The Effect of Predictive Analytics-Driven Interventions on Healthcare Utilization
Among high-risk Medicare Advantage members with congestive heart failure, a proactive outreach program driven by a claims-based predictive algorithm reduced the likelihood of an emergency department (ED) or specialist visit in one year by 20% and 21%, respectively. The average number of visits dropped as well, with a 40% reduction in the volume of ED visits and a 27% reduction in the volume of cardiology visits after the first year.
Rise of the Machines in Health Care?
Report from a Penn Machine Learning Workshop for HSR Researchers
Online Ratings of the Patient Experience: Emergency Departments Versus Urgent Care Centers
Anish K. Agarwal, Kevin Mahoney, Amy L. Lanza, Elissa V. Klinger,...
ABSTRACT [FROM JOURNAL]
Study Objective: Individuals increasingly use online rating platforms to rate and review hospitals. We seek to describe and compare publicly available online review content and ratings of emergency departments (EDs) and urgent care centers.
Methods: We analyzed Yelp reviews of EDs and urgent care centers to
...Popular Science Spotlights Penn Medicine Nudge Unit Chief
Comparison of Pharmacy Claims and Electronic Pill Bottles for Measurement of Medication Adherence Among Myocardial Infarction Patients
Shivan J. Mehta, David A. Asch, Andrea B. Troxel, Raymond Lim, Jennifer Lewey, Wenli Wang, Jingsan Zhu, Laurie Norton, Noora Marcus, ...
ABSTRACT [FROM JOURNAL]
Background: Medication adherence after myocardial infarction remains low. Pharmacy claims have typically been used to measure medication adherence, but electronic pill bottles may offer additional information.
Objective: The main objectives of this study were to compare the association of adherence measured by prescription claims and remote monitoring technologies with cardiovascular events.
Research Design: This study was a secondary analysis of a remote monitoring intervention to increase...
A Randomized Controlled Trial of Opt-in Versus Opt-Out Colorectal Cancer Screening Outreach
Shivan J. Mehta, Tanya Khan, Carmen Guerra, Catherine Reitz, Timothy McAuliffe, ...
ABSTRACT [FROM JOURNAL]
Objectives: Colorectal cancer (CRC) screening uptake is suboptimal, despite national efforts to increase screening rates. Behavioral economic approaches such as changing defaults may increase participation. We compare response rates to opt-in or opt-out messaging in mailed fecal immunochemical test (FIT) outreach.
Methods: This is a two-arm randomized controlled trial among 314 patients aged 50–74 years who had at least two primary care visits in the 2...
Advances in Alzheimer's Imaging are Changing the Experience of Alzheimer's Disease
Shana D. Stites, Richard Milne, and Jason Karlawish
ABSTRACT [FROM JOURNAL]
Neuroimaging is advancing a new definition of Alzheimer's disease (AD). Using imaging biomarkers, clinicians may begin to diagnose the disease by identifying pathology and neurodegeneration in either cognitively impaired or unimpaired adults. This “biomarker-based” diagnosis may allow clinicians novel opportunities to use interventions that either delay the onset or slow the progression of cognitive decline, but it will also bring novel challenges. How will changing the definition of AD from a clinical to a biomarker construct change the experience of...
Nurse Generated EHR Data Supports Post-Acute Care Referral Decision Making: Development and Validation of a Two-step Algorithm
Kathryn H. Bowles, Sarah J. Ratcliffe, Mary D. Naylor, ...
ABSTRACT [FROM JOURNAL]
Objective: Build and validate a clinical decision support (CDS) algorithm for discharge decisions regarding referral for post-acute care (PAC) and to what site of care.
Materials and Methods: Case studies derived from EHR data were judged by 171 interdisciplinary experts and prediction models were generated.
Results: A two-step algorithm emerged with area under the curve (AUC) in validation of 91.5% (yes/no refer) and AUC 89.7% (where to refer).
Discussion: CDS for...
Inclusion of Unstructured Clinical Text Improves Early Prediction of Death or Prolonged ICU Stay
Gary E. Weissman, Rebecca Hubbard, Lyle H. Ungar, Michael O. Harhay, Casey S. Greene, Blanca E. Himes, ...
Abstract [from journal]
Objectives: Early prediction of undesired outcomes among newly hospitalized patients could improve patient triage and prompt conversations about patients' goals of care. We evaluated the performance of logistic regression, gradient boosting machine, random forest, and elastic net regression models, with and without unstructured clinical text data, to predict a binary composite outcome of in-hospital death or ICU length of stay greater than or equal to 7 days using data from the first 48 hours of hospitalization.
Design...
The Use of Individual Provider Performance Reports by US Hospitals
In the Journal of Hospital Medicine, Joshua Rolnick and Kira Ryskina analyze trends in how hospitals use the electronic health record to track and provide feedback on provider performance. The authors use data from a nationally representative survey of US hospitals from 2013 to 2015 to identify hospitals that have used electronic data to create individual provider performance profiles (IPPs). They model how the odds of IPP use differ as a function of hospital characteristics, including ownership (non-profit, for-profit, or government), geographic region, teaching versus...
Aravind Menon Joins LDI as Associate Director of Research
Accuracy of Billing Codes Used in the Therapeutic Care of Diabetic Retinopathy
Marisa Lau, Jonathan L. Prenner, Alexander J. Brucker, Brian L. VanderBeek
In JAMA Ophthalmology, Marisa Lau and colleagues, including Brian VanderBeek, seek to determine the accuracy of diagnostic, procedural, and therapeutic billing codes used in the treatment of diabetic retinopathy. As insurance billing claim databases represent a growing field of scientific inquiry within ophthalmology, validating the accuracy of billing claim codes is of increasing importance. The authors conducted a retrospective medical record review at three clinical practices, examining each patient’s billing data and medical record, measuring positive and negative predictive...
Video Analysis of Factors Associated With Response Time to Physiologic Monitor Alarms in a Children’s Hospital
Christopher P. Bonafide, A. Russell Localio, John H. Holmes, Vinay M. Nadkarni, Shannon Stemler, Matthew MacMurchy, Miriam Zander, Kathryn E. Roberts, Richard Lin, Ron Keren
In JAMA Pediatrics, Christopher Bonafide and colleagues, including John Holmes and Ron Keren, seek to identify factors associated with nurses' response time to physiologic monitor alarms at the bedside. As nurse response time to bed alarms remains slow, the authors examine patient- and nurse-related factors that affect responses to alarms. The authors video recorded 551 hours of care administered by 38 nurses to 100 children. They find several variables that shorten nurses’ response time to alarms. These include if the patient was on complex care service, if family members were...
Treatment seeking as a mechanism of change in a randomized controlled trial of a mobile health intervention to support recovery from alcohol use disorders
Joseph E. Glass, James R. McKay, David H. Gustafson, Rachel Kornfield, Paul J. Rathouz, Fiona M. McTavish, Amy K. Atwood, Andrew Isham, Andrew Quanbeck, Dhavan Shah
In Journal of Substance Abuse Treatment, Joseph Glass and colleagues, including James McKay, assessed the efficacy of an Addiction-Comprehensive Health Enhancement Support System (A-CHESS) in increasing the use of services for addiction. This model is an electronic health framework that has been applied to a diverse set of health problems, including pediatric asthma and breast and lung cancer, and was previously adopted to alcohol addiction in a randomized-controlled trial. The authors conducted secondary data analyses of this trial, including 349 adults with alcohol use disorders...