AHA/HRET HEN: Data and Coaching Webinar: Reducing Readmissions Data Review June 4, :00 – 2:00 PM, CDT

Welcome and Overview Welcome, thank you for joining us today! Housekeeping: – This webinar is being recorded and will be archived. – You will receive a PDF of today’s presentation, later this week, as well as a link to fill-out the evaluation, a summary of Q&A and a link for the recording. – For questions: please reach out to your state lead – or us: Agenda: – Readmission Measures – Content Review – Hospital Story – Teach Back 2

Polling Questions (#1 and #2) How Many of You are Joining Us From: Hospital type? – A. General Medical / Surgical – B. Teaching – C. Rural – D. Children’s – E. Long-term Care – F. Psychiatric Hospital size? – A. CAH – B. Not CHA, <100 beds – C. Not CAH, beds – D. Not CAH, 300+ beds 3

Objectives: Readmission Data and Measures Review data requirements Discuss measures listed in the HRET Encyclopedia of Measures Review measure definitions and interpretation examples Discuss options for organization- defined measures 4

Introductions Janine Douglass, MPH, CIC, HRET Bruce Spurlock, MD, Cynosure Health Charisse Coulombe, MS, MBA, HRET Denise Remus, PhD, RN, Cynosure Health 5

Readmission Data Management Strategy Charisse Coulombe, MS, MBA Data Director, HRET

7 Readmission Opportunity? Do you know what your readmission rates are? Overall? For AMI, HF and/or Pneumonia? Compared to other hospitals in your area? State? National?

8 What is Your AIMS Statement? Reduce AMI readmissions by 20% by December 31, 2013 By the end of 2013, reduce readmissions for heart failure patients by 30% Reduce same-hospital readmissions by 20% by December 31, 2013

9 Why is Readmission Data Needed? Measures are used to assess the impact of changes To demonstrate hospitals have reduced their readmissions rates over the two-year period To monitor that interventions to reduce readmissions are working Part of the PDSA cycle

What Readmission Data is Needed? At a minimum, 1 process measure and 1 outcome measure: -Process: Measures interactions between healthcare practitioner and patient; a series of actions, changes or functions bringing about a result -Outcome: Measures change or the end result of healthcare intervention 10

11 Encyclopedia of Measures Technical manual to ensure the hospital's measure definitions align with the comprehensive data system (CDS) Comprehensive details about measure characteristics: -Topic -Measure Name -Definition -Numerator, Denominator -Calculation specifications -Source(s)

12 Preventable Readmissions Process Measures – recommended: -Completion of Discharge Bundle (Project BOOST) -Completion of Patient Care Plan (Project RED) -Formal Assessment of Patients Risk of Readmission (Project RED / BOOST) -Patients Receiving Complete Discharge Education Verified by Teach-Back or Other Means (Project Red / BOOST)

13 Preventable Readmissions Process Measures – alternate: -Evaluation of LVS Function (HF-2) -Heart Failure (HF) Discharge Instructions (HF-1) (Readmission) -Pediatric Asthma: Home Management Plan of Care (HMPC) Document Given to Patient/Caregiver (CAC- 3) -Psychiatric Patients - Post Discharge Continuing Care Plan Transmitted- Overall Rate (HBIPS-7) -Psychiatric Patients with Post Discharge Plan - Overall Rate (HBIPS-6) -Timely Transmission of Transition Record (Inpatients)

14 Readmission Process Measures Identification of high risk patients by using a formal risk assessment tool -Process Measure: Formal Assessment of Patient Risk of Readmission Numerator: Number of patients who were assessed for risk of readmission using the formal tool Denominator: All eligible patients

15 Readmission Process Measures Completion of Bundle/Care Plan/Discharge Instructions -Research has shown that all elements of the discharge process must be completed -Readmission process measures look to see if all elements are completed If elements are consistently being missed/skipped, opportunity to have the readmission team review the process to see where issues are.

