The value of self-monitoring of blood glucose (SMBG) in patients with type 2 diabetes treated with oral hypoglycaemic agents is controversial.
Haidar, E.A.C. et al (2008) Self-monitoring of blood glucose in type 2 diabetes This is an extended version of the article published in Nursing Times; 104: 3, 32-33.
BACKGROUND: The value of self-monitoring of blood glucose (SMBG) in patients with type 2 diabetes treated with oral hypoglycaemic agents is controversial.
AIM: To assess the effect of SMBG on glycaemic control in type 2 diabetes.
METHOD: An observational prospective study with historical controls was carried out in a socially deprived general practice. Participants were 100 patients of South Asian origin with type 2 diabetes. Sixty-seven patients completed: 32 females, 35 males. Patients were educated and performed SMBG.
MAIN OUTCOME MEASURES: HbA1c and fasting blood glucose (FBG) were checked at three, six and nine months and compared with baseline data.
RESULTS: There was a significant difference between pre-SMBG HbA1c (mean 9.8%, SD +/-2.4) and post-SMBG (mean 8.5%, SD+/-2.1). One-way ANOVA confirmed a significant difference between pre and post results of HbA1c and FBG.
CONCLUSION: SMBG improved glycaemic control in patients with type 2 diabetes.
- Figures and tables can be seen in the attached print-friendly PDF file of the complete article in the ‘Files’ section of this page
Elizabeth A.C. Haidar, BSc, MSc, AHEA, RN, is lecturer and advanced nurse practitioner, King’s College, London; Andrew C. Burden, MD FRCP, is community diabetologist, Heart of Birmingham teaching PCT; John R. Skelton, BA, MA, RSA, MRCGP, is professor of clinical communication, University of Birmingham.
Since many patients diagnosed with type 2 diabetes have had the condition for several years, means micro/macrovascular diabetic complications are commonly found at this time. It is therefore vital to achieve and maintain normoglycaemia - a 1% reduction in HbA1c will give a 21% decrease in any diabetes-related complication (UKPDS, 2000). Although the amended consensus guidelines for type 2 diabetes (Owens et al, 2004; American Diabetes Association, 1997) confirm that diabetic control can be aided by self-monitoring of blood glucose (SMBG) , in literature the value of SMBG in patients with type 2 diabetes treated with oral hypoglycaemic agents is controversial (Harris, 2001; UKPDS, 2000; Coster et al, 2000). There is a lack of evidence from randomised controlled clinical trials (RCTs) to support SMBG in the treatment of type 2 diabetes, and results from meta-analyses are inconclusive (Harris, 2001). The purpose of this study was to investigate the effect of SMBG on glycaemic control.
SMBG is proven beneficial for patients with type 1 diabetes (Evana et al, 1999) and for those using insulin to treat type 2 diabetes (Karter et al, 2001), as patients’ glycaemic control is monitored and adjusted accordingly resulting in improved outcomes. It has been suggested that patients with type 2 diabetes who are on oral hypoglycaemic therapy may also benefit from SMBG (American Diabetes Association, 1997). These patients may benefit from a better perceived quality of life by improving their understanding of their disease adjusting lifestyle and medications accordingly (Karter et al, 2001).
A meta-analysis of RCTs (Jansen, 2006) showed that SMBG improves HbA1c by 0.4% when compared with no SMBG. However, the level of evidence was ‘moderate’ (Palmer et al, 2006). This was also supported by a retrospective study (Martin et al, 2006), which followed 3,268 patients from between 1995 and 1999 until the end of 2003. SMBG was associated with decreased diabetes-related morbidity and all-cause mortality in type 2 diabetes. This association remained in a subgroup of patients who were not receiving insulin therapy (Martin et al, 2006). It was suggested that SMBG may be associated with a healthier lifestyle and/or better disease management (Martin et al, 2006).
Another study (Guerci et al, 2003) looked at 988 patients over six months and found that SMBG was statistically associated with better quality of metabolic control than usual traditional recommendations alone in type 2 diabetes.
A meta-analysis from the Netherlands (Jansen, 2006) found that RCTs performed to date provided positive results of the effectiveness of interventions with SMBG in type 2 diabetes. Likewise, a meta-analysis from the Philippines (Sarol et al, 2005) found that a multi-component diabetes management programme with SMBG resulted in better glycaemic control among non-insulin using patients with type 2 diabetes.
Nevertheless, there were studies with negative feedback (Latalski et al, 2005) suggesting there was no association between glycaemic control and SMBG in type 2 diabetes. These results were taken from only 64 patients, of whom only 70% had type 2 diabetes. It was also noted that patients only answered a questionnaire and that there were no recommendations on how patients should use the results from SMBG effectively.
