Performance Monitoring and

Dashboards for Hospitalists

Leslie Flores MHA, SFHM
April 29 and 30, 2014
 2

Housekeeping
• Questions?
– Type them into the “Questions” box in the
GoToWebinar panel on the right side of your screen at
any time.
– We will wait and address questions at the end of the
session.
• Copies of the slide set will be available via the
CHMB website at www.chmbinc.com
• For questions, contact Lacey Buquet at
ron@chmbinc.com
 3

Leslie Flores MHA, SFHM

• Former hospital executive in

Southern California
• Partner, Nelson Flores
Hospital Medicine Consultants
• Advisor to the Society of
Hospital Medicine for practice
management issues
 4

Agenda
• Why is it important to have a formal
performance monitoring process?
• What types of metrics should you be
measuring?
• Key data and analysis considerations
• Steps in developing a dashboard
• Sample reports and dashboards
 5

Why Have a Dashboard, Report Card,

Performance Report, etc.?

• Understand how you’re performing

• Reduce variation
• Demonstrate value
• Identify trends
• External comparisons
• Reward good performance
 6

Why Have a Dashboard, Report Card,

Performance Report, etc.?

• To drive change
– Identify areas for improvement
– Hawthorne effect
 7

Suggested Approach

Develop an
Distill key action plan
indicators
Generate into a
and analyze dashboard
Set targets reports

Decide
what to
measure
 8

WHAT TO MEASURE?

Take a Balanced
Approach
 9

Key Hospitalist Performance Domains

Descriptive Metrics

Work Effort and Productivity

Clinical Quality

Resource Management

Service and Satisfaction

Financial
 10

In Reality, There’s Lots of Overlap

Quality

Productivity Resources

Service Financial
 11

Descriptive Metrics
• Not performance per se, but these metrics
inform discussions about performance
– Volume
• Number and types of services
– Acuity
• CMI
• Top diagnoses or DRGs
– Payor mix
12
 13

Work Effort and Productivity

– Shifts worked per physician
• Number and type
– Clinical productivity
• Encounters and wRVUs
• Number of patients seen per shift
– Other work effort
• Committee meetings
• Academic work
• Performance improvement projects
14
Management Reports – RVU Metrics
 16

Quality

• What to measure here is evolving quickly

– Hospital Value-Based Purchasing metrics
• Clinical Process of Care domain
– Heart failure discharge instructions
– Pneumonia initial antibiotic selection
• Patient Experience of Care domain
– Communication with doctors
• Outcome domain
– 30-day O/E mortality (AMI/HF/pneumonia)
 17

Quality
– Readmission rates
• 72-hour
– Did focus on LOS management result in patients being
discharged too early?
• 30-day
– How good are care transitions and post-discharge follow-up?
– Other TJC core measures
• e.g. stroke core measures
18
 19

Quality
• Care transitions measures
– PCP notification of admissions and discharges
– Percent of patients with follow-up appointment
scheduled prior to discharge
– Proportion of discharge summaries dictated or
entered on the date of discharge
– Percent of time the discharge summary
medication list matches that given to the patient
 20

Quality
– Percent of patients with more than one attending
hospitalist
• A measure of physician-patient continuity
– Compliance with order sets and pathways
– PQRS measures
– Percent of required VTE risk assessments
performed on admission
– Percent of diabetes patients managed within
target glucose range
21
 22

Resource Management
– Severity-adjusted ALOS
• Comparison to non-hospitalist peer group, external
peer group (e.g., Premier, Crimson, etc.) or Medicare
GMLOS
– Severity-adjusted average cost per discharge
• Major ancillary categories like imaging, clinical
laboratory and pharmaceutical costs
– Avoidable/denied days as a percent of total days
– Utilization of consultants
 23

Resource Management
• Patient flow variables
– ED admission notification to initial hospitalist order time
– ED admission notification to hospitalist in-person visit
– Time elapsed between ED call/page & hospitalist call-back
– Percent of discharge orders entered before 10:00 a.m.
24
25
 26

Service and Satisfaction

• Citizenship
– Attendance at hospitalist group meetings
– Participation on hospital/medical staff committees
and performance improvement initiatives
– Working extra shifts or otherwise helping out
when needed
• Patient complaints
• Satisfaction surveys
– PCPs, ED physicians, specialists, nursing staff
 27

