plenary i donald berwick

Can Health Care Ever Be Safe?Ideas from the “5 Million Lives Campaign”: 

Can Health Care Ever Be Safe? Ideas from the '5 Million Lives Campaign' Donald M. Berwick, MD, MPP Institute for Healthcare Improvement International Forum on Quality and Safety in Health Care Palau de Congressos de Catalunya Barcelona, Spain: April 18, 2007

Slide2: 

100,000 Lives Campaign Objectives(December 2004 – June 2006): 

100,000 Lives Campaign Objectives (December 2004 – June 2006) Save 100,000 Lives Enroll more than 2,000 hospitals Build a reusable national infrastructure for change Raise the profile of the problem - and our proactive response

Components of a Campaign: 

Components of a Campaign Platform Communications Field Operations Measurement Fund-Raising

The Campaign Platform: 

The Campaign Platform The 100,000 Lives Planks Rapid Response Teams Reliable Care for Acute Myocardial Infarction Medical Reconciliation Prevent Central Line Infections Prevent Ventilator Associated Pneumonias Prevent Surgical Site Infections

Slide6: 

Campaign Field Operations Structure: “Nodes”: 

Campaign Field Operations Structure: 'Nodes' FACILITIES (2000-plus) NODES (approx. 75) *Each Node Chairs 1 Network *30 to 60 Facilities per Network Introduction, expert support/science, ongoing orientation, learning network development, national environment for change Ongoing communication IHI and Campaign Leadership Local recruitment and support of a smaller network through communication/collaboratives Implementation (with roles for each stakeholder in hospital and use of existing spread strategies) Mentor Hospitals

Slide8: 

Rapid Response Results: Henry Ford Hospital: 

Rapid Response Results: Henry Ford Hospital

Rapid Response Results: Benedictine Hospital: 

Rapid Response Results: Benedictine Hospital 43% Reduction

Slide11: 

P andlt; 0.05 MRT Preventable Code Events 73% decrease Rapid Response Results: Cincinnati Children’s Hospital and University of Cincinnati Results at One Year

Slide12: 

Ascension Health Mortality Reduction

The 100,000 Lives Campaign Scorecard: 

The 100,000 Lives Campaign Scorecard An estimated 122,000 lives saved by participating hospitals (through work on the Campaign but also through other improvements and work on complementary initiatives) Over 3,100 hospitals enrolled Over 78% of all discharges Over 78% of all acute care beds Over 85% of participating hospitals sending IHI mortality data

The 100,000 Lives Campaign Scorecard: 

The 100,000 Lives Campaign Scorecard Participation in Campaign Interventions: Rapid Response Teams: 60% AMI Care Reliability: 77% Medication Reconciliation: 73% Surgical Site Infection Bundles: 72% Ventilator Bundles: 67% Central Venous Line Bundles: 65% All Six: 42%

We Aim to Achieve Care That Is…: 

We Aim to Achieve Care That Is… Safe Effective Patient-centered Timely Efficient Equitable

IHI’s “No Needless” List: 

IHI’s 'No Needless' List No needless deaths No needless pain No helplessness No unwanted waiting No waste …for anyone

The Next Campaign: 

The Next Campaign Reduce Harm

The Next Campaign: 

The Next Campaign Reduce Harm… but what do we mean by 'harm?'

Our Definition of Medical Harm: 

Our Definition of Medical Harm Unintended physical injury resulting from or contributed to by medical care (including the absence of indicated medical treatment), that requires additional monitoring, treatment or hospitalization, or that results in death. Such injury is considered harm whether or not it is considered preventable, whether or not it resulted from a medical error, and whether or not it occurred within a hospital.

The Next Campaign: 

The Next Campaign Reduce Harm… but how much harm will we reduce?

Logic Chain: Step 1: 

Logic Chain: Step 1 How Many Admissions per Year in the US? 37 Million Admissions Source: The AHA National Hospital Survey for 2005

Logic Chain: Step 2: 

Logic Chain: Step 2 40 to 50 Patient Injuries per 100 Hospital Admissions Source: IHI 'Global Trigger Tool' Guiding Record Reviews How Often Are Patients Injured by Care?