A larger study (Franciosi et al, 2005) looked at 1,896 patients at six monthly intervals for three years and found the performance and frequency of SMBG did not predict better metabolic control over the study period. However, in this study, SMBG was only performed to avoid hypoglycaemic episodes and again there were no recommendations on how to use the results from SMBG.
Invasiveness is a serious barrier to SMBG. Wagner et al (2005) found that patients would perform SMBG more frequently and improve quality of life with a non-invasive mechanism. This feedback came from 339 patients and was obtained through a tick box questionnaire which addressed issues such as SMBG, anxiety and burden. It could be said to provoke negative comments and again there was no mention of education on how to act upon the results of SMBG.
A qualitative study of patients’ perspectives on SMBG (Peel et al, 2004) elicited a mixed response - the technique can heighten patients’ awareness of the impact of lifestyle but it can also amplify a sense of failure about self-management, leading to anxiety and self-reproach if blood glucose levels remain high. This study comprised qualitative interviews with 40 patients from 16 general practices and hospitals.
One 12-month study suggested that SMBG can improve metabolic control (Oria-Pino et al, 2006) but requires careful selection of patients. The researchers managed to carry out an RCT with 100 participants in total, although again participants were offered no education. Those patients who were educated to a higher level (degree, masters degree) monitored their glycaemia more often than those who were not educated to this level (Skelly et al, 2005) but the diabetes control did not differ. No patient education was mentioned.
Patient education is of paramount importance (Skelly et al, 2005). This was shown to be the case in a study undertaken in a rural region among older people with diabetes, which found that the effect of healthcare providers’ recommendations on SMBG significantly affected practising SMBG and SMBG frequency. These 698 patients were a mixture of African/American and Caucasians over 65 years of age. The idea was to educate them to use the information from SMBG to encourage them to alter their lifestyle/diet in order to gain glycaemic control.
Materials and methods
This study was carried out between February 2004 and February 2005. Participants were recruited from a new diabetes clinic in an inner-city GP practice serving patients of South Asian origin in Birmingham. The clinic was run by an advanced nurse practitioner (ANP). Participants were each given a glucometer (MediSense Optium Plus), educated and shown how to perform SMBG. Their HbA1c and fasting blood glucose (FBG) were checked at three, six and nine months following the introduction of SMBG. The outcome was compared with baseline tests that had been performed three, six and nine months before participants commenced SMBG. It was an observational prospective study (quasi-experimental design) with historical controls.
Patients were recruited through the diabetes clinic as a convenience sample, with the following inclusion criteria:
Type 2 diabetes;
Taking sulphonylureas and/or biguanides;
Poorly controlled with HbA1c above 7%;
The diabetes clinic had been set up recently in response to an audit showing a high number of patients with poor glycaemic control. An educational programme and a multidisciplinary team were introduced.
Patients were not given specific times or frequency to monitor blood glucose, but were advised to take a first reading in the morning before eating or drinking, then a post-prandial reading two hours later, which could provide them with information to reduce their dietary intake and regulate their glycaemic control (Oria-Pino et al, 2006).
The results of HbA1c and fasting blood glucose were analysed using equipment in the local hospital. In order to establish validity and reliability, it was processed by high performance liquid chromatography (HPLC) using a Tosoh G7 analyser (Tosoh Bioscience, Redditch, Worcs); precision 2.9% at 5.9% HbA1c (normal control) and 2.1% at 10.2% HbA1c.
The sample size was based on the clinically significant reference that suggests a 1% reduction in HbA1c gives a 21% decrease in diabetes-related complications, such as microvascular/macrovascular disease (UKPDS, 2000). To detect a mean difference of 1% in HbA1c and assuming a standard deviation (SD) of 2.0 with a 0.05 significance level, the sample size had to be 63 patients in order to have 80% power. No pilot study was performed due to time constraints. The student’s t-test, one-way analysis of variance (ANOVA) and Pearson
correlation coefficient were used to compare the pre and post results. The chosen significance level was p=0.05.
Although 100 patients were recruited to the study, 33 dropped out: seven converted to insulin, one became pregnant and 25 did not continue with SMBG, dropping out of the study at different stages. Of those who completed the study, 32 were female and 35 were male. The average age was 60 years (range 33-87).
Table 1 presents the average FBG results achieved during the study, their SDs and ranges, which are also illustrated in Fig 1. The overall average FBG was 13.95mmol/L pre-SMBG and 9.94mmol/L post-SMBG.
