Ongoing Pilot Projects
In the initial round of pilot funding , five pilot projects were funded in 2022 with the goal of using data from these seed projects as preliminary data for proposals for external funding to support a broader array of studies focused around a central theme in medical communication. In 2023, two additional pilots were funded. Read about the pilots below.
2024 Pilot Project:
Content Popularity and Trust
Testing the effect of content popularity on trust in health information within a clinical setting
PIs: Sharath Chandra Guntuku, Assistant Professor, Department of Computer and Information Science, University of Pennsylvania; and Anish Agarwal, MD, MPH, Assistant Professor, Perelman School of Medicine
Summary: This project will empirically test the role of popularity on perceived trust in online health information. Participants engaged in social media between the ages of 18 and 35 will be exposed to medically inaccurate information from medical professionals and influencers with metadata masked or unmasked in each case. They will then be asked about their intention to follow up on the health information presented in the message. Authors hypothesize that social media metadata (likes, shares, views, and comments) drive perceived trust and that influencers will have a disproportionately higher level of trust as compared to clinicians.
2023 Pilot Projects:
Online Health Information Searches
Online health information seeking: Capturing naturalistic search strategies and test strategies to restrict encounters with misinformation
PIs: David Lydon-Staley, Assistant Professor, Annenberg; and Melissa Mercincavage, Research Assistant Professor of Psychiatry, SOM.
Summary: Participants will be provided with 5 medical topics to search and randomized to a naturalistic search group or an experimental search group in which curated websites endorsed by the medical community will be provided. Online search behavior will be recorded to determine which websites participants visit, how long participants spend searching before coming to an answer, and to quantify the diversity of viewpoints encountered during their searches.
Augmented Reality for Health Messaging
Project CARE—Evaluating Community-based Augmented Reality Education posters to promote pediatric influenza vaccination
PIs: Katerina Girginova, PhD, Faculty, Co-Director of Annenberg VR Lab, Annenberg; Melanie Kornides, ScD, MPH, APRN, Assistant Professor of Nursing and Pediatrics, SON; Co-I: Jeffrey Vadala, PhD, Director Penn Neurology VR Lab, SOM; Andy Tan, PhD, M.P.H., M.B.A., M.B.B.S., Associate Professor, Director of Health Communication & Equity Lab, Annenberg; Terri Lipman, PhD, CRNP, FAAN, Professor Emerita, SON.
Summary: Augmented reality (AR) is a mobile-focused technology that allows users to see interactive web content overlaid onto real-world settings. AR can boost user engagement while providing them valuable contextual information, but it remains a largely untested communication medium for thedissemination of public medical messaging and specifically, vaccine hesitancy campaigns. This proposal will assess the development, feasibility, and acceptability of an augmented reality promoting the pediatric flu vaccine amongst marginalized communities of color in Philadelphia. in collaboration with community members, assesses the development, feasibility, and acceptability of an AR poster campaign promoting the pediatric influenza (flu) vaccine amongst marginalized communities of color – those with the lowest uptake – in Philadelphia
2022 Pilot Projects:
Assessing Misinformation in Targeted Health Care Advertising in a Pilot RCT
PIs: Matthew McCoy, PhD, Assistant Professor of Medical Ethics and Health Policy, SOM; Ari Friedman, MD, PhD, Assistant Professor of Emergency Medicine, SOM
Summary: When patients seek health information online, nearly all webpages that they visit contain tracking code that reports the specific pages they visit to online ad brokers. This process of “web tracking” allows ad brokers to infer information about patients’ health conditions and concerns based on their browsing histories. This information can be used by ad brokers to deliver targeted health care advertising to patients. Targeted advertising is minimally regulated and may be a vector for false or misleading health information. This proposal will develop and test an assessment tool designed to assess misinformation across domains of health care advertising. These domains include implantable devices, drugs/biologics, supplements, hospitals/health systems, clinical service providers, insurance products, and assistive devices.
