Improving Patient Flow by Managing Variability

Improving Patient Flow by Managing Variability

Eugene Litvak, PhDProgram for Management of Variability

In Health Care Delivery, Boston University

Eugene Litvak, PhDProgram for Management of Variability

In Health Care Delivery, Boston University

2nd Annual Ellison Pierce Symposium Positioning Your ORs For The Future

Boston University School of Medicine

May 19, 2006

Boston University School of Medicine

May 19, 20068:00-8:30am8:00-8:30am

20%

10%

20%

20%

1 2 3 4

What do you think is the largest source of your hospital’s census variability? What do you think is the largest source of your hospital’s census variability?

1. the emergency room

2. the elective OR schedule

3. the ED and elective OR schedules impact is equal

4. No idea

1. the emergency room

2. the elective OR schedule

3. the ED and elective OR schedules impact is equal

4. No idea

QUESTION:QUESTION:

50%

30%

20%

1 2 3

Does your hospital try to smooth scheduled patient flow? Does your hospital try to smooth scheduled patient flow?

1. Yes

2. No

3. Don’t know

1. Yes

2. No

3. Don’t know

QUESTION:QUESTION:

Why should we smooth scheduled patient flow?

Can we afford not to smooth scheduled patient flow?

Why should we smooth scheduled patient flow?

Can we afford not to smooth scheduled patient flow?

How unsmooth census looks like?How unsmooth census looks like?

Time

# of

Pat

ien

ts

Time

# of

Pat

ien

ts

Systemic Effects of Peak LoadsSystemic Effects of Peak Loads

• Internal Divert –Patients sent to alternative floors\Intensive Care locations

• Internal Delays – PACU backs up

• External Divert - ED divert

• Staff overload – medical errors and inability to retain staff

• System Gridlock – Increase in LOS

• Decreased throughput and revenue

• Internal Divert –Patients sent to alternative floors\Intensive Care locations

• Internal Delays – PACU backs up

• External Divert - ED divert

• Staff overload – medical errors and inability to retain staff

• System Gridlock – Increase in LOS

• Decreased throughput and revenue

Litvak E. & Long MC. Cost and Quality Under Managed Care: Irreconcilable Differences? American Journal of Managed Care, 2000; 6 (3): 305-312.

http://www.ajmc.com/files/articlefiles/AJMC2000MarLitvak305_312.pdf

Litvak E. "Optimizing patient flow by managing its variability". In Berman S. (ed.): Front Office to Front Line: Essential Issues for Health Care Leaders. Oakbrook Terrace, IL: Joint Commission Resources, 2005, pp. 91-111.

The Ideal Healthcare System (100% efficiency)

The Ideal Healthcare System (100% efficiency)

1. All patients have the same disease with the same severity.

2. All patients arrive at the same rate.

3. All providers (physicians, nurses) are equal in their ability to provide quality care.

1. All patients have the same disease with the same severity.

2. All patients arrive at the same rate.

3. All providers (physicians, nurses) are equal in their ability to provide quality care.

Variability as the source of system stress

Variability as the source of system stress

1. Clinical stress.

2. Patient flow stress.

3. Stress by variaton in proffesional abilities or teaching responsibilities.

1. Clinical stress.

2. Patient flow stress.

3. Stress by variaton in proffesional abilities or teaching responsibilities.

Natural Variability

Natural Variability

I) Clinical Variability

II) Flow Variability

III) Professional Variability

I) Clinical Variability

II) Flow Variability

III) Professional Variability

• Random• Can not be eliminated (or even

reduced)• Must be optimally managed

• Random• Can not be eliminated (or even

reduced)• Must be optimally managed

}

Why managing variability today is more important than

before?

Why managing variability today is more important than

before?

Designing and Testing Complex Mechanical Systems: Family CarDesigning and Testing Complex Mechanical Systems: Family Car

• Hitting a pothole vs. high speed impact against the wall

• Health care “financial bumper”

• Are the stresses an intrinsic part of health care delivery?

• Hitting a pothole vs. high speed impact against the wall

• Health care “financial bumper”

• Are the stresses an intrinsic part of health care delivery?

What makes hospital census variable?

What makes hospital census variable?

Time

# o

f P

ati

ents

-

• If ED cases are 50% of admissions

and…

• Elective-scheduled OR cases are 35% of admissions

then…

• Which would you expect to be the largest source of census variability?

• If ED cases are 50% of admissions

and…

• Elective-scheduled OR cases are 35% of admissions

then…

• Which would you expect to be the largest source of census variability?

What makes hospital census variable?

What makes hospital census variable?

The answer is…The answer is…

The ED and Elective-Scheduled OR have approximately equal effects on census variability.

Why?

Because of another (hidden) type of variability...

The ED and Elective-Scheduled OR have approximately equal effects on census variability.

Why?

Because of another (hidden) type of variability...

