Self-monitoring of blood glucose in type 2 diabetes: Results of the 2011 Survey on Living with Chronic Diseases in Canada
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
by Calypse B. Agborsangaya, Cynthia Robitaille, Peggy Dunbar, Marie-France Langlois, Lawrence A. Leiter, Sulan Dai, Catherine Pelletier and Jeffrey A. Johnson
Self-monitoring of blood glucose (SMBG) is a useful component of diabetes management. SMBG entails collecting information on blood glucose levels at different times of the day. SMBG is, therefore, important for patients with type 2 diabetes mellitus (T2DM) using insulin, because it enables them to detect hypoglycemia and self-administer doses of insulin for glycemic control.1-3
However, for T2DM patients who do not use insulin, evidence supporting the clinical relevance and cost-effectiveness of SMBG is equivocal.4-9 Hypoglycemia in these patients is rare,10 and the extent to which they can adjust anti-hyperglycemic agents in response to SMBG readings is generally limited.11 Nonetheless, some Clinical Practice Guidelines (CPGs) recommend routine SMBG in T2DM patients who do not use insulin.11CPGs in Canada recommend that SMBG be individualized for each patient,11 but the Canadian Agency for Drugs and Technologies in Health does not recommend the routine use of blood glucose test strips for this patient population.1
Studies evaluating the prevalence of SMBG among T2DM patients have had different outcome measures and have focused on specific subgroups.4,12-16 Only a few reports have been based on nationally representative data.17,18 Because self-monitoring may vary with population characteristics,19 a broader examination of the extent of SMBG is needed to inform discussions about its utility, and perhaps, to update CPGs. As well, associations between the frequency of SMBG and factors such as self-management support and insurance coverage should be considered.
This study examines the prevalence, frequency and correlates of SMBG, based on a nationally representative sample of patients with T2DM.
The data are from the 2011 Survey on Living with Chronic Diseases in Canada - Diabetes Component (SLCDC-DM).20 Respondents aged 20 or older to the 2010 Canadian Community Health Survey (CCHS) who reported that a health care professional had diagnosed them as having diabetes were eligible for participation in the 2011 SLCDC-DM. The CCHS excludes members of the Canadian Forces and residents of First Nations Reserves, Crown lands, institutions, and the three territories. Of the 3,590 CCHS respondents contacted, 2,933 agreed to participate in the SLCDC, for a response rate of 81.7%. Additional information about the 2011 SLCDC is available online.20
This analysis excluded respondents who reported having type 1 diabetes (n = 211), those with another type of diabetes (n = 19), those who did not know the type (n = 14), and women who did not state whether they had been diagnosed with diabetes other than during pregnancy (n = 7). This left 2,682 respondents.
The socio-demographic characteristics of the population examined in this analysis are sex, age, racial background, education and household income. Except for age and sex, socio-demographic information came from linkage of the 2010 CCHS and the 2011 SLCDC.
To determine the prevalence and frequency of SMBG, respondents were asked, “Do you or a family member or friend check your blood sugar?” Those who answered “yes” were asked, “How often do you....check your blood sugar level?” Patients reported frequency in different units (per day, week, month or year); for this analysis, all values were converted to “per week” (for example, per dayvalues were multiplied by 7).
For the health care correlates of SMBG, patients were asked if a doctor or other health professional had ever recommended monitoring their blood sugar level and/or had given them an A1C test (a simple lab test that reflects average blood glucose level over the last three months). Patients who replied affirmatively to the latter were asked, “The last time your A1C was measured by a health professional, . . . your blood sugar was: normal, borderline, high, low, did not say?” For this analysis, borderline and normal were combined. Respondents were also asked: the type of medication they used to control their glucose level; if they had insurance coverage for glucose monitoring; and if they had visited an emergency room in the past 12 months because of hypoglycemia.
The weighted prevalence and mean frequency of SMBG were determined by socio-demographic and health care characteristics. Z statistics were used to identify differences in the prevalence and frequency of SMBG across descriptors, receipt of health professional support, clinical monitoring, insurance coverage, hypoglycemia-related emergency room visits, and medication use. Multivariate prevalence rate ratios for the associations between these clinically relevant characteristics and SMBG were tested using binomial regression models.
Point estimates were weighted using the survey sample weights to reflect the Canadian population aged 20 or older living in private dwellings. To account for the complex survey design, 95% confidence intervals were calculated using exact standard errors generated through bootstrap re-sampling methods.21 Missing data were excluded from the analyses. Data were analyzed using SAS Enterprise Guide 4.1 (Cary, North Carolina).
Prevalence of self-monitoring of blood glucose
The majority of the 2,682 respondents aged 20 or older with T2DM were men (58%), White (81%), and had completed postsecondary education (67%). Close to half (48%) were aged 65 or older (Table 1).
Most of these T2DM patients (88%) reported having used SMBG during the past 12 months. No significant differences in the prevalence of SMBG were observed by sex, age, education, household income, racial background or time since diagnosis (Table 2). When these characteristics were taken into account, health care factors associated with SMBG were health professional recommendations, having insurance coverage for the procedure, and having had an A1C test from a health professional. As expected, patients not using any medication were less likely to report SMBG than were those using insulin only. However, no significant difference was apparent between those using insulin only and those taking insulin plus oral medication, or an oral agent only.
