J. Marc Overhage - Academia.edu (original) (raw)

Uploads

Papers by J. Marc Overhage

Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems

Artifcial intelligence (AI) supported clinical decision support (CDS) technologies can parse vast... more Artifcial intelligence (AI) supported clinical decision support (CDS) technologies can parse vast quantities of patient data into meaningful insights for healthcare providers. Much work is underway to determine the technical feasibility and the accuracy of AI-driven insights. Much less is known about what insights are considered useful and actionable by healthcare providers, their trust in the insights, and clinical workfow integration challenges. Our research team used a conceptual prototype based on AI-generated treatment insights for type 2 diabetes medications to elicit feedback from 41 U.S.-based clinicians, including primary care and internal medicine physicians, endocrinologists, nurse practitioners, physician assistants, and pharmacists. We contribute to the human-computer interaction (HCI) community by describing decision optimization and design objective tensions between population-level and personalized insights, and patterns of use and trust of AI systems. We also contribute a set of 6 design principles for AI-supported CDS.

Merged Result Retrieval Merged Result Retrieval ED Charge Data ED Charge Data

Pediatrics, 2020

BACKGROUND: The time providers spend using their electronic health records (EHRs) delivering care... more BACKGROUND: The time providers spend using their electronic health records (EHRs) delivering care and its potential impact on patient care are of concern for the health care system. In studies to date, researchers have focused on providers who primarily care for adults. Scant information exists for pediatricians. Given this gap, it is important to quantify EHR activity for this group. METHODS: We studied pediatricians practicing in US-based ambulatory practices using the Cerner Millennium EHR by extracting data from software log files in the Lights On Network for the calendar year 2018 and summarizing the time spent on each of 13 clinically-focused EHR functions according to clinical specialty. RESULTS: Our data included >20 million encounters by almost 30 thousand physicians from 417 health systems. Pediatric physicians spent an average of 16 minutes per encounter using their EHR. Chart review (31%), documentation (31%), and ordering (13%) functions accounted for most of the tim...

NEJM Catalyst, 2020

Traditional health care providers can make the digital revolution work for them by creating new w... more Traditional health care providers can make the digital revolution work for them by creating new ways to improve convenience and lower costs for routine primary care.

American Journal of Health-System Pharmacy, 1999

A method for rating the value of pharmacists' clinical services was studied. An instrument was de... more A method for rating the value of pharmacists' clinical services was studied. An instrument was developed to measure the severity of medication errors and the value of pharmacists' clinical interventions. Pharmacists at a hospital pharmacy department used the instrument at the time they made an intervention. A single pharmacist reviewed and adjusted the scores assigned by the pharmacist who made the intervention. An expert panel consisting of two clinical

AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium, 2011

Formularies are highly variable, which limits physicians ability to prescribe cost effective medi... more Formularies are highly variable, which limits physicians ability to prescribe cost effective medications for their patients. In this study we explore the composition of 3,346 formularies in terms of outpatient prescription coverage, medication class coverage, and cost implications. Our analysis revealed that 42% of formularies are duplicative and that the unique formularies contain variability for medications that contribute little in terms of cost or overall use. These results lead us to believe the number and complexities of formularies can be dramatically reduced leading to more intuitive decision support for physicians when writing prescriptions.

AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium, Jan 11, 2007

Over one hundred Regional Health Information Organizations (RHIOs) are under development in the U... more Over one hundred Regional Health Information Organizations (RHIOs) are under development in the United States. Many of these will fail but many will become a vital part of the Nationwide Health Information Network (NHIN). Documentation was reviewed and summarized and a core group of Vermont Information Technology Leaders (VITL, Inc.) were interviewed to ascertained lessons learned in the development of Vermont's RHIO. Issues were grouped into five major categories: early planning, organization, education and marketing, technology, and financial sustainability. There are a number of commonalities about all RHIOs but also a number of differences predicated on location. RHIOs must remain dynamic and learn from others in order to survive.

