that patients enrolled in a drug treatment program with onsite medical care are less likely to use an ED than are patients enrolled in a program without medical care.20 However, in this study, only 7% of respondents identified lack of a primary care provider as a reason contributing to their taking an ambulance that day. Those developing alternatives to ambulance transport must consider the predominant reasons that patients take an ambulance, because the likelihood of patients accepting alternatives is higher if those alternatives specifically address the reasons that the patient chose to take an ambulance in the first place.
LIMITATIONS This study was conducted in a single ED and may not be generalizable to other populations and settings. In addition, hours of enrollment were limited to 7:00 AM to 11:00 PM; there may be a significant difference in the reasons that patients call the ambulance after 11:00 PM or in their willingness to accept alternatives to ambulance transport in this time period. Although our research assistants attempted to enroll all patients transported by ambulance, they were unable to interview some patients, either because of their acute medical condition or because they were unwilling to participate. We cannot determine whether the responses of that group would have been different. We were unable to collect information on patients who were not enrolled either because they met exclusion criteria or because they declined to participate, and we therefore cannot compare the enrolled to the unenrolled group, nor rule out the potential for selection bias. In addition, no attempt was made to determine the necessity of ambulance transport. However, many of the patients most likely to need the ambulance (trauma system entries, code 3 ambulances, and patients too ill to be interviewed in the first 2 hours) were intentionally excluded. Nonetheless, we cannot comment on whether alternatives to transport would have been appropriate in the group that would be willing to accept them. Finally, this study presented patients who had already been transported by ambulance to a hospital ED with a hypothetical question: If an alternative were available, would you have been willing to accept it? Although we believe that this justifies concluding that these patients would indeed be willing to consider alternatives to ambulance transport, we recognize that post hoc willingness to consider alternatives may not equate to prospective choice. Because our system does not currently offer
ACAD EMERG MED
April 2006, Vol. 13, No. 4
www.aemj.org
such alternatives, it was not feasible to conduct this study prospectively. Such a study would be a logical next step. CONCLUSIONS A majority of respondents in this study who took an ambulance to the ED would be willing to consider an alternative to ambulance transport to the ED if one were offered. Although further studies are needed to establish the safety and cost-effectiveness of alternatives to ambulance transport to the ED, our results indicate that patients would be interested in transportation alternatives if they were offered. The authors thank their OHSU Clinical Research Investigative Studies Program (CRISP) research volunteers, Terri Davis from OHSU Department of Emergency Medicine, and Jon Shields from American Medical Response for their contributions to this investigation.
References 1. Gratton MC, Ellison SR, Hunt J, Ma OJ. Prospective determination of medical necessity for ambulance transport by paramedics. Prehosp Emerg Care. 2003; 7:466–9. 2. Schaefer RA, Rea TD, Plorde M, Peiguss K, Goldberg P, Murray JA. An Emergency Medical Services program of alternate destination of patient care. Prehosp Emerg Care. 2002; 6:309–14. 3. Grumbach K, Keane D, Bindman A. Primary care and public emergency department overcrowding. Am J Public Health. 1993; 83:372–8. 4. Schmidt TA, Neely K, Adams AL, et al. Is it possible to safely triage callers to EMS dispatch centers to alternative resources? Prehosp Emerg Care. 2003; 7:368–74. 5. Schmidt TA, Atcheson R, Federiuk C, et al. Hospital follow-up of patients categorized as not needing an ambulance using a set of emergency medical technician protocols. Prehosp Emerg Care. 2001; 5:366–70. 6. Schmidt TA, Atcheson R, Federiuk C, et al. Evaluation of protocols allowing EMTs to determine need for treatment and transport. Acad Emerg Med. 2000; 7:663–9. 7. Neely KW, Norton RL, Schmidt TA. The strength of specific EMS questions for identifying patients with important clinical field findings. Prehosp Emerg Care. 2000; 4:322–6. 8. Schmidt TA, Mann NC, Federiuk CS, Atcheson CS, Fuller D, Christie MJ. Do patients refusing transport
405
remember descriptions of risks after initial ALS assessment? Acad Emerg Med. 1998; 5:796–801. 9. Kamper M, Mohoney BD, Nelson S, Peterson J. Feasibility of paramedic treatment and referral of minor illnesses and injuries. Prehosp Emerg Care. 2001; 5: 371–8. 10. Silvestri S, Rothrock SG, Kennedy D, Ladde J, Bryant M, Pagane J. Can paramedics accurately identify patients who do not require emergency department care? Prehosp Emerg Care. 2002; 6:387–90. 11. Pointer JE, Levitt MA, Young JC, Promes SB, Messana BJ, Ader MEJ. Can paramedics using guidelines accurately triage patients? Ann Emerg Med. 2001; 38: 268–77. 12. Dunne RB, Compton S, Welch RD, Zalenski RJ, Bock BF. Prehospital onsite triaging. Prehosp Emerg Care. 2003; 7:85–8. 13. Mann NC, Schmidt TA, Cone DC. Defining research criteria to characterize medical necessity in emergency medical services: a consensus among experts at the Neely conference. Prehosp Emerg Care. 2004; 8:138–53. 14. Rothman KJ, Greenland S. Applications of stratified analysis methods. In: Modern Epidemiology. 2nd ed. Philadelphia: Lippincott-Raven, 1998. 15. Cone DC, Benson R, Schmidt TA, Mann NC. Field triage systems: methodologies from the literature. Prehosp Emerg Care. 2004; 8:138–53. 16. Wei HG, Camargo CA. Patient education in the emergency department. Acad Emerg Med. 2000; 7: 710–7. 17. Osganian SK, Zapka JG, Feldman HA, et al., for the REACT Study Group. Use of emergency medical services for suspected acute cardiac ischemia among demographic and clinical subgroups: the REACT Trial Rapid Early Action for Coronary Treatment. Prehosp Emerg Care. 2002; 6:175–85. 18. Ertz-Berger B, Edlund T, Brooks H, Lowe R. Oregon Health Plan cutbacks and access to care: a survey of emergency department patients [abstract]. Ann Emerg Med. 2004; 44(4):S7. 19. Lowe RA, Localio AR, Schwarz DF, et al. Association between primary care practice characteristics and emergency department use in a Medicaid managed care organization. Med Care. 2005; 43:792–800. 20. Laine C, Lin YT, Hauck WW, Turner BJ. Availability of medical care services in drug treatment clinics associated with lower repeat emergency department use. Med Care. 2005; 43:985–95.