Slide23: 

Sample Size Injury Rate % Judged Preventable HMPS (1984 data) ~34,000 3.7% 58% 13.6% Utah-Colorado (1992 data) 15,000 2.9% 53% 6.6% % Life Threatening or Fatal Canada AE (2000 data) 3,745 7.5% 36.9% 20.8% (fatal) Australian AE at LDSH 10.2% (?) Australian AE (1992 data) 2,353 16.6% 69.8% 22.3% How Often Are Patients Injured by Care? Review by Brent James

The “NCC – MERP” Framework: 

The 'NCC – MERP' Framework The capacity to cause error Did not reach the patient Did not cause the patient harm Required monitoring to confirm it resulted in no harm to the patient and/or required intervention to preclude harm Required intervention Required hospitalization Permanent patient harm Sustain life Patient’s death Source: Index of the National Coordinating Council for Medication Error and Reporting and Prevention http://www.nccmerp.org/pdf/indexColor2001-06-12.pdf

Category “E”: 

Category 'E' EXAMPLE OF AN 'E' 'An elderly woman was started on antibiotics for a skin infection without taking into consideration she was on an anticoagulant. She got an injection, and that led to a large and painful bleed into her thigh muscle.' Temporary Injury from Care Requiring Intervention

Category “F”: 

Category 'F' EXAMPLE OF AN 'F' 'A retired farmer had a hip replacement. On the second night after the operation he got confused and fell out of the bed, and dislocated his new hip. He was taken back to the operating room for repair and he went home a few days later than originally planned.' Temporary Injury from Care Requiring Initial or Prolonged Hospitalization

Category “G”: 

Category 'G' EXAMPLE OF A 'G' 'A 59 year old man had elective heart bypass surgery. He went home in three days, but he came back to the office with a fever and infection in his chest incision. This required several additional operations and weeks of antibiotics. He was left with a markedly deformed chest.' Injury from Care Leading to Permanent Patient Harm

Category “H”: 

Category 'H' EXAMPLE OF AN 'H' 'A 64 year old lung cancer patient had elective surgery. One hour after the surgery, he was found unresponsive and with a very low blood pressure. He was resuscitated and brought to the operating room to fix a bleeding artery.' Injury from Care Requiring Intervention to Sustain Life

Category “I”: 

Category 'I' EXAMPLE OF AN 'I' 'A 55 year old bus driver needed anticoagulation for atrial fibrillation. Three days after starting, he suffered a massive bleed into his brain – a stroke. He died six days later.' Injury from Care Contributing to or Causing the Patient’s Death

“Global Trigger Tool”: 

'Global Trigger Tool' Finding Injuries of Severity 'E' through 'I'

Slide31: 

Confirmed events: IHI found: Utah-Missouri found: Utah-Missouri false (+): 171 160 72 21 (93.6%) (42.1%) (22.6%) IHI Global Triggers vs Utah-Missouri The Sensitivity of the Trigger Tool Review by Brent James

Why Do We Find So Many? : 

Why Do We Find So Many? 40 to 50 Injuries per 100 Admissions Include Levels 'E' through 'I' Most others start at 'F' Global Trigger Tool increases efficiency of search Do not distinguish 'preventable' from 'non-preventable' given current knowledge Include out-of-hospital events that lead to admission

The Global Trigger Tool at LDS Hospital Review by Brent James: 

The Global Trigger Tool at LDS Hospital Review by Brent James LDS Hospital; 325 patients; October 2004; Seven trained abstractors; all charts independently reviewed twice 35.1% of all admissions had at least one adverse event 9.1% of all hospital admissions resulted from outpatient care-associated adverse events

Logic Chain: Step 3: 

Logic Chain: Step 3 37 Million Admissions X 40 Injuries per 100 Admissions = 15 Million Injuries per Year How Many Injuries in the US?

Logic Chain: Step 4: 

Logic Chain: Step 4 Approximately 3.5 Million If we could replicate best performance across the existing Campaign population, how many injuries might we expect to avoid?

Slide36: 

“The 5 Million Lives Campaign”: 

'The 5 Million Lives Campaign' Lives 'Bettered' Lives 'Unharmed' 'Safer' Lives

The 5 Million Lives Campaign: 

The 5 Million Lives Campaign Campaign Objectives: Avoid five million incidents of harm over 24 months; Enroll more than 4,000 hospitals and their communities; Strengthen the Campaign’s national infrastructure for change, and transform it into a national asset; Raise the profile of the problem (and of hospitals’ proactive response) with a larger, public audience.