16 Polling Question (#3) Has your hospital selected your readmission process measure(s): -A. Yes, selected and actively tracking -B. Yes, only selected (the measure) -C. No, still researching

17 Preventable Readmissions Outcome – recommended: -Readmission within 15 days (All Cause) -Readmission within 30 days (All Cause) Outcome – alternate: -Acute Myocardial Infarction (AMI) Patients - Readmissions within 30 days (All Cause) -Heart Failure (HF) Patients - Readmissions within 30 days (All Cause) -Pneumonia (PN) Patients - Readmissions within 30 days (All Cause)

18 Readmission Outcome Numerator and Denominator Numerator: -Number of patients who were readmitted to the facility within 30 days of their index discharge Denominator: -Number of patients who were discharged alive from the facility for that month (all those patients that have the potential for returning to a hospital within 30 days)

19 Readmission Terms Admission: Patient is admitted to hospital with an inpatient status Index discharge: Patient is discharged from the hospital (alive) Readmission: Patient comes back to hospital (within a pre-determined amount of time) and is admitted with an inpatient status All Cause Readmission: When patient is readmitted to the hospital for any reason (related or unrelated to index discharge diagnosis/DRG)

20 Patient Encounters Inpatient Readmissions: Patient’s index discharge and subsequent readmission to the inpatient setting Patient admitted to observation within 30 days of discharge -Not counted as a readmission but should evaluate Patient seen in Emergency Room within 30 days of discharge -Not counted as a readmission but should evaluate

21 What Month are Readmissions Counted In? Readmissions are counted in the index discharge month. If the patient’s index discharge is March 31, 2012 and they return to the hospital on April 15 th, the readmission counts in the March numerator, not April. This is why the data collection deadline are 2 months out (you need to give the patient 30 days to come back to the hospital).

22 How are Readmissions Counted? Index Discharge is counted in the March denominator Readmission is counted in the March numerator Index Discharge March 15, 2012 Readmission April 4, days from index discharge to readmission

23 How are Readmissions Counted? Index Discharge is counted in the March denominator Readmission is not counted in the March numerator Index Discharge March 15, 2012 Readmission April 17, 2012 > 30 days from index discharge to readmission

24 How are Readmissions Counted? Index Discharge counted in March denominator Only 1 of the readmissions would count in the March numerator (answers the CMS question – was the patient readmitted within 30 days: Yes/No) Index Discharge March 15, 2012 Readmission April 10, days from index discharge to readmission 26 days from index discharge to readmission Readmission April 4, 2012

25 How are Readmissions Counted? Index Discharge counted in March denominator Only 1 of the readmissions would count in the March numerator (answers the CMS question – was the patient readmitted within 30 days: Yes/No) Index Discharge March 15, 2012 Readmission April 10, days from index discharge to readmission 26 days from index discharge to readmission Readmission April 4, 2012

26 Readmission Rate 50 patients discharged alive in the month of March – Denominator 5 patients readmitted within 30 days - Numerator 5/50 * 100 = 10% Readmission Rate

27 Scenarios Bob’s index discharge is March 1 st and is readmitted on March 15 th -He is in March numerator and denominator. Allie’s index discharge is March 15 th and is readmitted on May 2 nd -She is in the March denominator but not in the numerator. Rita’s index discharge is March 20 th and is readmitted on April 14 th -She is in the March numerator and denominator.

28 Scenarios Sheila’s index discharge is March 12 th and is readmitted on March 15 th and April 14 th -She is in March numerator and denominator. The April 14 th admission is not counted towards March. Tori’s index discharge for heart failure is January 1 st and was readmitted on January 15 th, January 25 th and February 10 th -She is counted as a readmission based on the 1/1 since 1/15 and 1/25 fall within the 30 days. 2/10 is the admission that “resets” the readmission count. If Tori is readmitted within 30 day, she will be counted as a readmitted patient.

