ORIGINAL CONTRIBUTION Reasons for Frequent Emergency
ORIGINAL CONTRIBUTION Reasons for Frequent Emergency Department Use by Medicaid Enrollees: A Qualitative Study Roberta Capp, MD, MHS, Lauren Kelley, MSW, MS, Peter Ellis, MD, Juan Carmona, MPH, Adrienne Lofton, BSN, Darcey Cobbs-Lomax, MBA, and Gail D’Onofrio, MD, MS Abstract Background: The Affordable Care Act initiated several care coordination programs tailored to reduce emergency department (ED) use for Medicaid-enrolled frequent ED users. It is important to clarify from the patient’s perspective why Medicaid enrollees who want to receive care coordination services to improve primary care utilization frequently use the ED. Methods: We conducted a qualitative data analysis of patient summary reports obtained from Medicaid enrolled frequent ED users who agreed to participate in a randomized control trial (RCT) evaluating the impact of patient navigation intervention compared with standard of care on ED use and hospital admissions. We defined frequent ED users as those who used the ED four to 18 times in the past year. The study was conducted at an urban, teaching hospital ED with approximately 90,000 visits per year. The research staff conducted interviews (~30–40 minutes), regarding the patient’s medical history, reasons for ED visits, health care access issues, and social distresses. The aforementioned findings were summarized in a 1- to 2-page report and presented to the RCT’s project team (social worker, emergency medicine physician, primary care physician, and patient navigators) on a weekly basis to further understand the needs of this patient population. A diverse team of researchers (program staff and physicians) coded all reports and reached consensus using reflexive team analysis. We reconciled differences in code interpretations and generated themes. Results: One-hundred patients enrolled in the RCT from March 2013 to February 2014, and all 100 patient summary reports were evaluated. We identified three key themes associated with Medicaid enrollee frequent ED use: 1) negative personal experiences with the healthcare system, 2) challenges associated with having low socioeconomic status, and 3) significant chronic mental and physical disease burden. Conclusions: Medicaid frequent ED users engaged in receiving patient navigation services with the goal to reduce ED use and hospital admissions describe barriers that go beyond timely primary care access issues. These include sociodeterminants of health, lack of trust in primary care providers, and healthcare system. ACADEMIC EMERGENCY MEDICINE 2016;23:476–481 © 2016 by the Society for Academic Emergency Medicine The Affordable Care Act added millions of adults to the healthcare system via Medicaid enrollment.1 Medicaid patients, at high risk for using the emergency department (ED) four or more times per year, are sometimes called “frequent ED users.”2–5 In a recent bulletin, the Centers for Medicare and Medicaid Services (CMS) asked states to develop care coordination programs aimed at reducing avoidable ED utilization.6 However, many of those programs are developed by clinical providers with very little patient input.7–9 Previous survey studies show that greater convenience, lack of transportation, and inability to get an From the Department of Emergency Medicine, University of Colorado, School of Medicine (RC), Aurora, CO; Project Access-New Haven (LK, JC, AL, DCL), New Haven, CT; and the Department of Medicine (PE) and the Department of Emergency Medicine (GD), Yale University School of Medicine, New Haven, CT. Received June 17, 2015; revision received December 27, 2015; accepted January 5, 2016. Dr. Capp is funded by KL2 TR001080. All authors listed have contributed sufficiently to the project to be included as authors, and all those who are qualified to be authors are listed in the author byline. The authors have no potential conflicts to disclose. Supervising Editor: Zachary F. Meisel, MD, MPH, MSc. Address for correspondence and reprints: Roberta Capp, MD; e-mail: Roberta.capp@ucdenver.edu. 476 ISSN 1069-6563 476 © 2016 by the Society for Academic Emergency Medicine 476 PII ISSN 1069-6563583 doi: 10.1111/acem.12952 urgent appointment with a primary care provider (PCP) lead patients to use an ED rather than a primary care clinic.10–12 However, those studies are limited by survey methodology and focus their questions on researchers’ preconceptions about why patients use the ED frequently.10–12 In a recent qualitative study, Mautner and colleagues identified three previously unknown factors associated with frequent ED use: early life instability and trauma, difficult interactions with health care providers, and the belief that a positive relationship with a provider is important.13 However, that study included Medicare patients, was conducted up to 4 years after the intervention, and