into Big Data for Diagnostic Error David E. Newman-Toker, MD PhD - - PowerPoint PPT Presentation

▶
into big data for diagnostic error
SMART_READER_LITE
LIVE PREVIEW

into Big Data for Diagnostic Error David E. Newman-Toker, MD PhD - - PowerPoint PPT Presentation

Jun 25, 2023 477 likes •790 views

Academy Health Annual Research Meeting, Boston 2016 Diagnostic Research Methods: Brief Overview plus a Deeper Dive into Big Data for Diagnostic Error David E. Newman-Toker, MD PhD Associate Professor of Neurology Johns Hopkins University

slide-1
SLIDE 1

Academy Health Annual Research Meeting, Boston 2016

Diagnostic Research Methods: Brief Overview plus a Deeper Dive into Big Data for Diagnostic Error

David E. Newman-Toker, MD PhD

Associate Professor of Neurology Johns Hopkins University School of Medicine Johns Hopkins Bloomberg School of Public Health Johns Hopkins Armstrong Institute for Patient Safety & Quality

slide-2
SLIDE 2

DISCLOSURES

1.

Grant support

NIH U01 DC013778-01A1 (NIDCD), 5U01NS080824, (NINDS), U24TR001609-01 (NCATS), AHRQ (pending)

Siemens/SIDM, Brainscope, Kaiser Permanente

2.

Research VOG devices loaned by

GN Otometrics

Autronics-Interacoustics

3.

Founding Board Member SIDM (unpaid)

4.

‘Diagnosis’ career focus (academic COI)

Investigational Use – Device

slide-3
SLIDE 3

OBJECTIVES

1) Discuss the link between different diagnostic research contexts and choice of research methods 2) List at least one method for each diagnostic safety measurement objective (burden, causes, solutions) 3) Describe analytic methods and techniques to enhance ‘big data for diagnostic error’ research

Newman-Toker

slide-4
SLIDE 4

OUTLINE

1) Conceptual Model Defining Research Space 2) Linking Conceptual Model to Research Methods 3) Deep(er) Dive into Big Data for Diagnostic Error 4) Final Thoughts & Take Home Messages 5) Questions/Panel Discussion

Newman-Toker

slide-5
SLIDE 5

CONCEPTUAL MODEL

FOR RESEARCH SPACE

slide-6
SLIDE 6

IOM Definition of Diagnostic Error

DIAGNOSTIC ERROR is the failure to…

(a) establish an accurate and timely explanation

  • f the patient’s health problem(s) or

(b) communicate that explanation to the patient

slide-7
SLIDE 7

Diagnostic Process Failure Diagnosis Label Failure Preventable Diagnostic Error “No Fault” Misdiagnosis “Near Miss” Process Failure

Newman-T

  • ker, Diagnosis, 2014

Opportunity for… Quality Assurance

Safety

slide-8
SLIDE 8

Diagnostic Process Failure

N O H A R M

Suboptimal Diagnostic Process Diagnosis Label Failure Optimal Diagnostic Process UNDIAGNOSED & UNDIAGNOSABLE Standard yet Suboptimal Care Preventable Diagnostic Error “NEAR MISS” PROCESS PROBLEM* HARM FROM OVERTESTING & OVERDIAGNOSIS*

H A R M

Opportunity for… Quality Improvement

Dissemination

Opportunity for… Quality Assurance

Safety

Opportunity for… New Science

Discovery

Newman-T

  • ker, Diagnosis, 2014
slide-9
SLIDE 9

Diagnostic Process Failure

N O H A R M

Suboptimal Diagnostic Process Diagnosis Label Failure Optimal Diagnostic Process UNDIAGNOSED & UNDIAGNOSABLE Standard yet Suboptimal Care Preventable Diagnostic Error “NEAR MISS” PROCESS PROBLEM* HARM FROM OVERTESTING & OVERDIAGNOSIS*

H A R M

Opportunity for… Quality Improvement

Dissemination

Opportunity for… Quality Assurance

Safety

Opportunity for… New Science

Discovery

Newman-T

  • ker, Diagnosis, 2014
slide-10
SLIDE 10

LINKING MODEL TO RESEARCH METHODS

slide-11
SLIDE 11

LINKING MODELS & METHODS DISCOVERY RESEARCH IN DIAGNOSIS

Newman-Toker

Validation Cross-Sectional E2E, Cluster RCT Works? Accurate? Δ Tests? Δ Treatment? Δ Outcomes? Phase III RCT Phase II RCT

Opportunity for… New Science

Discovery

slide-12
SLIDE 12

Pronovost, BMJ 2008

TRIP Conceptual Model (v2)

