A Human(e) Factor in Clinical Decision Support Systems.

The overwhelming amount, production speed, multidimensionality, and potential value of data currently available-often simplified and referred to as big data -exceed the limits of understanding of the human brain. At the same time, developments in data analytics and computational power provide the opportunity to obtain new insights and transfer data-provided added value to clinical practice in real time. What is the role of the health care professional in collaboration with the data scientist in the changing landscape of modern care? We discuss how health care professionals should provide expert knowledge in each of the stages of clinical decision support design: data level, algorithm level, and decision support level. Including various ethical considerations, we advocate for health care professionals to responsibly initiate and guide interprofessional teams, including patients, and embrace novel analytic technologies to translate big data into patient benefit driven by human(e) values.

Fragile neutrophils in surgical patients: A phenomenon associated with critical illness.

Leukocyte viability (determined by e.g. propidium iodide [PI] staining) is automatically measured by hematology analyzers to check for delayed bench time. Incidental findings in fresh blood samples revealed the existence of leukocytes with decreased viability in critically ill surgical patients. Not much is known about these cells and their functional and/or clinical implications. Therefore, we investigated the incidence of decreased leukocyte viability, the implications for leukocyte functioning and its relation with clinical outcomes. An automated alarm was set in a routine hematology analyzer (Cell-Dyn Sapphire) for the presence of non-viable leukocytes characterized by increased fluorescence in the PI-channel (FL3:630±30nm). Patients with non-viable leukocytes were prospectively included and blood samples were drawn to investigate leukocyte viability in detail and to investigate leukocyte functioning (phagocytosis and responsiveness to a bacterial stimulus). Then, a retrospective analysis was conducted to investigate the incidence of fragile neutrophils in the circulation and clinical outcomes of surgical patients with fragile neutrophils hospitalized between 2013-2017. A high FL3 signal was either caused by 1) neutrophil autofluorescence which was considered false positive, or by 2) actual non-viable PI-positive neutrophils in the blood sample. These two causes could be distinguished using automatically generated data from the hematology analyzer. The non-viable (PI-positive) neutrophils proved to be viable (PI-negative) in non-lysed blood samples, and were therefore referred to as 'fragile neutrophils'. Overall leukocyte functioning was not impaired in patients with fragile neutrophils. Of the 11 872 retrospectively included surgical patients, 75 (0.63%) were identified to have fragile neutrophils during hospitalization. Of all patients with fragile neutrophils, 75.7% developed an infection, 70.3% were admitted to the ICU and 31.3% died during hospitalization. In conclusion, fragile neutrophils occur in the circulation of critically ill surgical patients. These cells can be automatically detected during routine blood analyses and are an indicator of critical illness.

Data mining information from electronic health records produced high yield and accuracy for current smoking status.

OBJECTIVES: Researchers are increasingly using routine clinical data for care evaluations and feedback to patients and clinicians. The quality of these evaluations depends on the quality and completeness of the input data. STUDY DESIGN AND SETTING: We assessed the performance of an electronic health record (EHR)-based data mining algorithm, using the example of the smoking status in a cardiovascular population. As a reference standard, we used the questionnaire from the Utrecht Cardiovascular Cohort (UCC). To assess diagnostic accuracy, we calculated sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). RESULTS: We analyzed 1,661 patients included in the UCC to January 18, 2019. Of those, 14% (n = 238) had missing information on smoking status in the UCC questionnaire. Data mining provided information on smoking status in 99% of the 1,661 participants. Diagnostic accuracy for current smoking was sensitivity 88%, specificity 92%, NPV 98%, and PPV 63%. From false positives, 85% reported they had quit smoking at the time of the UCC. CONCLUSION: Data mining showed great potential in retrieving information on smoking (a near complete yield). Its diagnostic performance is good for negative smoking statuses. The implications of misclassification with data mining are dependent on the application of the data.