Development of Emergency Department Quality Indicators QI

1
Development of Emergency Department Quality
Indicators (QI)

• Kathryn M. McDonald
• Center for Health Policy / Center for Primary
  Care and Outcomes Research, Stanford University
• September 14, 2009

2
Overview

• Development of AHRQ QI Using ED Data
• Patient Safety Indicators (PSI)
• Prevention Quality Indicators (PQI)
• AHRQ QI Development Methodology
• Adapting the AHRQ QI to the ED
• Potential QI ED Indicators
• Patient Safety Events
• Avoidable ED visits
• Data Issues
• Summary

2
3
Development of AHRQ ED QI

• Purpose of this New Task
• Develop a set of quality indicators that is
  applicable to the emergency department setting
• Incorporate set into the publicly available AHRQ
  QI software
• Implement the established AHRQ QI measurement
  development process
• Adapt existing AHRQ QI to ED setting
• Conduct a review of new candidate indicators

3
4
Existing AHRQ QIsPatient Safety Indicators (PSI)

• Death in low mortality DRGs
• Decubitus ulcer
• Failure to rescue
• Foreign body left during procedure
• Iatrogenic pneumothorax
• Selected infections due to medical care
• Postoperative hemorrhage or hematoma
• Postoperative hip fracture
• Postoperative physiological and metabolic
  derangement
• Postoperative PE or DVT

The indicators marked with are also provided as
area level indicators.
5
Patient Safety Indicators (cont.)

• Postoperative respiratory failure
• Postoperative sepsis
• Postoperative wound dehiscence
• Technical difficulty with procedure
• Transfusion reaction
• Birth trauma injury to neonate
• OB trauma vaginal delivery with instrument (w/
  and w/o 3rd degree lacerations)
• OB trauma vaginal delivery without instrument
  (w/ and w/o 3rd degree lacerations)

The indicators marked with are also provided as
area level indicators.
6
Existing AHRQ QIsPrevention Quality Indicators

• Bacterial pneumonia
• Dehydration
• Urinary tract infection
• Perforated appendix
• Low birth weight
• Angina without procedure
• Congestive heart failure

• Hypertension
• Adult asthma
• COPD
• Diabetes cx - short term
• Diabetes cx - long term
• Uncontrolled diabetes
• Lower extremity amputation

7
Measure Development and Validation Process
SOURCES
8
Starting Point 1 Adapting the AHRQ QI

• Developing emergency department PSI shall involve
  several challenges. Conceptually patient safety
  issues may manifest themselves at different
  points in time and in different settings
• Within the ER visit, or
• In an admission to the same hospital (promising
  area of focus) or
• Outside of the ER of interest or the same
  hospital (e.g, home, another ER, another
  hospital)
• Development effort might consider the
  relationship between the PQIs and potential PSIs
  in the ED context
• For example, a patient that presents at the
  emergency department with a Urinary Tract
  Infection (UTI) and then gets admitted as an
  inpatient. This patient will flag as a PQI at the
  ED level and at the inpatient admission level.
• However, another patient that presents with a UTI
  and then is sent home, but returns the following
  day and needs to be admitted might be considered
  a patient safety problem.
• Our development effort will consider these issues
  and relationships to develop indicators of two
  types potentially preventable ED visits, and
  potentially preventable patient safety events.

8
9
Starting Point 2 Conceptualizing Quality Issues

• What types of adverse events might occur during
  an ED admission/encounter?
• Conditions worsens while waiting (related to ED
  crowding)
• Occult GI bleeds, sepsis, hypoglycemia, atypical
  heart attacks
• Cardiac arrest in waiting room, death of asthma
  patient
• Air embolism from IV lines
• Medication error
• Wrong medication to wrong patient (e.g., with too
  many patients in one treatment room)
• Incorrect dose (e.g. patient not weighed,
  transcription errors)
• Medication give when patient has known allergy
• Drug interaction reaction (e.g., coumadin with
  certain common antibiotics)
• Inadequate monitoring (e.g., hypogylcemia after
  insulin, resp distress with narcotics)
• Hemolytic reaction due to administration of
  incompatible blood or blood products
• Falls
• From Stretcher, Bed, Bathroom, Wet floor

