2

Quality Council 2017

OR Anesthesia

Preoperative clinic/CCT

Critical Care

Echocardiography/Pacemaker service

Acute pain service/Regional

Comprehensive Pain Medicine

Center

UCLA Community ASCs

Community Pain Centers

Informatics and Analytics Division

Remote/ Tele Pre-op

West LACommunity

Anesthesiology & Perioperative Medicine: Clinical Services

MLK HospitalOlive View

VA Medical Center

Main Strategies and Operations• Perioperative Care Redesign

• Bring Expertise for Patients Safety, Efficiency, Quality, IT, Metrics from the Intraoperative Period to the Perioperative Period

• Develop a Rationale and Systematic Implementation of Quality Improvements Processes

• Main focuses on:• 1. Developing the TEAM culture and multidisciplinary projects with Departments across

the Healthcare System

• 2. Care coordination of the surgical patients

• 3. Leveraging Technologies to Improve Quality of Care

4

REDUCING RISK-ADJUSTED

MORTALITY

••Complex Care Team

••Rapid Response

IMPROVING PROCESS AND

OUTCOMESMEASURES

••PEPC

••Periop Pathways

••Enhanced Recovery After

Surgery

IMPLEMENTING VALUE-BASED

REDESIGN

••CCJR••Enhanced Recovery After

Surgery••Cataract

Surgery: Preopredesign

••Pre-habilitation Programs

ENHANCING THE PATIENT EXPERIENCE

••PEPC

••Telemedicine

••Regional Service

REDUCING PREVENTALBE READMISSIONS

• CCJR

• Enhanced Recovery After

Surgery

STRENGTHENINGPATIENT SAFETY

••eARS System

••Peer Support

••Safe Handoff

Anesthesia Quality Council

Quality Improvement & Innovation Team Initiatives

Slide adapted from UCLA MOVERS strategy

Clinical Operations

MD Champions:

N. Kamdar; A Edwards A Dhillon

MD Champions:

N. Kamdar

MD Champions:

M. CannessonV. Duval

MD Champions:

S. Rahman M. Ferrante

V. Duval

MD Champions:

V. Duval A. Dhillon

MD Champions:

K. Kuchta,E. Methangkool

Alignment with Nursing & Interdisciplinary teams

Surgical Service Partnerships ValU Team, QMS, Performance Excellence

Data Analytics (Bioinformatics, Hospital QIA, Decision Support)Division Chiefs

Site Directors

Perioperative Models of Care

6

Our Model for the Search of Value

7

8

Differentiating Complexity to Maximize Throughput

9

Mortality ReductionOutcome Improvement

Value-Based Care

CCT

Complex Care Team10

Dr. Nirav Kamdar Dr. Alex Edwards

11

UCLA Co-management by Complex Care Team

0

2

4

6

8

10

LOS

Traditional care

Complex Care Team

190 ASA 3/4 patients managed by the Complex Care Team with-

Decrease in length of stay (LOS) from 9.2 days to 5.4 days (56% reduction)

The average case delay was reduced by 28%.NO same day of surgery cancellations.

Day

s

12

Focus on Complexity in and out of the OR

13

Outcome: 18% De-escalation from ICU Care

14

Outcome: AKI is our largest complication

15

PEPC

Preoperative Evaluation and Planning Center

16

Evolution of Preoperative Evaluation

Gather and summarize all available and relevant

informationOptimize OR throughput

Identify high risk patients

Coordinate preoperative process

• Recommendations for perioperative care• Reduce silo-driven care• Early involvement of operative and perioperative

teams

Understand the patient’s goals, needs, values, lifestyle, and make their health care work within that framework

Prepare patients for surgery• Allergy de-labeling• OSA Dx & Rx• Smoking cessation • Obesity Management

Improve overall health• Quality of life• Behaviors• Medical conditionsNe

xtPh

ase:

Pr

ehab

ilita

tion

Rec

ent

Prog

ress

Orig

inal

D

esig

n

17

Can

cella

tion

rate

(%)

