Electronic health records (EHR) represent a rich and relatively untapped resource for characterizing the true nature of clinical practice and for quantifying the degree of inter-relatedness of medical entities such as drugs, diseases, procedures and devices. We provide a unique set of co-occurrence matrices, quantifying the pairwise mentions of 3 million terms mapped onto 1 million clinical concepts, calculated from the raw text of 20 million clinical notes spanning 19 years of data. Co-frequencies were computed by means of a parallelized annotation, hashing, and counting pipeline that was applied over clinical notes from Stanford Hospitals and Clinics. The co-occurrence matrix quantifies the relatedness among medical concepts which can serve as the basis for many statistical tests, and can be used to directly compute Bayesian conditional probabilities, association rules, as well as a range of test statistics such as relative risks and odds ratios. This dataset can be leveraged to quantitatively assess comorbidity, drug-drug, and drug-disease patterns for a range of clinical, epidemiological, and financial applications.

BACKGROUND: Risk adjustment is an important component of surgical outcomes and quality analyses. Current models include numerous preoperative variables; however, the relative contribution of these variables may be limited. This research seeks to identify a model with the fewest number of variables necessary to perform an adequate risk adjustment to predict any inpatient adverse event for use in resource-limited settings. METHODS: All patients from the National Surgical Quality Improvement Program (NSQIP) database from 2005 to 2010 were included. Outcomes were inpatient mortality or any surgical complication captured by NSQIP. Models were built by sequential addition of preoperative risk variables selected by their area under the receiver operator characteristic curve (AUC). RESULTS: Among 863,349 patients, the single variable with the highest AUC was American Society of Anesthesiologists (ASA) classification (AUC = 0.7127). AUC values reached 0.7923 with five variables (ASA classification, wound classification, functional status prior to surgery, albumin, and age) and 0.7945 with six variables. The sixth variable was one of the following: alkaline phosphatase, weight loss, principal anesthesia technique, gender, or emergency status. The model with the highest discrimination that did not require laboratories included ASA classification, functional status prior to surgery, wound classification, and age (AUC = 0.7810). Including all 66 preoperative variables produced little additional gain (AUC = 0.8006). CONCLUSIONS: Six variables are sufficient to develop a risk adjustment tool for inpatient surgical mortality and morbidity. This research has important implications for the field of surgical outcomes research by improving efficiency of data collection. This limited model can aid the expansion of risk-adjusted analyses to resource-limited settings worldwide.

The 21st Century Cures Act passed by the United States Congress mandates the US Food and Drug Administration to develop guidance to evaluate the use of real-world evidence (RWE) to support the regulatory process. RWE has generated important medical discoveries, especially in areas where traditional clinical trials would be unethical or infeasible. However, RWE suffers from several issues that hinder its ability to provide proof of treatment efficacy at a level comparable to randomized controlled trials. In this review article, we summarized the advantages and limitations of RWE, identified the key opportunities for RWE, and pointed the way forward to maximize the potential of RWE for regulatory purposes.

Individualized genetic therapies-medicines that precisely target a genetic variant that may only be found in a small number of individuals, as few as only one-offer promise for addressing unmet needs in genetic disease, but present unique challenges for trial design. By nature these new individualized medicines require testing in individualized N-of-1 trials. Here, we provide a framework for maintaining scientific rigor in N-of-1 trials. Building upon best practices from traditional clinical trial design, recent guidance from the United States Food and Drug Administration, and our own clinical research experience, we suggest key considerations including comprehensive baseline natural history, selection of appropriate clinical outcome assessments (COAs) individualized to the patient genotype-phenotype for safety and efficacy assessment over time, and specific statistical considerations. Standardization of N-of-1 trial designs in this fashion will maximize efficient learning from this next generation of targeted individualized therapeutics.