29 Polling Question (#4) Has your hospital selected your readmission outcome measure(s): -A. Yes, selected and actively tracking -B. Yes, only selected (the measure) -C. No, still researching

30 How Does CMS Count Readmissions on Hospital Compare? The 30-day readmission measures focus on Medicare fee-for-service patients and patients admitted to Veterans Administration (VA) hospitals, at least 65 years of age, with a principal diagnosis of AMI, HF, or PN on the index discharge CMS has access to hospitals’ administrative Medicare claims so if your Medicare patient is discharged and readmitted to a different inpatient hospital, CMS counts that readmission

31 Why is CMS Looking at All Cause AMI, HF and PN Readmissions on Hospital Compare? Medicare Payment Advisory Committee identified those 3 diagnosis readmissions as common, costly and often preventable According to CMS: -Readmission for any cause is an adverse event -Hard to exclude quality issues and accountability based on diagnosis (e.g. HF patient that gets a Hospital Acquired Infection during the index stay and is readmitted with sepsis)

32 CMS Readmissions on Hospital Compare CMS uses a 3 year period for readmission calculation CMS uses risk adjusted methodology CMS counts multiple readmissions within the 30 day period as 1 readmission (answering the question – was the patient readmitted within 30 days, Yes/No) CMS counts your patients that are readmitted to other hospitals Planned/elective readmissions not counted for AMI patient population if they are readmitted for CABG or PTCA procedures

33 CMS Risk-Adjustment Readmission Methodology on Hospital Compare CMS uses risk standardized readmission rates -Number of Readmissions within 30 days predicted based on the hospital’s performance with its observed case-mix -Number of Readmissions expected based on the nation’s performance with that hospital’s case- mix Hospitals are not able to replicate the RSRR independently Resource:

34 Readmission Data Update to Hospital Compare CMS plans to release updated data to Hospital Compare in July, 2012 Will include data from July 1, 2008 – June 30, 2011 CMS updated measure code to accommodate 25 diagnosis codes (increased from 10); accommodate 25 procedure codes (increased from 6) and includes 12 additional e-code slots 34

35 Organization-Defined Readmission Measure A hospital collects readmission measures that are not included in the Encyclopedia of Measures or use different operational definitions Data system allows the hospital to create an organization-defined measure -Hospital specifies the numerator and denominator definitions in addition to entering their data

36 What Readmission Data is Submitted? Baseline -Timeframe flexible -Can submit 1 year, 6 months, or whatever is available -Data set that will be used for comparison to the measurement period(s) Measurement (2 years) -Submitted in monthly increments -Data set that will be compared to the baseline to determine if improvement is occurring

37 Readmission Data Collection & Submission Current: -Hospital directly enters all readmission data into CDS In Progress: -State-level data warehouse readmission data gets uploaded to CDS by HRET Note: -Only collecting aggregate hospital-level readmission data

Polling Question (#5) What type of challenges does your hospital have related to collecting data on readmissions? – A. Billing data not processed in a timely fashion – B. Chart review only available (no electronic system to calculate readmission) – C. Readmission data not given to front-line staff/team in a timely fashion (within 20 days of close of month) – D. Difficulty in tracking patient (master patient index number not available, medical record numbers differ from admission to admission) – E. A combination of A, B, C and D 38

Polling Question (#6) Who is currently reviewing readmission data on a monthly basis? – A. Board of Directors/Quality Committee of Board – B. Senior leadership of your hospital/system (e.g. CEO, VPs) – C. Case Management only – D. Nursing only – E. Interdisciplinary Readmission Team – F. All of the Above – G. Combination of A, B, C, D 39

Questions? 40 ? … Thank you for joining us! We will now transition into the Coaching session of the webinar.

