“ACEing” Complex Population Management Past, Present, and Possible Future for ACE Models of Care UNTHSC Geriatric Grand Rounds March 25, 2015 Kellie L. Flood, MD Associate Professor Geriatric Quality Officer, UAB Hospital Director, Geriatric Medicine Section Division of Gerontology, Geriatrics, and Palliative Care University of Alabama at Birmingham

Page 2 Learning Objectives  Define the components of an Acute Care for Elders (ACE) Unit  List outcomes from clinical trials evaluating the ACE Unit model of care  Discuss the new role of ACE Units in reducing unplanned readmissions and complex population management

Page 3 In what year did these statements appear in a health care administration publication?  “The overwhelming needs of the aging population have led to increasing expenditures for hospital care….”  “How this growing elderly population will obtain and pay for health care is emerging as a major social issue……”  “This changing financial and demographic trend, coupled with the limited resources provided for the elderly population, has been called the “Geriatric Imperative”” Bachman et al, Hospital and Health Services Administration 32(4):

Page Stock Market Crash Gallon of Gas 88¢ Simpsons Debut on TV BAD Album Released Movie Good Morning Vietnam Released

Page : Healthcare Did Not Heed the Warning Silver Tsunami is Here If you can’t stop the wave… Learn to SURF! 10,000 Baby Boomers will turn 65 years old every day until 2031

Is it just a numbers thing?

Page 7 Older Adults Are A Different Patient Population Just As Pediatric Patients Are Boult et al, The Permanente Journal Winter 2008;12:50-4; Boyd et al, Guided Care for Multimorbid Older Adults, Gerontol, 2007 Older adults experiencing multimorbidity consume 96% of the Medicare budget 62% of older Americans are experiencing multimorbidity

Page 8 Older Adults are More Likely to Experience Geriatric Syndromes  Dementia  Delirium  Depression  Gait and balance abnormalities/Falls  Frailty/Functional Decline  Malnutrition  Pressure ulcers  Polypharmacy  Incontinence  Caregiver Stress Kresevic et al, Ger Nursing, 1998 Geriatric Syndromes = Increased Risk for Adverse Outcomes

Is it just an age thing?

Page 10 Older Adults Are a More Heterogeneous Patient Population Than Younger Adults Functionally and Cognitively Intact (some - maybe not much - room to spare) Functionally or Cognitively Impaired (no margin for error = vulnerable)

Page 11 Why do We Need Evidence-Based Geriatric Care Models?? Baby Boomers Lack of Geriatric Training for all Providers Reduced Reimbursement for Care Slow Economy/ Deficit Slow Economy/ Deficit Perfect Storm

Page 12 We Must Think Outside the Box!!

Page 13 Coordinating Person-Centered Elder Care Requires an Inter(Trans)disciplinary Team

Page 14 Types of Teams in Healthcare  Uniprofessional:  Group of people all from the same discipline working together  Multiprofessional:  Group of people from different disciplines who develop a treatment plan independently GITT Curriculum: Teams and Teamwork; Klarare A. et al, J Pall Med 2013;16(9): Transprofessional:  Although roles are specialized, everyone is prepared to step in/replace each other when necessary; Team leadership varies with the situation – OK to get outside your lane a bit Interprofessional:  Group of people from different disciplines assess and plan care in a collaborative manner

Page 15 Transprofessional “I have learned the importance of the effects of polypharmacy in the care and treatment of UAB's geriatric patients…….. A patient's life may be changed due to medications.” - UAB Trauma Unit Occupational Therapist, 2013

Page 16 Transprofessional “One of the best things I have learned was about the different routes and half-life of IV compared to po pain meds. Last week I was able to counsel a patient and her daughter on the benefits of transitioning off IV pain meds.” - UAB ACE Unit Social Worker, 2014

Page 17 Adapted from slide by SUMMA Health Care What is an ACE Unit? A Model of Inter/Trans- professional Coordinated Care in the Hospital Functional Older Person Acute Illness, Possible Impairment Hospitalization: ACE Unit Depressed Mood Negative Expectations ACE Acute Care for Elders Prehab Program: Specialized environment Patient-centered, interdisciplinary care Multi-dimensional geriatric assessment and non- pharmacologic management with nurse driven care Daily medical review Care transition planning from day 1 Reduced Impairment Decreased Iatrogenic Risk Factors Improved Mood Positive Expectations Functional Older Person

Page 18 Participants: UAB ACE Interdisciplinary Team Meeting  Geriatrician/Geriatric NP  ACE Unit Coordinator  Nurses  Rehabilitation Services (PT, OT)  Pharmacist  Dietician (intermittently)  Social Worker  Pastoral Care (intermittently)  Psychology Interns (intermittently)  Trainees from all disciplines