had a high nonresponse rate. To our knowledge, no studies to date have focused on understanding the why adult Medicaid enrollees frequently use the ED when they want to receive help from a patient navigator to avoid future ED visits and hospital admissions. That assessment is essential to developing and implementing patient-centered care coordination programs, such as those provided by patient navigators. In this study, we conducted a qualitative analysis of patient summaries obtained from Medicaid frequent ED users who agreed to participate in a patient navigation versus standard of care randomized control trial (RCT). Our study uniquely focused on a population that is willing to engage in using primary care services and avoid future ED use. METHODS Study Setting and Population This was a secondary data analysis study; we conducted a qualitative analysis of 100 patient summary reports generated by our research staff (two patient navigators and two trained research assistants [RAs]) on Medicaid enrolled frequent ED users who agreed to participate in a RCT of patient navigation for 12 months versus standard of care using qualitative methods (NCT01797068).14 The research staff screened and enrolled patients into the RCT at one ED site of a large, urban, 1,500-bed teaching hospital with approximately 90,000 ED visits per year. This teaching hospital is located in a small city with approximately 131,000 residents, of which 35% are blacks, 27% are Hispanics, and 32% are whites, and 27% of all residents live below poverty level. There are two hospitals located in this city. Eligible patients included patients 21–62 years of age with active Medicaid insurance who lived in the urban area described above and were frequent ED users. For the purposes of the RCT, frequent ED users were defined as those who had between four and 18 ED visits in the past year at any of two hospitals located in the city where the study took place. Patients were excluded if more than 50% of their ED visits were primarily for mental health or substance abuse conditions, incarcerated, or unable to consent to participate in the study. This study was approved by the Yale Institutional Review Board. Data Collection We approached patients eligible to participate in the RCT study while they were receiving care in the ED (Figure 1). The research staff was composed of two patient navigators (one lay navigator and one nurse navigator) and two RAs (one public health graduate student and one premedical student) was present in the ED for approximately 20 hours a week (including evening and weekends) from March 2013 until February 2014. After confirming eligibility and obtaining written informed consent, the staff proceeded to conduct a brief health screening and asked unstructured questions about the patient’s barriers in accessing primary care services, feelings toward primary care use and providers, social situations, and life stressors that could be impacting their health and services patients wanted to receive from their patient navigators. The goal of writing these 1- to 2-page patient summary reports was to help educate the patient navigation project staff on frequent ED users’ characteristics and health and social needs and inform the team on how to improve the delivery of services associated with the patient navigation intervention. The interviews were conducted in a private location and lasted for approximately 30–40 minutes. Immediately upon completing the interview, the research staff compiled a 1- to 2-page patient summary report describing each RCT participant (regardless of whether enrolled in the control group or intervention). These patient summaries were unique in that the information obtained from the patients were synthesized by the patient navigators and RAs. The patient summary reports were then presented by the patient navigators and RAs to the rest of the RCT intervention team, which included the program manager (social work background), program staff, and an emergency and primary care physicians. The team opted to review all 100 case reports to educate all members about this population and their health and social needs. Data Analysis A core team of three researchers, an emergency medicine physician, a primary care physician, and the patient navigation program manager (social work background), reviewed all 100 case reports. Please note that the three coders also attended all multidisciplinary team meetings described above, prior to coding the transcripts. The three coders met on a biweekly basis and evaluated all case reports using reflexive team analysis.15,16 The process began by reviewing the first 10 transcripts to achieve immersion and further develop the coding structure. Due to the nature of how the data was collected, the team knew that the transcripts were focused on key domains: patient’s barriers in accessing