Table 1. Average FBG results over the course of the study
|9 pre-SMBG||14.38mmol/L||+/- 4.5||27.3-6.1|
|6 pre-SMBG||13.96mmol/L||+/- 4.0||25.2-6.2|
|3 pre-SMBG||13.51mmol/L||+/- 3.9||25.7-5.5|
|3 post-SMBG||10.3mmol/L||+/- 4.4||25.9-4.9|
|6 post-SMBG||9.95mmol/L||+/- 4.3||24.9-4.4|
|9 post-SMBG||9.59mmol/L||+/- 3.4||18.4-5.1|
Table 2 presents average HbA1c results during the course of the study, their SDs and ranges, which are also illustrated in Fig 3. The overall average HbA1c was 9.82% pre-SMBG and 8.5% post-SMBG (Fig 4).
Table 2. Average of HbA1c readings over the course of the study
|9 pre SMBG||10.07%||+/- 2.5||18.2 - 7.1|
|6 pre SMBG||9.85%||+/- 2.4||17 - 6.7|
|3 pre SMBG||9.51%||+/- 2.3||17.2 - 6.5|
|3 post SMBG||8.9%||+/- 2.1||15.8 - 6|
|6 post SMBG||8.5%||+/- 1.8||14 - 5.6|
|9 post SMBG||8.1%||+/- 1.6||12.1 - 4.9|
The student’s t-test showed a significant difference between the averages of the three pre-SMBG measurements for FBG and those of the three post- SMBG measurements; the p value was 3.91E-13. Also, there was a significant difference between the averages of the three pre-SMBG measurements of HbA1c and those of the three post-SMBG measurements; the p value was 3.27E-10.
The one-way ANOVA was performed for HbA1c measurements over the six occasions. The chosen a was 0.05, F (found in F-distribution table) was 2.21 and the calculated F was 8.87. Since the calculated F was larger than 2.21, the null hypothesis was rejected and there was significant difference between the pre and post results.
The one-way ANOVA was also performed for FBG measurements over the six occasions. The chosen a was 0.05, F (found in F-distribution table) was 2.21 and the calculated F was 19.65. Again, since the calculated F was larger than 2.21 the null hypothesis was rejected and there was significant difference between the pre and post results.
Pearson correlation coefficient did not reveal any correlation between age and the change of FBG (the difference between the pre and post averages) R=0.11. Also, there was no correlation between age and the change of HbA1c (the difference between the pre and post averages) R=0.27.
With regard to gender, males showed a higher response than females. At the end of the study, 23 patients achieved HbA1c of =/<7% (Fig 5). Of these, 14 were male (40% of the total of 35 male participants) and nine were female (28% of the total of 32 female participants).
It is worth noting that in the nine months before the intervention, five patients had achieved that outcome.
In this study, HbA1c showed continuous decrease over the nine months following the intervention. However, the measurements from the pre- intervention period showed the HbA1c to be slowly improving (Fig 4). This could be attributed to the influence of education through the diabetes clinic and the role of the ANP. The improvement in HbA1c following SMBG was more significant (Fig 4 and Fig 5). Males faired marginally better in this study. This could be due to the invasiveness of the SMBG procedure making females more reluctant to perform SMBG frequently. Although no difference regarding age was found in this study, it is known that younger people and those with higher levels of education are more likely to perform SMBG (Latalski et al, 2005).
This study focused on SMBG and explained to patients when they could perform SMBG but they were not restricted to specified frequency. The requirement for minimal disruption in this busy inner-city general practice and the limited number of glucometers introduced in the diabetes clinic did not allow for random allocation.
The participants in this study were patients who were ‘regular Fig were related and could keep each other motivated to control their blood glucose. Furthermore, all were willing to participate. Since all the above factors can cause selection bias, it is doubtful that all patients would have the same enthusiasm for SMBG. Patients also received regular feedback on their outcome and this probably helped to improve their results. This is in accordance with other studies (Jansen, 2006).
Another pitfall in this study is that there was a slight improvement in the glycaemic control in the pre-intervention period (Fig 1 and Fig 3); this was attributed to the new diabetes clinic, education and the role of the ANP. Therefore, although there was a statistically significant difference between the pre and post improvements, it is difficult to attribute the post-intervention improvement solely to the SMBG. It may have been a good idea to compare the post-intervention measurements with those taken before the diabetes clinic was set up, but patients were not seen regularly before this and no data was collected.
It would have been helpful to produce a questionnaire to identify what specifically patients felt helped them to control their glycaemia, such as the education, SMBG, dietary advice, diabetes clinic, effect of the medication, or a combination of all these factors. Similarly, it is hoped that the improved glycaemic control achieved by the patients is not due to the ‘Hawthorne effect’, in which a change such as the introduction of the diabetes clinic or SMBG leads to a temporary change in behaviour.
