Developing Enhanced Prediction Models (PREDICT)

University of Pennsylvania

Status: Completed

Conditions: Diabetes; Pneumonia; Acute Myocardial Infraction or Chest Pain; Congestive Heart Failure (CHF); Chronic Obstructive Pulmonary Disease (COPD)

Study type: Observational

Funder types: Other

Identifiers

NCT02983812

825189

Details and patient eligibility

About

In this study, patients will be prospectively enrolled for data collection to design prediction models that integrate claims data (inpatient, outpatient, and pharmacy), electronic health record data (on clinical, social, and behavioral indicators), and patient-generated activity data. Patients will be randomized to use either a smartphone or a wearable activity tracking device to capture patient-generated health data.

Full description

Many hospital readmissions could be prevented if higher risk patients were identified and effective interventions then targeted towards these individuals. However, most existing claims-based predictive models perform poorly and do not provide timely and actionable information. In this study, researchers will prospectively enroll patients for data collection to design prediction models that integrate claims data (inpatient, outpatient, and pharmacy), electronic health record data (on clinical, social, and behavioral indicators), and use wearable devices or smartphones to collect patient-generated data (physical activity and sleep patterns). Patients will be randomized to use either a smartphone or a wearable activity tracking device to capture patient-generated health data.

Enrollment: 500 patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

1. Be 18 years or older
2. Be able to provide informed consent
3. Be admitted to the Hospital of the University of Pennsylvania or Penn Presbyterian Medical Center
4. Have a smartphone or tablet compatible with activity tracking devices
5. Plan to be discharged to home

Exclusion criteria

Have no medical condition which prohibits them from ambulating or plan for any medical procedure over the next 6 months that would prohibit them from ambulating.

Trial design

500 participants in 2 patient groups

Activity Monitoring - Smartphone

Description:

Patients in the activity monitoring - smartphone group will be randomly assigned to track their data using a smartphone app (which collects step counts) for 6 months. There will be no intervention for either group, both are being passively monitored.

Activity Monitoring - Wearable

Description:

Patients in the activity monitoring - wearable device group will be randomly assigned to track their data using a wearable activity tracker (which collects step counts and sleep patterns/duration). There will be no intervention for either group, both are being passively monitored.