Re-evaluating Home Monitoring for Diabetes: Science-Based Medicine at Work

There is no question that patients on insulin benefit from home monitoring. They need to adjust their insulin dose based on their blood glucose readings to avoid ketoacidosis or insulin shock. But what about patients with non-insulin dependent diabetes, those who are being treated with diet and lifestyle changes or oral medication? Do they benefit from home monitoring? Does it improve their blood glucose levels? Does it make them feel more in control of their disease?

This has been an area of considerable controversy. Various studies have given conflicting results. Those studies have been criticized for various flaws: some were retrospective, non-randomized, not designed to rule out confounding factors, high drop-out rate, subjects already had well-controlled diabetes, etc. A systematic review showed no benefit from monitoring. So a new prospective, randomized, controlled, community based study was designed to help resolve the conflict.

O’Kane et al studied 184 newly diagnosed patients with type 2 diabetes who had never used insulin or had any previous experience with blood glucose monitoring. They were under the age of 70 and recruited from community referrals to hospital outpatient clinics, so they were likely representative of patients commonly seen in practice. They were randomized to monitoring or no monitoring. Patients in the monitoring group were given glucose meters and were instructed in their use and in appropriate responses to high or low readings, such as dietary review or exercise. They were asked to take four fasting and four postprandial readings every week for a year. Patients in the no monitoring group were specifically asked NOT to acquire a glucose monitor or do any kind of self-testing. Otherwise, the two groups were treated alike with diabetes education and an identical treatment algorithm based on HgbA1C levels.

Their findings:

We were unable to identify any significant effect of self monitoring over one year on HbA1c, BMI, use of oral hypoglycaemic drugs, or reported incidence of hypoglycaemia. Furthermore, monitoring was associated with a 6% higher score on the well-being depression subscale.

So home monitoring not only did no good but it made patients feel worse. Why? Perhaps because they were constantly reminded that they had a disease and worried when blood glucose levels rose, especially when the recommended responses of dietary review and exercise didn’t rapidly lead to lower readings.

We would not accept the results of one isolated study without replication, but in this case the new study adds significantly to the weight of previous evidence and arguably tips the balance enough to justify a change in practice.

The American Diabetes Association still says “Experts feel that anyone with diabetes can benefit from checking their blood glucose.” But they only recommend blood glucose checks if you have diabetes and are:
• taking insulin or diabetes pills
• on intensive insulin therapy
• pregnant
• having a hard time controlling your blood glucose levels
• having severe low blood glucose levels or ketones from high blood glucose levels
• having low blood glucose levels without the usual warning signs

Diabetes experts see the severe, complicated cases and have a different perspective from that of the family physician seeing mostly mild and uncomplicated cases. An article in American Family Physician said

Except in patients taking multiple insulin injections, home monitoring of blood glucose levels has questionable utility, especially in relatively well-controlled patients. Its use should be tailored to the needs of the individual patient.

An editorial in the BMJ pointed out that

Home blood glucose monitoring is a big business. The main profit for the manufacturing industry comes from the blood glucose testing strips. Some £90m was spent on testing strips in the United Kingdom in 2001, 40% more than was spent on oral hypoglycaemic agents.2 New types of meters are usually not subject to the same rigorous evaluation of cost effectiveness, compared with existing models, as new pharmaceutical agents are.
If the scientific evidence supporting the role of home blood glucose monitoring in type 2 diabetes was subject to the same critical evaluation that is applied to new pharmaceutical agents, then it would perhaps not have been approved for use by patients.


Home glucose monitoring in type 2 diabetes is not justified by the evidence. It does not improve outcome, it is expensive, and it may decrease the quality of life of patients.

Common sense suggested monitoring should improve outcome. We had assumed it would work. Scientists thought to question that assumption. The found a way to test that assumption. New evidence showed that it was a false assumption. In response to that evidence, the practice is now being abandoned. This is how science is supposed to work. Another small triumph for science-based medicine.

Posted in: Clinical Trials, Science and Medicine

Leave a Comment (18) ↓

18 thoughts on “Re-evaluating Home Monitoring for Diabetes: Science-Based Medicine at Work

  1. storkdok says:

    My mother, a type 2 diabetic, says that the monitoring at home is depressing. She is pretty well controlled. I am going to send her this study to discuss with her doctor. I hope there will be more studies to see if there is replication confirming this finding. It would be great if she could stop doing so much monitoring in the future.

