Accurately identifying drug exposure is critical in pharmacoepidemiology studies. However, in the setting of sleep medications, there is no empirical evidence on the consistency of providers in using the ‘as needed’ designation (PRN) when prescribing these drugs or the impact of PRN prescribing on actual usage patterns by patients. Identifying patient characteristics associated with PRN prescribing and use will help us better understand the impact of PRN instructions, patterns of use and may inform the management of patients prescribed sleep medications.

The primary objectives of this study are to generate descriptive statistics regarding the proportion of sleep medications prescribed PRN, actual usage patterns of PRN vs non-PRN prescription and to characterize demographics, co-morbidities, co-medications of patients prescribed medications PRN vs. non-PRN.

PI: Paul Dexter, MD

RI Team Members: Nate Ring, Rachel Fuhr

Merck Team Members: Zhiwen Liu, Mark Marscio

Partners: Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc.

This is a retrospective cohort study using structured data (e.g., claims) and unstructured data (e.g., provider notes). The study population will be a cohort of patients extracted from the Indiana Network for Patient Care (INPC); patients will be limited to those with both structured and unstructured data in the INPC between January 1st 2009 and December 31st 2014.

This study does not have traditional exposure and outcomes but rather characterizes prescribing and usage patterns of sleep medications. Only descriptive data will be generated from this exploratory analysis. Patients with and without PRN prescriptions will be characterized and metrics on the performance of algorithms developed to identify patients will be reported. No formal statistical comparisons or hypothesis tests will be performed.

Start date: 2/1/2016 End date: 1/31/2017

The study team protocol has been reviewed and approved by the Investigational Review Board (IRB) and is being reviewed by the Merck Document Review Committee (DRC).
Updates on the project can be found at the study’s Wiki page:

Center for Biomedical Informatics