AHA/HRET HEN: Data and Coaching Webinar: Reducing Readmissions Coaching Session June 4, :00 – 3:00 PM, CDT

Welcome and Overview 42 Welcome, thank you for joining us today! Housekeeping: – This webinar is being recorded and will be archived. – You will receive a PDF of today’s presentation, later this week, as well as a link to fill-out the evaluation, a summary of Q&A and a link for the recording. – For questions: please reach out to your state lead – or us: Agenda: – Content Review – Hospital Story – Teach Back 42

Objectives Discuss best practices in reducing readmissions Lean how other hospitals have implemented tests of change and lessons learned Identify improvement strategies to test in your organization 43

Introductions Janine Douglass, MPH, CIC, HRET Bruce Spurlock, MD, Cynosure Health Denise Remus, PhD, RN, Cynosure Health Dean Schillinger, MD, UCSF Professor of Medicine and Director, Center for Vulnerable Populations, San Francisco General Hospital Joan Carroll, RN, BA, CDMS, CCM, Director of Care Transitions, Lee Memorial Health System 44

45 Dean Schillinger, MD UCSF Professor of Medicine Director, Center for Vulnerable Populations San Francisco General Hospital Bridging the Communication Gap to Prevent Readmissions: The “Teach Back Method” (a.k.a. “Closing the Loop”)

Objectives 46 Provide 3 actionable tips to improve communication at discharge: – Reduce jargon – Assess for medication discordance – Use teach-back technique (aka teach-to-goal, closing the loop, show-me-approach)

47

Medical Jargon 48 GLUCOMETER HEMOGLOBIN A1c DIALYSIS ANGINA RISK FACTORS CREATININE

Function of Jargon 49

… unclarified Glucometer Immunizations Weight is stable Microvascular complication System of nerves HbA1c EKG abnormalities Dialysis Wide range Risk factors Kidney function Interact Jargon Terms …from Patient’s own visit: benign blood drawn blood count CAT scan blood count correlate stool was negative stool baseline respiratory tract polyp washed out of your system receptors short course renal clinic blood cells increase your R screening vaccine …clarified Angina Microalbuminuria Ophthalmology Genetic Creatinine Symptoms

Dialysis Dialysis “Do you know what the number one cause for people in this country being on dialysis is? Diabetes” 51 Would you please tell me in your own words what dialysis means? In your own words, what do you think the doctor was trying to tell the patient? “ Check something every day. ” 1 “ Sugar is too high. ” 1 “ What? Is that about you toes? ” 1 “ I can't say it. ” 1 “ It means that your diabetes is going worse that you have to exercise to make diabetes. ” 1 “ Means that more people are getting diabetes. ” 1 “ You got to get on machine to pump.. redo blood to come up to par. ” 4 “ That the sugar was not … hmm. ” 1 “… regarding kidney. ” 2 “ Diabetes is one cause of kidney problems. ” 3 “ That is a warning … about the kidney … my doctor told me about those side effects of the diabetes. ” 3 “ About dialysis, because they are warning us, they are telling me about the complications … that if I'm having problems in my kidney, I'm going to have dialysis. ” 4 “ It ’ s a way to clean blood get off toxins out the blood. ” 4 “ That you need to be on dialysis to cleanse blood or gonna die. ” 4

Unclarified / Own Visit Unclarified Jargon Clarified Jargon Patient Comprehension of Jargon (% Some /Total Understanding)

Why Teach Back? 53 Clinicians frequently overwhelm patients with information and advice, and patients only recall or comprehend 1/2 what was said (probably less in hospital settings). Physician’s advice and instruction is often delivered out of context, is based on assumptions of shared meaning, and rarely is tailored to the individual patient’s needs. The “teach-back” method, if used early and often, can - – Ensure information is understood/integrated into memory – Check for lapses in communication – Open dialogue re health beliefs and unanticipated barriers to “action plans,” and self-mgmt – Reinforce and tailor health messages – Promote a common understanding or “shared meaning” – Elicit patient participation/activation – Maintain your curiosity in the patient as a unique person, with unique stories to tell-

What is it? 54 I employ the “teach-back method” in all of my encounters with patients (and families) at discharge, particularly for those in whom self-management is a central component in preventing readmission. In this interactive technique, the clinician prioritizes amongst the information exchange and explicitly asks the patient to “teach- back” what he/she has recalled and understood re those high- priority domains. Similarly, clinicians can use the strategy to assess patient’s perceptions of the information or advice given. The technique can be used toward the end of a visit or during the course of the visit, so as to tailor communication earlier. Teach-back is NQF Safe Practice #10 for informed consent discussions, and is gaining momentum as a Safe Practice for Discharge

