A Rapid Two-Stage Screening Protocol for Palliative Care in the Emergency Department: A Quality Improvement Initiative

Transcription

1 Vol. 42 No. 5 November 2011 Journal of Pain and Symptom Management 657 Special Section: Quality Improvement Projects A Rapid Two-Stage Screening Protocol for Palliative Care in the Emergency Department: A Quality Improvement Initiative Myra Glajchen, DSW, Robin Lawson, LMSW, Peter Homel, PhD, Paul DeSandre, DO, and Knox H. Todd, MD Department of Pain Medicine and Palliative Care (M.G., R.L., P.H.), Beth Israel Medical Center, New York, New York; Department of Emergency Medicine (P.D.), Emory University, Atlanta, Georgia; and Department of Emergency Medicine (K.H.T.), University of Texas M. D. Anderson Cancer Center, Houston, Texas, USA Abstract Background. A rapid two-stage screening protocol was developed to improve referral for palliative care needs among frail elderly in the emergency department (ED). Measures. A new triage tool was administered, with assessment tools for activities of daily living, performance, functional staging, symptom burden, and caregiver distress. Intervention. Stage One identified elderly patients meeting criteria for lifelimiting conditions. Stage Two referred patients with crescendo losses in activities of daily living, high symptom burden, and caregiver distress to palliative care or hospice. Outcomes. Over eight months, 1587 patients were screened, representing 22% of ED visits made by patients older than 65 years during this time period. Of these, 140 met functional decline criteria, and 51 of these needed palliative care consultation. Five patients were referred to hospice, 20 received palliative care, and 26 received no further service. Conclusions/Lessons Learned. The project shows unmet needs among elderly ED patients, and the feasibility of rapid screening and referral using a quality improvement approach. At its peak, the project accounted for half the referrals to the palliative care consultation service. J Pain Symptom Manage 2011;42:657e662. Ó 2011 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Rapid QI, palliative care screening, hospice referral, emergency department Address correspondence to: Myra Glajchen, DSW, Department of Pain Medicine and Palliative Care, Beth Israel Medical Center, 1st Avenue at 16th Street, New York, NY 10003, USA. mglajchen@ chpnet.org Accepted for publication: June 23, Ó 2011 U.S. Cancer Pain Relief Committee Published by Elsevier Inc. All rights reserved. Background The U.S. emergency department (ED) plays an important role as the provider of acute emergency medical care and the gateway to different sites and levels of care. 1 Approximately 10% of ED visitsdmore than 10 million annuallydare made by people older than /$ - see front matter doi: /j.jpainsymman

2 658 Glajchen et al. Vol. 42 No. 5 November years. This group is more likely to be hospitalized, receive a greater number of tests, and be admitted to a critical care unit. 2 Yet research suggests that elderly patients suffering from life-limiting illnesses get caught in a revolving door of emergency care that wastes resources and fails to address their goals of care. 3 All too often, frail elderly patients cycle from one acute episode to the next without being offered palliative care or service coordination that could ease distressing symptoms and improve follow-up care. In addition, the experience of elderly patients in the ED has been neglected in both clinical care and research. 4 Decisions made in the ED often determine subsequent medical treatments for lifelimiting illnesses. For this reason, the ED is a crucial setting for initiating end-of-life discussions with patients, families, and primary care physicians. Yet there are major obstacles preventing the practice of good end-of-life care in the ED setting, including acuity of medical problems, poor physical design, and lack of privacy. 5,6 Additional barriers to end-of-life-care in the ED include rapid decision making by a new physician, absence of a complete history, an ED culture that prioritizes aggressive diseasemodifying therapy, and problem-focused assessment. These factors can lead to neglect of complex palliative care issues, such as assessment of goals of care, communication and shared decision making, 7,8 and identification of unmet palliative care needs. 9 To address the needs of elderly patients and explore the feasibility of rapid screening and referral, BriefPal was developed as a joint project between the ED and palliative care service in a large academic medical center with 940 beds and 60,000 annual ED visits. The center serves the lower east side of Manhattan and Brooklyn, with a high percentage of elderly and lowincome patientsdmedicaid (44%), Medicare (26%)dand an ethnically diverse populationd Caucasian (37%), Hispanic (25%), African American (25%), and Asian (10%). Measures First, a full-time project social worker was hired and a team was established. The BriefPal team created a rapid standardized two-stage screening protocol to identify unmet needs among patients 65 years and older seen in the ED. The BriefPal protocol included a screening tool specifically designed for the project, the Karnofsky Performance Scale Index, the Functional Assessment Staging Tool, the Memorial Symptom Assessment Scale, the Katz Activities of Daily Living (ADLs) Scale, and the Brief Assessment Scale for Caregivers. A copy of the screening tool may be downloaded from Intervention In Stage One, elderly patients with specific life-limiting conditions were identified by the project social worker, who was stationed fulltime in the ED. Conditions included advanced dementia, severe congestive heart failure, chronic obstructive pulmonary disease, advanced malignancy, and AIDS, with at least moderate functional status limitations. In Stage Two, subgroups with recent losses in ADLs, high symptom distress, poor functional status, and high levels of caregiver burden were identified. The social worker administered the tools face-to-face during regular weekday daytime hours and identified appropriate patients to the ED resident or attending. Referrals for palliative care or hospice were made by the ED physicians from the subgroup of patients identified in Stage Two. The BriefPal project was implemented over eight months. During this period, referral processes were streamlined, a laminated card listing referral criteria was developed, and a referral hotline was installed. Palliative care training was concurrently provided to 79 physicians, physician assistants, nurses, and social workers, and local ED champions were identified. ED physicians then did their own screening and referral in consultation with the patient s primary care physician. Outcomes A total of 1587 patients older than 65 years were screened in the ED over the eightmonth active intervention project period (Fig. 1). This sample represents 22% of patients older than 65 years who visited the ED during daytime hours Monday through Friday. Of these patients, 144 met illness criteria, 140