Slide39: 

Reduction in Injuries: 

Reduction in Injuries 5 Million Lives 17% 100% 0 Now 1 Year 2 Years Percent Reduction in Injuries Time

Reduction in Injuries: 

Reduction in Injuries 34% 100% 0 Now 1 Year 2 Years Percent Reduction in Injuries 5 Million Lives Time 17%

Slide42: 

Improving Patient Safety at Mayo Clinic (Adverse Events per 1000 Patient Days – All Sites)

“Some Is Not a Number; Soon Is Not A Time”: 

'Some Is Not a Number; Soon Is Not A Time' Five Million Better Lives December 9, 2008 9:00a.m. How?

The Campaign Platform: 

The Campaign Platform The 100,000 Lives Planks Rapid Response Teams Acute Myocardial Infarction Medical Reconciliation Central Line Infections Ventilator Associated Pneumonia Surgical Site Infection The 5 Million Lives Planks Pressure Ulcers Congestive Heart Failure High Alert Medications Surgical Complications ('SCIP') Methicillin-Resistant Staphylococcus aureus 'Boards on Board'

The Campaign Platform: 

The Campaign Platform …plus numerous other interventions that hospitals must introduce in order to contribute to meeting our aim.

5 Million Lives Campaign The “Planks” – Starter Set: 

5 Million Lives Campaign The 'Planks' – Starter Set Prevent Pressure Ulcers

Pressure Ulcers: 

Pressure Ulcers

Burden of Pressure Ulcers: 

Burden of Pressure Ulcers Prevalence in acute care = 15 % Incidence in acute care = 7 % 5-7% of all acute hospital admissions 2.5 million patients treated each year Nearly 60,000 die each year from complications $11 billion dollars per year Sources: How-to-guide andamp; JAMA systematic review by Reddy 2006, referenced a national pressure ulcer Advisory panel (2001) 'Pressure Ulcers in America: Prevalence, Incidence, and Implications for the Future: An Executive Summary Of the National Pressure Ulcer Advisory Panel Monograph

An Example of What Is Possible… St. Vincent’s Medical Center: 

An Example of What Is Possible… St. Vincent’s Medical Center Source: Joint Comisision Journal on Quality and Patient Safety The Clinical Transformation of Ascension Health: Eliminating All Preventable Injuries and Deaths Clinical Excellence Series David B. Pryor, M.D. Sanford F. Tolchin, M.D. Ann Hendrich, M.S., R.N. Clarence S. Thomas, M.D. Anthony R. Tersigni, Ed.D. Decrease of 71%

Reducing Pressure Ulcers: 

Reducing Pressure Ulcers Conduct a Pressure Ulcer Admission Assessment for All Patients Reassess Risk for All Patients Daily Inspect Skin Daily Manage Moisture – Keep the Patient Dry and Moisturize Skin Optimize Nutrition and Hydration Minimize Pressure For High Risk Patients: For All Patients:

5 Million Lives Campaign The “Planks” – Starter Set: 

5 Million Lives Campaign The 'Planks' – Starter Set Prevent Pressure Ulcers Prevent Methicillin-Resistant Staphylococcus aureus (MRSA) Infections

MRSA: 

MRSA

Burden of MRSA Infections: 

Burden of MRSA Infections Over 126,000 infected by MRSA annually 3.95 MRSA cases per 1,000 hospital discharges Over 5,000 patients die as a result of these infections* Over $2.5 billion in excess health care costs For each patient with MRSA infection this means: 9.1 days excess cost per case Over $20,000 excess cost per case 4% excess in-hospital mortality* *These data are based on a multivariate analysis matching for potential cofounders, and apply toall staphylococcal infections, Not just hospital-acquired infections or MRSA infections. However, sub-analysis by the authors suggest that the mortality rate for hospital acquired infections is similar, and other studies document higher mortality for MRSA than MSSA.

Reducing MRSA: 

Reducing MRSA 1. Hand Hygiene 2. Decontamination of the Environment and Equipment 3. Active Surveillance Cultures 4. Contact Precautions for Infected and Colonized Patients 5. Device Bundles (Central Line and Ventilator)

5 Million Lives Campaign The “Planks” – Starter Set: 

5 Million Lives Campaign The 'Planks' – Starter Set Prevent Pressure Ulcers Prevent MRSA Infections Prevent Harm from High-Alert Medications

Burden of High-Alert Medications: 

Burden of High-Alert Medications Anticoagulants Sedatives Narcotics Insulin 40 to 50 Injuries per 100 Admissions Overall 'F' to 'I' - 25% Are Due to Medications 'E' to 'I' – 50% Are Due to Medications 58% of Medication Injuries Are Due to High-Alert Medications!

