The American journal of medicine
-
Review Historical Article
Calculated Medicine: Seven Decades of Accelerating Growth.
The field of Calculated Medicine has grown substantially over the last 7 decades. Comprised of objective, evidence-based medical decision tools, Calculated Medicine has broad application in medical practice, medical research, and health care management. This article reviews the history and varied methodologies of Calculated Medicine, starting with the 1953 Apgar score and concluding with a look into modern computational tools of the field: machine learning, natural language processing, artificial intelligence, and in silico research techniques. ⋯ Using natural language processing, we examine and analyze this burgeoning database. Lastly, we examine an important new direction of Calculated Medicine: self-reflection on its potential effect on racial and ethnic disparities in health care. Our field is making great strides promoting health care egality, and some of the most prominent contributions will be reviewed.
-
Primary care in the United States is undergoing bursts of evolution in response to health system stresses, changing demographics, and expansion of risk and value-based reimbursement structures. The impact of primary care remains substantive and associated with improved population health. ⋯ Evolutionary bursts yield new traits and in primary care, they are spawning new care models with significant implications for general internal medicine, internal medicine/pediatrics trained individuals and medicine subspecialties given the focus of these models on Medicare Advantage. Ultimately, changes in reimbursement and creative incentives will be two factors among many that will solidify the next stage of primary care in the United States.
-
Higher total serum cholesterol is associated with lower mortality in heart failure. Evaluating associations between lipoprotein subfractions and mortality among people with heart failure may provide insights into this observation. ⋯ LP-IR score was inversely associated with mortality among patients with heart failure and may be driven by smaller HDL particle size.
-
Readmission within 30 days is a prevalent issue among elderly patients, linked to unfavorable health outcomes. Our objective was to develop and validate multimodal machine learning models for predicting 30-day readmission risk in elderly patients discharged from internal medicine departments. ⋯ Leveraging multimodal machine learning offers a promising strategy for identifying elderly patients who are at high risk for 30-day readmission. By identifying these patients, machine learning models may facilitate the effective execution of preventive actions to reduce avoidable readmission incidents.