3 Vol. 42 No. 5 November 2011 Rapid Screening Protocol for Palliative Care in the ED 659 Screened > 64 Years of Age n = 1,587 Met Illness Criteria: CNS, COPD, CHF, CA, AIDS, Other n = 144 Karnofsky <80 or loss of ADL within three months n = 140 Current hospice patient, no service needed n = 1 No service needed n = 88 Palliative care consultation needed n = 51 Referred to hospice n = 5 Referred to palliative care n = 46 Palliative care consultation not pursued n = 26 Received palliative care consultation n = Fig. 1. Flow of patient recruitment. met further functional decline criteria, and 51 were identified as needing palliative care. Five patients were referred to hospice, and 46 were referred for specialist level palliative care. Table 1 shows the characteristics of the patients referred for consultation and the patients not referred. The mean age of patients in the referral group was 80 years. Of these, 35% of the patients spoke English, 30% spoke Spanish, 22% spoke Chinese, and 13% spoke another language. The most common primary diagnoses were cancer (43%), dementia (22%), and congestive heart failure (16%). The mean Karnofsky score of referred patients was 46 (of 100), and patients had lost an average of 18% of ADLs within three months. Pain was the most common symptom (14%), followed by shortness of breath (12%), nausea/diarrhea (8%), lack of appetite (8%), and weight loss (8%). Sixty-seven percent of the patients referred for palliative care consultation were accompanied by a caregiver. Although there was no significant difference in the rate of patients being accompanied by a caregiver, the caregivers of patients in the referral group tended to show significantly higher levels of burden. In comparing patients in the referral and non-referral groups, the two groups tended to be similar but patients in the referral group

4 660 Glajchen et al. Vol. 42 No. 5 November 2011 Table 1 Characteristics of Patients Referred to BriefPal vs. Patients Not Referred Characteristics Referred (n ¼ 51) n (%) Not Referred (n ¼ 88) P-value Age, y (mean SD) Male gender, 18 (35) 34 (39) 0.75 frequency (%) Dominant language a English 16 (35) 34 (46) 0.08 Spanish 14 (30) 29 (39) Chinese 10 (22) 6 (8) Other 6 (13) 5 (7) Accompanied by caregiver 34 (67) 57 (65) 0.86 Karnofsky score, mean SD Karnofsky <80 51 (100) 87 (99) 0.99 Loss of ADL within 3 9 (18) 3 (3) months Primary diagnosis CHF 8 (16) 16 (18) 0.71 COPD 2 (4) 10 (11) 0.13 Dementia 11 (22) 25 (28) 0.79 Cancer 22 (43) 20 (23) 0.01 Other CNS disease 5 (10) 9 (10) 1.00 Symptom distress Pain 7 (14) 11 (13) 0.84 Shortness of 6 (12) 20 (23) 0.11 breath Nausea/diarrhea 4 (8) 4 (5) 0.73 Lack of appetite 4 (8) 2 (2) 0.27 Weight loss 4 (8) 7 (8) 0.98 Died during 41 (80) NA d intervention period BASC, mean SD SD ¼ standard deviation; CHF ¼ congestive heart failure; COPD ¼ chronic obstructive pulmonary disease; CNS ¼ central nervous system; BASC ¼ Brief Assessment Scale for Caregivers. a Language information available for 46 referred patients and 74 nonreferred patients. had higher rates of cancer (P ¼ 0.01), higher rates of functional loss within three months (P ¼ 0.04), and significantly higher caregiver burden (P ¼ 0.04). There was also a trend toward lower Karnofsky scores among referred patients (P ¼ 0.06). Follow-up telephone calls were made to 112 patients who were referred, and their medical records were reviewed. Based on this followup, those patients who had received palliative care services reported symptom reduction in pain, shortness of breath, nausea, and anxiety. Forty-one patients (80%) from the referral group died during the intervention period, mostly in the hospital (n ¼ 25), but also on hospice (n ¼ 8), hospice home care (n ¼ 5), and in a nursing home (n ¼ 3). An additional 16 patients were unreachable because of telephone disconnection, suggesting an even higher rate of death among this group. Data regarding death in the non-referral group were more difficult to collect, and are, therefore, too incomplete to be reported. At the peak of implementation, the BriefPal project accounted for half of all referrals to the palliative care consultation service. Fig. 2 shows the rate of referral to palliative care and hospice during the implementation phase, with referral peaks at the midpoint, after staff education and during the months when the project social worker was in the ED full time, with a marked decline when she was away. The consultation process led to five new hospice admissions, including three direct admissions from the ED. Of the remaining 46 patients with unmet palliative care needs, 26 did not pursue palliative care consultation because of primary care physician reluctance (22) or patient/family reluctance. 4 Twenty patients went on to receive specialist level palliative care. Through the project, provider, patient, and system barriers to palliative consultations in the ED were identified by direct observation and field notes. Reluctance to refer by the primary care physician was the most common barrier, accounting for one-half of the nonreferred cases. Reasons included fear of involving palliative care, misunderstanding palliative care and hospice, and belief that they were already doing palliative care. System barriers included lack of understanding about palliative care and hospice and hesitation from ED staff about involving another medical specialty. The palliative care service itself was in a state of flux, with a gap in attendings for the duration of the project. Patient and family barriers May June July August September October November December Fig. 2. Monthly referrals to BriefPal from May 2007 to December 2007 (n ¼ 51).