Frequency of self-monitoring of blood glucose
Patients who self-monitored their blood glucose did so an average of more than 10 times a week, a figure that did not differ significantly by sex, age, education or racial background (Table 3). However, frequency tended to decrease at higher household income levels. For example, those whose household income was less than $15,000 reported that they self-monitored their blood glucose around 12 times a week, compared with about 8 times a week for those whose household income was $80,000 or more.
The average frequency of SMBG was higher among patients who reported having had an A1C test by a health professional than among those who did not (10.6 versus 8.5 times a week). As well, the average frequency was significantly higher among respondents reporting high A1C levels, compared with those reporting normal levels (12.6 versus 10.1). SMBG was significantly more frequent among patients who had a hypoglycemia-related emergency room visit, compared with those who did not (18.4 versus 10.2 times a week). SMBG frequency was also related to medication use: 18.7 times a week for patients taking insulin only, 8.4 times for those taking oral medication only, and 6.3 times for those not taking insulin or oral medications.
According to the results of this study, SMBG is common and frequent among adults with T2DM. The estimated prevalence was almost 88%, close to the 91% cited in a previous Canadian report.18 Factors significantly associated with SMBG included a health professional recommendation, having had an A1C test by a health professional, and health insurance coverage for glucose monitoring. The prevalence of SMBG did not differ for patients using only insulin, compared with those using insulin and oral medication, or oral medication only.
In fact, the prevalence of SMBG was 89% in patients using only oral medication. Hypoglycemia is rare in such patients,10 and the extent to which doses of oral anti-hyperglycemic agents can be adjusted in response to SMBG is limited.2 A systematic review conducted by the Canadian Agency for Drugs and Technology in Health3 reported that SMBG in patients who do not use insulin results in only marginal improvement in glycated hemoglobin (A1C) levels. Moreover, SMBG among these patients may entail substantial financial outlays—by the patients themselves and by payers.12,22,23
Having insurance coverage for all or part of the cost of glucose monitoring was associated with a higher prevalence of SMBG. Although the causal direction of the associations cannot be ascertained, insurance and other reimbursement policies may encourage SMBG in patients with T2DM.24,25 In Canada, reimbursement for SMBG varies widely by province, ranging from full coverage under a pharmaceutical benefits program with co-payments and deductibles to no coverage.26
The advice of health professionals is important for the management of T2DM.27 This analysis shows that patients were more likely to report SMBG if a health professional had recommended it, or if they had received an A1C test from a health professional.
In this study, the frequency of SMBG among patients using insulin is consistent with national and international CPGs,2,11,28 including those of the Canadian Agency for Drugs and Technology in Health.1,29 Patients on insulin-only treatment used SMBG an average 19 times per week, which is close to the Canadian CPGs’ recommendation of a minimum of three times a day.11
However, the frequency of SMBG by medication use in this analysis was about twice the values in previous Canadian reports based on claims data12,30 and in some international studies.14,31 Claims data are usually derived from fee-for-service administrative records, which potentially exclude individuals whose physicians do not use that billing method. In fact, the findings of the present analysis are similar to patterns reported in a study using a combined Canadian aggregated and claims level dataset.22
The higher mean frequency of SMBG among patients with lower household incomes is noteworthy. For people without comprehensive coverage for SMBG, out-of-pocket expenditures may result in income-related disparities in SMBG frequency.32 In Canada, however, the extent of coverage for glucose monitoring differs by province. Whether the differences by household income in the frequency of SMBG are due to differences in provincial coverage policies, age, literacy level, or disease severity remains to be tested.
This study has several limitations. The cross-sectional design precludes determination of causality. The data were self-reported and so may be subject to social desirability bias, in that patients may report diabetes self-care in a way that they think is socially acceptable. No objective measures were available against which to assess the accuracy of patient reports or the recommendations and support patients received from health professionals. Sampling bias may limit the generalizability of the findings; for example, residents of the three territories and of First Nations reserves were not included in the study. Finally, other factors that may confound the associations, such as access to care, patients’ language proficiency and health literacy, and health professionals’ practice style, were not captured in this analysis.
This study presents comprehensive, nationally representative data on the prevalence, correlates and frequency of SMBG among adults with T2DM. SMBG is common and is associated with receipt of recommendation for SMBG and an A1C test from a health professional, as well as having insurance coverage for glucose monitoring. The prevalence of SMBG did not differ among patients using only insulin and those taking medication only. These findings have implications for the development of clinical recommendations for SMBG in persons with T2DM.
Calypse B. Agborsangaya is a post-doctoral fellow supported by a Canadian Institutes for Health Research (CIHR) Team Grant to the Alliance for Canadian Health Outcomes Research in Diabetes (reference #: OTG- 88588), sponsored by the CIHR Institute of Nutrition, Metabolism and Diabetes. Jeffrey A. Johnson is a Senior Scholar with Alberta Innovates—Health Solutions and holds a Canada Research Chair in Diabetes Health Outcomes.