primary care services, feelings toward primary care use and providers, social situations, and life stressors that could be impacting their health and services patients wanted to receive from this type of a program.16 Nonetheless, after reading the transcripts, the team was able to identify several new codes (i.e., lack of PCP trust). The team maximized the chance that codes and themes would be reflective of the study participant’s perspectives by developing a set of codes after reading the transcripts. The remaining 90 case reports were split in half, each 45 case reports were coded by two team members, with each team having a ACADEMIC EMERGENCY MEDICINE • April 2016, Vol. 23, No. 4 • www.aemj.org 477 constant person (an emergency medicine physician). All coding team members were trained in qualitative research methods. Other research team members, including those who developed the summary reports, provided input during the code analysis and theme development. We organized our analysis by adopting key elements of the integrated behavioral model.17 The integrated behavioral model has three main components that affect behavior: 1) attitude (experiential and instrumental), 2) perceived norm (injunctive and descriptive), and 3) personal agency (perceived control and self-efficacy). These three components are responsible for intent or decision to perform behavior. Other factors affecting behavior include knowledge and skills to perform the behavior, salience of the behavior, environmental constraints, and habit.17 Transcripts were reviewed in Atlas.ti (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany) by all three coding team members. The transcripts were analyzed using constant comparison method to identify new codes and refine existing ones while adjusting the coding structure accordingly. Prior to coding analysis, all patient summaries were deidentified. All coding and thematic analysis were conducted in Atlas.ti. RESULTS Study Population A total of 248 patients who met the inclusion criteria were approached; 100 (40.3%) agreed to participate in the RCT study and had patient summary reports written by research staff. Of those patients who declined to Eligible patients to participate in the randomized control trial (RCT) evaluating the impact of patient navigation services on ED use and hospital admissions: • Ages 18-62 years • Medicaid insurance • 4-18 ED visits/year at the two local hospitals • Not have =50% ED visits for mental health or substance abuse as a primary complaint • Not be incarcerated • • Residence within the city limit Able to consent to participate in the study Patient agrees to participate in the RCT of patient navigation intervention versus standard of care? Yes No • Written informed consent signed. • Participate in 30-40 minute interview with research staff (patient navigators or research assistants). Upon completion of interview: • Patient randomized to receive standard of care or intervention. • Research staff summarizes findings into a 1-2 patient summary report. Research staff presents patient summary reports to RCT project multidisciplinary team. • Research staff investigates reasons for not wanting to participate in RCT. Patient navigator or research assistant approaches eligible patients Patient summary reports used in this study. Figure 1. Consort study flow diagram. 478 Capp et al. • UNDERSTANDING FREQUENT ED USE participate in the RCT, ~50% were “not interested” or “refused,” while others provided reasons for refusal (i.e., “We will still be coming to the ED a lot anyways,” “I don’t need help with my medical care,” and “I don’t want people in my business”). Study participant’s characteristics are outlined in Table 1. The mean age was 40.3 years, and the majority of enrollees were minorities: 48% blacks, 33% Hispanic, 18% whites, and 1% other. Interestingly, 83 patients identified having a PCP at the time of enrollment. Seventy-seven patients had one or more chronic diseases and 66 had at least one mental illness. Sixty-four patients described their health as either poor or very poor. A total of three themes emerged from the patient navigator’s summary of the patients enrolled in the study (Table 2): 1) personal experiences with the health care system, 2) challenges associated with having low socio-economic status, and 3) significant mental and physical disease burden. Personal Experiences With the Health Care System Several of the patient summaries highlighted how the patients’ previous experiences with the healthcare system shaped their current utilization practice. In particular, patients described negative experiences with their PCPs and/or primary care clinic staff. Although several patients described having an established primary care clinic for several years, they often did not see the same provider for every visit. Furthermore, several patients felt that