Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Disclosures Improving the Quality and Safety of Patient Care Lessons Learned from Hand Hygiene

• I have nothing to disclose.

W. Matthew Linam, MD, MS

archildrens.org archildrens.org

arpediatrics.org

uams.edu

archildrens.org archildrens.org

arpediatrics.org

uams.edu

Objectives • Understand the basic components of the Model for Improvement • Effectively use data to guide the improvement process • Be able to design and run test a change

WHY DO WE NEED TO IMPROVE?

• List the steps needed to sustain improvement archildrens.org archildrens.org

arpediatrics.org

uams.edu

archildrens.org archildrens.org

arpediatrics.org

uams.edu

Medical Errors Mortality • 44,000-98,000 Americans die each year from medical errors (1999 est.)

Prevalence • 2.9%-3.7% hospitalized patients have an adverse event (annual estimates) • Adverse drug reactions occur during 2.0%-6.7% of hospital admissions (annual estimates)

Cost • Cost attributable to medical errors (2008 est.) is $19.5 billion • Total cost per error (2008 est.) is $13,000 • Annual cost attributable to surgical errors (2008 est.) is $1.5 billion Institute of Medicine Committee on the Quality of Health Care in America . To Err is Human: Building a Safer Health System. November 1999.

archildrens.org archildrens.org

arpediatrics.org

uams.edu

1

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Healthcare--Associated Infections Healthcare Each year in U.S. hospitals alone: • 1.7 million healthcare-associated infections • 99,000 associated deaths

Why do We Fail? “One of the report’s main conclusions is that the majority of medical errors do not result from individual recklessness or the actions of a particular group--this is not a “bad apple” problem. More commonly, errors are caused by faulty systems, processes, and conditions that lead people to make mistakes or fail to prevent them.”

Institute of Medicine Committee on the Quality of Health Care in America . To Err is Human: Building a Safer Health System. November 1999.

www.cdc.gov

archildrens.org archildrens.org

arpediatrics.org

uams.edu

archildrens.org archildrens.org

So Why Hand Hygiene?

The Improvement Model What are we trying to accomplish? How will we know that a change is an improvement?

What changes can we make that will result in improvement?

Act

Study

• Hand hygiene is one of the most important ways to prevent healthcare-associated infections (HAIs) • How we treat a simple safety measure like hand hygiene says a lot about our overall attention to patient safety. • Education and supply availability have not proven to be enough • Complex behavior – – – –

Plan

Do

Self efficacy Behavior of role models and peers Expectations of leadership Long-standing habits

archildrens.org archildrens.org

Evidence Supporting the Effectiveness of Hand Hygiene • Hospitals that have improved the hand hygiene practices of their healthcare workers have: – Decreased their rates of (HAIs) by almost 50% Pittet, D. et al. The Lancet. Vol 356. October 14, 2000: p1307-12.

Lam BC et al. Pediatrics. 2004 Nov;114(5): e565-17.

– Decreased the HAI rate of multi-drug resistant bacteria like MRSA from 0.52 to 0.24 per 1,000 patient-days Lederer J.W. et al. The Joint Commission Journal on Quality and Patient Safety. Vol 35. No 4 April 2009: p180-5.

– Decreased the HAI rate of hospital-associated viral infections such as rotavirus from 5.9 to 2.2 episodes per 1,000 discharged patients Zerr DM et al. PIDJ. 2005 May;24(5): 397-403. archildrens.org archildrens.org

arpediatrics.org

uams.edu

uams.edu

arpediatrics.org

arpediatrics.org

uams.edu

What Actually Happens • Overall median hand hygiene compliance rate: 40% • ICU settings: 40-50% • Non-ICU settings: 50-60% • Physicians: 32% • Nurses: 48% • Before care: 21% • After care: 47%

Erasmus et al. ICHE. March 2010, vol. 31, no. 3. p283-94.

archildrens.org archildrens.org

arpediatrics.org

uams.edu

2

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Defining a Problem

The Improvement Model What are we trying to accomplish? How will we know that a change is an improvement?

