Reducing the Unnecessary Use of Heavily Marketed Medications: A Randomized Controlled Trial

Summary

Prescribing decisions by clinicians are often thought to be simple: a patient's clinical problem leads a prescriber to choose the optimal treatment. However, many factors other than the patient's condition affect prescribing decisions, including the marketing of pharmaceuticals. Clinicians are subjected to direct "detailing" by representatives of the pharmaceutical industry, advertisements in medical journals and requests for specific treatments from patients, who are increasingly exposed to direct-to-consumer (DTC) advertising. These influences, often based on biased or inaccurate information, contribute to a variety of problems in prescribing, including the unnecessary use of expensive, heavily marketed medications.

Overcoming these influences requires innovative approaches. The movement toward widespread adoption of electronic health records (EHRs) and electronic prescribing presents new opportunities to educate both clinicians and patients at the time of medication prescribing. This project, endorsed by the AHRQ-supported Centers for Education and Research on Therapeutics (CERTs; www.certs.hhs.gov) and the U.S. Food and Drug Administration (FDA), aims to test the effectiveness of computerized prescribing alerts and state-of-the-art educational outreach to reduce the unnecessary use of heavily marketed medications. A second goal is to improve clinicians' knowledge of industry marketing practices, so that they can more effectively assess information provided by drug companies. Thus, the study has two specific aims:

Specific Aim 1: To assess whether computerized prescribing alerts linked electronically to patient educational material can reduce prescribing of heavily marketed medications.

Specific Aim 2: To assess whether group academic detailing increases clinicians' knowledge about industry marketing practices and increases the effect of prescribing alerts.

Conditions

• Use of Sleep Medications

Interventions

BEHAVIORAL

Computerized alerts

Computerized Clinical Decision Support to clinician at the time of prescribing

BEHAVIORAL

Alerts Plus Detailing

Computerized Clinical Decision Support to clinician at the time of prescribing PLUS one group academic detailing session

Sponsors & Collaborators

• Oregon

  collaborator OTHER

• VA Boston Healthcare System

  lead FED

Principal Investigators

• Steven R Simon, MD · Harvard Medical School and Harvard Pilgrim Health Care

Study Design

Allocation

RANDOMIZED

Purpose

HEALTH_SERVICES_RESEARCH

Masking

NONE

Model

PARALLEL

Eligibility

Sex

ALL

Healthy Volunteers

Yes

Timeline & Regulatory

Start

2007-03-31

Primary Completion

2008-03-31

Completion

2008-03-31

Countries

• United States

Study Locations