What is the Evidence that It Can Work? 55 Informed Consent Studies Diabetes Management in Ambulatory Care Asthma Education in Hospitalized Patients CHF Self Management Education/Diuretic Self-Titration

56 Ensures info understood/integrated into memory; checks for lapses Opens dialogue re health beliefs; reinforces and tailors health messages Promotes a common understanding; elicits patient participation

57

Closing the Loop 58 Physicians assessed recall or comprehension for 15/124 new concepts (12%) When new concepts included patient assessment, patient provided incorrect response half the time (7/15=47%) Visits using interactive communication loop not longer (20.3 min. vs min) Application of loop associated with better HbA1c (AOR 9.0, p=.02)

Characteristic% 1 pass% 2 passes% >2 passes Adequate literacy Marginal literacy Inadequate literacy US born Born outside US The number of passes required through consent process to obtain informed consent, by participant characteristics Sudore, Schillinger 2006 JGIM

Provider-Patient Concordance in Medication Regimen 60 Patients with atrial fibrillation at high risk of stroke Treatment with warfarin (blood-thinner) reduces risk of stroke by 70% Requires close monitoring and frequent dose adjustments Miscommunication/ inappropriate dosing can lead to poor outcomes (stroke or bleeding) Studies have shown miscommunication rates (discordance) as high as 50%

Literacy, Discordance and Safety 61 Anticoagulant regimen concordance lower for patients with inadequate vs adequate literacy (42 % vs 64 %, OR = 0.41, P<0.01), Anticoagulant discordance associated with being out of therapeutic range: – under-anticoagulation (AOR 1.67, p=.05) – over-anticoagulation (AOR 3.44, p=.01)

Computerized Visual Medication Schedule Linked to Brief Scripted “Teach-Back” 62

Intervention Reduces Time To Therapeutic Range (N=142) 63

64 Intervention Reduces Time To Therapeutic Range (N=142)

HOW? Example 1 Medication Change 65 Doctor (to patient): “ I want to make sure I did a good job explaining your heart medications, because this can sometimes be confusing. Can you tell me what changes we decided to make and how you NOW will take the medications? “ Note especially how the physician places the onus of any possible mis-communication on him/herself. In other words, the “teach-back” task is conveyed not as a test of the patient, but of how well the physician explained the concept.

HOW? Example 2 Behavior Change 66 Nurse (to patient): “ We’ve spent the last few minutes discussing how you are going to exercise and how you are going to change what you eat. These things can be heard to change. Can you repeat back to me these new plans on exercise and eating? And can you tell me how easy or difficult these will be for you to do and what problems you might have in doing them? This will help me give you the best advice? “ Note how the nurse normatizes any possible dis-agreement re the plan or future non-adherence to the agreed-upon plan by framing such disclosures as one means to improve on the nature of any advice. It is important to give the patient time and space to respond; avoid interrupting the patient before he/she has a chance to respond.

How Not To… Example #3: Taking the Easy Way Out 67 Doctor (to patient): “Do you understand what we just talked about? “ or “Do you understand the plan regarding your blood pressure medications?” “Did that makes sense?” These routine queries, which do not require explicit articulation of recall, comprehension, or perceptions on the part of the patient, will universally be met with an uninformative (and possible falsely re-assuring) “Yes, doctor”.