the PCPs did not listen to their complaints, which incentivized the patients to seek ED care to get a second opinion and tests performed. This was often the case when patients had complaints that were fully worked up, with diagnostics, but no etiology for the patient’s symptoms (Table 2). Many of the case reports stated that because the patient had “state insurance” and because of that the patient felt that he or she was bothering the primary care clinic with questions and was treated differently than those who have private insurance. Furthermore, several patients stated the clinic staff members made negative comments regarding the patient and for that reason the patient felt uneasy utilizing the primary care clinic: “What frustrates the patient the most is that when the patient calls the PCP clinic repeatedly the people who pick up the phone get annoyed with the patient. The patient feels ‘because I have state insurance, I get treated less than a human being’.” Challenges Associated With Having Low Socioeconomic Status Several patients described complex living situations, ranging from being homeless to being a single parent and not having food to feed the children. Most patients reported that dealing with these situations was a priority, making their chronic and/or acute health issues a lower priority. Given the multitude of current and active events in these patients’ lives, several patients described the inability to remember appointments or follow up on referrals, especially given these tended to be far out into the future. Several patients admitted to having more than one no-show PCP appointment due to the aforementioned reasons. Other reasons for no-show appointments included having several providers caring for the person’s health. Most patients who chose to participate in the program felt that they could use help in remembering appointments: “The patient expressed once he leaves the hospital he usually doesn’t follow up with his primary care because he gets busy doing other things or forgets that he has appointments. He would like help remembering his appointments.” “In addition, most if not all patients mentioned transportation being one of the biggest barriers in accessing primary care services. The patients who did not feel that transportation was an issue mentioned living close to the ED and being able to walk to the ED for services when needed. Interestingly, the state Medicaid office pays for cabs to pick up and drop off patients at the primary care clinic, but two patients mentioned that this service was not available for urgent appointments (i.e., same day or next day) and said that it was easier to call an ambulance than to take several buses to get to a PCP clinic. Significant Mental and Physical Disease Burden None of the patients described preventive care and its value; instead patients often appear to wait for health problems to arise, or for acute exacerbation of diseases to occur, to then seek health care services, at which time they are unable to get a same-day appointment. Although study participants were young (mean age was 40 years), 77% of patients had at least one chronic disease, and 66% had at least one mental health diagnosis. Several patients described traumatic life events during their conversation with the patient navigator. Many Table 1 Patient Characteristics Characteristics Percent (N = 100) Gender (female) 73 Mean age (y) 40.3 (mean) Race/ethnicity White 16 Black 50 Hispanic 28 Other 6 Past medical history = 1 chronic disease 77 = 1 mental illness 66 Self-rating of health Excellent 7 Very good 11 Good 18 Poor 41 Very poor 23 Social determinants of health Food insecurity 84 Receives government assistance 89 Homeless 22 Identifies PCP 83 Seen PCP for ? 2 y 31 Seen PCP for = 2 y 52 PCP = primary care provider. ACADEMIC EMERGENCY MEDICINE • April 2016, Vol. 23, No. 4 • www.aemj.org 479 of the participants expressed anxiety and/or depression associated with the events. Although some life stressors described were from childhood or young age, many were recent life stressors (Table 2). DISCUSSION In this study, we found three key reasons why those patients frequently use the ED, beyond lack of access to primary care services: 1) negative personal experiences with the healthcare system, 2) the challenges of low socioeconomic status, and 3) significant chronic mental and physical disease burden. Contrary to recent media portrayals of frequent ED users as “abusers of the system” or “seekers of drugs,” our study shows that chronic illness, mental illness, and complex social situations prevent these patients from accessing their PCPs in a timely fashion. Our findings should inform care coordination programs and healthcare delivery reforms that aim to reduce avoidable ED use and hospital admissions by Medicaid enrollees. Although eight of 10 participants in our study reported