Management of type 2 diabetes is an ongoing challenge for patients and their doctors. In order to prevent complications, patients need to monitor and control their blood sugar levels. In addition, they may need to have an ongoing communication with their doctor in order to modify treatment. In this study the investigators wish to compare the benefits of continuous glucose monitoring system (CGMS) to an internet-based glucose monitoring system (IBGMS). The investigators also want to determine whether the frequency of blood glucose testing affects blood glucose control and to see whether the benefits of CGMS and IBGMS are independent of testing frequency.

In this current study, the investigators wish to test several hypotheses: whether the frequency of SMBG is correlated with HbA1c, whether the knowledge that strips are being counted affects HbA1c, whether the reduction of HbA1c when using an IBGMS is independent of the frequency of testing, and whether there is a difference between IBGMS and CGMS in HbA1c reduction.

The investigators will recruit 100 patients with type 2 diabetes who satisfy the inclusion/exclusion criteria. For the first visit, the patients will be immediately randomized into four groups. One group will be testing 3 times a day (SMBG). The second group will be similar to the SMBG group except they will know that we are counting the frequency of testing (SMBG With Knowledge). The third group will use an IBGMS and the fourth group will use a CGMS. During the first visit, we will teach the patient how to use the programs and machinery for their group. All four groups will be asked to follow the protocol and to visit their doctor in 3 month intervals for 6 months along with completed laboratory blood test results.

For the SMBG and the SMBG With Knowledge group, each will be given a new meter and be asked to test 3 times a day. They will also be required to keep a diary of their SMBG levels. The difference between the two is that the SMBG With Knowledge group will know that we are counting strips. The SMBG group will believe that we are testing a new meter and will not know that the SMBG With Knowledge group exists. This deception is necessary to test the experimental effect of if the patient knows that we are counting the frequency of testing, does that affect the frequency that they are testing. The investigators will tell the SMBG group that we are testing the accuracy of a new meter and they should test 3 times a day.

For the CGMS and IBGMS group, they will also be counting the frequency of blood glucose testing to see whether the benefits from both are independent of the number of tests. Furthermore, we wish to compare the these two methods of blood glucose monitoring. Both groups will have to upload their data onto the internet and communicate with their doctor every 2 weeks using a secure online web site.

At the end of 3 and 6 months, we will collect the frequency of testing from each group to determine whether increased glycemic control correlates with increased testing. We will also collect their SMBG diaries to see whether the written data correlates with the data on their meter. The investigators will also be comparing a measure of their blood glucose levels for each group and determining the effectiveness of CGMS and IBGMS.

Inclusion Criteria:

• Type 2 diabetes patients being treated with insulin for at least 3 months HbA1c > 7%
• 25 years of age
• Willingness to test blood glucose levels a minimum of 3 times daily
• Willingness to be trained on using the Continuous Glucose Monitoring System
• Willingness to be randomized
• Trained in self-blood glucose monitoring
• Internet Access on a Windows Personal Computer
• No prior use or training on CGMS in the past 6 months
• No prior use or training on IBGMS in the past 6 months

Exclusion Criteria:

Patients who do not meet the above criteria or are not willing to participate will not be included in the study. Additional exclusion criteria include:

• Patient with medical conditions that may affect their study participation or - - results will be excluded.
• Patients with the potential to become pregnant
• Patients using medications known to influence control of diabetes (eg steroids systemic or inhaled)
• Liver disease (AST or ALT levels > 2.5 times the reference level)
• Renal insufficient with a serum creatinine level > 200 μmol/L