Supporting Post Myocardial Infarction (MI) Risk Modification Intervention Via Telemedicine Evaluation

PURPOSE OF STUDY Cardiovascular disease (CVD) is the leading cause of death in the U.S., accounting for over one third of deaths. Of these, coronary heart disease (CHD) accounts for the greatest proportion of morbidity and mortality with an estimated 770,000 Americans having a first myocardial infarction (MI) and an additional 430,000 having a recurrent MI in 2008.

To date, much of the focus on improvement of evidence-based medical therapy has been on the 'in-hospital' care phase, yet the greater challenge in improving CHD outcomes is in the transition from hospital to home.

We are testing two novel CVD risk reduction interventions for providing tailored disease-management support for patients who recently had a MI. The first intervention will utilize home BP monitors for all in this intervention arm, an interactive web-based tool based on Microsoft's HealthVault technology (for data transmission, tracking and communication), and nurse telephone disease-management to expand CVD risk management beyond the walls of the hospital and clinics. This intervention builds on our prior extensive experience in hypertension management including randomized interventional evaluation in over 2000 patients. This proven approach is now being enhanced with novel technology and a broader application namely secondary prevention. Combined, it has the potential to enhance post MI access, and quality of care, and ultimately, to improve patient outcomes. The second intervention, similar to the other invention arm, will utilize home BP monitors the same interactive web-based tool based on Microsoft's HealthVault technology (for data transmission, tracking, and communication), however, the second intervention arm will include a web-based disease-management to expand CVD risk management. Thus, the two intervention arms will be similar with the difference being whether individuals receive feedback from a nurse using a tailored behavioral telephone based intervention or use a tailored web-based intervention without direct communication with a nurse. The two arms will allow us to test the clinical impact and cost of using a tailored web-intervention as compared to having nurses implement a tailored intervention.

DESIGN AND PROCEDURES The study is a three-arm design of up to 600 patients who were hospitalized post-MI are recruited from a large hospital and randomized to either the education group (control group) or one of the two intervention groups. Patients randomized to one of the two intervention groups will receive a home BP monitor, trained in it's use and will have access to an interactive web-based tool based on Microsoft's HealthVault technology (for data transmission, tracking and communication). Those in the intervention arms will receive either a nurse telephone disease-management intervention or web-based disease-management intervention. Those randomized to the educational control arm, will receive usual care and information on CHD. The 12 months effects of the intervention will be evaluated.

For baseline assessments of the outcomes, the study will obtain BP, nonfasting LDL, and Hb A1c. Patients will also be surveyed about demographics and health behaviors during the baseline and 12 months. After the patient has completed the measurement battery, patients will be randomly assigned to the educational control group or the intervention group, respectively. The project's masters-level statistician will perform the entire study randomization before patient enrollment begins. Participants will be randomized using a computerized random number generator in blocks (size <10) within a single stratification factor, diabetic status. Additionally, because there are two nurse interventionists delivering the intervention in the nurse-intervention arm, patients will be randomly allocated to a nurse in a blocked fashion to ensure approximate even distribution of patients between nurses. At all follow-up measures, the RA assigned to measure BP and obtain survey information will not be blinded as to which group the patient is randomized.

If patients are not able to complete baseline assessments including BP, LDL-Direct and/or Hb A1c, the RA will reschedule individuals' appointment to complete the interview at a later date. Comorbidities will be obtained from individuals' medical records over the 12 months of the study.

All patients randomized to the two intervention arms will be provided an easy-to-use home BP monitor . The primary means of BP monitoring will be through the plug and play uploads of their readings to the Microsoft HealthVault application. Participants are asked to take blood pressure readings daily and to upload values to HealthVault weekly. If diabetic, participants are asked to upload glucose values when taken (as recommended by their provider). If participants are unable to upload their readings to the Microsoft HealthVault application, the RA will instruct participants on how to enter values into the application manually. If participants are unable or unwilling to use HealthVault, the study will provide the participant with a document to record their home monitor blood pressure and glucose readings and a self addressed stamped envelope. Every two weeks (bi-weekly), the study will ask the participant to send the readings back to the study staff. The RA will then enter home monitor blood pressure and glucose readings into the Microsoft HealthVault application. In addition to patients' home monitoring information, patients' lab values for non fasting LDL, Hb A1c as well as important safety monitoring parameters will be manually entered into the study's data repository.

SUBJECT RECRUITMENT AND COMPENSATION We will enroll up to 600 subjects who were admitted to Duke University Medical Center (DUHS) for a MI. Recruitment will occur within approximately a week to 36 months after hospital discharge. Subjects will be randomly assigned to be in the educational control group or one of the two intervention groups respectively. At least 50% of our sample will be women and 40% will be African American based upon our prior hypertension studies using similar enrollment strategies. Subjects will receive compensation for participation in the study.

CONSENT PROCESS The Research Assistant (RA) will contact the subject within approximately a week to 36 months after hospital discharge. Patients will receive a letter detailing the study after hospital discharge and will have the opportunity to opt out of the study. If the patient does not opt out, the RA will contact the patient after discharge from the hospital, screen for eligibility, describe the study, obtain informed consent and interview the patient.

SUBJECT'S CAPACITY TO GIVE LEGALLY EFFECTIVE CONSENT Subjects not competent to give consent are not included in the study.

STUDY INTERVENTIONS In the nurse intervention arm, the research nurse (RN) will contact the patient within one week of enrollment for the purpose of introduction. The schedule of intervention modules will begin at the initial phone call from the nurse. In the web intervention arm, participants will receive an e-mail within one week of enrolling with a link to a secure web-page were the intervention can be activated. A unique aspect of the intervention is that while some modules are activated at every encounter, other modules will only be activated at specific encounters when needed.

The activation frequency of each module (multiple modules make up an encounter) can vary. The first modules (medication and side effects) can be potentially activated for all intervention patients every month if a problem/issue arises (e.g., a medication change has occurred or a patient reports a new medication side effect). The remaining modules (patient/provider communication, CVD knowledge/risk perception, and health behaviors) can only be activated for individuals at specific encounters. Patients are also able to telephone the nurse with questions related to the management of their diseases. Individuals in the web-based intervention can e-mail questions to the study team. Should major emergent health care issues arise during these, or any calls or e-mail communications, the nurse can immediately notify patients' primary care provider. In the case of e-mail communications, information will be forwarded to participants' providers.

A multibehavioral, comprehensive approach is proposed because no one factor has been shown to consistently improve CVD outcomes. An important feature of the intervention is that it incorporates tailored information and feedback that is specifically relevant to a particular patient. Previous studies have shown effectiveness using an early version of this intervention, addressing up to 13 health behaviors focused on improving patient management of CVD and related health behaviors. Patients will be provided evidence-based recommendations regarding lifestyle behaviors and will be advised on how to achieve their goals with respect to these behaviors. In the nurse arm, verbal information will be reinforced with written and visual material mailed to the patient. In the web-based arm, these materials will be sent to patients via e-mail. All intervention components are designed to be culturally sensitive. The behavioral modules include, diet, exercise, smoking, alcohol, stress reduction, memory, patient-provider relationship, medication update, side effects, and knowledge/risk perception. In addition, those patients using the Microsoft HealthVault system will have access to further on line patient educational materials prepared by the America Heart Association (AHA).