Article Text


Systematic review
The additional value of self-monitoring of blood glucose in patients with type 2 diabetes not using insulin is ‘not proven’
  1. Nanne Kleefstra1,2,3,
  2. Henk J G Bilo1,2
  1. 1Diabetes Centre, Isala clinics, Zwolle, The Netherlands
  2. 2University Hospital Groningen, Groningen, The Netherlands
  3. 3Langerhans Medical Research Group, Zwolle, The Netherlands
  1. Correspondence to Nanne Kleefstra, Dokter van Heesweg
    Dokter van Heesweg 2, 8025 AB Zwolle, Postbus 10400, 8000 GK Zwolle, The Netherlands; kleefstra{at}

Statistics from

Commentary on:


For patients with type 1 diabetes mellitus and patients with type 2 diabetes mellitus (T2DM), using insulin self-monitoring of blood glucose (SMBG) helps to improve glycaemic control.1 However, there is still much debate about the use and effectiveness of SMBG in non-insulin-treated T2DM: ‘The SMBG controversy’. In this study, Clar and colleagues have undertaken a review aiming to investigate the effects of SMBG on (among others) glycemic control, hypoglycaemia, quality of life and the cost of SMBG in patients with T2DM not using insulin or using basal insulin in combination with oral blood glucose lowering agents.


This was a systematic review investigating the effects of SMBG in patients with T2DM. The Cochrane Library, MEDLINE, EMBASE, PsycINFO and the Web of Science were searched for systematic reviews, meta-analyses, randomised controlled trials (RCT), observational studies (500 participants or more), qualitative studies and economic studies published since 1996. Non-English language papers were excluded.

Results of the RCTs were meta-analysed (weighted mean differences, random effects model, inverse variance method) and heterogeneity was calculated (χ2 and I2 methods). The outcomes of interest were: HbA1c, hypoglycaemia, quality of life, anxiety, depression, costs, treatment satisfaction, weight, treatment change, lipid profile and blood pressure.


It was only possible to perform meta-analyses for HbA1c. In patients performing SMBG versus no SMBG or self-monitoring of urine glucose, HbA1c changed by −0.21% (95% CI −0.31% to −0.10%) or −0.06% (95% CI −0.69% to 0.56%), respectively. In patients using SMBG in combination with an educational/feedback component, versus SMBG without these components or versus no SMBG, HbA1c was reduced by −0.20% (95% CI −0.44% to 0.03%) and −0.52% (95% CI −0.98% to −0.06%), respectively. For the latter, significant heterogeneity was found. From the study by Farmer and colleagues, it was concluded that SMBG was not cost-effective.2

Virtually no significant differences were found in outcomes such as quality of life and treatment satisfaction. Four trials found conflicting results concerning depression. There were no differences between groups in the studies investigating weight and blood pressure parameters. Results were inconsistent concerning hypoglycaemia and lipid profile.


Clar and colleagues concluded that SMBG can provide only slight benefits in terms of glycemic control, can have psychological disbenefits and is not cost-effective. They have performed a thorough literature search which yielded more studies than previous systematic reviews. Their results are in line with other recent systematic reviews with SMBG resulting in a benefit in terms of HbA1c change of around −0.2% on average.

Although the methodological quality of the selected trials was assessed, this was not used to nuance the results and conclusions. In the comparison of SMBG versus no SMBG, only 3 out of 10 trials were of high quality (one study was not even an RCT). The mean effect of SMBG versus no SMBG in the high quality trials would probably not have changed much. However, for the analyses of the effects on HbA1c of SMBG in combination with an educational/feedback component versus SMBG without these components or versus no SMBG, the only trial of high quality was the trial with the least (ie, no) beneficial effects. Focusing on quality of life and depression, there are four trials with significant results: the two that were of high quality reported worsening, while the two that were of poor quality reported improvement.

A criticism of most of the high quality trials is that the baseline or end of study HbA1c was too low in order to be able to find any effects of SMBG. However, at present, good metabolic control is the norm in this patient group.

We agree with the author's final conclusion, that the value of SMBG in patients with T2DM not using insulin is ‘not proven’. If there is any, the benefit of SMBG on glycemic control is not clinically relevant and the patient's quality of life can deteriorate. The authors' additional conclusion that SMBG can clearly yield benefits if used appropriately is not justified by their own results. We have to realise that it is possible that SMBG in this patient group does not have the beneficial effects hoped for. Finally, an individual patient data meta-analysis of several SMBG trials is currently under way that will try to identify subgroups of patients that might benefit from SMBG and should also try to identify subgroups that might be harmed.3


View Abstract


  • Competing interests NK and HJGB have conducted a diabetes study which was partly funded by Roche diagnostics.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.