Reducing Harm from High-Alert Medications: 

Reducing Harm from High-Alert Medications General principles: Prevent Identify Mitigate For each category of high-alert medications: Most effective changes Standardization Monitoring Partnering with patients and families

5 Million Lives Campaign The “Planks” – Starter Set: 

5 Million Lives Campaign The 'Planks' – Starter Set Prevent Pressure Ulcers Prevent MRSA Infections Prevent harm from High-Alert Medications Reduce Surgical Complications – Adopt 'SCIP'

Slide59: 

www.medqic.org/SCIP

Burden of Surgical Complications: 

Burden of Surgical Complications 15,000 patient records from 1992 3,794 operative events occurred =15% 30 million inpatient surgeries in the United States every year 2.5 to 3.5 million surgical patients per year experiencing unintended harm A Conservative Estimate: 8-12% of 30 million Sources: , (Thomas. BMJ. 2000;320:741-744). SCIP Improving Surgical Care

Reducing Surgical Complications : 

Reducing Surgical Complications Surgical Site Infection Prevention Beta Blockers for Patients on Beta Blockers prior to Admission Venous Thromboembolism (VTE) Prophylaxis Ventilator-Associated Pneumonia Prevention Teamwork and Culture

Our Goal: 

Our Goal Reduce surgical complications by 25 percent by December 2008 by reliably implementing the changes in care recommended by SCIP

Proportion of Adverse EventsMost Frequent Categories: 

Proportion of Adverse Events Most Frequent Categories Brennan. N Engl J Med. 1991;324:370-376. Non-surgical Surgical

Impact of SSI: 

Impact of SSI *Pairs matched for procedure, NNIS index, age *General inpatient surgical population; 22, 742 procedures included Kirkland. Infect Control Hosp Epidemiol. 1999;20:725.

Reduce Surgical Site Infections : 

Reduce Surgical Site Infections Appropriate use of antibiotics Appropriate hair removal Postoperative glucose control (major cardiac surgery patients cared for in an ICU)* Postoperative normothermia (colorectal surgery patients)* * These components of care are supported by clinical trials and experimental evidence in the specified populations; they may prove valuable for other surgical patients as well.

Appropriate Antibiotics: 

Appropriate Antibiotics Common Sense Science: Not all surgery is the same. Not all infections are the same. Certain surgeries need certain antibiotics to be given to prevent infection. Antibiotics should be present in the tissue to be operated on at the time incision is made and throughout time the wound is open. It takes a certain amount of antibiotic in order to work

Protocols, Protocols, Protocols: 

Protocols, Protocols, Protocols Design protocols based on surgery type Initiate protocol as a standard Nursing and/or pharmacy drives protocol No reliance on individual physician memory Include guidance for exceptions Beta lactam allergy Use your own formulary to narrow choices Makes protocol easier and saves costs

Appropriate Antibiotics:Timing of the First Dose: 

Appropriate Antibiotics: Timing of the First Dose Identify owners clearly: Who starts it? and Who documents it? Set a narrower performance margin If goal is 0-60, strive for 10-45 A few strays will still be within the goal Take advantage of habits and patterns Dose of antibiotic started when staff 'hit button' to open door to OR – easy to remember

Appropriate Antibiotics:Timing of the First Dose: 

Appropriate Antibiotics: Timing of the First Dose Verify prior to incision time Final check at pre-procedural briefing or time-out Write the dose time on a 'white board' in OR Reliable procedures take coordination between preoperative nursing and anesthesia services.

Appropriate Antibiotics:Discontinuation of Antibiotics: 

Appropriate Antibiotics: Discontinuation of Antibiotics Opt-out vs. Opt-in Discontinuation of antibiotics automatic If doses required, times set by nursing or pharmacy to end within 24 hours Allow options for appropriate clinical exceptions – UTI, fever, etc Review all cases with 'opt out'

What Is Happening Here?: 

What Is Happening Here? Unprecedented Reliability Unprecedented Teamwork Unprecedented Transparency These are transformative principles that reach far beyond the notion of a 'Campaign.'

5 Million Lives Campaign The “Planks” – Starter Set: 

5 Million Lives Campaign The 'Planks' – Starter Set Prevent Pressure Ulcers Prevent MRSA Infections Prevent Harm from High-Alert Medications Reduce Surgical Complications – Adopt 'SCIP' Deliver Reliable, Evidence-Based Care for Congestive Heart Failure

Burden of CHF: 

Burden of CHF 4.9 million Americans have CHF 12-15 millions office visits, 6.5 million hospital days $29.6 billion in 2006 One of the leading causes of re-hospitalization: Readmission Rates: 27% in 30 days 39% in 60 days. 47% in 90 days.