Financial

• Hospitalist program cost center

– Performance to budget
– Financial support/stipend/loss per FTE
• Revenue cycle performance
– Charge capture rate and/or charge lag
– Total charges and collections by provider
– CPT code utilization
– Average net collections per wRVU
– Days in A/R
– Claim edits, rejection and denial rates
– PQRS performance
 28

Source: Society of Hospital Medicine’s 2012 State of Hospital Medicine Report

 30

Coding Intensity
Operational Reports - E&M Utilization

Andrews, James
Brandon, Kim
Davidson, Tom
Garcia, Fred
Liget, Vicki
Marnet, Stewart
Rodriquez, Mary
Thompson, Ed
Wynn, David
Yasini, Shabar
 32

CPT Distribution
Management Reports – Key Performance Indicators
Operational Reports – Rejections and Denials Analysis
 35

DATA/ANALYSIS CONSIDERATIONS
 36

Understand Your Environment

• Each organization has a unique culture,
goals, priorities, operational habits
– Terminology
– Analytical methods
 37

Understand Data Sources and

Limitations
• Common sources of data
– Hospital ADT, clinical, EHR, and financial systems
– Practice management and revenue cycle software
– Third-party data warehouses
• Premier, Crimson, Truven, UHC, CHMB
– Medicare data
– Third party survey data
• MGMA, AMGA, Sullivan Cotter, ECG, SHM
 38

Understand Data Sources and

Limitations
• Limitations
– Completeness and accuracy of inputs
– Reliability of reporting methodologies
• Attribution issues
– Availability and timeliness
– Sample size
– Sheer volume of data
 39

Decide What Types of Analyses

• Individual vs. group?
• Snapshot vs. trend?
• Comparison to . . .
– Internal peer group? External peer group? Survey
data? Established target?
• Statistical analysis options
– Average vs. median
– Arithmetic mean vs. geometric mean
 40

The Problem of Attribution

• Which hospitalist? Hospitalist or consultant?
• Many metrics are best reported at the group level
– Mortality and readmission rates
• Some metrics best reported by admitting provider
– Initial antibiotic selection for pneumonia
• Some metrics best reported by discharging physician
– HF discharge instructions
• Some practices allocate credit based on the proportion
of days each hospitalist cared for the patient
– Patient satisfaction or LOS
 41

Blinded or Un-blinded?
• Usually best to present performance data
about individual hospitalists un-blinded

– Example:
• Each doctor sees every other doctor’s wRVU reports
with names attached

Note: where attribution is an issue, it’s usually better to blind the data
or report it at the group level
 42

What To Do With All This Information?

• High-level assessment
– Is this a plausible representation?
• What does this information
mean for your practice?
– Opportunities for improvement
– Is the information actionable?
• Distill key metrics into a
dashboard or report card
 43

CREATING YOUR DASHBOARD

 44

Creating Your Dashboard

 45

Steps in Creating Your Dashboard

Choose Dashboard Metrics

Set Performance Targets
Of all the information
available to you, which Design Dashboard Format
few metrics should be Who/what is the
presented in the comparison group? Assign
monthly dashboard? What is the range of How often will the Responsibility
acceptable dashboard be
distributed? Who is responsible for
performance?
How best to show producing source data?
performance against Who is responsible for
targets? preparing and
distributing the
monthly dashboard?
Who is responsible for
following up?
 46

Creating a Dashboard

• Pick a handful of key indicators (10 – 15)

– Important to hospitalists AND stakeholders
– Readily measurable
– Consistently available
– Seen as valid
– Actionable
 47

Creating a Dashboard

• Make it simple, short and attractive

– Show results graphically where possible
• Ensure the dashboard is regularly produced
– Routinely distributed to all hospitalists and key
stakeholders
• “Push” vs. “pull”
 48

Just Do It!
• Precise metrics and format are important –
but the most important thing is to have a
dashboard
– And that it is updated and distributed regularly