If I had only one hour to save the world, I would spend 55 minutes defining the problem and only five minutes finding the solution.

What changes can we make that will result in improvement?

Act

- Albert Einstein

Plan

Study

Do

archildrens.org archildrens.org

Hand Hygiene Improvement Project Design

Defining the Problem

• This was a quasi-experimental study using a staggered intervention design.

• Collect data about the problem – Key outcomes – Processes related to the problem

• Review the literature • Talk with people intimately related to the problem – Frontline staff – Patients and families archildrens.org archildrens.org

arpediatrics.org

uams.edu

Baseline Descriptive Characteristics of Unit A and Unit B Unit A Unit B 26 44 adolescents neurology pulmonary/ cystic fibrosis neurosurgery orthopedic general pediatric psychiatric child abuse Percentage of patients in isolation other 23.8% (n= 1668) 16.8% (n= 2542) than standard precautions Location of sinks Inside each patient door Inside each patient door Inside and outside Inside each patient door Location of alcohol gel dispensers each patient door with 6 spaced along the hallways Characteristic Number of single patient rooms Primary patient population

Location of personal protective equipment (gowns, gloves, and masks)

Multiple cabinets spaced throughout the unit

Multiple cabinets spaced throughout the unit

1.7% (7 / 420)

1.6% (12 / 761)

n = 1668

n = 2553

Physician

15.6%

13.2%

Nurse

49.2%

53.7%

Patient care attendant

23.3%

22.8%

Respiratory therapist

3.6%

Percentage of hand hygiene supplies partially available Healthcare worker type

1. Online module and presentation Hand hygiene improvement interventions on unit 2. Alcohol gel dispensers placed outside patient doors

uams.edu

arpediatrics.org

1.2% 1. Multidimensional improvement 2. Online module

• Interventions were sequentially tested on 2 general pediatric units using the model for improvement. • 12 months of baseline hand hygiene compliance data was obtained prior to interventions. • Measurement of HCW hand hygiene compliance was performed by covert observations made during routine patient care.

archildrens.org archildrens.org

uams.edu

arpediatrics.org

Percentage of Healthcare Worker Hand Hygiene Compliance Both Before and After Patient Care by Unit from November 1, 2007 through November 9, 2008 Unit A Variable Overall Hand hygiene before and after Hand hygiene before care only Hand hygiene after care only Isolation Standard precautions

Percent

Unit B

n

Percent

n

OR

95% CI

P

64.6%

1595

73.8%

2434

0.65

0.56- 0.74

< 0.001

70.7%

1635

78.7%

2487

0.65

0.57- 0.75

< 0.001

76.1%

1626

82.3%

2489

0.68

0.59- 0.80

< 0.001

59.1%

1217

72.5%

2020

0.55

0.47- 0.64

< 0.001

Other isolation Hand hygiene supplies All supplies available

82.5%

378

80.9%

403

1.12

0.78- 1.61

0.552

67.8%

398

71.3%

721

0.85

0.65- 1.11

0.228

Supplies partially available HCW type Physician Nurse Patient care attendant Respiratory therapist

28.6%

7

83.3%

12

0.08

0.005- 1.06

0.045

49.0% 64.5% 71.9%

253 785 370

67.5% 75.5% 76.3%

323 1310 556

0.46 0.59 0.8

0.33- 0.65 0.48- 0.71 0.59- 1.07

< 0.001 < 0.001 0.135

90.9%

55

67.9%

28

4.74

1.41- 15.95

0.008

3

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Statistical Process Control Chart Showing Percentage of Hand Hygiene Compliance Both Before and After Patient Care by Month from HAND HYGIENEOctober COMPLIANCE January 1, 2008 through 31, 2008 on Unit A A6N 100% 90% 80%