Outcome: PEPC reduces Surgical Cancellations

Average: 2.8%

Average: 6.4%

18

SMART Screen: Maximizing our LEAN Process

• Algorithms implemented as pilot on MP200 GI Suite • Pilot launch in SEI – Nov 2017

• Identify those patients likely to benefit from MOC screening based on

18

Decision Support Screening CriteriaDiabetes ESRD

CAD Elevated BMICHF Previous ASA Score >3

Fewer than 2 previous visits to UCLA

AUTOMATEDASSESSMENTOFPATIENT'SREVISEDCARDIACRISKINDEXUSINGALGORITHMICSOFTWAREHoferI,ChengD,FujimotoY,Cannesson M,MahajanA

SMART Screen: Improves our Bandwidth

19

20

CJR

Comprehensive Care for Joint Replacement

Dr. Neesa Patel Dr. Natale Naim

21

CCJR: Protocols for Predictable Outcomes

21

Pre-Op Intra-Op Post-Op

TelemedicineConsult

• History Gathering

• Expectation Setting & Education

• Regional Consent

Regional Anesthesia

• Adductor Canal Block

• Spinal Anesthetic

Improved Recovery

• mLOS = 2.66

• Time to mobility

22

CCJR: Focus on LEAN operations using analytics

Outcome: Reducing LOS within 1 Quarter

23

2.4

2.5

2.6

2.7

2.8

2.9

3

3.1

3.2

Sep Aug Oct Nov

LOS

Data courtesy of

24

Enhanced Recovery After Surgery Collaborative

Dr. Aviva RegevCarol Lee, RN-BC

Dr. Siamak Rahman

25

Special thanks to our multispecialty partnerships Dr. Lin, Sack, Kazanjian, Hallie Chung, RN

Dr, Litwin, Chamie, SaigalDr. Cohen

NursingPharmacy

376 patientstreated by our ERAS

teams (Nov 2017)

Outcome: Reduced LOS, Opioids & PONV

26

Post ERASJune 2016- July 2017

7.5%Reduction in LOS in 1 year

for Non-IBD ERAS-Colorectal patients

PONV (Median)

33%Reduction in

27

Patient ExperiencePatient Satisfaction

Wellness Bundle : Health System Alignment

• Comfort• Opioid sparing strategies• Early recognition of high risk patients & interventions by pain management services

• Nutrition• Optimal fasting times• Pre-op Carb loading beverage • Gum chewing/sham feeding

• Mobility• Adequate pain management to promote early ambulation

28

Sample department wide campaign

Telemedicine: Growing numbers and satisfaction

29

“Convenient and saved me 120 miles of driving. It beats driving down [to Westwood] in traffic.”

30

Pain ManagementMultimodal Analgesia

& Opioid Sparing Techniques

Dr. Neesa Patel Dr. Siamak Rahman

31

020406080

100120140160

4/20

135/

2013

6/20

137/

2013

8/20

139/

2013

10/2

013

11/2

013

12/2

013

1/20

142/

2014

3/20

144/

2014

5/20

146/

2014

7/20

148/

2014

9/20

1410

/201

411

/201

412

/201

41/

2015

2/20

153/

2015

4/20

155/

2015

6/20

157/

2015

8/20

159/

2015

10/2

015

11/2

015

12/2

015

1/20

162/

2016

3/20

164/

2016

5/20

166/

2016

7/20

168/

2016

9/20

1610

/201

611

/201

6

Orthopedic ServiceBlocks Placed Per Month at SMH

00.5

11.5

22.5

33.5

4/20

135/

2013

6/20

137/

2013

8/20

139/

2013

10/2

013

11/2

013

12/2

013

1/20

142/

2014

3/20

144/

2014

5/20

146/

2014

7/20

148/

2014

9/20

1410

/201

411

/201

412

/201

41/

2015

2/20

153/

2015

4/20

155/

2015

6/20

157/

2015

8/20

159/

2015

10/2

015

11/2

015

12/2

015

1/20

162/

2016

3/20

164/

2016

5/20

166/

2016

7/20

168/

2016

9/20

1610

/201

611

/201

6

Aver

age

LOS

in d

ays

Average LOS in Days for All Orthopedic Patients at SMH

Ortho Patients With Blocks Linear (Ortho Patients With Blocks)

