Interventions to Reduce Diagnostic Errors in Ambulatory Care (Text Version)

On September 28, 2010, Mark Graber and Hardeep Singh made this presentation at the 2010 Annual Conference. Select to access the PowerPoint® presentation (348 KB).

---

Slide 1

Interventions to Reduce Diagnostic Errors in Ambulatory Care

Mark Graber, MD, Stephanie Kissam, MPH, Hardeep Singh, MD, MPH, Asta Sorensen, MA, Nancy Lenfestey, MHA, Elizabeth Tant, BA, Ken LaBresh, MD, and Kerm Henriksen, PhD

Collaborators: RTI International, Northport Veterans Affairs Medical Center, Michael E. DeBakey Veterans Affairs Medical Center and the Houston VA HSR&D Center of Excellence

AHRQ Action I Master Task Order Mechanism, Contract Number HHSA290200600001 Task 8

Slide 2

Diagnostic Error in Medicine 2010

October 25-27, 2010 | Toronto, Canada
www.smdm.org/diagnostic_errors.shtml

Sponsored by AHRQ
 

Slide 3

Project Goals

• Perform a comprehensive literature review of interventions that could reduce diagnostic errors.
• Identify and pilot test an intervention targeting diagnostic errors in an ambulatory care setting.

Slide 4

Framework for Adverse Events

Image of a framework.

Notes: This is the basic framework we use to understand adverse events in medicine. If a patient is injured, one can think of the root causes as reflecting one of two possible problems (or both): the provider erred, usually a cognitive mistake or slip, or there were inherent flaws in the healthcare system that contributed to the error. System-related problems include communications breakdowns, problems coordinating care, insufficient training, weak policies, problems in the work environment, and many other factors. So the solutions to diagnostic errors could focus on the cognitive skills of the provider, on the characteristics of the healthcare system, or conceivable on the patient as a possible collaborator in reducing error.

Slide 5

Etiology of Diagnostic Errors

• Cognitive Error Only: 28%
• System Error Only: 19%
• No Fault Error Only: 7%
• Both System and Cognitive Errors: 46%

Slide 6

Methods

PubMed database search: 2000-2010

Handpicked articles:

• Non-medical databases (business, psychology, military, engineering).
• Recommendations from experts.

Analysis:

• All articles reviewed by one of three health service researchers.
• Any questionable inclusions reviewed by collaborating physicians.

Slide 7

Inclusion Criteria

• Articles describing tested interventions to reduce error in medical diagnostic settings.
• Studies demonstrating outcome measures in the field of diagnostic errors.
• Articles providing a theoretical basis on how to reduce diagnostic errors (from any field).

Slide 8

Exclusion Criteria

• Studies describing inter-rater or observer variation.
• Articles describing validations of screening instruments, tests, case reports, or techniques to enhance diagnosis.
• Articles describing screening instruments, tests, or technology aides.
• Studies reporting diagnostic error frequency; etiology; or assessments of provider satisfaction, preference, or acceptance of interventions.

Slide 9

Results

Total number of articles: 949

Articles meeting inclusion criteria: 157

Tested interventions: 37
Cognitive: 32
System-related: 5
Engaging patients: 0

Suggested interventions: 120

Slide 10

Systems Interventions

Hardeep Singh, MD MPH
Houston HSR&D Center of Excellence,
Michael E. DeBakey VA Medical Center

Result of collaboration between RTI International, Northport Veterans Affairs Medical Center, Michael E. DeBakey Veterans Affairs Medical Center and the Houston VA HSR&D Center of Excellence

AHRQ Action I Master Task Order Mechanism, Contract Number HHSA290200600001 Task 8.

RTI International is a trade name of Research Triangle Institute

Slide 11

Systems Factors

• Communication and coordination of care issues (transitions)
• Teamwork/Supervision
• Technology/equipment related issues
• Organizational features:
  • Safety culture
  • Policy, processes and procedure related issues
  • Leadership, management, or personnel problems
  • Inadequate resources or available expertise
  • Training issues

Slide 12

Looking for Interventions in These "Process" Dimensions...

Patient-Provider Encounter: History, exam or ordering diagnostic tests for further work-up

Diagnostic Tests: Ordered tests either not performed or performed/interpreted incorrectly

Follow-up and Tracking: Problems with follow-up of abnormal test results or scheduling of follow-up visits

Referrals: Lack of appropriate actions on requested consultation or communication breakdown from consultant to referring provider

Patient Related: Delay in seeking care or adherence to appointments

Slide 13

General Results

• Only 1 of 5 controlled study:
  • 2 were only post-test evaluations.
  • All effective.
• Most interventions in the literature were "conceptual".
• Lack of standardization in process or outcome measures.

Slide 14

Patient-Provider Encounter

2 tested interventions

Change the process of care delivery

• Form designated trauma response team in ER.
• Conduct comprehensive reexamination in ER.

Establish educational programs (suggested only)

• Reinforce history-taking skills.
• Provide teamwork training in medical setting

Perno JF, et al. (2005)
Howard J, et al. (2006)

Slide 15

Diagnostic Tests

One tested intervention

• Implementation of Picture Archiving and Communication System (PACS) for radiology images

Weatherburn, G et al. (2000)

Slide 16

Follow-Up and Tracking

Improving delivery of test results through electronic means

2 tested interventions

Other suggested interventions:

• Establish criteria for communication of abnormal test results.
• Standardize steps involved in the flow of test result information.
• Improve management and presentation of test result data.
• Use an ER manager to monitor radiology test results reporting.
• Create processes to ensure easy retrieval of test result information.
• Develop highly structured hand-off processes that are performed systematically.

