Utilizers of the Emergency Department (ED) : What Is the Impact on - PowerPoint PPT Presentation

Community Health Worker Outreach Among Frequent Utilizers of the Emergency Department (ED) : What Is the Impact on Primary Care Follow-Up? Erica Shelton, MD, MPH, MHS Assistant Professor Department of Emergency Medicine Johns Hopkins University @EricaSheltonMD

Rise of CHW Programs  CHW concept gained prominence in the 1960s with emergence of CHW programs in many LMIC  Limitations of Western medical model to reach needs of large rural and impoverished populations  WHO publication Health by the People (1975)  Significant impact in areas of communicable diseases (TB, HIV), nutrition, and maternal & child health _________________________________________________________________________________ Perry HB, et al. Annu Review Publ Health 2014. 35:399-421. Newell KW, ed. World Health Orgn.(WHO) Health by the People 1975.

Rise of CHW Programs…cont’d  Later in 1980s & 1990s, CHW programs gained attention in US for achieving positive health outcomes  Limited health literacy and socioeconomically disadvantaged populations in urban and rural settings  Self-care for chronic diseases (DM, HTN) and overcoming barriers to health care experienced in underserved communities ________________________________________________________ Perry HB, et al. Annu Review Publ Health 2014. 35:399-421. Newell KW, ed. World Health Orgn.(WHO) Health by the People 1975.

Background  From 1997 to 2007, annual number of U.S. ED visits increased by 23% from 96 million to 117 million visits  From 2006 to 2014, this number increased by 14.8%  4.5 – 8% of patients visiting the ED are frequent utilizers  Yet frequent ED utilizers account for 21 – 28% of all ED visits  ED utilization by this group is often perceived as nonemergent and contributing toward ED overcrowding ________________________________________________ Jencks SF, et al. N Engl J Med . 2009 Apr 2;360(14):1418-28 Moore BJ, et al. HCUP Statistical Brief #227 , AHRQ. Sept 2017 Shulan M, et al. Health Care Manag Sci. 2013 Jun;16(2):167-75. Epub 2013 Jan 27 Kansagara D, et al. JAMA 2011 Oct 19: 306(15):1688 – 1698

The Perception… Insert ‘boo boo ’ cartoon

The Reality…  Research of frequent ED users in the US has found these patients to have  higher overall mortality  to be admitted to the hospital more frequently  to incur larger heath care costs when compared to occasional ED users ___________________________ LaCalle E, Rabin E. Ann Emerg Med . 2010 Jul;56(1):42-8 Mandelberg JH, et al. Acad Emerg Med , 2000, July: 637-646 Ruger JP, et al. Acad Emerg Med . 2004 Dec;11(12):1311-7

ED Resource Utilization & Costs  Care for this group of patients within the emergency department represents approximately 50% of the Medicaid dollars utilized for ED care  Given finite resources to address ED healthcare, interventions directed towards these patients may  reduce ED utilization rates  improve provider-to-patient ratios for the critically ill ___________________________ LaCalle E, Rabin E. Ann Emerg Med . 2010 Jul;56(1):42-8 Ruger JP, et al. Acad Emerg Med . 2004 Dec;11(12):1311-7

CHW Impact in the US Inpatient Setting  Community health worker (CHW) models within the inpatient setting have demonstrated  Decreased 30-day hospital readmissions  Increased adherence to primary care follow-up after hospital discharge  Increased patient activation and medication adherence _______________________________________________ Kangovi S, et al. JAMA Intern Med. 2014 Apr;174(4):535-43 Gaskin DJ, et al. Health Affairs. 2018; 37(10): 1546-1554

Hypothesis  CHW intervention for frequent utilizers of the ED increases adherence to recommended follow-up primary care and may potentially reduce repeat ED encounters for non-urgent presentations.

