September 15, 2017
5 min read

Should self-monitoring blood glucose be encouraged for adults with type 2 diabetes?

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Click here to read the Cover Story, "Diabetes data hold hidden clues, but time, costs complicate clinic downloads."



Structured SMBG can help people with type 2 diabetes not using insulin therapy to manage their blood glucose levels.

In a key study published in 1983, Cohen and Zimmet showed that blood glucose control improves when SMBG is used by adults with type 2 diabetes who are well-informed as to when to monitor and why they are monitoring, what their blood glucose values mean and what needs to be done about them (Med J Aust. 1983;2:377-380). SMBG also increased motivation and fostered healthier food choices. Therefore, we’ve known for more than 30 years that, in the right circumstances, SMBG can be beneficial for this population.

David R. Owens
Jane Speight

In 2009, Owens was involved in writing the guidelines for SMBG in noninsulin-treated type 2 diabetes for the International Diabetes Federation (available at, which introduced the concept of “structured” SMBG, in which people measure their blood glucose before and after a meal. In this structured way, they can see the effect of different meals (composition and portion sizes) on their glucose levels, and this provides a firm basis for discussion with their clinicians of their blood glucose patterns.

Even now, there is still huge debate as to whether SMBG is beneficial to patients with type 2 diabetes. While a Cochrane review suggested that SMBG lacks clinical benefit in this group (Malanda UL, et al. Cochrane Database Syst Rev. 2012;doi:10.1002/14651858.CD005060.pub3), a closer critique suggests considerable heterogeneity among the studies included and that structured SMBG is far more beneficial than suggested by the review (Speight J, et al. Curr Med Res Opin. 2013;doi:10.1185/03007995.2012.761957). A recent study once again casts doubt on the value of SMBG for people with noninsulin-treated type 2 diabetes, but did not actually adopt a structured approach (Young LA, et al. JAMA Intern Med. 2017;doi:10.1001/jamainternmed.2017.1233).

When blood glucose is checked just once a day, at different times of day and with limited advice about how to respond to the value, then SMBG is unstructured and relatively meaningless. Thus, it is unsurprising when this approach does not lead to better HbA1c and even less surprising when it causes frustration.

A structured approach to SMBG can reduce the number of blood glucose checks that are carried out, as the SMBG readings obtained are more meaningful and actionable. Therefore, the frequency of SMBG can be varied according to the individual’s needs and preferences, particularly whenever management changes are introduced.

The majority of people with type 2 diabetes have an unhealthy BMI and would benefit enormously from lifestyle changes. However, such changes are difficult to initiate and sustain. One clear way to understand dietary impact on blood glucose is to measure it before and after meals, which enables the person to see the impact of the type and quantity of food consumed. This is known as “discovery learning.” When the blood glucose is noted to be too high or too low, then the informed patient can take appropriate action to adjust their food or drink and physical activity. Often the person can reason out what it means for ongoing self-care. If there is any doubt, they can discuss with their clinician whether there is any need to change their food choices or physical activity levels or to adjust medications.

Unfortunately, SMBG conducted in an unstructured way is of limited value and can, indeed, be a waste of money. However, with structured SMBG, people can begin to recognize and understand their blood glucose profiles, and this increases their self-confidence in managing their diabetes. It’s almost like driving a car: You wouldn’t drive a car without a speedometer, but if you can see that you’re driving too slowly or too fast, then you will adjust your reactions.

David R. Owens, CBE, MD, FRCP, is professor of diabetes at Swansea University Medical School, U.K. Disclosure: Owens reports no relevant financial disclosures. Jane Speight, MSc PhD CPsychol AFBPsS, is foundation director for The Australian Centre for Behavioural Research in Diabetes, a partnership for better health between Diabetes Victoria and Deakin University. Disclosure: Speight reports financial relationships with Abbott Diabetes Care, Medtronic, Roche Diabetes Care and Sanofi.



Current evidence suggests SMBG should not be routine for most patients with noninsulin-treated type 2 diabetes.

Katrina Donohue

There has been a lot of debate in the literature about home blood glucose testing, specifically, in patients with noninsulin-treated type 2 diabetes. Some studies have shown modest improvements in glycemic control with SMBG, although these improvements have not typically been clinically significant. There have also been some intriguing studies on enhanced SMBG, in which the patient and the provider are more actively engaged in interpreting blood glucose values. We used that model for the basis of our study, employing enhanced SMBG in a real-world, daily setting. We recruited patients who had noninsulin-treated type 2 diabetes and their providers from 15 internal medicine and family medicine practices, and randomly assigned these patients to one of three groups: no home blood glucose testing, once-daily blood glucose testing (what typically goes on in the real world, with a wireless meter), and enhanced blood glucose testing. Patients in the enhanced group would test once daily and would receive a preprogrammed message after each test, either with encouragement or a recommendation for the patient. Messages would also be sent to the patient’s provider. We followed patients for 1 year, powered on two primary outcomes: HbA1c and health-related quality of life at 1 year.

After 1 year, we found no significant differences in blood glucose control across all three arms, and no significant differences in health-related quality of life. Our team was split on how enhanced testing might affect patient HbA1c and quality of life — would it make a difference? When we analyzed the data, initially, it did. There were some improvements in HbA1c early on, at around 6 months, although these were HbA1c values caught in electronic health records. But, at 1 year, in the real world, we couldn’t see a difference.

There are several reasons why we may not see results with enhanced SMBG in these patients. Many behavioral studies suggest that with new technology, like the Fitbit, for example, there is an initial excitement with using the device, and then that excitement wears off. Blood glucose testing is one component of diabetes management, but it is certainly not all of it. In addition, patients with diabetes have many other comorbidities and conditions that they are trying to manage at the same time.

After a review of the data, I would caveat the answer. In our study, we did not find any differences. We did analyze controlling for baseline HbA1c and diabetes duration, but our study was not powered for subgroups. For some subgroups, enhanced testing might be beneficial for patients with noninsulin-treated type 2 diabetes. At the end of the day, I think patients and their providers need to sit down and discuss whether to test or not. What is most important is that it is a shared decision-making process.

Katrina Donohue, MD, MPH, is professor and research director in the department of family medicine at University of North Carolina at Chapel Hill and co-director of the North Carolina Network Consortium. Disclosure: Donohue reports copyright work for a glucose monitoring treatment messaging algorithm that is licensed to Telcare.