Telemetric Glucose Data Acquisition During Initiation of Insulin Pump Therapy

The study is focused on the following objectives:

1. Determine whether the use of the GlucoDYNAMIXâ„¢ diabetes intervention system and the
GlucoMONâ„¢ telemetry device will facilitate faster regulation of blood glucose
excursions into a well-controlled range, compared to conventional monitoring and
reporting.

- hemoglobin A1c at the end of the 3 month study period and again after an
additional 3 month follow-up period to account for a possible for an extended
difference between the groups;

- magnitude of glucose excursions and fluctuations as indicated by the standard
deviation between glucose tests to indicate relative glucose control and the mean
daily glucose value over the 3 month period following the initiation of pump
therapy; and,

- frequency of severe hypoglycemia and extended hyperglycemia.

2. Determine whether use of the GlucoDYNAMIX intervention system, during initiation of
adult insulin pump starts, results in a measured saving of staff time in support of
data collection, analysis and patient feedback including delivery and follow-up to
monitor effects on glucose control from patient implementation of prescribed changes.
This objective will be measured by:

- tracking the number of interventions each day;

- estimating the amount of time spent collecting, analyzing and communicating with
each study patient; and,

- daily logs will be tallied and verified weekly.

3. To determine through use of questionnaires administered to the patients and to the
professional staff whether use of the technology results in perceived improvement in
the quality of care and improved satisfaction for all involved. This objective is
primarily concerned with people's willingness to commit to using a similar technology
as standard practice for future initiation of insulin pump starts for adult patient
with diabetes mellitus.

Study Design and Description The study is designed to run for a period of 3 months of
intensive management for each enrolled patient as they begin insulin pump therapy plus
detailed glucose records beginning 4 weeks prior to the initiation of insulin pump therapy.
Understanding relative glucose control following intensive management requires that we
consider the results of the patient's HbA1c 3 months after conclusion of intensive
management thus extending the time frame to 6 months.

Group A The Intervention Technology Group will be issued a GlucoMON device, a GlucoMON
wall-mount recharging unit, a OneTouch Ultra glucose meter, paper logbook forms and test
strips to support frequent testing for the 3 months following initiation of insulin pump
therapy. The patient will be instructed on use and care of the GlucoMON as well as the
OneTouch Ultra during a one-on-one session at the meeting immediately prior to initial
delivery of insulin via the pump start.

Patients are also issued a specially configured low-cost 2way wireless email device which
supports sending and receiving individual messages of up to 500 characters including the
ability to send canned response options for easy data marking. Real-time messaging as part
of the system is described in Addendum B. The intelligent logic behind the patient
interaction is managed on a server in Diabetech's remote data center thus allowing for easy
modification and centralized management of the patient interaction.

The TJU diabetes staff will initially review the patient record daily and will document the
results of their review including any changes prescribed to the patient and the approximate
amount of time involved in the handling of the encounter.

Group B (Control) The Conventional Technology Group will be issued a OneTouch Ultra glucose
meter, paper logbook forms and test strips to support frequent testing for the 3 months
following initiation of insulin pump therapy. The patient will be instructed on use and care
of the OneTouch Ultra during a one-on-one session at the meeting to begin initial delivery
of insulin via the pump start.

Patients will be instructed to complete the logbooks and fax daily to the TJU diabetes
center. The TJU diabetes staff are available to review patient logs Monday thru Friday 8-5.
Emergency after hours help is also available. The staff will initially review the patient
logbook and will document the results of their review including any changes prescribed to
the patient and the approximate amount of time involved in the handling of the encounter.

Data Analysis The data collected will include the patient's self-monitoring blood glucose
(SMBG) test data, time-event logging annotations to the glucose tests, baseline and
follow-up hemoglobin A1c test results, clinical events including hypoglycemia and
hyperglycemia, and results of satisfaction questionnaires and time spent for analysis by the
clinical staff for each patient.

The endpoints to be measured include frequent blood glucose testing, as is usual when a
patient is initiating therapy with an insulin infusion pump, as well as HbA1c. The glucose
levels will be analyzed with regard to the magnitude of glucose excursions and fluctuations
and the mean daily glucose value over the 6 month period following the initiation of pump
therapy.

The mean and standard deviation of each treatment group's (intervention and control) HbA1c
will be calculated for each time point (0, 3, and 6 months post enrollment). Statistical
differences in the HbA1c values between treatment groups at each time point and between
different time points for each group will be sought. In addition, the change in HbA1c
between time points (0 to 3, 3 to 6, and 0 to 6) will be calculated. The mean and standard
deviation of these HbA1c changes will be calculated for each treatment group and analyzed
for statistical significance.

