Three Approaches to Glucose Monitoring in Non-insulin Treated Diabetes

Within the context of a controlled and randomized but pragmatic trial, we seek to answer the
following question: does SMBG testing make sense for Non-insulin treated (NIT) diabetes
mellitus (DM) patients in terms of either A1c values or Health-related Quality of Life (QOL).
We will also examine whether or not patients with different baseline characteristics might
see different outcomes from three SMBG testing approaches. To achieve our goal of testing
SMBG options in community-based primary care practices, the study has been designed to
minimize as much as possible any interruptions in or alterations to standard daily patient
care. We will recruit 450 patients from five primary care practices. Patients will be
followed for one year. During routine clinic visits, their health care providers will be
guided to modify therapies based on American Diabetes Association (ADA) guidelines, which
focus on A1c values and SMBG values if available.

Study Arms Two of the three study arms were chosen to reflect existing, widely used options
for patients with NIT DM. Arm 3, described below, of this trial is particularly novel in that
it tests the potential benefit of new technology now available to these patients; namely,
wireless transmission of blood glucose test results accompanied by instantaneous tailored
feedback. While this option may hold great promise for patients and their providers, data
supporting this are lacking. The results of this study will inform consumers and providers
about whether this new technology can support efforts to alter behaviors in a way that
results in improved A1c values or better quality of life. Particularly appealing is that this
approach employs technology that supports more efficient use of SMBG values by both patients
and their providers. In terms of their pragmatic 'fit', all three groups fit easily into
current practice at the participating practices. For example, they do not require extra
clinic visits (other than possibly the initial recruitment assessment), and while staff will
need to download glucose reports for patients in Arms 2 and 3, this will be a very simple,
streamlined procedure. In terms of risks to patients, none of the three groups puts patients
at any additional risk, because all three are versions of 'standard practice' for this
patient population and doctors are free to alter therapies and testing schedules as they
normally would.

- Arm 1: No SMBG Testing This group represents an approach to SMBG that is supported by
research and followed by many primary care providers. Patients randomized to this group
will not be provided a meter, but will receive standardized education detailing the
signs and symptoms of hyper- and hypoglycemia and information regarding lifestyle
approaches for the maintenance of normoglycemia. If they have previously been obtaining
SMBG values, they will be asked to discontinue. Healthcare providers and staff will be
alerted that these patients are in the no testing arm and will be encouraged to maintain
treatment fidelity except where patient safety is a concern. Study coordinators will
provide patients with both written and oral educational information supporting the
approach of no SMBG testing, with a view toward reconciling any messages they may have
received in the past or will receive in the future regarding SMBG. This education will
recur every 3-6 months at routine clinic visits (the recommended follow-up frequency for
persons with T2DM) by their health care providers. A1c will be obtained during these
visits as per usual care and providers will adjust medication and lifestyle
recommendations as they see fit. Providers will be given an algorithm based on current
ADA recommendations relating to A1c results to use as a guide for treatment
modification.

- Arm 2: Once Daily Testing with Standard Feedback This group represents an approach taken
by a large proportion of NIT DM patients in the US. At the assessment visit, patients
will be given the study glucose meter and testing supplies, along with instruction for
their use. They will be asked to test once daily and at different times of the day on a
rotating basis, in order to obtain a representative distribution of blood glucose values
over time. The meter will have wireless capability such that it will transmit both the
SMBG result and the time of testing to a secure server. Subjects will also note the
circumstances of their test times on the glucose meter (e.g., fasting, 2 hours post
meal, etc.) and this information will also be sent to the server. As with Arm 1, Arm 2
patients will receive training in hyper- and hypoglycemia and approaches for maintaining
normoglycemia. They will also receive training and written material instructing them in
how to interpret their SMBG readings and actions they might take in response to those
values. SMBG results will be reviewed at all clinic visits (3-6 month intervals) by the
primary care provider using a print out of glucose meter values downloaded and analyzed
wirelessly. During the early phase of the project, a streamlined system will be
developed to allow participating clinic staff to download the SMBG values with minimal
effort and create the accompanying SMBG report for the patient's provider. This
customized report, detailing blood glucose averages, ranges and standard deviations,
will be generated for all group 2 participants and given to their provider prior to each
routine clinic visit. The report will identify problematic glucose values (hypo- and
hyperglycemia) and suggested approaches for addressing these problems. Providers will
review the results with the participant and make appropriate treatment recommendations
based upon blood glucose trends. The recommendations in the report will be driven by an
algorithm for treatment modification based upon ADA guidelines which will include
suggestions for both lifestyle and pharmacotherapy adjustments. At the end of each
clinic visit, patients will be instructed by their primary care provider to continue
once daily testing. Using shared decision-making and standardized testing
recommendations, the health care provider and the patient will jointly determine the
testing approach (e.g., timing of once daily testing) the patient will employ over the
next monitoring period.