Medical Care, 1997

The purpose of the study was to estimate the 8-year rate of hospitalization for congestive heart ... more The purpose of the study was to estimate the 8-year rate of hospitalization for congestive heart failure (CHF), to report the resources consumed, and to evaluate previously reported risk factors in a nationally representative sample of 7,286 older white and black adults. Secondary analysis of baseline interview data was linked to Medicare hospitalization and death records for 1984 to 1991. Hospitalization for CHF was defined as having one or more episodes with an International Classification of Diseases (ninth revision, clinical modification) discharge code of 428. Combined and separate analyses of first-listed and second-through fifth-listed CHF discharge diagnoses were conducted. Multivariable proportional hazards models were used to evaluate the risks in pooled analyses of all white and black men and women and in separate stratified analyses of white men and white women. Over the 8-year period, 1,102 or 15.1% of the 7,286 older white and black adults were hospitalized for CHF (7.1% with first-listed and 8.1% with second- through fifth-listed diagnoses). The 1- and 5-year combined postdischarge mortality rates were 34.7% and 69.0%, respectively. In descending order, the major risk factors for being hospitalized for CHF in the combined, pooled analysis were age, being a white man, having lower body functional limitations, and having self-reported medical histories of coronary heart disease, heart attack, diabetes, and angina. The increased risk associated with age was not linear, and it diminished significantly over the course of life. Some significant differences were observed in the risk factors for hospitalization for first-listed versus second- through fifth-listed CHF and in the risk factors for white women versus white men. Hospitalization for CHF among older adults is a common, costly event with a poor prognosis. The differential risk for white men remains unexplained and warrants further study.

American Journal of Epidemiology, 2013

Clinical studies that use observational databases to evaluate the effects of medical products hav... more Clinical studies that use observational databases to evaluate the effects of medical products have become commonplace. Such studies begin by selecting a particular database, a decision that published papers invariably report but do not discuss. Studies of the same issue in different databases, however, can and do generate different results, sometimes with strikingly different clinical implications. In this paper, we systematically study heterogeneity among databases, holding other study methods constant, by exploring relative risk estimates for 53 drugoutcome pairs and 2 widely used study designs (cohort studies and self-controlled case series) across 10 observational databases. When holding the study design constant, our analysis shows that estimated relative risks range from a statistically significant decreased risk to a statistically significant increased risk in 11 of 53 (21%) of drug-outcome pairs that use a cohort design and 19 of 53 (36%) of drug-outcome pairs that use a self-controlled case series design. This exceeds the proportion of pairs that were consistent across databases in both direction and statistical significance, which was 9 of 53 (17%) for cohort studies and 5 of 53 (9%) for self-controlled case series. Our findings show that clinical studies that use observational databases can be sensitive to the choice of database. More attention is needed to consider how the choice of data source may be affecting results. database; heterogeneity; methods; population characteristics; reproducibility of results; surveillance

Advances in Pharmacoepidemiology & Drug Safety, 2014

We evaluated and compared different methods for measuring adherence to Oral Antihyperglycemic Age... more We evaluated and compared different methods for measuring adherence to Oral Antihyperglycemic Agents (OHA), based on the correlation between these measures and glycated hemoglobin A1C (HbA1c) levels in Medicaid patients with Type 2 diabetes. An observational sample of 831 Medicaid patients with Type 2 diabetes who had HbA1c test results recorded between

Archives of Internal Medicine, 1996

Hospitalizations are an opportunity to provide preventive care. To determine if computer reminder... more Hospitalizations are an opportunity to provide preventive care. To determine if computer reminders, which we have shown to be effective in our ambulatory care setting, increasing the provision of inpatient preventive care. Randomized, controlled trial on the general medicine inpatient service of an urban, university-affiliated public hospital. Study subjects were 78 house staff rotating on the 6 general medicine services. The intervention was reminders to physicians printed on daily rounds reports about preventive care for which their patients were eligible, and suggested orders for preventive care provided through the physicians' workstations. The preventive care guidelines were derived from the US Preventive Care Task Force recommendations. Compliance with preventive care guidelines and house staff attitudes toward providing preventive care to hospitalized patients were the main outcome measures. No significant differences were seen between intervention and control physicians in compliance with preventive care guidelines in the aggregate or when individual preventive care actions individual preventive care actions were analyzed. This was true even though most physicians endorsed providing most kinds of preventive care for hospitalized patients. Despite past success in increasing preventive care in the outpatient setting, we were unable, using a moderately intensive intervention, to increase the provision of preventive care during hospitalizations. The physicians providing care during the hospitalization were not the patients' primary care physicians, which proved to be an important barrier. More intensive interventions, or more direct linkages between inpatient and outpatient care providers, may be required to overcome this resistance.