60% 50% 40% 30% 20% 10% 0% 12 /3 1 /2 00 7 1/1 4/ 20 08 1/2 8/ 20 08 2/1 1/ 20 08 2/2 5/ 20 08 3/1 0/ 20 08 3/2 4/ 20 08 4/7 /2 00 8 4/2 1/ 20 08 5/5 /2 00 8 5/1 9/ 20 08 6/2 /2 00 8 6/1 6/ 20 08 6/3 0/ 20 08 7/1 4/ 20 08 7/2 8/ 20 08 8/1 1/ 20 08 8/2 5/ 20 08 9/8 /2 00 8 9/2 2/ 20 08 10 /6 /2 00 8 10 /2 0 /2 00 8

Percent Compliant

70%

Week Beginning

Control Limits

Percent

Center Line

Special Cause

Last update: 11-06-08 by M. Linam, Data source: Hand Hygiene Database

Literature Review • HCWs knowledgeable about when, how and why to perform hand hygiene correctly • Hand hygiene supplies consistently available at the point of care • Leadership committed to improving hand hygiene compliance • HCW behavior had to be addressed – Create a change in culture – Change long-standing habits

archildrens.org archildrens.org

Setting Project Goals

S

SPECIFIC

M

MEASURABLE

Lederer J.W. et al. The Joint Commission Journal on Quality and Patient Safety. Vol 35. No 4 April 2009: p180-5. How-to Guide: Improving Hand Hygiene A Guide for Improving Practices among Health Care Workers.

arpediatrics.org

uams.edu

Goals vs. SMART Goals Goal: Improve the hand hygiene compliance of healthcare workers (HCWs) throughout the hospital

A

ATTAINABLE

R

RELEVANT

T

TIME-BOUND

SMART Goal: To improve hand hygiene compliance among HCWs on units A and B from a baseline compliance of 65% and 74%, respectively, to > 90% within 6 months.

archildrens.org archildrens.org

How will we know that a change is an improvement?

What changes can we make that will result in improvement?

uams.edu

Effective Measurement

The Improvement Model What are we trying to accomplish?

arpediatrics.org

• Outcome Measures – Measure the performance of the system under study – Directly relate to the aim of the project

• Process Measures – Measure whether the parts/steps in the system performing as planned

• Balancing Measures Act

Study

Plan

Do

– Measure whether changes designed to improve one part of the system cause problems in other parts of the system archildrens.org archildrens.org

arpediatrics.org

uams.edu

4

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Measuring Quality Vs. Research

Measuring Quality Vs. Research

Research Vs.

Measurement for Research Measurement for Learning and Process Improvement To discover new knowledge To bring new knowledge into daily practice One large "blind" test Many sequential, observable tests Control for as many biases as Stabilize the biases from test possible to test Gather as much data as Gather "just enough" data to possible, "just in case" learn and complete another cycle Can take long periods of "Small tests of significant time to obtain results changes"

Purpose

Quality

Tests Biases Data

Duration

www.IHI.org

Variation in Healthcare Systems INTENDED VARIATION

COMMON CAUSE

• Part of effective, patient-centered care • One size does not fit all • Often called purposeful, planned, guided or considered UNINTENDED VARIATION • Due to changes introduced into healthcare process that are not purposeful, planned or guided • Creates inefficiencies, waste, rework, ineffective care, errors and injuries in our system

archildrens.org archildrens.org

arpediatrics.org

uams.edu

Hand Hygiene Compliance • Performance of acceptable hand hygiene both BEFORE and AFTER encountering the patient or the patient’s immediate care environment (bed, overthe-bed table, IV pump or pole, etc.).

archildrens.org archildrens.org

Sources of Variation

arpediatrics.org

uams.edu

• • • • •

Inherent in the process over time Affect all outcomes of the process Seen by chance Stable Process Process is in statistical control