  2. daedalus2u says:

    Very nice article and analysis which shows very nicely the different approach to health that SBM and EBM take compared to CAM. The hypothesis that at home testing would help is quite reasonable, and intuitively makes sense. It turns out it was wrong. These are the most important kinds of studies to do, the ones where our intuition is wrong. They are also the most difficult to get funding to test.

    What is interesting is the rapid responses there were a number of complex ideas put forward as to why there was no effect, ideas which distorted what was actually done in the study.

    CAM wouldn’t even bother to test something like this.

    It also puts the lie to statements that SBM blindly favors interventions.

  3. weing says:

    I have not been telling my patients to monitor the glucose at home if it’s well controlled based on glycohemoglobin measurements. It’s good to have some backing for this. My approach has been to have the patients focus on living their lives and not have their lives revolve around a disease. There are some patients that get some kick out of monitoring their sugars and want to do it even though I don’t see any need for it. So, it takes all kinds.

  4. DeanVenture says:

    Couldn’t help but think that studies like this show CAMmers that we’re not in the corporations’ pockets after all…

    Good review!

  5. Basiorana says:

    My dad monitors his Type 2 diabetes at home. He can’t do anything if it’s high except not eat any more sugar the rest of the day. It kind of is ridiculous for him– he eats a big meal, checks his sugar, realizes it spiked, then eats salted peanuts all day to avoid sugars. Not exactly healthy. But then again, getting him to follow a diabetic diet completely flopped too…

  6. Calli Arcale says:

    Having had relatives who depended on home glucose monitoring (they were in the category of people where they really had problems if they didn’t keep track of the glucose level), I can say that in addition to it being depressing and costly, it’s *painful*. Not having to do it would be a mercy.

  7. mr. grieves says:

    This is a very interesting study, thanks for posting it. I am curious if these results would continue beyond the 1 year mark. I would imagine (and have observed) that many patients (but not all!) are very motivated to take control of their disease early after diagnosis. However, as time goes on they become less and less interested in both monitoring and lifestyle management (diet and exercise). I wonder if continuing to monitor at home would reinforce the need for patients to lead a healthy lifestyle in these patients with a longer history of diabetes. On the other hand, I have also observed numerous patients who experience pain with home monitoring, difficulty utilizing glucose meters and those who face economic hardship due to the cost of lancets and strips.

  8. Newcoaster says:

    As Daedalus pointed out, this kind of study shows the difference between EBM and sCAM. We actually test things that are common clinical practice, and adjust our practice accordingly if the evidence shows it doesn’t work.
    Another recent study (I think in JAMA) also showed that tight glucose control in the ICU was counter-productive.

  9. Those with Type 1 diabetes tend to be obsessive about monitoring themselves, because it is a life or death situation. Those with Type 2 diabetes tend to be less obsessive (or even annoyed) about monitoring, because the effects tend to be more long-term.

    I also believe, with just a little bit of evidence, that most of the glucose monitors are just plain difficult to use, counterintuitive, and the lancets that are used to make the fingerstick are a bit medieval in design.

    I know many of you want to beat up on Big Pharma (or in this case Big Diagnostic), but maybe someone needs to send these results to managed care or third-party payors. They are authorizing and paying for the use of these devices by anyone who is diagnosed with Type 2 diabetes to manage this disease in the hope of reducing the risk from other diseases like CVD. Many of the managed care facilities are trying to teach patients how to use the devices.

    Anyways, I used to think there are lots of meters sitting in houses gathering dust. I guess now I’m convinced there are. What a waste of money.

  10. I want to amend my comments slightly. Is blood glucose monitoring for Type II diabetes useful? What I’m reading here is that there is a compliance issue, not a science issue. The studies did not indicate that blood glucose monitoring was useless in managing the disease, more that it’s a waste of money because compliance is so low.

    Even though A1C testing is more accurate and gives better information for chronic levels of blood glucose, the information is a delayed effect. In other words, it may take a few months of lowering blood glucose to elicit a reduction in A1C levels.

    If you’re trying to reduce your blood glucose from 160 to 110 (for example), I don’t know if there’s a way to observe that small of a reduction without a diagnostic test. You’re probably not going to notice anything short-term in your health (maybe you’ll urinate more glucose, but I’m not sure you can even tell that).

    So, blood glucose monitoring may be a waste of money, but is it useless? I can think of several cases where it isn’t. A Type 2 diabetic who exercises heavily and frequently needs to know their pre and post-exercise glucose. Someone is truly trying to manage their diet (talk about positive feedback).