Patient Activation Measure (PAM) … Selected Excerpts 1) ‘When all is said and done, I am the person who is responsible for managing my health condition.’ Do you  Strongly agree Agree Disagree Strongly disagree 2) ‘Taking an active role in my own health care is the most important factor in determining my health and ability to function.’ Do you  Strongly agree Agree Disagree Strongly disagree 3) ‘I am confident that I can take actions that will help prevent or minimize some symptoms or problems associated with my health condition.’ Do you  Strongly agree Agree Disagree Strongly disagree 4) ‘I know what each of my prescribed medications does.’ Do you  Strongly agree Agree Disagree Strongly disagree 5) ‘I am confident that I can tell when I need to go get medical care and when I can handle a health problem myself.’ Do you  Strongly agree Agree Disagree Strongly disagree ______________________________________________________________________________________ Hibbard JH, et al. Development of the Patient Activation Measure (PAM). Health Serv Res. 2004;39(4, pt 1):1005-1026

Methods  Performed a block randomization of 191 patients presenting to an urban ED in Baltimore, MD from July 2016-August 2017  Patients were followed over time through June 2018  All participants had >4 ED visits in the past year, described in the literature as indicative of frequent ED utilization.  Enrolled low acuity (Emergency Severity Index Level 3-5) medical and trauma patients discharged from the ED to home within 24 hours  Demographic information collected (age, race, gender); patient intrinsic factors of medication adherence & patient activation measured  ITT analysis completed using Poisson regression ____________________________________________ Morisky DE, et al. Med Care. 1986 Jan;24(1):67-74. Ware JE, et al. New England Medical Center, Health Institute ; 1995.

Baseline Study Participant Characteristics Characteristics Number (%) Enhanced Standard CHW Intervention of Care (n=113) (n=82) Age Group 18-36 32 (28.3) 28 (34.1) 37-60 65 (57.5) 47 (57.3) 60+ 16 (14.2) 3(2.7) Gender Male 55 (48.7) 39 (47.6) Female 58 (51.3) 43 (52.4) Race White 14 (12.4) 15 (18.3) Black 99 (87.6) 67 (81.7) Morisky Medication Adherence Score 0 7 (6.19) 8 (9.76) 1 17 (15.0) 10 (12.2) 2 20 (17.7) 23 (28.0) 3 34 (30.1) 8 (9.76) 4 33 (29.2) 20 (24.4) Baseline PAM Level 1 12 (10.6) 4 (4.88) 2 27 (23.9) 27 (32.9) 3 36 (31.9) 29 (35.4) 4 38 (33.6) 22 (26.8)

Participant Poisson Regression of Primary Care Visits Characteristics IRR (95% CI) 1.12 1.17 1.10 1.16 1.36 1.12 1.56 CHW Group (0.94-1.33) (0.98-1.40) (0.92-1.30) (0.98-1.38) (1.14-1.64) (0.95-1.34) (1.29-1.88) Age Group 18-36 Reference Reference 2.14 2.36 37-60 (1.71-2.68) (1.87-2.97) 2.42 2.13 >60 (1.81-3.24) (1.57-2.89) 1.72 1.76 Female (1.44-2.05) (1.46-2.12) African-American or 3.12 2.75 African Descent (2.10-4.62) (1.84-4.10) Morisky Medication Adherence Score 0 Reference Reference 0.82 0.67 1 (0.55-1.21) (0.45-1.00) 0.98 0.89 2 (0.70-1.38) (0.63-1.27) 1.42 1.18 3 (1.02-1.99) (0.84-1.68) 1.02 0.85 4 (0.73-1.42) (0.60-1.19) Baseline PAM Level 1 Reference 0.49 0.42 2 (0.37-0.64) (0.32-0.56) 0.46 0.36 3 (0.35-0.60) (0.27-0.48) 0.55 0.36 4 (0.42-0.71) (0.28-0.48)

Study Implications CHW Intervention  Associated with increased adherence to primary care follow-up post ED encounter  May have a potential role for decreasing frequent ED utilization for non-emergent conditions  Larger sample size and longer follow-up duration can help better characterize impact

….But What do Our Patients Think?  “She followed up with me, over the entire thing — about 3 months. She met with me at some of my appointments. She was really great.”  “…thoughtful for the medical profession to follow up behind the patients and see how they’re doing after their visit. Very helpful.”

….But What do Our Patients Think?  “Really wonderful. She came with us to different appointments and when the doctor said something we didn’t understand, she was able to ask the question in a way so that the doctor could clarify for us and say things in a way where we could understand.”

Acknowledgements Tynetta (Tye) Lane, CHW Eili Klein, PhD Yu-Hsiang Hsieh, PhD Scott Levin, PhD Felicia Hill-Briggs, PhD Richard Rothman, MD, PhD