• Don’t let uncertainty about metrics and

format paralyze you
– Plan to revise metrics and format periodically
 49

Common Challenges
• Consistent access to meaningful, reliable,
timely data
• Who “owns” dashboard production?
– Manual work to produce the dashboard
• Look for IT solutions
• Ensuring the dashboard serves as a stimulus
to action
– Build in accountability mechanisms
50
51
 52

XYZ Hospitalist Group Page 1 - Productivity

ABC Hospital For the month of: Jan-10
Current Month Encounter-Equivalents vs. Target Total Encounter-Equivalents Trend
250
2,500

230
200
1,916

210

204
2,000
197

189

188
150
1,412
155

144
1,500
100
192

192

192

192

192

192

192

192

100
138
94

88

83
50 1,000

96

96
82

82
35
0 500
0 0 0 0 0 0 0 0 0 0
0
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

Current Month Actual Monthly Target Target Total Enc-Equiv

Current Month wRVUs vs. Target Total wRVUs Trend

450
400 4,000
3,419 3,298
410

404

350 3,500
388

365
360

300
344

3,000
250
2,500
255

200
345

345

345

345

345

345

345

345
203

150 2,000
175
248
168

152

100
145
173

173

1,500
148

148

50
50

1,000
0
500 0 0 0 0 0 0 0 0 0 0
0
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec
Current Month Actual Monthly Target Target Total Enc-Equiv

183 Total EKG interpretations 7.8% % of total encounters 148 Total shifts worked during the month
337 Total stress tests 14.4% % of total encounters 12.9 Average billable encounter-equivalents per shift this month
26 Total bedside procedures 1.1% % of total encounters 11.0 Target billable encunter-equivalents per shift
1,802 Total E&M and other encs 76.7% % of total encounters
2348 Total encounters of all types
 53
XYZ Hospitalist Group Page 2 - Revenue Cycle
ABC Hospital For the month of: Jan-10
Quarterly CPT Code Distribution - Admissions Quarterly CPT Code Distribution - Subsequent Visits
Last Year 26% 57% 17% Last Year 38% 56% 6%
Total This Qtr 18% 46% 37% Total This Qtr 34% 38% 28%
Mark 10% 55% 35% Mark 27% 31% 43%
Lenny 26% 49% 26% Lenny 49% 26% 26%
Kareem 11% 40% 49% Kareem 33% 40% 26%
Jack 44% 32% 24% Jack 29% 40% 31%
Irene 13% 54% 33% Irene 54% 33% 13%
Hank 26% 60% 14% Hank 52% 17% 31%
Geetha 18% 38% 45% Geetha 33% 40% 27%
Freda 19% 33% 48% Freda 33% 48% 19%
Edgar 19% 28% 53% Edgar 28% 53% 19%
Diana 4% 57% 39% Diana 24% 68% 8%
Charlie 15% 15% 69% Charlie 15% 69% 15%
Bruce 22% 64% 14% Bruce 35% 14% 51%
Anne 12% 59% 29% Anne 59% 29% 12%
0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100%
99221 99222 99223 99231 99232 99233

Quarterly CPT Code Distribution - Discharges

Monthly Statistics:
Last Year 76% 24%
Total This Qtr 52% 48% 1.78 Average wRVUs per encounter-equivalent
Mark 38% 62% 1.80 Target wRVUs per encounter-equivalent
Lenny 65% 35%
Kareem 54% 46% 15 Total "No Charge" or un-billed encounters
Jack 47% 53% 0 Target "No Charge" or un-billed encounters
Irene 48% 52%
Hank 21% 79% Quarterly Statistics:
Geetha 63% 37%
Freda 36% 64% Target Actual
Edgar 49% 51% < 10% 16.1% Submitted claims that were rejected
Diana 73% 27% < 2% 1.8% "Clean" claims that were denied
Charlie 85% 15% > 85% 89.0% Denied claims paid upon appeal
Bruce 60% 40%
Anne 81% 19% $48.37 Average net professional fee collections per wRVU
0% 20% 40% 60% 80% 100% $50.00 Target net professional fee collections per wRVU
99238 99239
 54
XYZ Hospitalist Group Page 3 - Quality Indicators
ABC Hospital For the month of: Jan-10
DRG Assurance Query Response Trend
1.28 This month's case mix index 100%
82.0%
74.2% This month's proportion of Medicare patients
80% 64.0%
58.0%
89% Order set usage this month 60% 45.0%
> 95% Target order set usage
40%