Translating Research Into Practice Opportunity for… Quality Improvement

Dissemination Implementation Science Outcome Methods (e.g., Quasi-Experimental, Stepped Wedge)

slide-13
SLIDE 13

Newman-Toker

Burden Cause

Methods (‘Meta’)

Solution

LINKING MODELS & METHODS SAFETY DOMAINS

Opportunity for… Quality Assurance

Safety Simulations, Experiments Qualitative (RCA, process) Case-Control, Cohort Pre-Post, Stepped Wedge Diagnostic Strategy RCT Meta-Analysis, Modeling Patient, Provider Surveys Concordance, Spectrum (OverDx) Surveillance for Unplanned Events

slide-14
SLIDE 14

DEEP(ER) DIVE INTO BIG DATA FOR DX ERROR

slide-15
SLIDE 15

Newman-Toker

BIG DATA FOR DX ERROR CASE STUDY: MISSED STROKE IN DIZZINESS

A 30 year-old woman presents with new vertigo and vomiting to the ED. Woke with symptoms this morning and still has them 12 hours later. Associated with nausea, vomiting, head motion intolerance, gait

  • unsteadiness. No other neurologic symptoms.

Does the patient have a stroke? ED physician orders a CT scan of the brain. When it returns with a normal result, the patient is discharged with medication (meclizine) for “labyrinthitis” and told to follow up with their primary care provider. The patient returns 48 hours later herniating from a large posterior fossa stroke, and ends up disabled in a nursing home.

slide-16
SLIDE 16

LARGE CEREBELLAR INFARCTION

medial PICA-territory stroke, dimensions 3.0 x 5.0 x 4.4 cm

slide-17
SLIDE 17

Newman-Toker

BIG DATA FOR DX ERROR CASE STUDY: MISSED STROKE IN DIZZINESS

We know this happens, but… How often? and… How can we monitor it operationally for performance feedback and to measure impact of solutions?

slide-18
SLIDE 18

BIG DATA FOR DX ERROR

MISSED STROKE IN “BENIGN” DIZZINESS

Look Back Approach: Stroke patients more likely to have been discharged from ED with “benign” dizziness prior ~14 days (N = ~180,000 strokes) Look Forward Approach: ‘Benign’ dizziness sent home from ED more likely to return with a stroke within ~30 days, but not heart attack (N = ~30,000 ED dizzy discharges)

slide-19
SLIDE 19

Look Back (Outcome to Symptoms)

Stroke

  • Dizziness
  • Headache
  • Numbness

Myocardial Infarction

  • Chest pain
  • Back pain
  • Dyspnea

Pulmonary Embolus

  • Dyspnea
  • Chest pain
  • Back pain
slide-20
SLIDE 20

Look Forward (Symptoms to Outcome)

Dizziness Vertigo

  • Stroke
  • Myocardial

infarction

  • Pulmonary embolus

Headache

  • Subarachnoid

hemorrhage

  • Meningitis
  • Stroke

Back Pain

  • Myocardial

infarction

  • Aortic dissection
  • Spinal cord

compression

slide-21
SLIDE 21

BIG DATA FOR DX ERROR

PEARL #1 – LUMP FOR LOOK BACK

PROBLEM: Variable Granularity Administrative Data SOLUTION: ‘Lump’ Sensibly Using HCUP-CCS

slide-22
SLIDE 22

BIG DATA FOR DX ERROR

PEARL #2 – USE O/E, TEMPORAL ANALYSIS

PROBLEM: Risk of Spurious Association SOLUTION: Use O/E and temporal profile

slide-23
SLIDE 23

BIG DATA FOR DX ERROR

PEARL #3 – USE CONTROL GROUPS

PROBLEM: Risk of Non-Specific Association SOLUTION: Use Clinical Comparison Groups

slide-24
SLIDE 24

FINALTHOUGHTS & TAKE HOME MESSAGES

slide-25
SLIDE 25

Diagnostics Development Pipeline

slide-26
SLIDE 26

Gaps in the Diagnostic Research Translational Continuum

slide-27
SLIDE 27

TAKE HOME POINTS

1) Match methods to research context (discovery, dissemination, or safety) 2) Consider your measurement objectives in the safety space (burdens, causes, solutions) 3) With ‘big data for diagnostic error’ combine ‘look back’ with ‘look forward’ and use analytic pearls (lump, O/E-time, control)

Newman-Toker

slide-28
SLIDE 28

QUESTIONS & PANEL DISCUSSION

slide-29
SLIDE 29

DIAGNOSTIC STRATEGY RCT

SOLUTIONS THROUGH INNOVATION

Portable Video-oculography: The “Eye ECG” AVERT Trial (NIH-sponsored $5.5M Phase II RCT)

slide-30
SLIDE 30