SPECIAL CAUSE • • • •

Not part of the process, arise from specific problems Assignable Unstable Process Langley GL, et al. The Improvement Guide: A Practical Approach to Enhancing Organizational Process is not in statistical control Performance (2nd Edition); 2009

archildrens.org archildrens.org

arpediatrics.org

uams.edu

Acceptable Hand Hygiene • Application of an alcohol hand rub product with complete coverage of hands and fingers • Hand washing with soap followed by turning off the faucet without using fingers or the palm of the hand • For patients on isolation, appropriate personal protective equipment (including gloves) donned upon room entry then removed and hand hygiene performed before exiting the room

archildrens.org archildrens.org

arpediatrics.org

uams.edu

5

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Statistical Process Control Chart Showing Percentage of Hand Hygiene Compliance Both Before and After Patient Care by Month from HAND HYGIENEOctober COMPLIANCE January 1, 2008 through 31, 2008 on Unit A A6N

The Data Collection Process

– Make up to 8 observations during an 8-hour shift – No more than 2 observations on any one HCW during a shift

80% 70% 60% 50% 40% 30% 20%

12 /3 1 /2 00 7 1/1 4/ 20 08 1/2 8/ 20 08 2/1 1/ 20 08 2/2 5/ 20 08 3/1 0/ 20 08 3/2 4/ 20 08 4/7 /2 00 8 4/2 1/ 20 08 5/5 /2 00 8 5/1 9/ 20 08 6/2 /2 00 8 6/1 6/ 20 08 6/3 0/ 20 08 7/1 4/ 20 08 7/2 8/ 20 08 8/1 1/ 20 08 8/2 5/ 20 08 9/8 /2 00 8 9/2 2/ 20 08 10 /6 /2 00 8 10 /2 0 /2 00 8

0%

Week Beginning

Control Limits

– handbook was created – infection control staff met with the data collection team monthly

arpediatrics.org

90%

10%

• Entered into a database by a single person • To ensure consistency of the data collection

archildrens.org archildrens.org

100%

Percent Compliant

• Cohort of 15 patient attendants (12% of the 128 patient attendants) • Patient attendants provide 24-hour bedside observation of patients who have medical or behavioral conditions that could result in self-injury, removal of medical devices, flight or abduction. • Trained by the infection control staff • Hand hygiene observations were recorded on a standard data collection tool

Percent

Center Line

Special Cause

Last update: 11-06-08 by M. Linam, Data source: Hand Hygiene Database

uams.edu

What is a Test?

The Improvement Model What are we trying to accomplish? How will we know that a change is an improvement?

• Putting a change into effect on a temporary basis and learning about its potential impact

What changes can we make that will result in improvement?

Act

Plan

Study

Do

archildrens.org archildrens.org

PDSA Model

Reasons to Test Changes • To increase your belief that the change will result in improvement • To evaluate how much improvement can be expected from the change • To test whether the proposed change will work in the environment of interest • To evaluate costs, social impact, and side effects from a proposed change • To minimize resistance to the change archildrens.org archildrens.org

arpediatrics.org

uams.edu

arpediatrics.org

Act

Study

Plan

Do

uams.edu

6

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Appropriate Scope for a PDSA Cycle

PDSA Pitfalls

Staff Readiness to Make Change Current Situation

Low Confidence that change idea will lead to Improvement

High Confidence that change idea will lead to Improvement

Resistant

Indifferent

Ready

Very Small Scale Test

Very Small Scale Test

Very Small Scale Test

Very Small Scale Test

Very Small Scale Test

Small Scale Test

Cost of failure large

Very Small Scale Test

Small Scale Test

Cost of failure small

Small Scale Test

Cost of failure large Cost of failure small

Large Scale Test

Large Scale Test

• Tests that are miss-sized for the stage of the project (too big or too small) • No prediction or hypothesis before testing the change • Lack of a detailed execution plan • Failure to perform the entire cycle • Failure to use qualitative measures • Not learning from “failures”

Implement

Langley GL, et al. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance (2nd Edition); 2009

Testing vs. Implementing

archildrens.org archildrens.org

uams.edu

arpediatrics.org

Use of PDSA cycles

• Testing: Putting a change in place on a temporary basis

A P

Changes That Result in Improvement

S D

• • • •

Test on a small scale first Test under as many different conditions as possible Learn from successes and failures Adapt the change based on what is learned

• Implementing: Making the change a part of the day-to-day operation of the system archildrens.org archildrens.org

arpediatrics.org

Implementation of Change Wide-Scale Tests of Change

A P

Evidence Best Practice Testable Ideas

S D

Follow-up Tests

Very Small Scale Test

uams.edu

Choosing a Test Does the choice of intervention matter?