having a PCP, barriers such as transportation, insurance insecurity, and poor recall of primary care appointments prevented them from seeing that provider. Patient navigators are trained to help patients overcome barriers associated with healthcare and social care utilization.9,18,19,20 Therefore, we hypothesize that these types of care coordination programs will prove to be successful as they address the needs of a subset of patients. Nevertheless, the Medicaid enrollees in our study mentioned several factors influencing frequent ED use that a patient navigation program would not address. These include having their mental health needs met, strengthening trust in PCPs/clinics, and being able to secure primary social needs (i.e., stable housing). Previous studies have documented the importance of “housing first” programs and behavioral health services for this patient population.21–23 In fact, the Affordable Care Act facilitated the integration of physical and behavioral health services, which will help to meet these patients’ needs.24 However, not many current programs focus on fostering trust in PCPs or healthcare systems. Many of the Medicaid enrollees in our study reported being treated differently because they have “public insurance.” They note that front desk and telephone staff often dismiss their requests, and several patients mentioned overhearing staff making negative personal comments about them. This negative past experience with the primary care clinic staff appears to influence future utilization of that particular primary care clinic in the setting of needing immediate assistance. This type of healthcare delivery concern cannot be addressed by care coordination programs; it requires physician and healthcare system engagement. LIMITATIONS Our findings have several limitations. First, these responses apply to Medicaid enrollees who voluntarily chose to receive help in improving primary care utilization; we did not interview patients who refused to participate in the program. Second, the patient case summaries were developed by the enrollment research staff, not the patients themselves, and therefore may be biased. The research staff were instructed to ask patients how they perceived their own challenges, but the staff’s subjective interpretation cannot be ruled out. Third, these case reports were obtained through open-ended dialogue with patients, without an interview guide. As a result, some qualitative elements may have appeared in some summaries and not others, and some features may have been missed entirely because patients were not asked about them. Nonetheless, our large sample of 100 patients yielded a diverse set of responses. Finally, this qualitative, hypothesis-generating study reflects local practices; further studies must investigate the themes that our study has foregrounded. Table 2 Reason for Frequent ED Use Among Adult Medicaid Enrollees Theme Representative Quotes Personal Experiences with the Health Care System Experiential and instrumental attitudes “The patient thinks that when she goes to the primary care clinic, the doctor does not seriously address the patient’s concerns.” Negative feelings about how others perceive Medicaid frequent ED users “The patient claims that the people behind the counter ‘were ghetto’ and didn’t treat the patient with respect. The patient heard one of them saying ‘Oh God, what is she here for!’ .” Challenges Associated with Having Low Socio-Economic Status Lack of self-efficacy when using the primary care system She says she has a hard time keeping up with all of her appointments which include, Nephrology, Urology, and Gastroenterology. She sometimes misses her appointments and it is very hard to reschedule them.” Challenges with social environment when seeking health care “When asked about his biggest barrier to health care, he answered transportation. He reports he walks to all his appointments and sometimes he does not feel weel when doing it.” Significant Mental and Physical Disease Burden Personal life distresses “The patient recently had gall bladder surgery, at which point she coded.” Significant personal disease burden “Her medical history includes: Grave’s disease, hypertension, myocardial infarction, congestive heart failure, chronic kidney disease, stroke, and morbid obesity.” 