Regional Block and Ortho LOS at SM93 % of patients extremely satisfied with post op pain

management

> 400 ambulatory nerve catheters in 2016

32

Total Blocks – SM OR

*

* Blocks calculated until 11/21/17

33

Patient SafetyData Driven Improvement…

A Focus on Patient and Provider

eARS: Improving patient safety reporting

• Mandatory Reporting for Quality Improvement & Safety• External to Epic to comply with confidentiality

34

eARS: Data driven patient safety impact

35

Monthly Average~2% incidence of reported events

141 reported QI events 121 cases with events

Peer Support – Another Anesthesia First

Peer Support Project

conception collaboration

November 2017

VolunteerPeer Supporter

Training

December 2017

Peer Support and Burnout Survey

December 18th

Official Launch Peer Support Program

Pilot

2018: UC Health

SystemwidePeer Support

Program

20152017

20172017

2018

37

InformaticsProviding Operations Analysis to our Collaborators

Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.1880 www.anesthesia-analgesia.org June 2016 ڇ Volume 122 ڇ Number 6

The 2009 American Recovery and Reinvestment Act cre-ated “meaningful use criteria” for the adoption and implementation of electronic medical records (EMRs)1

assuming that increased adoption of EMRs would improve the quality of care and reduce costs.2,3 Unfortunately, these cost savings have yet to be realized,4 and some have found that EMRs have paradoxically increased the cost of care by allowing improved billing capture.5,6 One barrier to improved care quality and associated cost savings is the difficulties asso-ciated with turning EMR data into actionable information that can be used to improve health care delivery and outcomes.7

The transition from volume-based payments to value-based payments encouraged by the Affordable Care Act as a way to realize these savings requires consistent and reliable extraction of data from the EMR for both measurement and use in quality improvement programs.8 The Perioperative Surgical Home,9 the American Society of Anesthesiologists’ implementation of an accountable care organization, specifically targets getting these data10 through the early use of a data registry as part of its rollout. Despite the urgency of this need, as well as signifi-cant effort, these data remain difficult to obtain.

Currently, EPIC Systems’ EMR is the largest EMR plat-form, with more than half of the US population now having

a patient record in an EPIC system.11 Although there has been some success in extracting data into uniform data models from other systems,12 EPIC EMRs have been par-ticularly challenging, given their expansiveness and large number of tables (>15,000).

A well-established method to combine disparate, and often unstructured, data in such a way as to make it more easily accessible to the end users is to create a data warehouse (Appendix 1). We present our experience and describe the methodology for successfully extracting clini-cal data around the entire perioperative period from our EPIC EMR (Epic Systems, Verona, WI) into an indepen-dently designed data warehouse designed for business intelligence that simplifies access to the data and standard-izes definitions so as to allow multiple groups to report off of the same data.

METHODSAfter obtaining exemption from informed consent from the University of California, Los Angeles IRB, a review of clinical and operational metrics desired for reporting was undertaken. The necessary raw clinical data were located in Clarity, the relational database created by EPIC for data analytics and reporting. Given the array of metrics needed and the complexity of the data structure in Clarity, a 2-stage data warehouse was constructed to reduce the need to join and optimize multiple tables.

The first stage, termed “Base Tables,” was designed to serve as a middle layer decreasing the number of tables and simplifying the joins between them. Conceptually, the tables coalesced into 3 groups: (1) patient-centered information (laboratories, allergies, medial history, etc.); (2) encounter- centered information (admission, discharge, and transfer [ADT], orders, notes, laboratories, etc.); and (3) operative procedure-centered information (staffing, scheduling, room times, etc.). Tables are joined by 1 of 3 of the following fields: (1) a patient identifier (pat_id); (2) a case identifier (case_id); or (3) an encounter identifier (enc_id). In creating these tables,

Copyright © 2016 International Anesthesia Research SocietyDOI: 10.1213/ANE.0000000000001201