• Prevent: Most improvement efforts begin with an intent to follow a uniform standard process or guideline. This can improve reliability to roughly 90%. – Basic standardization – Memory aids such as checklists – Feedback mechanisms regarding compliance with standards – Awareness-raising and training Nolan T, et al. IHI Innovation Series: Improving the Reliability of Health Care. 2004

archildrens.org archildrens.org

arpediatrics.org

uams.edu

archildrens.org archildrens.org

arpediatrics.org

uams.edu

7

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

• Identify and Mitigate: Strategies to reduce the opportunities for humans to make mistakes or “error-proofing” can improve reliability to at least 95%. – – – – –

Building decision aids and reminders into the system Making the desired action the default (based on evidence) Creating intentional redundancy Scheduling key tasks such as discharges Taking advantage of existing habits and patterns Nolan T, et al. IHI Innovation Series: Improving the Reliability of Health Care. 2004

archildrens.org archildrens.org

uams.edu

arpediatrics.org

• Passive interventions, such as increasing availability of alcohol-based hand products: – Did not improve compliance unless they were coupled to interventions that also changed behavior

• Multimodal interventions that focused only on education and supply availability: – Improved hand hygiene compliance to 55%-80%

• Multimodal interventions that addressed HCW behavior: – Improved hand hygiene compliance to 80%-98% archildrens.org archildrens.org

uams.edu

arpediatrics.org

Design Changes Key Drivers/ Processes

• Hand hygiene habits are developed by around 9 years of age • Hand hygiene driven by sense of self-preservation • Our long-standing habits do not match up with what is expected in the healthcare setting • When faced with time constraints HCWs make judgment calls based on the perceived risk • Perceived hand hygiene is consistently better than what is actually measured Whitby M, et al. Infect Control Hosp Epidemiol. 2006;27(5):484-492

archildrens.org archildrens.org

arpediatrics.org

Project Aim

Aim : To show sustainable improvement in hand hygiene compliance among healthcare workers on units A and B from a baseline compliance of 65% and 74%, respectively, to > 90%.

Measure: Performance of acceptable hand hygiene both BEFORE and AFTER encountering the patient or the patient’s immediate care environment (bed, over-thebed table, IV pump or pole, etc.).

1.

Leadership committed to improving hand hygiene compliance

Leadership 1. 2.

Education 1. 2.

2.

HCWs knowledgeable about when, how and why to perform hand hygiene correctly

Hand hygiene supplies consistently available at the point of care

4.

Correct hand hygiene is the social norm

Online hand hygiene learning module Multimodal education using unit education rounds

System Change 1. 2.

3.

Unit leadership committed Unit leaders serve as role models and educators

Reliable stocking and repair process for supplies Placement of additional alcohol hand-rub dispensers outside all patient rooms

Behavior Change 1. 2. 3. 4. 5.

Visibly posting unit hand hygiene data Presentation to increase awareness Reminder signs on patient doors Identification and mitigation of incorrect behavior Facilitate HCWs reminding each other

uams.edu

Annotated Statistical Process Control Charts Showing Percentage of Hand Hygiene Compliance Both Before and After Patient Care by Month from November 1, 2007 through April 30, 2010 on Unit A

