Identification of undiagnosed diabetes and quality of diabetes care in the United States: cross-sectional study of 11.5 million primary care electronic records

NLP based congestive heart failure case finding: A prospective analysis on statewide electronic medical records

“…10,11,12 Alternatively, the applications of NLP to clinical notes, e.g. identification of pneumonia, 13 diabetes, 14 and CHF, 15,16 have shown promise as a case finding method. The reported NLP based case finding studies to date have achieved good Fmeasures.…”

“…The prospective outcomes were visualized in groups of true positive, false positive, false negative and true negative. To understand the unique patterns associated with the false positive and true positive samples, these patients' notes were analyzed.Learning Transfer14 In the retrospective study, we had two gold standard datasets, one for cutoff point finding and the other for blind-testing purposes. The cutoff point finding dataset was constructed by the notes from the same care facilities as the training dataset.The blind-testing dataset, on the other hand, was constructed by the notes from other care facilities independent from the ones used in the training and cutoff point finding dataset.…”

Did I Tell You That? Ethical Issues Related to Using Computational Methods to Discover Non-Disclosed Patient Characteristics

“…A number of phenotyping applications have already been developed to discover characteristics of patients from their EHRs. For example, Holt and colleagues described methods to search EHRs for laboratory results and clinical diagnoses to classify diabetic patients (Holt, Gunnarsson, Cload, & Ross, 2014). Similarly, other researchers have developed phenotypes to identify patients at risk for suicide (Tran et al, 2014).…”

Implementation of data management and effect on chronic disease coding in a primary care organisation: A parallel cohort observational study

“…Diagnostic coding may be missing 25. Free text may be used instead of structured data and data may be entered in inconsistent fields 26 – 34.…”

“…A recent analysis of 11.5 million primary care electronic records in the U.S. found significantly better quality of care for patients when a coded diagnosis of diabetes was present in the problem list 25. Lack of standardization and coding in EMRs is associated with challenges in benchmarking and comparisons, which are important activities for primary care clinical quality improvement efforts 37 , 38.…”