5 Vol. 42 No. 5 November 2011 Rapid Screening Protocol for Palliative Care in the ED 661 included uncertainty about the patient s prognosis and the symbolic meaning of palliative care and hospice. Conclusions/Lessons Learned The main focus of the ED is to manage people with acute illness or trauma, but people with chronic and advanced disease, and those facing end-of-life emergencies, present to EDs across the country in increasing numbers. 10 The proportion of elderly patients admitted to the ED is likely to increase dramatically over the coming decade as the population ages. Rapid screening for unmet palliative care needs in the ED is, therefore, critically important, now and in the future. The BriefPal project was designed to explore whether precisely targeted palliative care education in the ED, coupled with the crucial presence of local champions, could expedite the rate of referrals to palliative care and hospice. In this regard, the project was very successful. We found unmet symptom, functional, and caregiver needs among this group of elderly patients, suggesting that ED visits provide an unrealized opportunity to address these needs. Referrals peaked when the project social worker was in the ED, clearly demonstrating the importance of trained champions even in the rapidly changing ED environment. Other worthwhile lessons were learned through the BriefPal project. Ideally, the rate of referral should be sustained regardless of staff assignments, but referrals dropped when the social worker was away. Staff embraced the role of staff champions, but high staff turnover meant loss of these champions, leaving a referral void. In addition, funding and staffing can affect communication and slow response time to urgent consultations. The palliative care service was both understaffed and in flux for the duration of the BriefPal project, which could have affected the management of the three to five new referrals per week, as well as the rate of future referrals. With a full complement of three attendings, a nurse practitioner, and four fellows in place, the more mature service now manages 10 new weekly referrals with ease. Although the staff tried to demonstrate cost savings, it proved difficult to develop a financially sustainable model as there were too few hospice admissions to support a full-time coordinator. Nonetheless, the BriefPal project showed the feasibility of successful screening, the value of staff education, and the importance of trained champions, especially for a newly developing palliative care service. In addition, the project clearly demonstrates that although palliative care and ED medicine operate within very different cultures, a successful collaboration can be developed and referral processes can be streamlined within the environment of acute clinical needs and rapid response times. Acknowledgments This research project was supported by a grant from The Fan Fox and Leslie R. Samuels Foundation, Inc. The authors declare no conflicts of interest. References 1. Pitts SR, Niska RW, Xu J, Burt CW. National hospital ambulatory medical care survey: 2006 emergency department summary. Natl Health Stat Report 2008;7:1e Singal BM, Hedges JR, Rousseau EW, et al. Geriatric patient emergency visits. Part I: comparison of visits by geriatric and younger patients. Ann Emerg Med 1992;21:802e Todd KH, Ducharme J, Choiniere M, et al. Pain in the emergency department: results of the Pain and Emergency Medicine Initiative (PEMI) multicenter study. J Pain 2007;8:460e Chan GK. End-of-life and palliative care in the emergency department: a call for research, education, policy and improved practice in this frontier area. J Emerg Nurs 2006;32:101e LaMantia MA, Platts-Mills TF, Biese K, et al. Predicting hospital admission and returns to the emergency department for elderly patients. Acad Emerg Med 2010;17:252e Rogers D. The increasing geriatric population and overcrowding in the emergency department: one hospital s approach. J Emerg Nurs 2009;35: 447e Chan GK. End-of-life models and emergency department care. Acad Emerg Med 2004;11:79e Smith AK, Fisher J, Schonberg MA, et al. Am I doing the right thing? Provider perspectives on improving palliative care in the emergency department. Ann Emerg Med 2009;54:86e93.

6 662 Glajchen et al. Vol. 42 No. 5 November Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with metastatic nonsmall-cell lung cancer. N Engl J Med 2010;363: 733e Rondeau DF, Schmidt TA. Treating cancer patients who are near the end of life in the emergency department. Emerg Med Clin North Am 2009;27: 341e354.