    I accept the fact that it’s depressing for patients to be reminded of their diabetes. But, and since in my life I require no bedside manner, type 2 diabetes is almost always a consequence of poor lifestyle choices. Appropriate management of blood glucose can be the first step in changing that.

    I’m of two minds about blood glucose monitoring. It can be a valuable tool. But what a waste of money for our health care system. And the studies just show that it’s a waste of money, not where it is a valuable tool.

  11. Sara says:

    I’d definitely like to know what kind of dietary advice was given to these patients. If, for instance, they were advised to eat something like the ADA suggested diet, which suggests 45-60g of carbohydrate per meal, and the suggested response to experiencing highs or lows was to try to stick more closely to that diet, then it’s no wonder they didn’t find metering helpful. If I ate like that, I’d find my glucose meter useless, distressing, and depressing, too. I’d like to see a version of this study where the participants were advised to limit carbs, and to test before and after meals to see how a particular meal affected them, and then see if they found the meters useless, or instead empowering.

  12. I am really concerned about the conclusions given in both the paper and the review here. Is there a subset of patients who are compliant, find the monitoring to be a positive experience, or have managed their disease positively? Further, is there something about England’s monitors or diabetes education that’s different from say, the US (I’m going to say, maybe)?

    I’m not convinced that BGM is a complete waste of money.

  13. xwolp says:

    In my experience as a type I diabetic, the difference between blood glucose levels of 160 and 110 should be very noticeable unless said person suffers from chronically higher levels (the perception changes a lot depending on your mean levels).
    But that is besides the point.
    Having a mother who suffers from chronic pancreatitis and thus periodically develops symptoms of diabetes I can attest to how much home monitoring can throw a therapy off course.
    A relatively minor hyperglycemia (170-190) can easily turn into more severe hypoglycemias since most type II medication is not very precise. In overall these are much more dangerous and a higher stress for the system overall.

  14. ned says:

    I use a glucose monitor and I think it helps a lot. I use a notation method to keep track of my carbs daily, I write it all down with other comments—–how I feel both emotionally and physically. Sometimes I feel like eating something off diet I think of the morning number I will be facing. I gives me a little incentive to stay on the straight and narrow.

    People who diet find recording their food consumed to be helpful; I feel this is just an extension of that method. Been at this for 12 years, still controlling with diet and exercise.

  15. Harriet Hall says:

    ned perceives a psychological benefit from home monitoring, but the study suggests that the few like ned are outweighed by the many who get more depressed when monitoring. And ned can’t really be sure he would have done worse without monitoring. Perhaps those who are diligent about monitoring are those who would naturally be more diligent about diet and life style changes.

    The AFP advice “Its use should be tailored to the needs of the individual patient.” makes allowances for people like ned.

  16. The Blind Watchmaker says:

    I was happy to see this study published last month. I recommend all of my diabetic patients to see a dietitian. Most diabetic dietitians are still insisting on frequent glucose testing even in patients not on any meds, let alone insulin. It seems that some dietitians were trained with Type 1 diabetics in mind. It is time to focus on what is important in Type 2 diabetics: nutrition, exercise and reduction of abdominal fat.

    Many patients have been conditioned to check their sugars often and become upset when the doctor checks the “test <2 times a day” box on the prior authorization form for their diabetic supplies. I often feel frustrated that some people focus so much on testing, but focus so little on lifestyle changes.

  17. beadle says:

    I agree with Michael Simpson concerning what it is this study really shows. I lean towards not having my DMII patients not test that often, because of the seeming cost/benefit mismatch shown in this study. Admittedly, this attitude has always been a “gut feeling” of mine more than anything. This study initially seems to validate that tack, but what is it really studying? Compliance? UK-brand BGM accuracy?

    The study’s stated end-points were “…group differences in HbA1c, psychological indices, use of oral hypoglycaemic drugs, body mass index (BMI), and reported hypoglycaemia rates” over a one year period. Fine. But DM is a life-long disease with some consequences of poor glycemic control not showing up for many years.

    It logically follows that no differences in a year (especially A1c levels and hypoglycemic events), multiplied by many years, equals no differences in many years. But that’s logic, not science. This study does not attempt to address long-term complications of DM between patients who monitor and those who don’t. Further research is needed.

  18. 21stCenturyRox says:

    It’s an interesting finding – but could the problem be not enough information? If patients are tracking diet, exercise and glucose, using a tool like Microsoft HealthVault (, and can see positive trends over time, maybe that could correct the emotional reaction and improve compliance with lifestyle changes. Of course, speculation w/o data is just a call for more research…

Comments are closed.