Extraction of data from the electronic medical record is becoming increasingly important for quality improvement initiatives such as the American Society of Anesthesiologists Perioperative Surgical Home. To meet this need, the authors have built a robust and scalable data mart based on their implementation of EPIC containing data from across the perioperative period. The data mart is structured in such a way so as to first simplify the overall EPIC reporting structure into a series of Base Tables and then create several Reporting Schemas each around a specific concept (operating room cases, obstetrics, hospital admission, etc.), which contain all of the data required for reporting on various metrics. This structure allows centralized definitions with simplified reporting by a large number of individuals who access only the Reporting Schemas. In creating the database, the authors were able to significantly reduce the number of required table identifiers from >10 to 3, as well as to correct errors in linkages affecting up to 18.4% of cases. In addition, the data mart greatly simplified the code required to extract data, making the data accessible to individuals who lacked a strong coding background. Overall, this infrastruc-ture represents a scalable way to successfully report on perioperative EPIC data while standard-izing the definitions and improving access for end users. (Anesth Analg 2016;122:1880–4)

From the Departments of *Anesthesiology and Perioperative Medicine and †Medicine, David Geffen School of Medicine at UCLA, Los Angeles, Califor-nia; and ‡Office of Health Informatics and Analytics, David Geffen School of Medicine at UCLA, Los Angeles, California.Accepted for publication December 23, 2015.Funding: None.The authors declare no conflicts of interest.Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (www.anesthesia-analgesia.org).Reprints will not be available from the authors.Address correspondence to Ira S. Hofer, MD, Department of Anesthesiology and Perioperative Medicine, David Geffen School of Medicine at UCLA, 757 Westwood Blvd., Los Angeles, CA 90095. Address e-mail to ihofer@mednet.ucla.edu.

A Systematic Approach to Creation of a Perioperative Data WarehouseIra S. Hofer, MD,* Eilon Gabel, MD, MS,* Michael Pfeffer, MD,† Mohammed Mahbouba, MD, MS,‡ and Aman Mahajan, MD, PhD*

E TECHNICAL COMMUNICATION

Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.June 2016 ڇ Volume 122 ڇ Number 6 www.anesthesia-analgesia.org 1883

Creation of a Perioperative Data Warehouse

The increasing use of data to drive organizational deci-sions has created the need to develop tools allowing those with less technical knowledge access to increasingly sophis-ticated data. Traditional methods or having a report writer create separate reports for each business case can be quite cumbersome, has long turnaround times, and requires a strong background in database query and design. The goal of this data warehouse was to make these data accessible to those who might lack these skills.

The metrics contained in the Reporting Schemas were dictated by organizational needs. Once a metric was defined, it was reported on for all operative cases, not just those in a specific cohort of cases (as would be the case with traditional independent queries for each report). Instead, the scope of the report is limited by the report writer using their analytics soft-ware. This results in an ever increasing library of metrics for reporting and a dramatic decrease in the technical skill needed to generate reports, as seen in Table  2 and Supplemental Digital Content 2 (http://links.lww.com/AA/B369).

The limitations of creating this data mart reflect the underlying complexity and fluidity of the EPIC data

structure. First, because each EPIC implementation is different, the overall structure and concepts described here can be replicated at another institution, but the detailed code and data validation would need to be developed based on the workflow at that institution. Second, although the final data extraction from the Reporting Schema may be straightforward, implementa-tion of the data mart requires personnel who have the technical ability to build a database and enough clini-cal knowledge to help drive the metric creation and data validation. These resources may be beyond the scope of smaller institutions.

The rapid adoption of EMRs over the past 5 years, combined with the transition to a value-based model of care, has resulted in a rapidly growing need to improve extraction of data from the EMR. The data model pre-sented here has provided the authors and their institu-tion with a dramatically improved ability to rapidly find and report on all aspects of the perioperative encounter, and, thus, it is a valuable tool in improving perioperative patient care. E

Figure 2. A visual depiction at the tables involved in calculating case cancellation data at each state of the data warehouse.

The Perioperative Data Warehouse (PDW)

PDW Reporting• Nearly 800 highly validated metrics

• LOS• OR Volume• Readmission• Clinical Outcomes

• Renal Failure

Dashboards: Individual insights – Systemic Success

40