SMBG data will be analyzed using glycemic control indices to compare the efficacy of both
treatments. The following numerical indices will be used to describe the quality of
metabolic control:

1. The mean and standard deviation of patient SMBG data

2. The percentage of SMBG data points that are above and below a threshold for
hyperglycemia (e.g., 200mg/dl) and hypoglycemia (e.g., 60mg/dl)

3. The M-value is a quantitative index of the lack of treatment efficacy of an individual
patient. The index describes both mean glycemic control and glucose excursions. The
M-value is a logarithmic transformation of the deviation from an arbitrary standard
(selected by the physician) that gives extra weight to hypoglycemic values. The
original concept of Schlichtkrull et al. (1965)3 has been modified by Service et al.
(1970) 4 into the formula:

/10 log10 (x/standard)/3 / n where x is a SMBG data point, standard is aM = glucose
value chosen to define the degree of metabolic control (typically 100 mg/dl), and n is
the number of samples. BG values below 100mg/dl cause the M-value to rise rapidly while
BG values above 100 mg/dl cause the M-value to rise more slowly. M-values less than 18
correspond to good metabolic control [Schlichtkrull J 1965].

4. The mean amplitude of glycemic excursions (MAGE) quantifies the amplitude and duration
of wide swings of BG values, diurnal variability of mean BG values, and episodes of
hypoglycemia as an index of diabetic instability. MAGE is calculated by taking the
arithmetic mean of the changes in BG when they exceed one standard deviation of the BG
mean over the preceding 24-hour period. Excursions less than 1 standard deviation are
ignored. MAGE is a valuable index of metabolic control in healthy and diabetic subjects
[Service 1970]. MAGE averaged 42 mg/dl in healthy subjects, 147 mg/dl in type 1
diabetic patients treated with multiple injections of short-acting insulin, and 161
mg/dl in type 1 diabetic patients treated with a single daily injection of
intermediate-acting insulin.

These indices will be computed using a moving window of data. The window will be stepped
across each patient's SMBG data set through the intensive management period. Those data
points that fall within the window will be used to compute the indices. The steps and widths
of the window will be set at 1, 7, 14, and 28 days. For example, with a step and width of 7
days, each calculation of the M-value of a patient's SMBG data would encompass 7 days and
there would be 12 different calculations for each week of the 3-month intensive management
period. Statistical differences in the indices for a given window for each treatment group
will be sought.

The amount of time taken by nurses and physicians to manage each patient will be tabulated.
These data will be used to compute a cost associated with the management of the insulin pump
start for each patient. The mean and standard deviation of this cost will be calculated for
each treatment group and analyzed for a statistical significance.

A paired Student t-test or a one-way analysis of variance will be used to calculate a
significant difference. A p value less than 0.05 will be considered statistically
significant.

Standard Care Guidelines typically involve the diabetes center distributing a handbook of
sorts to the patient upon diagnosis or as part of the initial visit to your center. In terms
of ongoing care, patients are responsible for acquiring tools on their own for managing
diabetes. In addition to insulin, patients must purchase their own glucose meters, test
strips, lancing device, lancets, alcohol, fast acting sugar of some sort to treat low blood
sugar, glucagon for severe hypoglycemia, etc... Depending on the form of insulin therapy,
the patient must also purchase syringes or an insulin pump with disposable cartridges and
infusion sets.

For the duration of the study, patients enrolled as Standard Care (Group B) will be given a
glucose meter and test strips at no charge.

To facilitate communication of blood sugars and other diabetes relevant data with the
center, the diabetes center shares one page of a standardized logbook form on 8.5" x 11"
paper. It is the patient's responsibility to make copies of this form for their own use and
also to deliver these completed forms to the diabetes center as needed. Methods for
delivering this logbook include faxing (a patient must purchase their own fax machine or
pay/borrow as well as any costs associated with fax transmission), US Mail (using their own
envelope and postage) or hand deliver (using their own transportation).

Study Population As a patient elects to participate they will be assigned alternatively to
either Group A or Group B beginning with assignment of the first patient enrolled into Group
A.

Informed Consent Informed consent will be obtained from each participant.

Documentation Requirements All data and documentation will be held private and secure in
compliance with HIPAA guidelines.

References

1. Wireless Diabetes Management System; Poster Presentation 2003 Diabetes Technology
Society: December 2002 - April 2003; Stephen Ponder MD CDE, Driscoll Children's
Hospital with support from Diabetech.

2. Wojcicki JM, Analysis: Transfer of the Data Collected by the Patient to the Health Care
Providers: Problems and Solutions, Diabetes Technology & Therapeutics. Apr 2005, Vol.
7, No. 2: 248-252

3. Schlichtkrull J. Munck O. Jersild M. The M-value, an index of blood-sugar control in
diabetics. Acto Med Scand. 177(1):95-102, 1965.

4. Service FJ. Molnar GD. Rosevear JW. Ackerman E. Gatewood LC. Taylor WF. Mean amplitude
of glycemic excursions, a measure of diabetic instability. Diabetes. 19(9):644-55, 1970