Page 19 UAB Hospitalist ACE Unit Process Discussed in daily IDT Care transition planning begins Day 1 based on screens Existing/new/risks for geriatric syndromes identified Geri care and transition planning revised daily ACE Coordinator ensures plan implemented Formal geriatric consult for complex cases Admit to ACE Bedside Fxn and Cogn Screen - Katz ADLs - Lawton IADLs - Six Item Screener

Acute Care for Elders (ACE) Units are a team model of coordinated geriatric care in the hospital setting originally designed to maintain patient functional status during hospitalization

Page 21 Change in ADL performance from admission to discharge (p=0.009) Landefeld et al, NEJM, 1995 Secondary Outcome SNF/rehab/LTC placement: 14% ACE Unit vs 22% Usual Care (p=0.01) ACE Unit: Randomized Controlled Trial

Page 22 ACE in a Community Hospital  1531 community-dwelling patients age ≥ 70 admitted for acute medical illness  Randomized to ACE vs Usual Care  Demonstrated improved processes of care in the intervention unit  Reduced use of restraints  Fewer high risk meds  Earlier and more frequent involvement of physical therapy and social work  Improved patient and provider satisfaction Counsell et al, JAGS, 2000

Page 23 Health Care Utilization and ACE  Retrospective, case-control study  Academic urban hospital  680 ACE vs 680 non-ACE patients age ≥ 65  Matched for age, ethnicity, comorbidity, and DRG (CHF, pneumonia, UTI)  ACE patients:  Shorter mean LOS (4.9 ± 4.3 vs 5.9 ± 4.5, p=0.01)  9.7 % reduced unadjusted mean costs ($13,586 vs $15,040; p=0.012)  No difference in mean number of unadjusted readmissions  11% reduced readmission rate after controlling for age, race, comorbidity, and pre-admission rate Jayadevappa et al, Value in Health, 2006;9:

Page 24

Page 25 UAB ACE Study Comparison of ACE vs Usual Care: FY 10 ACE UnitUsual Care Number of beds % patients age  % % Unit nursing staff allotment (WHPPD) 9.75 Physical therapists FTE: bed ratio 1:191:26 Attending PhysicianHospitalists Formal Geriatric Consultation available upon request Yes Evidence-based delirium prevention care processes YesNo Volunteer mealtime assistance program YesNo Daily Geriatrician led IDT Rounds for Geriatric Care Management YesNo Counselor for patients/families YesNo Flood et al, JAMA Int Med 2013;173:981-7.

Page 26 Patient Characteristics ACE vs UC FY 10: Age ≥ 70 who spent entire hospital stay on ACE or UC Variables Mean (SD) or % ACE (N=428)UC (N=390)P value Age (years)81.6 (6.9)80.9 (6.8)0.11 Gender (Female)69.4%65.9%0.29 Race (White)64.5%59.2%0.30 Comorbidity Score3.4 (3.2)3.1 (3.0)0.14 Case Mix Index1.1 (0.5)1.1 (0.6)1.00 No significant differences in patient characteristics between groups Flood et al, JAMA Int Med 2013;173:981-7.

Page 27 Cost and Readmission Outcomes ACE vs Usual Care FY 10: Age ≥ 70 who spent entire hospital stay on ACE or UC Variables Mean (SD) or % All DRGsTop 25 DRGs ACE (N=428) UC (N=390) P Value ACE (N=260) UC (N=214) P Value LOS (days); Mean (SD) 4.0 (2.7)4.2 (2.8) (2.4)4.1 (2.8)0.11 Variable Direct Cost/ Case ($); Mean (SD) $2,109 ($1,870) $2,480 ($2,113) $1693 ($1063) $2138 ($1431) <.001 Daily Variable Direct Cost/Case ($); Mean (SD) $542 ($383) $595 ($227) 0.01 $484 ($162) $545 ($120) <.001 Patients readmitted to UAB within 30 days of discharge 7.9%12.8% Flood et al, JAMA Int Med 2013;173:981-7.

Page 28 Cost Savings from ACE Model Variable Direct Cost Savings = $371/case ~ $371,000 savings in variable direct cost for every 1000 patients If UC patients experienced ACE model Number of patients age ≥ 65 discharged from ACE Unit

So how can the ACE model, originally designed to maintain patient functional status, possibly impact readmissions?

Page 30 Readmission Patterns for Older Adults with AMI, CHF, and Pneumonia  Medicare claims data from to determine patterns  Mean age of readmitted patients = 80 yrs for all DRGs studied  Most readmits within first 15 days for all studied DRGs Dharmarajan et al, JAMA 2013;309(4):

Page 31 Study Authors’ Thoughts:  “The broad range of acute conditions responsible for readmission may reflect post-hospitalization syndrome – a generalized vulnerability to illness among recently discharged patients, many of whom have developed new impairments both during and after hospitalization.”  Losses in mobility/functional status, nutritional status, delirium, adverse drug events, etc.  “The heightened vulnerability to a diversity of illnesses may explain why interventions that are broadly applicable to many conditions with multiple components or are delivered by a multidisciplinary team are more likely to reduce readmissions.” Dharmarajan et al, JAMA 2013;309(4): Aren’t these what ACE Units address?

