Organization

Administrative Organization

The study will be administered by a steering committee which includes Drs. Ira Ockene, James Hebert, Judith Ockene, Abigail Adams, and Robert Nicolosi, and Mr. Philip Merriam. This committee will meet weekly during year 1 and bi-weekly during years 2 and 3.

Project Management, Pt and Physician Tracking, and Data Management Procedures (Summary)

As noted in section D.8.4.1., this study will use Lotus Notes for tracking all pt and physician data. This system is ideal for monitoring the progress of study pts and physicians, and alerts the project manager and other study personnel to all study data collection points. The software will also be used to gather and store data which can be shared by individuals both on-site and in remote locations. This includes 24HR call assignments and the status of these and other data which are collected at remote sites and downloaded directly to the Lotus Notes system (see appendix O).  It also provides reports of data such as those from the 7DDR, other pt questionnaires, and blood lipids. Storage of such data provides instant access to the pt’s diet and other critical information for interventions by the DLMCs. Multiple levels of password protection are utilized to ensure data security. This type of tracking system has been in use for 3 years in other division studies and has proven to be easily learned, reliable, and efficient. (74)

Systems for Patient and Provider Contact

Most of the contact required will be fixed to a time of enrollment or related to an event (e.g., completion of a form or attendance at a class). The tracking system will allow notification of study staff to carry out such activities as scheduling of pts for endpoint  measurements, or identification of physicians due to receive post-training questionnaires. Additionally, the tracking system will identify pts who have completed  assessments or attended classes. This information is necessary in order for the DLMC and the site coordinator, both of whom have access to the tracking database, to make informed and timely contact with study participants.

Time Line

Phases of the Investigation

There are six phases in the sequence of the investigation, as shown in table D.3.

Table D.3. – Study Timeline

Recruitment

The internists see large numbers of pts, usually 60-70/week. A pool of some 1740 pts/wk (60x29) results in a potential population of 49 eligible pts/wk by the most conservative calculation (1740 x.25 (highest 25% LDL) x.8 (allowing for 20% refusals) x .85 (%pts age 20-75) ÷ 3.0 (avg. # visits/ yr/individual) x .5 (conservative estimate of those excluded for conditions listed in D.5.1.). Thus recruitment (12.5/wk) should not be limited by the number of available pts. However, the screening effort required to recruit the necessary number of pts is considerable. Based on our experience in WATCH (which was exactly in line with our projections) we estimate it will be nec­essary to do 7850 initial fingerstick cholesterol measurements to recruit 1200 subjects (15%). The projections are based on 25% of people being LDL-eligible, and additional  losses related to regression towards the mean and refusal to participate/no show on the 2nd visit). Some 10,000 telephone contacts (requiring 33,800 call attempts) will need to be made to result in 7850 initial cholesterol screens, allowing for loss from refusals, ineligible pts, and no-shows. We will utilize automatic dialing equipment to expedite this process. The study is screening-intensive, and we believe that a 24-month recruitment period is necessary to carry out such an effort.

Patient Follow-up, Potential Patient Losses, and Measures To Ensure Complete Data Collection

PCHIS obtains additional data via electronic links with other UMMC systems. Demographic updates are obtained from the Meditech registration module. Future appointments are obtained from the IDX scheduling system. Lab results are obtained through Meditech’s Patient Care Inquiry module. We will utilize PCHIS as the front end into appointment, registration, and nutrition referral information to be used both by study recruitment personnel and by the SBLMP. PCHIS can also validate and cross-check pt information, ensuring data quality.

Patient Losses:  Loss of pts can occur via illness or death, or refusal to participate further. As this is a primary care population without CHD, the recruitment is from routine, non-emergent visits, and pts over age 75 are excluded, loss to death or serious illness should be minimal (in WATCH 7 pts (.05%) died during the study). Refusal to finish the study also should be modest, as the two initial visits required to become a participant should uncover most pts who will not follow-up appropriately (in WATCH 91 pts dropped out during the study). Placement of pts on lipid-lowering medication requires special consideration. Few pts are likely to be placed on such therapy, and Pts with serum total cholesterol >300 mg/dl or LDL-C >200 mg/dl are excluded, and the guidelines taught and provided to physicians suggests consideration of drug therapy during the study year only for pts with LDL-C over the 90th percentile, and then only after the 6-month mark if dietary treatment has failed. Despite marketing pressure to increase utilization of lipid-lowering agents, there is considerable resistance to their use. (71, 108) In WATCH only 10.7% of pts in condition III (the control state in the present proposal) were placed on lipid-lowering medications during the study year, despite multiple reminders of the pt’s lipid levels. Pt placement on such medication will be considered as a control variable in analyses.

Human Subjects

Two consent processes are required in this study. Pts will initially be invited for a screening fingerstick cholesterol via a telephone call from a member of the research staff. The research assistant will briefly ex­plain the nature of the study, obtaining oral consent, and will also explain that a high cholesterol level will result in further information. If the pt's screening cholesterol is elevated, he or she will have an immediate repeat fingerstick which will be analyzed using the Cholestech lipid profile cassette. If the pt remains eligible by LDL-C and other criteria, full informed consent will be obtained by the research site coordinator, who will explain the project, note the need for several blood tests and for the pt to fill out a number of survey instruments, and explain potential risks (the small risk of a venipuncture) and benefits (improved management of cholesterol, both for the individual (half randomized to the systems-based intervention) and for society.

In general, the study is low-risk, with most of the necessary study blood tests (two fingerstick and three venous measurements over the 12 month period of a patient's involvement) being studies which are clini­cally indicated by the National Cholesterol Education Program guidelines. Institutional Review Board approval for this continuation of our previously approved WATCH study is presently pending.