The strength of this study lies in its prospective nature, its power of 80% and the large sample size of 67 patients. The regular attendance of participants was remarkable and the follow-up was good (18 months). Only 33 patients were lost from the study; seven of these were converted to insulin and one became pregnant.
There are significant cost implications in performing SMBG with all patients with type 2 diabetes. However, considering the quality-of-life-years gained, when comparing this cost directly with the cost of complications, the difference was well within the current UK ‘willingness to pay’ limits (Palmer et al, 2006).
SMBG was effective in improving glycaemic control in South Asian patients with type 2 diabetes who are on oral hypoglycaemic agents. The effect included improvements in both HbA1c and FBG. SMBG was beneficial to be used along with other methods such as education, diet, exercise and medications. Nevertheless, to fully assess its beneficial effects in the community as a whole, a large and well-designed RCT is required. Additional work is also needed to establish optimal frequency and timing of SMBG.
The authors wish to acknowledge the support received from Dr Saed Haque and Dr Tim Marshall, both statisticians from the University of Birmingham, regarding the statistical tests carried out in this study.
American Diabetes Association (1997) Self-monitoring of blood glucose (Consensus Statement). Diabetes Care; 19: Supplement 1, S62-S66.
Coster, S. et al (2000) Monitoring blood glucose control in diabetes mellitus: a systematic review. Health Technology Assessment; 4: 12.
Evana, J.M.M. et al (1999) Frequency of blood glucose monitoring in relation to glycaemic control; observational study with diabetes database. British Medical Journal; 319: 83-86.
Franciosi, M. et al (2005) Self-monitoring of blood glucose in non-insulin treated diabetic patients; a longitudinal evaluation of its impact on metabolic control. Diabetic Medicine; 22: 7, 900-906.
Guerci, B. et al (2003) Self-monitoring of blood glucose significantly improves metabolic control in patients with type 2 diabetes mellitus: the Auto- Surveillance Intervention Active (ASIA) study. Diabetes and Metabolism; 29: 6, 587-594.
Harris, M.I. (2001) Frequency of blood glucose monitoring in relation to glycaemic control in patients with type 2 diabetes. Diabetes Care; 24: 979-982.
Jansen, J.P. (2006) Self-monitoring of glucose in type 2 diabetes mellitus: a Bayesian meta-analysis of direct and indirect comparisons. Current Medical Research and Opinion; 22: 4, 671-681.
Karter, A.J. et al (2001) Self-monitoring of blood glucose levels and glycaemic control; the Northern California Kaiser Permanente diabetes registry. American Journal of Medicine; 111: 1-9.
Latalski, M. et al (2005) Frequency of self-monitoring in relation to metabolic control in patients with type 1 and type 2 diabetes treated at the diabetic clinic of the institute of agricultural medicine in Lublin. Wiadomosci Lekarskie; Supplement 1: 305-312.
Martin, S. et al (2006) Self-monitoring of blood glucose in type 2 diabetes and long-term outcome: an epidemiological cohort study. Diabetologia; 49: 2, 271-278.
Oria-Pino, A. et al (2006) Effectiveness and efficacy of self-measurement of capillary blood glucose in patients with type 2 diabetes mellitus. Medicina Clinica; 126: 19, 728-735.
Owens, D. et al (2004) Blood glucose self- monitoring in type 1 and type 2 diabetes: reaching a multidisciplinary consensus. Diabetes and Primary Care; 6: 1, 8-16.
Palmer, A.J. et al (2006) Cost utility analysis in a UK setting of self-monitoring of blood glucose in patients with type 2 diabetes. Current Medical Research and Opinion; 22: 5, 861-872.
Peel, E. et al (2004) Blood glucose self-monitoring in non insulin treated type 2 diabetes: a qualitative study of patients’ perspectives. British Journal of General Practice; 54: 500, 183-188.
Sarol, J.N. Jr et al (2005) Self-monitoring of blood glucose as part of a multi-component therapy among non-insulin requiring type 2 diabetes patients: a meta-analysis (1966-2004). Current Medical Research and Opinion; 21: 2, 173-184.
Skelly, A.H. et al (2005) Self-monitoring of blood glucose in a multiethnic population of rural older adults with diabetes. The Diabetes Educator; 31: 1, 84-90.
UK Prospective Diabetes Study (2000) Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. British Medical Journal; 321: 405-412.
Wagner, J. et al (2005) Invasiveness as a barrier to self-monitoring of blood glucose in diabetes. Diabetes Technology and Therapeutics; 7: 4, 612-619.