Identify and Mitigate • Patient attendants carry a handheld bottle of alcohol-based hand rub • If a HCW enters the room, and forgets to perform hand hygiene the patient attendant kindly offers the HCW hand gel • Initially tested only on unit A • Due to its success it was quickly spread hospital-wide • Observations were recorded based on the HCWs intended behavior archildrens.org archildrens.org

arpediatrics.org

uams.edu

8

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Annotated Statistical Process Control Charts Showing Percentage of Hand Hygiene Compliance Both Before and After Patient Care by Month from November 1, 2007 through April 30, 2010 on Unit B

Barriers and Keys to Success • Barriers – Trying to overcome long-standing habits – Lack of belief their hand hygiene was poor

• Keys to success – – – –

Support from the unit leadership was crucial Interventions built upon existing processes Worked to increase awareness and urgency The identify and mitigate process was able to improve hand hygiene compliance to 90% or greater across multiple HCW types

archildrens.org archildrens.org

Implementing a Successful Change • Implementation is a permanent change to the way work is done • If the change is not built into the process improvement often erodes over time • Key components necessary for sustainability – Process owner – Ongoing monitoring process – Redesign of the system

archildrens.org archildrens.org

arpediatrics.org

arpediatrics.org

uams.edu

Designing for Sustainability • Process owner: – who is going to oversee the new process? – Who is going to monitor the data collection?

• Monitoring system: – Integrate the data collection plan to monitor the process and continued improvement – Have a process in place to address special causes and trends in the data

uams.edu

archildrens.org archildrens.org

arpediatrics.org

uams.edu

Designing for Sustainability • Redesigning the system: – Update key stakeholders – Create a revised process map – Update policies and procedures – Redesign job responsibilities – Training employees in the new procedures and use of new equipment

archildrens.org archildrens.org

arpediatrics.org

uams.edu

SPREADING IMPROVEMENT

archildrens.org archildrens.org

arpediatrics.org

uams.edu

9

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Arkansas Children’s Hospital Hand Hygiene Improvement Project Goal • To improve hand hygiene compliance for each patient care unit to at least 90% and show sustained improvement for at least 6 months

archildrens.org archildrens.org

arpediatrics.org

uams.edu

Design Changes Key Drivers/ Processes 1.

Leadership committed to improving hand hygiene compliance

2.

HCWs knowledgeable about when, how and why to perform hand hygiene correctly

3.

Hand hygiene supplies consistently available at the point of care

Measure: Performance of acceptable hand hygiene upon ENTERING and LEAVING the patient’s care environment (zone).

4.

Correct hand hygiene is the social norm

1. Senior leadership priority 2. Unit leadership makes hand hygiene a priority 3. Hand hygiene Infomercial

Education and Training

Project Aim To improve hand hygiene compliance for each patient care unit to at least 90% and show sustained improvement for at least 6 months

Leadership

1. 2. 3.

ACH Moments for Hand Hygiene posters Online education module Update new employee orientation

Supply Availability 1. 2.

Reliable restocking process for supplies Placement of alcohol handrub dispensers in the path of care

Behavior Change 1. 2. 3.

Sharing compliance data with staff Immediate positive feedback to staff Resident physician teambased competition

HCW indicates healthcare worker.

The Improvement Model What are we trying to accomplish? How will we know that a change is an improvement?

What changes can we make that will result in improvement?

Act

Study

Plan

WHERE DO WE GO FROM HERE?

Do

archildrens.org archildrens.org

arpediatrics.org

uams.edu

10

Pediatrics Grand Rounds 23 March 2012

University of Texas Health Science Center at San Antonio

Additional References •

Whitby M, et al. Am J Infect Control. 2008;36(5):349-355

• Pittet D, et al. Lancet. 2000;356(9238):1307-1312 • Lam BC, et al. Pediatrics. 2004;114(5):e565-571 •

Lederer JW, et al. Jt Comm J Qual Patient Saf. 2009;35(4):180-185

•

Won SP, et al. Infect Control Hosp Epidemiol. 2004;25(9):742-746

• Zerr DM, et al. Pediatr Infect Dis J. 2005;24(5):397-403

archildrens.org archildrens.org

arpediatrics.org

uams.edu

11