Prevalence and Characterization of Diagnostic Error Among Patients With Bacteremia

Background and Rationale:

Diagnostic error is an important but often under-recognized source of adverse events in the hospital. It ranks second as the cause of adverse events1. Studies have indicated that major diagnostic discrepancies are found in 10-20% of autopsies, and that diagnostic error is the leading cause of malpractice litigation for physicians in Canada2. Furthermore, diagnostic error is of upmost importance in the eyes of our patients: Studies have cited that 55% of patients indicated diagnostic error as their chief concern when seeing a physician. Diagnostic error occurs most commonly in areas where there is a high degree of diagnostic uncertainty such as General Internal Medicine.

Diagnostic error can be classified as either systems or cognitive errors. Systems errors involve technical failure, equipment problems and organization flaws. Systems errors have been the focus of most quality improvement and patient safety research in the past few decades. In contrast, cognitive sources of errors have received little attention despite studies indicating that the cognitive factors play in a role in up to 74% of cases involving diagnostic error. Cognitive error is defined by faulty knowledge, data gathering and synthesis of information. They often occur as the result of biases and heuristics used by physicians during the diagnostic process.

Most diagnostic errors are associated with common conditions such as sepsis, pulmonary embolism, drug intoxication, myocardial infarction and appendicitis.

This study is focused on the delayed or missed diagnosis of sepsis associated with bacteremia among patients admitted to The Ottawa Hospital. Sepsis is a common and important diagnosis. Delayed treatment is associated with significant morbidity and mortality9. Therefore, timely and correct diagnosis of sepsis is of upmost importance. No research thus far has been conducted assess the incidence and etiology of diagnostic error among patients with sepsis.

Sepsis is defined as life-threatening organ dysfunction caused by dysregulated host response to infection and the definitive diagnosis of sepsis is often complex. Given this, this study will include only patients with a missed diagnosis of sepsis who also have a positive blood cultures given the need for an objective marker of the diagnosis.

Aims and Objectives:

The aim of this study is to determine the prevalence and characterization of diagnostic error among patients admitted to the at The Ottawa Hospital with sepsis associated with a positive blood culture.

Study Design and Methods:

This study will consist of a retrospective chart review of all patients who are greater than 18 years with positive blood culture taken within 12 hours of presenting to the Emergency Department of Ottawa Hospital Civic and General Campus between January 2016 to August 13th 2017.

For patients with a positive blood culture, charts will be reviewed only for those patients who did not receive treatment for sepsis on presentation by an Emergency Room (ER) physician or consultant within the first 24 hours.

Patients with positive blood cultures with an organism that is deemed a probable-contaminant organism from the study will be excluded. For this study, probable contaminant organisms will be defined as the following: (1) Coagulase-negative staphylococci (CoNs), (2) Bacillus species other than Bacillus anthracis, (3) Propionbacterium acnes and (4) Corynebacterium species. Coagulase negative staphylocci (CoNs) will be considered as a probable-contaminant if detected in only one bottle or the minority of bottles within a blood culture set in the absence of risk factors for invasive infection. Risk factors for CoNs invasive infection include intravascular catheters, hemodialysis catheters, vascular grafts, recent prosthetic joints or hardware, presence of pacemaker or prosthetic cardiac valves. Bacillus species, Proprionbacterium acnes and Corynebacterium species, will be considered contaminants if they are detected in one or the minority of bottles in a blood culture set.

For charts that are eligible for review for diagnostic errors, the qSOFA (Quick Sequential [Sepsis-related] Organ Failure Assessment) score at time at admission will be used to identify those patients who were at risk of serious harm because of a missed diagnosis of sepsis and lack of treatment. The qSOFA is a score included in the Sepsis-3 Recommendations published in 2016, and is a validated bedside tool used for out-of-hospital, emergency department or general hospital ward settings in which patients with a suspected infection can be rapidly identified as being more likely to have poor outcomes typical of sepsis. The score consists of the following criteria and a score of 2 or more criteria correlates to worse outcomes: (1) respiratory rate of 22/min or greater, (2) altered mentation and (3) systolic blood pressure of 100mm Hg or less.

Patient variables will be collected and compared between the following three groups: (1) patients who received antibiotics within 24 hours, (2) patients who did not receive antibiotics within 24 hours and who did not meet qSOFA criteria, and (3) patients who did not receive antibiotic within 24 hours and who met qSOFA criteria. Variables to be studied include age, gender, hospital campus, admitting service, day of admission (weekend vs. weekday), bacteria species, time of admission (night vs. day), length of stay, disposition and Elixhauser Comorbidity Index.

For charts eligible for review for diagnostic error, the DEER (Diagnostic Error Evaluation Research) Taxonomy tool will be used to classify cases per the location and type of error that occurred in the diagnostic process.