Reducing Medical Bias
An Experimental Implementation Study to Evaluate the Effects of Networked Collective Intelligence on Medical Bias in a Medical Training Center
PIs: Damon Centola, PhD, Professor of Communication, Sociology, and Engineering; Raina Merchant, MD, Department of Emergency Medicine, SOM
Summary: Bias is an enduring cause of healthcare disparities by race and gender. This proposal will test whether altering the structure of peer networks among physicians-in-training can significantly alter the quality of their medical decisions—and their implicit norms of practice—to reduce bias in medical decision making. The investigators will conduct a clinical experiment using a supervisory staff of network science experts and senior medical staff to conduct evaluations of clinical cases drawn from resident clinicians’ patient visits. The hypothesis is that the residents will learn to evaluate their own clinical performance against objective feedback from peer networks outside of the current feedback structure, and that biases in their care delivery will be revealed. The investigators also hypothesize that the residents in the program will learn to think about peer networks as a way of enhancing their clinical acumen.
Increasing Vaccination Rates
Comparing Strategies for Countering Misinformation and Spurring COVID-19 Vaccination
PIs: Jessica Fishman, PhD, Director of Message Effects Lab, SOM/Annenberg; Gregory Bisson, MD, MSCE, Associate Professor of Medicine and Epidemiology; Melanie Kornides, ScD, MPH, APRN, Assistant Professor of Nursing and Pediatrics; Co-I: Joseph Cappella, PhD, Annenberg
Summary: This proposal will use “myth busting” strategies to increase vaccination rates in lower-income communities. Four messaging strategies will be compared to a control strategy. The four strategies are myth then fact, sandwich of fact/myth/fact, fact only, and only flagging the myth. These will be tested first online, with any strategy performing worse than control eliminated. The other messaging strategies will move forward in a field experiment in collaboration with the Philadelphia Department of Health. Each messaging strategy will be disseminated through printed materials that will be distributed by community health workers canvasing door-to-door, randomized by neighborhood. Neighborhoods with similarly low COVID-19 vaccination rates and sociodemographic characteristics, such as poverty rates, will be selected. Each neighborhood will also have a public health clinic providing $100 cash incentives for COVID-19 vaccination, to reduce financial barriers, such as the cost of transportation or missed work, that are present in this population.
Communicating with Caregivers
Communicating with Parents about the COVID-19 Vaccine: A Natural Language Processing Analysis of Public Opinion to Inform Clinical Guidance
PIs: Alison Buttenheim, MBA, PhD, Associate Professor of Nursing and Health Policy; Sharath Guntuku, PhD, Assistant Professor, Department of Computer and Information Science, SEAS
Summary: Parents are particularly susceptible to misinformation about vaccine development and approval processes, and about the risk-benefit profile of the COVID-19 vaccine for children. The investigators will apply natural language processing and topic modeling techniques to two large datasets in order to characterize emerging misinformation about COVID-19 vaccines for children, and to inform the development of clinician communication strategies. They will then develop prototype messages for pediatric providers and test these messages in parents of children under 12 through online panel services. The investigators plan to submit an extramural grant application to build and deploy a dynamic algorithmic engine that can identify emerging vaccine misinformation and rapidly translate insights into clinical guidance.
Improving Diversity and Inclusion
Testing the Impact of Appeals to Diversity and Inclusion in Pediatric Clinical Research Study Enrollment Videos: A Randomized Pilot Study
PIs: Kevin Johnson MD, MS, Professor of Informatics, Computer and Information Science and Communication, VP of Applied Clinical Informatics; Andy Tan, PhD, Associate Professor, Annenberg; Jessica Fishman, PhD, Message Effects Lab; Susan Furth, MD, PhD, Professor of Pediatrics and Epidemiology
Summary: Data derived from clinical care or research among people from diverse backgrounds is critical for advancing health equity. Unfortunately, not all patients benefit from research discoveries, because they are inadequately represented in the data that are available to researchers and healthcare providers. The goal of this proposal is to develop and test the impact of a video-based inclusion appeal on enrolling African American children into the research notification system at CHOP. Parents of pediatric patients will be randomized to complete a survey only, to view a usual-care video about clinical trials, or to view the same video with a tailored explanation about why it is important to enroll African American families for this trial. Results will inform a larger study to assess the impact of using video-based inclusion appeals on clinical trial enrollment among diverse groups who are underrepresented in pediatric clinical research.