- Arm 3: Once Daily Testing with Enhanced Feedback This Arm represents an enhanced,
patient-centered approach. Participants randomized to Arm 3 will, like the Arm 2
participants, be instructed to test once daily and at different times of the day, and
their glucose values will be transmitted wirelessly to a server. However, for Arm 3
participants, the blood glucose values will be analyzed automatically each time a value
is transmitted. Based upon the value, instantaneous personalized feedback will be
provided via wireless messaging to the patients' glucose meters. These messages will not
include suggestions for pharmacological therapy modification but will be motivational in
nature, including suggestions to enhance efforts at lifestyle management and adherence
to pharmacologic therapy if prescribed. At weekly intervals, an analysis of the SMBG
values from the previous week will be sent to the glucose meter highlighting trends and
identifying specific problematic times. Subjects will also receive messages reminding
them of upcoming clinic visits (beginning 1 week prior to their scheduled visit). As
with Arm 2, SMBG results and customized reports will be downloaded and reviewed at all
clinic visits by the primary care provider. In addition, a summary of the messages sent
wirelessly to the patient since the most recent visit will be included. Like the glucose
values report, this message report will be given to all providers at the beginning of
the clinic visit and providers will be asked to also review these results with the
patient. Providers will make appropriate treatment recommendations based upon blood
glucose trends and messages. Report recommendations will be driven by an algorithm for
treatment modification based upon ADA treatment guidelines and include suggestions for
lifestyle and pharmacotherapy adjustments. At the end of the clinic visit, patients will
be instructed by their primary care provider to continue testing their blood glucose
once daily and collaboratively determine the testing approach (e.g., timing of once
daily testing) the patient will employ over the next monitoring period.

Potential Alterations to the Testing Recommendations: Providers will be strongly encouraged
to maintain SMBG testing fidelity based upon the treatment arm to which a patient is
randomized; however, if a provider believes that testing should occur more frequently due to
concern for serious unrecognized hypo- or hyperglycemia or for any other reason, testing
recommendations may be modified. This approach provides for patient safety and preserves the
pragmatic nature of the trial. Glucose testing strips will be provided to participants on a
quarterly basis. If a provider deems that SMBG testing should occur more frequently than what
is recommended to the patient based on the study arm to which he or she has been randomized,
the health care provider must write the prescription for the additional test strips and the
strips must be paid for out-of-pocket or through insurance. Following an intention to treat
model, patients who opt not to follow the testing regimen to which they were randomly
assigned will be encouraged to remain in the study. Any deviations from the assigned SMBG
testing protocol will be readily known via the glucose meter reports.

Mixed methods approach. A mixed methods approach in which both qualitative and quantitative
data are collected provides advantages when exploring complex research questions such as the
ones posed in this proposal. The quantitative data will allow us to assess changes in
objective measures, while the qualitative data will provide a deeper understanding of
patients' and providers' experiences with the various components of SMBG testing arms (e.g.,
the meters, SMBG results, personalized messaging, downloadable reports, treatment algorithms,
etc.) It is also the case that, while some patients are interested in seeing hard facts and
figures, others prefer testimonials from patients like themselves. The mixed methods approach
enables us to provide both types of information, increasing the effectiveness and uptake of
our dissemination efforts.

Recruitment and randomization procedures. Patient Recruitment: All potentially eligible
patients expressing interest in the study will complete an initial screening phone call. The
call will take about 15 minutes, during which a member of the study team will describe the
study, answer questions, and conduct a short set of simple screening to determine
eligibility. Eligible patients who remain interested in participation will complete an
assessment visit with a research coordinator. To decrease patient burden and further engage
participating practices, assessment visits will occur at the patient's primary care office.
Assessments will be separate from their appointment with their primary care provider, and may
or may not occur on the same day as a regularly scheduled clinic visit, though for patient
convenience we will make every effort to coordinate the assessments with a regular clinic
visits. During these assessments, the research coordinator will review the study details in
greater depth, verify all inclusion and exclusion criteria, and obtain written informed
consent.

Patient Randomization: After providing informed consent, baseline A1c, and completing all
baseline study questionnaires, participants will be randomized to one of the three arms using
sequentially numbered, opaque, sealed envelopes. The research coordinator will review the
treatment assignment with the patient, using a standardized script, provide the training and
supplies necessary for participation in that study arm and answer any remaining questions.

Inclusion Criteria:

- A diagnosis of type 2 diabetes

- Age > 30

- An established patient at the participating UNC Physicians Network (UNCPN) practice
who identifies a UNCPN health care provider within that practice as their primary
provider of diabetes care.

- A1c >6.5% but <9.5% within the 6 months preceding the screening call/visit, as
obtained from the patient's medical record.

- Willing to comply with the results of random assignment into a study group.

Exclusion Criteria:

- Currently sees or plans to see an endocrinologist or other diabetes specialist in the
next year.

- Use of insulin

- Is or plans to become pregnant in the next 12 months.

- Plans to relocate in the next 12 months.

- Has other conditions (e.g. renal or cardiovascular disease, poor visual acuity), other
factors (e.g. frailty,) or comorbidities (e.g. cancer) that might put the patient at
risk when following study protocols.

- No history of significant issues with known or suspected hypoglycemia or any history
of "severe" hypoglycemia (requiring assistance from a third party).