Conclusions: 68 A simple communication tool – the “teach-back method”, a.k.a. “Closing the Loop” - if used early, often, and at strategic moments, can help promote more effective two-way discourse between clinicians and patients without significantly lengthening the discharge communication time Critical component of Project RED When linked very clear prioritization re key information, it is a very promising practice to prevent re-hospitalization

Polling Question (#1) Are your clinicians actively using the “teach- back” method with patients to communicate medication changes during the hospital stay? Who is currently reviewing readmission data on a monthly basis? – A. Yes – all the time (100%) – B. Yes – most of the time (> 50% to < 100%) – C. Yes – sometimes (< 50%) – D. No or rarely 69

Polling Question (#2) Are your clinicians actively using the “teach- back” method with patients to communicate discharge instructions? – A. Yes – all the time (100%) – B. Yes – most of the time (> 50% to < 100%) – C. Yes – sometimes (< 50%) – D. No or rarely 70

Connecting the Dots Hospital to Home Joan Carroll, RN, BA, CDMS, CCM Lee Memorial Health System, FL

72 CTI Program Design Care Transitions Intervention (CTI) is a 4 week program to help patients transition from hospital to home, while learning how to manage their chronic condition Eric Coleman has been a leader in establishing an evidence based model called Care Transitions and this is the model we have chosen to adopt

73 Lee Memorial Health System 4 acute care hospitals in SW Florida Regional trauma Center, regional children’s hospital, inpatient rehab hospital and a full post acute service network # of residents over 65 is 23.5% or nearly double the national average of 13% CHF readmission rate has been 24% Care Transitions readmission rate is 6.9%

74 CTI Program Design Care Transition coach sees the patient in the hospital and completes the first Patient Activation Assessment (PAA). Follow up meeting within 3 days of discharge in the patient’s home to identify medication discrepancies, teach Medication Management, the use of a Personal Health Record, Red Flags and how to manage them, diet and fluid balance, and physician communications. This visit allows us the opportunity to evaluate psych-social needs and connect people to community services. Weekly phone calls for 3 weeks to determine if the patient has seen their PCP and all their questions have been adequately answered. The 2 nd PAA gives us assurance that the patient can self manage their condition.

CTI Lessons Patient Findings 70% of our patients said no one went over discharge instructions Patients may not be honest about their understanding Patients may not be realistic about their abilities Patients do not know anything about sodium Actions Began monthly meetings with discharge nurses/ added highlighters to DI/ shared stories Teach back is mandatory, cognition problems Observe their follow through with scales, meds We worked with the VP of nursing to change DI to include the word salt.

CTI Lessons Patient Findings: Patients do not understand “Activity as tolerated” and often do not know their diagnosis Many patients are not assessed adequately for home going assistance Example, PT, balance programs for frequent fallers /ADLs not IADLs, financial or transportation needs Actions: Quarterly meetings with Hospitalists and asked to change that instruction on DI/ share stories to demonstrate pt confusion Continual meetings with case managers to share stories and increase post acute referrals appropriately

CTI Findings Medication Discrepancies Brand/generic names Pt has given inaccurate list of meds on admission Dosage change not clearly explained New meds not highlighted and described Prescriptions missing Actions Work with pharmacy and have them available to review with patient Have both names on DI and on the bottles Met with nursing. They are working on improved listing of meds on admission. EMR will help with many of these issues

78 Suggestions for CTI Success Form a committee with VP of nursing, CAO, Medical Director, Post Acute administration, Case Management, Process Improvement leaders, Hospitalists, Cardiologist Pharmacy and Home Health Directors. Meet monthly to discuss data collected and share stories. Meet with the nursing staffs and the discharge nurses regularly.

79 Caution Encourage the CTI coaches to share stories so they can be part of the solution rather than the problem. Be careful with the risk assessments. They are only as good as the person doing the questioning. Most patients are not non- compliant. There are real reasons they did not do as they were instructed. You have to ask the right questions.

80 Polling Question (#3) Do you use a formal assessment instrument or tool to evaluate readmission risk? -A. Yes – all patients -B. Yes – only high risk patients -C. Sometimes to rarely, use is inconsistent -D. Not yet

81 Questions? 81 ?

Wrap Up and Next Steps Next TOC Reminder: Visit the HRET HEN website: for information, resources and events, such as the additional topic-specific Data and Coaching webinars throughout June and July. Thank you for joining us! 82