86% VTE Risk Assessments Performed on Admission 20%

85% VTE Risk Assessment Target
0%
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec
92% Medication Reconciliation Complete on Discharge
Target > 95% Query Response Rate
> 95% Medication Reconciliation Target
Severity-Adjusted ALOS Trend
Core Measures: 6 5.5

5 4.2
77% "Heart Failure Discharge Instructions" performance 3.8
4 3.6
100% "Heart Failure Discharge Instructions" target
3

0
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

Target < 3.9 Average Length of Stay (Sev. Adj.)

Readmission Rates Trend Severity-Adjusted Cost per Case Trend

20.0% $6,000 $5,216 $5,087
$4,898 $4,630
16.0% $5,000
15.0%
12.6% $4,000
10.0% 9.4% 8.8% $3,000

$2,000
5.0% 1.9% 2.2% 1.6% 1.7%
$1,000
0.0% $0
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

72-Hr Readmissions 30-Day Readmissions Target < 4,249 Average Cost per Disch (Sev. Adj.)
 55
XYZ Hospitalist Group Page 4 - Service Indicators
ABC Hospital For the month of: Jan-10
Percent of Discharge Orders Written by 10A

80.0%
68.0%
70.0% 61.0%
58.0%
60.0% 54.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

Disch Orders by 10A Target 60%

Percent of Discharge Summaries Complete at Discharge

100.0% 88.0% 90.0%

85.0%
80.0% 72.0%

60.0%

40.0%

20.0%

0.0%
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

D/S Complete @ Discharge Target 85%

Press Ganey Patient Satisfaction Scores

80% 4.8 Current Physician Satisfaction Survey score
62% > 4.5 Physician Satisfaction Survey score target
56%
60% 52%
48%

40% 4.4 Current Nursing Satisfaction Survey score

> 4.5 Nursing Satisfaction Survey score target
20%

0% 0 Number of patient complaints this month

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec 0 Patient complaints target
"Physician" Question %tile Rank Target
 56

Source: Measuring Hospitalist Performance: Metrics, Reports and Dashboards, Society of Hospital Medicine 2006
 57

Source: Crimson – a product of The Advisory Board

 58

How Can We Help?

• Hospitalist practice management consultants

• Leslie Flores, MHA and John Nelson, MD
• Helping clients build successful new hospitalist programs and enhance the
effectiveness and value of existing programs since 2004.
• Collectively we’ve worked with more than 300 sites
• Services:
– Start-ups, comprehensive practice assessments, compensation
plans, staffing/scheduling models, integration of APPs, team-
building and leadership development, patient experience
training
 59

How Can We Help?

• Founded in 1999 by physicians
• 25,000 users across 900 healthcare facilities
– 12,000 Hospitalist Users
• Patient encounter platform that increases quality and
revenue by streamlining and automating the following key
areas:
– Care Coordination and Communication
– Quality Enhancement and Cost Reduction
– Coding, Compliance, and Documentation
– Revenue Cycle Management
– Data Analytics and Business Intelligence
 How Can We Help?
60

• Since 1995, serving 4,000+ physicians nationwide

• Comprehensive RCM Solution for Hospitalists

– 11% Average Collections Increase

– 8 Days Decrease in Days Charges in AR (DAR)

– Integrated Electronic Charge Capture Solutions

– Advanced Reporting and Analytics Engine - CURVE

• Consulting, Credentialing and Group Formation

• Systems Integration, Interfaces, Data Conversions

• Coding, Education and Training

• Contact us to arrange for a comparative assessment of your

current RCM Results
• Deliverables include a complete practice Dashboard
 61

Contact Us
Leslie Flores Ron Anderson
Partner Director
Nelson Flores Hospital CHMB Inc.
Medicine Consultants 760-520-1340
760-771-3323 ron@chmbinc.com
leslie.flores@nelsonflores.com www.chmbinc.com
www.nelsonflores.com

Mimi Thornton
Regional Mgr., Southwest
Ingenious Med, Inc.
678-501-6237
mimi.thornton@ingeniousmed.com
www.ingeniousmed.com