Singh, H,et al. (2009)
Poon, EG, et al. (2002)

Slide 17

Patients

No tested interventions suggested only

Notify patients of test results

• Address patient preferences for receiving test results.
• Communicate normal test results.
• Use computerized test results management tool.
• Designate patient navigator.

Slide 18

Patients

Provide patient access to test results

• Use online portal.
• Provide access to entire medical record.

Improve patient-clinician communication

• Consider cognitive limitations when taking patient history.
• Consider communication strategies to optimize patient understanding of medical information.

Increase patient engagement in health care

• Involve patients to ensure the follow-up of test results.

Slide 19

General Interventions (No Specific Dimension)

• Manage error-producing conditions (suggested only)
  • Provide education on error-producing conditions like fatigue.
  • Address work-related conditions that could produce boredom, time pressure, etc.
• Establish systematic tracking of diagnostic error in organization (suggested only)
  • Downstream feedback.

Slide 20

Conclusions—System Issues

• Limited literature on systems interventions that reduced diagnostic error in ambulatory care:
  • Empiric data only for 3/5 dimensions of diagnostic process.
  • Many interventions well conceptualized but poorly operationalized as "testable" interventions.
  • Much discussion of methods to notify patients of test results, but little focused on abnormal results.
  • Health IT potential and workflow related issues .

Slide 21

Conclusions—System Issues

• Gaps in tested interventions aimed at patients:
  • Efficacy of patient and family engagement in preventing or reducing diagnostic error?
  • Multiple organizations and experts advocate for patient engagement in patient safety, yet limited studies successfully do so.
  • No studies report actual interventions engaging patients and families in the process of making medical diagnoses.

Slide 22

Open Discussion—System Issues

Question: How and when can we effectively engage patients and families in diagnostic error reduction?

Slide 23

Cognitive Interventions

Mark L Graber MD FACP
VA Medical Center, Northport NY & SUNY Stony Brook

Collaborators: RTI International, Northport Veterans Affairs Medical Center, Michael E. DeBakey Veterans Affairs Medical Center and the Houston VA HSR&D Center of Excellence

AHRQ Action I Master Task Order Mechanism, Contract Number HHSA290200600001 Task 8.

Slide 24

Cognitive Errors

• Faulty Data Gathering: 14%
• Faulty Sysntesis [sic]: 83%
• Faulty Knowledge: 3%

Slide 25

Cognitive Errors

Most cognitive errors involve breakdowns in synthesizing the available data, due to ...

• Faulty context assumptions.
• Premature closure.
• The inherent shortcomings of heuristic (intuitive) thinking.
• Affective biases and environmental factors that detract from optimal conditions: distractions, fatigue, stress, workload.

Slide 26

Interventions to Reduce Cognitive Error

• Increase medical knowledge and expertise.
• Improve clinical reasoning.
• Get help.

Slide 27

Increase Knowledge & Expertise

• Increase training time & events to increase experience
  (3 tested interventions)
• Use simulation to provide compacted experience
  (1 tested intervention)
• Increase feedback to improve calibration and reduce overconfidence
  (3 tested interventions)

Slide 28

Improve Clinical Reasoning

No tested interventions suggested only

• Improve evidence-based medicine skills, normative decision-making skills.
• Improve intuitive decision-making:
  • Teach heuristics & biases.
  • Use de-biasing techniques; improve metacognition.
    • Reflective practice; checklists; be comprehensive, consider the opposite.

Slide 29

Get Help

• Increase consultation, second opinions, fresh eyes
  10 tested interventions
• Use decision support tools; increase access to medical knowledge (Web access, texts, info buttons)
  12 tested interventions

Slide 30

Conclusions—Cognitive Factors

• A broad array of ideas for interventions (N=157), but few tested (N=37).
• Gaps:
  • Most interventions apply to diagnostic specialties (radiology, pathology, laboratory), not the ED or PC.
  • Tests have been done under artificial conditions.
  • Learning assessed only in the short term.
  • Tools developed aren't used.

Slide 31

Suggested Project: Checklist(s)

• Checklists are ideal in dealing with complexity.
• Checklists can combine system-based, patient-based, and cognitive interventions.
• Checklists are hot.

Slide 32

A General Checklist

• Obtain Your Own, Complete medical history.
• Perform a Focused and Purposeful physical examination.
• Generate some initial hypotheses and differentiate these with appropriate questions, examination, or diagnostic tests.
• Pause to reflect—Take a diagnostic "time out":
  • Was I comprehensive?
  • Did I consider the inherent flaws of heuristic thinking?
  • Was my judgment affected by any other bias?
  • Do I need to make the diagnosis now, or can I wait?
  • What's the worst case scenario? What are the 'don't miss' entities?
• Embark on a plan, but acknowledge uncertainty and Ensure a Pathway for Follow-Up.

Slide 33

A Syndrome-Specific Checklist

Chest Pain

__ MI

__ PE

__ Pneumonia

__ Pericarditis

__ Musculoskeletal

__ Gerd

__ Herpes Zoster

__ Pleurisy

__ Aortic stenosis

__ Tumors—lung, lymphoma, mediastinum

__ Spinal cord compression

__ Esophageal spasm

__ Psychiatric

Slide 34

Open Discussion—Cognitive Issues

• Question 1—Which would be more effective —a General checklist, or Syndrome Specific checklists?
• Question 2—What would it take to convince frontline providers to use a checklist?
• Question 3—Will they help reduce diagnostic errors, or are we better off just trusting our initial (intuitive) diagnoses?

Slide 35

Other Questions?

• Measurement of diagnostic errors?
• How to evaluate quality of clinical reasoning?
• How do you teach this stuff?

Slide 36

Acknowledgments: AHRQ, RTI, VA