One of the most common health care interventions in any healthcare setting is a medication prescription. Unfortunately, up to 25% of outpatient prescriptions are associated with adverse drug events (ADEs). ADEs decrease patient health directly and can lead to non-adherence, which in turn has negative consequences.

The investigators recently conducted a pilot project in which the feasibility and potential utility of an information technology enabled outreach program for monitoring patients receiving an outpatient prescription was tested. This intervention involved the use of an interactive voice response system programmed to automatically call ambulatory care patients following a prescription. If the system identified a potential medication problem, a pharmacist was notified who contacted the patient, modified the therapy accordingly, and informed a physician when necessary.

The pilot project included 568 patients with diverse illnesses in two Canadian cities. High levels of patient and provider acceptability of the system were determined from 21 day interviews. The program identified 56 of 125 (45%) ADEs and 10 of 26 (30%) of primary non-compliance events. Very few episodes of ameliorable ADEs were observed.

The investigators feel these results justify a randomized control trial to assess the effectiveness of the intervention for improving patient centered outcomes. Ambulatory care patients receiving incident prescriptions for one of four conditions (hypertension, diabetes mellitus, depression, and anxiety) will be randomized to the intervention or routine care. The investigators expect that the system will reduce the severity and duration of outpatient ADEs and improve adherence to medication care.

One of the most common health care interventions in any healthcare setting is a medication prescription. On an individual and population level, outpatient medication use has the potential to improve health. Unfortunately, up to 25% of outpatient prescriptions are associated with adverse drug events (ADEs). ADEs decrease patient health directly and can lead to non-adherence, which in turn has negative consequences.

The investigators recently conducted a pilot project in which the feasibility and potential utility of an information technology enabled outreach program for monitoring patients receiving an outpatient prescription was tested. This intervention involved the use of an interactive voice response system programmed to automatically call ambulatory care patients following a prescription. If the system identified a potential medication problem, a pharmacist was notified who subsequently contacted the patient, modified the therapy accordingly, and informed a physician when necessary.

The pilot project included 568 patients with diverse illnesses in two Canadian cities (Montreal and Quebec City). The system successfully contacted 477 patients (84%). High levels of patient and provider acceptability of the system were determined from 21 day interviews. The program identified 56 of 125 (45%) ADEs and 10 of 26 (30%) of primary non compliance events. Very few episodes of ameliorable ADEs were observed in which the ADE severity could have been diminished with more appropriate management.

The investigators feel these results justify a randomized control trial to assess the effectiveness of our intervention for improving patient-centered outcomes. To determine whether the system reduces the severity and duration of outpatient ADEs and improves adherence to medication care at three and 12 months, ambulatory care patients receiving incident prescriptions for one of four conditions (hypertension, diabetes mellitus, depression, and anxiety) will be randomized to the intervention or routine care. As an exploratory analysis, the impact of the intervention on patient self efficacy and health services use will also be examined.

This study will help to define whether this promising intervention can improve patient health from both the individual patient and population health perspective. There is a wide scope of potential applications for this approach. This study will provide important information to guide these future applications beyond the specific question asked in this protocol.

Inclusion Criteria:

• spoken language is French or English
• physician is a user of MOXXI (electronic health record system) and patient has consented to be captured by MOXXI
• receiving a high-risk incident prescription
• medication insurance through Regie de l'Assurances Maladie du Quebec (RAMQ) insurance program

Exclusion Criteria:

• patient not appropriate for inclusion (as determined by prescribing physician)