Page 32 ACE Unit Models of Care Have Been Shown to:  Improved functional performance at discharge  Improved likelihood of living at home after discharge  Reduced restraint use  Reduced high-risk medication use  Improved nutritional support during hospitalization  Improved patient and provider satisfaction  Reduced length of stay  Reduced health care utilization costs  Reduced 30-day readmissions Landefeld et al, N Engl J Med 1995; Counsell et al, JAGS 2000; Jayadevappa et al, Value in Health, 2006; Flood et al, Crit Rev Onc/Heme 2010; Flood, et al, Am J Geriatr Pharmacother 2009; Baztan et al, BMJ 2009; Flood et al, JAMA Int Med 2013.

What is the future for ACE? Helping hospitals address complex population management via: Higher Valued Care Quality Cost

Page 34 Possible Means of Leveraging ACE Model for Higher Valued Care Hospital-Wide  ACE for non-general medical patient populations  Oncology-ACE  Stroke-ACE  Ortho-ACE  ACE of Hearts  Etc, etc  “e-Geriatrician” using ACE Tracker  “Mobile ACE” Consultative Care  UAB “Virtual ACE” Pilot

“Acefying” a Hospital via “Virtual ACE” First UAB Virtual ACE Unit: Orthopedic Surgery

Page 36 Geriatric Info Now Feeds into the Unit ACE Tracker Report

Page 37 Key Geriatric Syndromes in Virtual ACE Training and Intervention  The “Why”  Function/Safe Mobility  Pain Management  Delirium  Care Transitions Delirium Toolbox

Page 38 Virtual ACE Ortho Unit Staff Feedback  “Before the ACE we had delirium cases so frequently, now cases have tremendously subsided.”  “Getting them moving early on has increased their satisfaction with the care at UAB and makes their pain much better. Appetite improves too.”  “Just the awareness of delirium prevention has opened our eyes to things we wouldn’t have noticed before. As a unit it seems everyone is working well together by implementing these initiatives. It’s easy to become complacent if you don’t know how to work effectively with a geriatric patient, but the ACE initiative has made us excited to make changes and actively see results.”  “Toolbox is a great thing to have ”

Page 39 Measuring Outcomes: Pre- and Post- Pilot Test of Virtual ACE Intervention Variable*Pre (N=31)Post (N=94)P-Value Age Mean Range 71.5 ± ± P=.055 Gender 55% F54%FNS H/o Fall in last 3 months 48% Yes NS Baseline Katz Score (Mean) 9.89 ± ± 3.5NS Current Katz Score (Mean) 6.55 ± ± 4.3NS % Abnormal Six Item Screen on Admission 21%20%NS *Variables have missing data for some patients

Page 40 Early Process Outcomes: 8 Weeks Pre- and 20-Weeks Post-Training

Page 41 Mobility in the Prior 24 Hours: All Patients Pre: 43 assessments in 31 patients; Post: 30 assessments in 26 patients Pre vs Post Baseline Katz 10.4 ± 3.2 vs ± 2.3, p=.278 Pre vs Post Current Katz 7.0 ± 5.1 vs 7.3 ± 4.3, p=.831

Page Ortho Unit Fall Rate for ≥ 65 years of age Virtual ACE Safe Mobility Training

Page 43 39% Relative Reduction in Delirium Prevalence Pre-Training: 38 NUDESC Screens in 31 patients Post-Training: 62 NUDESC Screens in 68 patients

Page Ortho Unit Restraint Usage Virtual ACE Delirium Training

Page 45 Potential Cost Savings from Delirium Prevention Reducing Delirium from 18% to 11% Reduces delirium cases from 3,578 to 2,187 + $2,500* cost/patient $3,477,500 saved JUST from delirium aspect of their care UAB Hospital Discharged 19,880 patients age ≥ 65 in FY 13 * Rubin JAGS 2006

Page 46 Learning Objectives Revisited  Define the components of an Acute Care for Elders (ACE) Unit  Interdisciplinary, patient-centered, multi-dimensional geriatric assessment, non-pharmacologic management, daily medical review, and care transition planning from day 1  List outcomes from clinical trials evaluating the ACE Unit model of care  Improved functional status, processes of care, med safety, likelihood of living at home after discharge, and reduced costs  Discuss the new role of ACE Units in reducing unplanned readmissions and complex population management  ACE appears to reduce readmissions via recognition and management of “post-hospitalization syndrome”  Improved outcomes provide leverage to disseminate ACE to non-medical patient populations and throughout an entire hospital

Page 47 QUESTIONS? UAB Hospital 1,156 beds of complex population management