9
10
Patient Safety Events

• What types of adverse events might occur during
  an ED admission/encounter or soon after?
• Alarm malfunctions
• Not audible to or not seen by nurses
• IV Pump issues
• Back up battery failures stops meds,
  unintentional change of settings, inadequate
  monitoring
• Patient specimen errors
• Patients not fully registered or identified when
  specimen leaves ED, leading to lags in results
  communications, or assigning results to wrong
  patient
• Infection control failures
• Incorrect diagnosis bounce back
• Patient returns to ED for abdominal pain (missed
  AAA)
• X rays interpreted as negative, fracture dx on
  later read or when patient returns
• Death in a psychiatric patient admitted to psych
  ward within 72 hours of ED visit (missed organic
  causes)
• Missed injuries (Traumatic brain injury)

10
11
Potentially Avoidable Visits

• What ED encounters are potentially preventable by
  high quality outpatient care?
• Diabetic Complications
• Hyperglycemia
• Infections
• Asthma
• Acute respiratory event
• Bronchitis
• Hypertension
• Hypertensive urgency
• Hemorrhagic stroke
• Colds, Flu Invasive Pneumococcal Disease
• Poor hand hygiene
• Lack of primary care
• Missed Flu shot
• Missed Pneumococcal Vaccine

11
12
Potentially Avoidable Visits

• What ED encounters are potentially preventable by
  public health interventions?
• Falls
• Hip Fractures, Long bone fractures, Pediatric
  (e.g., due to poor window guards)
• Brain injury secondary (not wearing helmets)
• Fireworks injuries
• Dehydration
• ETOH intoxication
• Overdose/substance abuse
• Sexually transmitted diseases
• Obstetric complications
• Motor vehicle collision
• Drowning
• Carbon monoxide poisoning (poorer quality
  furnaces)
• Accidental hypothermia (homeless sleeping in
  cold)
• Food poisoning (poor restaurant sanitation)

12
13
Potentially Avoidable Visits

• What ED encounters are potentially preventable
  with health care system changes?
• Primary care provider told patient to go to ED
  over the phone when office practice is closed
• Weekend
• Night Time
• Inadequate care outside ED for Aging population
• Unable to provide self care, and inadequate
  support available
• Incontinence issues
• Care given due to lack of adequate insurance for
  non-ED care
• Ingrown toe nails with infection
• Vaginitis
• Urethritis
• Otitis Media
• Running out of home oxygen
• Possibly due to lack of primary care doctor
• Sore throat
• Urinary tract infection

13
14
Potentially Avoidable Visits

• What ED encounters are potentially preventable
  with improvements to post-surgical care
  processes?
• Unable to care for self post-operatively (return
  visit)
• Outpatient Surgery Complications
• Wound dehiscence from outpatient surgery
• Post op Infections
• Hemorrhaging
• Other post op complications

14
15
Data Source Issues

• State Emergency Department Datasets (SEDD)
• Treat and release encounters from 27 states
• Encounters resulting in subsequent admission to
  the same hospital from 41 states
• How do records for the subsequent admissions
  relate to corresponding hospitalization records
  in the SID datasets?
• Diagnosis and Procedure Codes
• May be that ED diagnosis codes are dropped or
  incorporated into a longer list when the patient
  is admitted
• May be no effective method to evaluate the
  quality of ED care for patients who were
  hospitalized from the ED
• These issues will be evaluated in detail

15
16
Summary Project Philosophy

• Emergency medicine is particularly sensitive to
  guidelines and quality measures from other
  specialties (e.g., cardiology, infectious
  disease, pediatrics).
• The EM community is not always included in
  guideline and quality measure development at the
  initial stages, even though they are affected
  most.
• Therefore, a focus on ED quality needs to starts
  with the EM community (doctors, nurses,
  department managers).
• Use existing data sources available from AHRQ
• Therefore, this projects indicators will not
  cover all important ED quality concern (e.g.,
  ambulance diversions from overcrowding)
• Start with existing AHRQ measures and development
  approaches

17
Acknowledgments

• Funded by AHRQ
• Support for Quality Indicators II (Contract No.
  290-04-0020)
• Mamatha Pancholi, AHRQ Project Officer
• Jeffrey Geppert, Project Director, Battelle
  Memorial Institute Sciences
• Data used for analyses
• State Emergency Department Databases (SEDD),
  2002-2006. Healthcare Cost and Utilization
  Project (HCUP), Agency for Healthcare Research
  and Quality