480 Capp et al. • UNDERSTANDING FREQUENT ED USE CONCLUSIONS In summary, Medicaid-enrolled frequent ED users who wish to receive help from patient navigators to avoid future ED use and hospital admissions describe barriers that go beyond timely access to primary care, highlighting the need for care coordination services with concurrent improvements in patient experience. Our study suggests that the burden of chronic diseases that require ongoing management, combined with reluctance to use the primary care system because of previous negative personal experiences and environmental barriers in accessing primary care services, need to be addressed to prevent future avoidable ED use and hospital admissions. Patient navigation interventions address some of the aforementioned factors, but not all of them. Specifically, primary care provider trust and reducing negative patient experiences with primary care clinic staff members are key factors may not be addressed by patient navigators. We thank Dr. Karen Albright for her help and support in reviewing the manuscript and providing insightful feedback on the methodology of the paper. References 1. Wachino V, Artiga S, Rudowitz R. How Is the ACA Impacting Medicaid Enrollment? Available from: http:// kff.org/medicaid/issue-brief/how-is-the-aca-impacting-medicaid-enrollment/. Accessed May 15, 2015. 2. Capp R, Rosenthal MS, Desai MM, et al. Characteristics of Medicaid enrollees with frequent ED use. Am J Emerg Med 2013;31:1333–7. 3. Raven MC, Billings JC, Goldfrank LR, Manheimer ED, Gourevitch MN. Medicaid patients at high risk for frequent hospital admission: real-time identification and remediable risks. J Urban Health 2009;86:230–41. 4. Fuda KK, Immekus R. Frequent users of Massachusetts emergency departments: a statewide analysis. Ann Emerg Med 2006;48:9–16. 5. The Henry J Kaiser Family Foundation. Characteristics of Frequent Emergency Department Users. Available from: http://www.kff.org/insurance/ upload/7696.pdf. Accessed May 17, 2015. 6. Mann C. Targeting Medicaid Super-Utilizers to Decrease Costs and Improve Quality of Care. Available from: www.medicaid.gov/federal-policy-guidance/downloads/CIB-07-24-2013.pdf. Accessed May 15, 2015. 7. Shumway M, Boccellari A, O’Brien K, Okin RL. Cost-effectiveness of clinical case management for ED frequent users: results of a randomized trial. Am J Emerg Med 2008;26:155–64. 8. Raven MC, Doran KM, Kostrowski S, Gillespie CC, Elbel BD. An intervention to improve care and reduce costs for high-risk patients with frequent hospital admissions: a pilot study. BMC Health Serv Res 2011;11:270. 9. Enard KR, Ganelin DM. Reducing preventable emergency department utilization and costs by using community health workers as patient navigators. J Healthc Manag 2013;58:412–27; discussion 428. 10. Capp R, Rooks SP, Wiler JL, Zane RD, Ginde AA. National study of health insurance type and reasons for emergency department use. J Gen Intern Med 2014;29:621–7. 11. Cheung PT, Wiler JL, Lowe RA, Ginde AA. National study of barriers to timely primary care and emergency department utilization among Medicaid beneficiaries. Ann Emerg Med 2012;60:4–10.e12. 12. Young GP, Wagner MB, Kellermann AL, Ellis J, Bouley D. Ambulatory visits to hospital emergency departments. Patterns and reasons for use. 24 Hours in the ED Study Group. JAMA 1996;276:460–5. 13. Mautner DB, Pang H, Brenner JC, Shea JA, Gross KS, Frasso R, Cannuscio CC. Generating hypotheses about care needs of high utilizers: lessons from patient interviews. Popul Health Manag 2013;16 (Suppl 1):S26–33. 14. Clinical Trials Website. Patient Navigation for Medicaid Frequent ED Users. Available from: http://clinicaltrials.gov/ct2/show/NCT01797068. Accessed October 15, 2015. 15. Graneheim UH, Lundman B. Qualitative content analysis in nursing research: concepts, procedures and measures to achieve trustworthiness. Nurse Educ Today 2004;24:105–12. 16. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res 2005;15:1277–88. 17. Glanz K, Rimer BK, Viswanath K. Health Behavior and Health Education. Theory, Research, and Practice. Hoboken, NJ: John Wiley & Sons, Inc., 2008. 18. Billings J, Raven MC. Dispelling An urban legend: frequent emergency department users have substantial burden of disease. Health Aff (Millwood) 2013;32:2099–108. 19. Spatz ES, Phipps MS, Lagarde S, et al. Project Access-New Haven: improving access to specialty care for patients without insurance. Conn Med 2011;75:349–54. 20. Greysen SR, Allen R, Rosenthal MS, et al. Improving the quality of discharge care for the homeless: a patient-centered approach. J Health Care Poor Underserved 2013;24:444–55. 21. Larimer ME, Malone DK, Garner MD, et al. Health care and public service use and costs before and after provision of housing for chronically homeless persons with severe alcohol problems. JAMA 2009;301:1349–57. 22. Sadowski LS, Kee RA, VanderWeele TJ, Buchanan D. Effect of a housing and case management program on emergency department visits and hospitalizations among chronically ill homeless adults: a randomized trial. JAMA 2009;301:1771–8. 23. [No authors listed]. Intensive CM cuts ED visits, hospitalizations. Hosp Case Manag 2012;20:153–4. 24. Croft B, Parish SL. Care integration in the Patient Protection and Affordable Care Act: implications for behavioral health. Adm Policy Ment Health 2013;40:258–63. ACADEMIC EMERGENCY MEDICINE • April 2016, Vol. 23, No. 4 • www.aemj.org 481