Randi Connor-Schuler, BA, MD
Emory University School of Medicine
Atlanta, Georgia
Disclosure information not submitted.
Morgan Oskutis, MD
Dr
Emory University School of Medicine, United States
Disclosure information not submitted.
Ankita Agarwal, MD,
Emory University School of Medicine
Atlanta, Georgia
Disclosure information not submitted.
Ashley Binder, MD
PCCM Fellow
Emory University School of Medicine
Atlanta, Georgia, United States
Disclosure information not submitted.
David Murphy, MD, PhD, FCCM
Dr
Emory University School of Medicine, United States
Disclosure information not submitted.
Title: Clarity of Thought Before Rashness of Action: Differential Diagnosis and ICU Diagnostic Errors
Introduction: Differential diagnoses are integral to medicine, helping reduce cognitive failure and diagnostic error. Documentation demonstrates a healthcare provider’s thought process for diagnosis, workup, and management decisions. We evaluated the relationship between initial differential diagnosis documentation in provider charting and delayed diagnoses for patients’ primary intensive care unit (ICU) problems.
Methods: Charts from medical, surgical, and mixed medical-surgical ICU’s were reviewed for initial differential diagnoses (Ddx) and diagnostic delay. Two independent reviewers scored diagnosis documentation by initial provider (e.g. advanced practice provider, resident, fellow) and attending on a 5-point scale: 0-no explicit diagnosis, 1-single diagnosis, 2-limited Ddx, 3-Ddx with ≥3 diagnoses, 4-Ddx with rationale. Discrepancies were resolved by consensus after scoring by a third reviewer. Chi-square and Fisher’s exact test were used to compare proportions. We used a logistic regression model to calculate odds ratios (OR) and 95% confidence intervals (CI) between Ddx score and delayed diagnosis, adjusting for potential confounders based on univariate analyses.
Results: Of 100 charts, 55 were from medical ICU patients, 35 medical/surgical ICU, and 10 surgical ICU. Most charts were scored a 1 (39%) with 14% scored 0, 24% scored 2 (24%), 11% scored 3 and 12% scored 4. Delayed diagnoses were identified in 11 charts – 18% scored 0, 55% scored 1, 27% scored 2, and 0% scored 3 or 4. There was no difference in delayed diagnoses between problem- or systems-based notes or between ICU types. There was a significant relationship between chart rating and odds of delayed diagnosis for charts rated 2 or 1 compared to 0 (score 2 vs 0: OR 0.03, 95% CI 0.002-0.55 and score 1 vs 0: OR 0.07, 95% CI 0.004-0.98), adjusting for provider type and clinical significance.
Conclusions: Higher quality differential diagnoses on initial provider documentation was associated with lower rates of delayed diagnoses. Differential diagnosis documentation may help broaden the thought process for patients admitted to the ICU and help prevent early anchoring biases that may negatively impact patient outcomes. Further study is needed to examine barriers to effective differential diagnosis and identify improvement strategies.