CHF Success : 

CHF Success Source: Fonarow, G. C., Stevenson, L. W., Walden, J. A., Livingston, N. A., Steimle, A. E., Hamilton, M. A., et al. (1997). Impact of a comprehensive heart failure management program on hospital readmission and functional status of patients with advanced heart failure. Journal of the American College of Cardiology, 30 (3), 725–732. Distribution of hospital admissions for the 214 patients in the six months and after comprehensive heart failure management. There were 429 hospital admissions for heart failure in the six months before referral and 63 hospital admissions in the six months after undergoing the management program (p=0.0001). The numbers beneath the months indicate patients at risk for hospital admission during each interval. *This example includes a robust post-hospitalization support program

Reducing CHF Morbidity, Mortality and Readmissions : 

Reducing CHF Morbidity, Mortality and Readmissions Left ventricular (LV) systolic heart function assessment (CMS/JCAHO/ACC/AHA) ACEIs or ARBs at discharge for CHF patients with systolic dysfunction (CMS/JCAHO/ACC/AHA) Anticoagulant at discharge for heart failure patients with chronic/recurrent atrial fibrillation (ACC/AHA) Influenza immunization Pneumococcal immunization Smoking cessation counseling (CMS/JCAHO/ACC/AHA) Discharge instructions that address all of the following: activity level, diet, discharge medications, follow-up appointment, weight monitoring, and what to do if symptoms worsen (CMS/JCAHO/ACC/AHA)

5 Million Lives Campaign The “Planks” – Starter Set: 

5 Million Lives Campaign The 'Planks' – Starter Set Prevent Pressure Ulcers Prevent MRSA Infections Prevent Harm from High Alert Medications Reduce Surgical Complications – Adopt 'SCIP' Deliver Reliable, Evidence-Based Care for Congestive Heart Failure Get Boards on Board

The Governance Intervention: 

The Governance Intervention Setting Aims Getting Data and Hearing Stories Establishing and Monitoring System-Level Measures Changing the Environment, Policies, and Culture Learning… Starting with the Board Establishing Executive Accountability Six Things All Boards Should Do to Improve Quality and Reduce Harm

Two “Must Dos”: 

Two 'Must Dos' An Initial Chart Audit for Harm A review of 20 to 40 randomly chosen patient charts from the prior month to document all types and levels of injury. A 'Deep Dive' Case Study Conduct a detailed, personal investigation of a significant patient injury in the hospital, including interviewing the involved patient, family, and staff.

5 Million Lives Campaign: 

5 Million Lives Campaign Prevent Pressure Ulcers Prevent MRSA Infections Prevent Harm from High Alert Medications Reduce Surgical Complications – Adopt 'SCIP' Deliver Reliable, Evidence-Based Care for Congestive Heart Failure Get Boards on Board The 'Planks'

To Find Out More…: 

To Find Out More… www.ihi.org/IHI/Programs/Campaign/ OR Go to www.ihi.org, and click on the 'Campaign' logo

Some Tough Lessons about “Full Scale” Improvement: 

Some Tough Lessons about 'Full Scale' Improvement Going to full and public scale… …subjects you to criticism you may never hear about from people whom you will never meet. …leads to misinterpretation by well-meaning reporters that you cannot correct. …invites speculation about your motives and intentions to which you cannot reply. …impedes sincere and open shared exploration of subtle questions. …worries academicians, especially. …induces jealousy that you cannot assuage.

Skeptical Questions: 

Skeptical Questions Are there really so many harms from care? Is the measurement approach (record reviews guided by the Global Trigger Tool) valid and reliable? Is attribution of success to the Campaign possible? Is it overstated? Can hospitals (or IHI) properly evaluate their own success? Are the 'planks' sufficiently evidence-based to be spread widely? Is IHI a proper leader for a Campaign? Will this distract health care from other useful improvement work?

Some Good News about “Full Scale” Improvement: 

Some Good News about 'Full Scale' Improvement Going to full and public scale… …can open up enormous positive feeling, spirit and generosity. …can be inclusive, attractive, and affirming to thousands of people. …can invite creative local adaptations. …can use peer-to-peer support effectively. Generosity abounds People want to have hope

Eleanor Roosevelt: 

Eleanor Roosevelt 'You must do the thing you think you cannot do.'