Using Clinical Information Systems to Support Population Based Care

Paul Wallace MD

Care Management Institute

Kaiser Permanente

May 12, 2003

2

3

Managing Chronic Conditions: Delivering Health

• Process and experience oriented

• Local and tribal

• Clinician treating patients and curing acute conditions

• Access: to Clinicians and Visits

• Knowledge Management — Paper and Recall

• Outcome and knowledge oriented

• National and global

• Teams — including members — managing chronic conditions

• Access: to what you need, whenever you need it

• Knowledge Management — Electrons and Judgment

Healthcare Health

4

Systems of Care: Is there Evidence that They Work? Questions to be addressed...

What can you get by leveraging scale to acquire systems?– Types of system support– Impact on performance

Are systems affordable?

5

Kaiser Permanente: National Scope

8.4 million members Membership growth 2002: ~120,000 Services offered in California, Colorado, Georgia, Hawaii,

Maryland, Ohio, Oregon, Virginia, Washington, and the District of Columbia.

29 medical centers 423 medical offices (315 owned, 108 leased) 123,000 Kaiser HealthPlan and Hospitals employees 11,345 Permanente physicians (not including affiliated

networks) ~$21 billion 2002 operating revenues

6

Care Management Institute Strategy

Identify the right thing Promote and embed evidence-based medicine within systems

to support practice Leverage measurement to guide performance improvement Implement better and best practices Deliver member centered and culturally competent care

Make the right thing easier Hire and support people to lead and engage in local

implementation work Identify, develop, and implement effective and innovative

models of care Leverage technology to support population-based care

7

Incremental Cost of Chronic Conditions

“Incremental costs” are the annual “extra” healthcare costs per member with the chronic condition, over and above the healthcare costs of a member who does not have the chronic condition.

Est. mbrs. Total Estimated2001 affected Incremental Cost

prevalence 2002 ($ millions 2002)

Asthma 2.1% 142,654 $2,468 352.1

CAD 2.7% 183,040 $10,264 1,878.7

Depression 4.1% 365,267 $1,966 718.2

Diabetes 6.7% 451,575 $5,035 2,273.5Heart Failure 1.2% 83,646 $16,503 1,380.4

Total Incremental Cost of Chronic Conditions in "CMI Portfolio": 6,603.0

Estimated Incremental Cost($/year/member)

8

What can you get by leveraging scale to acquire systems? Types of system support Registries and databases Clinician support

Knowledge base access Alerts and reminders Rosters

Patient support Knowledge base access Alerts and reminders

System evaluation and improvement Outcomes reports and benchmarking Performance improvement Value demonstration

9

What can you get by leveraging scale to acquire systems? Types of system support Registries and databases Clinician support

Knowledge base access Alerts and reminders Rosters

Patient support Knowledge base access Alerts and reminders

System evaluation and improvement Outcomes reports and benchmarking Performance improvement Value demonstration

10

Registries: How many patients with diabetes are there? And, who are they?

A B C D E F G H I KP

11

The increasing prevalence of diabetes...

Maturation of the diabetes cohort (additional years to identify members)

National changes in biochemical thresholds for diagnosis of diabetes

Expanded inclusion criteria for identification of members with diabetes

Aging of the membership (older members have a higher likelihood of having or developing diabetes)

True increases in the prevalence of diabetes among KP membership.

12

Age Related Prevalence of Chronic Conditions Within Kaiser Permanente

02468

101214161820

18-44 45-54 55-64 65-74 75-84 85+

Age

Per

cen

tag

e o

f co

nti

nu

ou

sly

enro

lled

mem

ber

s in

ag

e g

rou

p

wit

h c

on

dit

ion

CAD

HF

Diabetes

Depression

13

Co-morbidities are common

14

Hospital Day Rates Among KP Members, 2001

500

1000

1500

2000

2500

Day

s p

er 1

000

mem

ber

s

Among KP Members with Diabetes

without Depression

Among KP Members with Diabetes

and Depression

Co-morbidities… implications

Among OverallKP Membership

Source: CMI 2002 Diabetes Outcomes Source: CMI 2002 Diabetes Outcomes Report Report

15

Leverage of Registry Information

Facility Performance Feedback

Panel Feedback

A B C D E F G H I J K L M N

16

What can you get by leveraging scale to acquire systems? Types of system support Registries and databases Clinician support

Knowledge base access Alerts and reminders Rosters

Patient support Knowledge base access Alerts and reminders

System evaluation and improvement Outcomes reports and benchmarking Performance improvement Value demonstration

17

Information-Seeking Activities of Kaiser Permanente CliniciansHolly Jimison, PhD, Mina Monroe, MPH, Mary-Anna Rae, PhD ABD, Aaron Snyder, MD

Methodology

Observed clinical day for 20 KP clinicians– 4 Regions (CO, NW, NC, SC)– 1 Physician consultant in each region– 1 expert computer user in each region– 1 Physician novice computer user in each region– 1 Physician Assistant or Nurse Practitioner in each region– 1 Medical specialist in each region

Data – Ethnographic Notes– Pictures– Artifacts

18

Desk Close-Up

19

Stickies wereubiquitous

Labels with patient Information and pre-visit summaries are also used as reminders

20

Need for Information at the Point of Care

Clinicianscarry

frequently used

informationresources with them

Information systems goal: Empty the lab coat pocket

21

Why Were Questions Unanswered?