Artificial VariabilityArtificial Variability

• Non-random

• Non-predictable (driven by unknown individual priorities)

• Should not be managed, must be identified and eliminated

• Non-random

• Non-predictable (driven by unknown individual priorities)

• Should not be managed, must be identified and eliminated

Variability in the Census - Rising Volume

Variability in the Census - Rising Volume

Time

# of

Pat

ient

s -

Floors

ICU

ED

Variability and access to careVariability and access to care

Scheduled demand

Variability and Quality of Care*Variability and Quality of Care*

Inadequate numbers of nursing staff contribute to 24% of all sentinel events in hospitals.  Inadequate orientation and in-service education of nursing staff are additional contributing factors in over 70% of sentinel events

Inadequate numbers of nursing staff contribute to 24% of all sentinel events in hospitals.  Inadequate orientation and in-service education of nursing staff are additional contributing factors in over 70% of sentinel events

* Dennis S. O’Leary, JCAHO (personal communication)

Source: Carol Haraden, Ph.D., IHI

Variability and mortalityVariability and mortality

“Each additional patient per nurse was associated with a 7% increase in the likelihood of dying within 30 days of admission and a 7% increase in the odds of failure-to-rescue”*

* Linda H. Aiken, Sean P. Clarke, Douglas M. Sloane, Julie Sochalski, and Jeffrey H. Silber. Hospital Nurse Staffing and Patient Mortality, Nurse Burnout, and Job Dissatisfaction. JAMA, 2002; 288: 1987:1993

“Each additional patient per nurse was associated with a 7% increase in the likelihood of dying within 30 days of admission and a 7% increase in the odds of failure-to-rescue”*

* Linda H. Aiken, Sean P. Clarke, Douglas M. Sloane, Julie Sochalski, and Jeffrey H. Silber. Hospital Nurse Staffing and Patient Mortality, Nurse Burnout, and Job Dissatisfaction. JAMA, 2002; 288: 1987:1993

Litvak E, Buerhaus PI, Davidoff F, Long MC, McManus ML, Berwick DM. “Managing Unnecessary Variability in Patient Demand to Reduce Nursing Stress and Improve Patient Safety,” Joint Commission Journal on Quality and Patient Safety, 2005; 31(6): 330-338.

Example:Example:

Assumptions:

• 200 surgical beds

• average census for surgical beds 160

• staffing level 40 nurses (1 nurse per 4 patients)

• average residual from 160 patients census is 20% or 32 patients

• patients are distributed evenly between the nurses

How the mortality rate will change with 20% increase in surgical demand?

Assumptions:

• 200 surgical beds

• average census for surgical beds 160

• staffing level 40 nurses (1 nurse per 4 patients)

• average residual from 160 patients census is 20% or 32 patients

• patients are distributed evenly between the nurses

How the mortality rate will change with 20% increase in surgical demand?

Results:Results:

• 32 additional patients will be distributed evenly between 32 nurses: 1 additional patient per nurse or 4 + 1 = 5 patient per nurse

• these 32 nurses now will take care of 160 patients, whose mortality rate increases by 7%

• if these additional 32 patients will be distributed evenly between 16 nurses, then each such nurse will take care of 4 + 2 = 6 patients

• these 16 nurses now will take care of 96 patients, whose mortality rate increases by 14%

• 32 additional patients will be distributed evenly between 32 nurses: 1 additional patient per nurse or 4 + 1 = 5 patient per nurse

• these 32 nurses now will take care of 160 patients, whose mortality rate increases by 7%

• if these additional 32 patients will be distributed evenly between 16 nurses, then each such nurse will take care of 4 + 2 = 6 patients

• these 16 nurses now will take care of 96 patients, whose mortality rate increases by 14%

Root Cause Analysis of Emergency Department Crowding and Ambulance Diversion in Massachusetts,

Boston University, 2002: ED diversions study under Department of Public Health grant

http://www.mass.gov/dph/dhcq/pdfs/final_report_exec_summary.pdf

When the scheduled demand is significant, there was much stronger correlation between

scheduled admissions and diversions than between ED demand and diversions

Elective Surgical Requests vs Total Refusals

Elective Surgical Requests vs Total Refusals

0

1

2

3

4

5

6

7

8

9

10

elective surgical patients seeking ICU admission patients diverted or rejected from the ICU

Michael L. McManus, M.D., M.P.H.; Michael C. Long, M.D.; Abbot Cooper; James Mandell, M.D.; Donald M. Berwick, MD; Marcello Pagano, Ph.D.; Eugene Litvak, Ph.D. Impact of Variability in Surgical Caseload on Access to Intensive Care Services, Anesthesiology 2003; 98: 1491-1496.

Smoothing elective admissions:Success story

Smoothing elective admissions:Success story

Managing Patient Flow: A Focus on Critical Processes       Managing Patient Flow: A Focus on Critical Processes      

http://store.trihost.com/jcaho/product.asp?dept%5Fid=34&catalog%5Fitem=712http://store.trihost.com/jcaho/product.asp?dept%5Fid=34&catalog%5Fitem=712

St. John’s Hospital (OR) St. John’s Hospital (OR)

•Increased surgical annual case volume by 33% in the last three years.

•Increased personal surgical revenue by 4.6%

•OR overtime is record low 2.9%

•Reduced waiting time for available OR by 45%

•Dramatically improved OR nurse retention

•Increased ED throughput by ≈ 60% with no patient boarding

•Increased surgical annual case volume by 33% in the last three years.

•Increased personal surgical revenue by 4.6%

•OR overtime is record low 2.9%

•Reduced waiting time for available OR by 45%

•Dramatically improved OR nurse retention

•Increased ED throughput by ≈ 60% with no patient boarding