Diffuse Intrinsic Pontine Glioma (DIPG) is a universally fatal pediatric brain tumor with an overall median survival of 9 to 11 months. As the molecular and genetic landscape of DIPG has become increasingly characterized over recent years, numerous clinical trials have explored a variety of targeted therapies. However, none have meaningfully prolonged survival.
Clinicians place orders for patients in the electronic health record (EHR). There is currently no internal mechanism to detect medical errors in EHR systems. Identifying and monitoring medical errors has relied on voluntary reporting and chart review, methods that are subject to substantial self-reporting bias. To quantify the magnitude of wrong-patient errors, I developed and validated the Wrong-Patient Retract-and-Reorder (RAR) Measure. The Wrong-Patient RAR measure overcomes limitations of voluntary reporting by using an electronic query to objectively detect wrong-patient orders in EHR data. Whereas previous data indicated an average of 9 wrong-patient medication errors per hospital per year based on voluntary reporting,2 the RAR measure identified 5,246 wrong-patient orders in a large healthcare system in 1 year. The vastly greater volume of errors detected provides insights into the epidemiology of wrong-patient orders, informs targeted intervention strategies, and yields sufficient numbers of events to power health IT safety intervention studies.
Background:
eMERGE is a NIH-funded national network of 10 sites that enroll participants into genetic research. It has had 3 earlier cycles with tens of thousands of participants recruited in each; the current cycle (eMERGE IV) plans to enroll 25,000 individuals, with a particular focus on patients from underserved communities. The network has developed several participant surveys that allow the collection of data across the sites. This study focuses on participants who self-identified as people with disabilities as well as those who have chronic conditions such as kidney disease or type 1 diabetes. Issues to be explored include diverse representation in cohort, recruitment and retention, identify stated needs for participation, health behaviors, and expectations surrounding genetic testing for common diseases. Tasks required to fulfill study aims are literature review, quantitative data analyses skills, familiarity with e-health records and ICD codes, and drafting of manuscript; the projected outcome is a manuscript in a peer-review journal.
Women identified as high-risk for breast cancer may benefit from personalized risk-reducing strategies; however, barriers exist, including the time required to conduct a risk assessment of each woman in a population. Electronic health records (EHRs), a common source for populating risk assessment models, present challenges, including missing data, and data types more accurate when provided by patients compared to EHRs. We previously extracted EHR data on age, race/ethnicity, family history of BC, benign breast disease, and breast density to calculate BC risk according to the Breast Cancer Surveillance Consortium (BCSC) model among 9,514 women. Comparing self-reported and EHR data, more women with a first-degree family history of BC (14.6% vs. 4.4%) and benign breast biopsies (21.3% vs. 11.3%) were identified with patient-reported data, but EHR data identified more women with atypia or lobular carcinoma in situ (1.1% vs. 2.3%). The EHR had missing data on race/ethnicity for 26.8% of women and on first-degree family history of BC for 87.2%. Opportunely, Fast Healthcare Interoperability Resources (FHIR), application programming interfaces (APIs), and new legislation offer an elegant solution for automated BC risk assessment that integrates both patient-generated health data and EHR data to harness the strengths of each approach. To increase the likelihood of developing disseminatable and equitable strategies that integrate EHR and patient-generated health data for risk assessment and personalized BC risk reduction, the focus of this project is to refine and test our approach among diverse multiethnic women. Our aims are: 1) upgrade FHIR integration of a patient decision aid, parse EHR data pulled by patients, develop a user interface to enable women to view and augment their EHR data, conduct user evaluations of the user interface; 2) assess the effect of the FHIR-enhanced patient decision aid on patient activation, risk perception, and usability in a pilot study of multiethnic high-risk women; and 3) identify multilevel barriers to implementing FHIR-enhanced patient decision aid into clinical care.
Alcohol use and smoking during pregnancy often occur in the context of maternal anxiety and depression. However, little is known regarding how various combinations of maternal alcohol consumption, smoking, and psychiatric symptoms interact to alter early central and autonomic nervous system (CNS, ANS) development and, in turn, how these changes underlie long-term neurobehavioral outcomes. The overall goals of the project are to a) comprehensively characterize maternal drinking, smoking, depression, and anxiety during pregnancy; b) perform deep phenotyping of electrocortical and physiological data from infants; c) relate maternal prenatal exposure clusters to deep phenotyping of CNS/ANS function and neurobehavioral outcomes; d) test the hypothesis that these measures mediate relationships between maternal prenatal exposures and subsequent child neurodevelopmental outcomes. To accomplish these goals, we will use existing data from an extensive study of mother-infant dyads recruited for the completed Safe Passage Study from recently completed follow-up studies which enrolled subsets of the PASS participants at 1 and 3 years of age. PASS participants were from geographically and socio-economically diverse regions in South Africa (SA) and the Northern Plains of the United States (US). The student(s) will be required to rapidly become familiar with a variety of data types, e.g., questionnaire data, physiological high-dimensional signals, neurobehavioral clinical examinations, neurodevelopmental assessments, etc. From a methodological point of view, the student(s) will support the development of: i) pipelines for the standardized and automated preprocessing and cleaning of physiological signals; ii) machine learning techniques (data-drive clustering, prediction) and iii) advanced statistical methodologies (inference, mediation, multidimensional data integration) tailored to the data available in this project. The student(s) will join a young and highly dynamic multidisciplinary team of experts, students, and RAs (some who originally joined the lab as DSI-students).