Features of the environment

Time Memory (short-term

and long-term) Pervasive irritation

with handling information

Risk-aversion (economics of time invested)

Barriers Design

Incomplete Data Incomplete Access difficulties Time issues Communications

22

23

The Computer in the Exam Room...

24

Member Summary Print-out Sheet

25

Member Summary Print-out Sheet…cont’d

26

27

28

Leveraging IT to Deliver Evidence Based Medicine

The allure... If you make it easier to do the right thing, the right thing

will probably happen

The additional implications… If you do make it easier, it had better be the right

thing...

29

Knee

Knee A+PKnee Series

30

If you do make it easier, it had better be the right thing...

Knee AP/Lat Knee Series

Knee Complete Knee Limited

Before... After...

31

What can you get by leveraging scale to acquire systems? Types of system support Registries and databases Clinician support

Knowledge base access Alerts and reminders Rosters

Patient support Knowledge base access Alerts and reminders

System evaluation and improvement Outcomes reports and benchmarking Performance improvement Value demonstration

32

Internet Tools: Self-Care

33

34

35

36

38

Automated Telephonic Outreach

000000_02050919332

4.wav

39

0%

20%

40%

60%

80%

100%

Current Smoker Taking Aspirin(among

members withDiabetes and

CVD)

Good toExcellent

Overall HealthStatus

Satisfactionwith Overall

Quality of Care

1997

2001

Among 6,000 randomly sampled KP adult members Among 6,000 randomly sampled KP adult members with diabetes with diabetes

Perc

en

tag

e o

f P

erc

en

tag

e o

f R

esp

on

den

tsR

esp

on

den

ts

Self-Reported Responses Self-Reported Responses Among Among

KP Adult Members with KP Adult Members with Diabetes Diabetes

40

What can you get by leveraging scale to acquire systems? Types of system support Registries and databases Clinician support

Knowledge base access Alerts and reminders Rosters

Patient support Knowledge base access Alerts and reminders

System evaluation and improvement Outcomes reports and benchmarking Performance improvement Value demonstration

41

0%

20%

40%

60%

80%

100%

HbA1c Testing HbA1c Control LDL-C Testing LDL-C Control

1996

1997

1998

1999

2000

2001

Among KP’s 420,000 adult members with Among KP’s 420,000 adult members with diabetes diabetes

1996-2001 Performance 1996-2001 Performance Measures for Measures for

KP Adult Members with KP Adult Members with Diabetes Diabetes

% o

f A

ll M

em

bers

wit

h

% o

f A

ll M

em

bers

wit

h

Dia

bete

s

Dia

bete

s

42

0

5

10

15

20

AMIHospitalization

StrokeHospitalization

MetabolicComplicationsHospitalization

Amputation

1998

1999

2000

2001

Among KP’s 420,000 adult members with Among KP’s 420,000 adult members with diabetes diabetes

Incidence Rates Among Incidence Rates Among KP Adult Members with KP Adult Members with

Diabetes Diabetes R

ate

s p

er

1000 K

P

Rate

s p

er

1000 K

P

Ad

ult

Mem

bers

wit

h

Ad

ult

Mem

bers

wit

h

Dia

bete

s

Dia

bete

s

43

Cost Trends in the Care of Diabetes and other Chronic Conditions

0%

20%

40%

60%

80%

100%

HbA1c Testing HbA1c Control LDL-C Testing LDL-C Control

1996

1997

1998

1999

2000

2001

0.0

1.0

2.0

3.0

4.0

5.0

6.0

1996 1997 1998 1999 2000

Incr

emen

tal c

ost r

atio

No Chronic Condition

CAD

Diabetes

Heart Failure

44

Evidence* indicates that performance improvements to date amongst KP’s 450,000 members with diabetes would be expected to yield: 1,300 fewer MIs1

1,800 fewer other cardiovascular events2

1,100 fewer cases of neuropathy3

1,500 fewer cases of retinopathy3

2,100 fewer cases of nephropathy3

* Including:1 LIPID Study Group NEJM 339:1349-57.2 4S Study Diabetes Care 20(4):614-620.3 DCCT Research Group including Diabetes 46:271-86.

Impact...

45

Summary: Is there Evidence that Systems of Care Work?

What can you get by leveraging scale to acquire systems?– Types of system support– Impact on performance

Are systems affordable?

46

Kaiser Permanente’s investment in the information enabled future...

The first people to cross the quality chasm will be

individuals.

The goal has to be to get entire populations across.