Self-monitoring of blood glucose significantly improves metabolic control in patients with type 2 diabetes mellitus: the Auto-Surveillance Intervention Active (ASIA) study

Diabetes Metab. 2003 Dec;29(6):587-94. doi: 10.1016/s1262-3636(07)70073-3.


Objective: Self monitoring of blood glucose (SMBG) in type 2 diabetes is a topic of current interest (imbalance between increased health-care costs and improvement in compliance with treatment and diet). An open label randomized prospective study was designed to compare changes in metabolic control over 6 months in patients managed with usual recommendations alone (conventional assessment group) or combined with SMBG.

Research design and methods: Patients not treated with insulin or previously self monitored, 40 to 75 years of age, with a diagnosis of type 2 diabetes > 1 year and standardized HbA(1c) level > =7.5 and< =11% were randomized to either a control group or SMBG group. They were followed up every 6 weeks over 24 weeks. Patients in the SMBG group were given the same device (Ascensia Esprit Discmeter, Bayer) and were required to perform at least 6 capillary assays a week (3 different days of the week, including weekend). Management of patients was standardized, including drugs, diet and physical activity. The primary efficacy criterion was change in HbA(1c) level in Intent To Treat (ITT) patients. Assays were performed at baseline, at 3 and 6 months using the calibrated DCA 2000(R) device (Bayer).

Results: Two hundred sixty five general practitioners randomized 988 patients (ITT Population), but 689 patients were evaluable for the primary criterion. At the endpoint, HbA(1c) was lower in the SMBG group (8.1 +/- 1.6%) than in the conventional treatment group (8.4 +/- 1.4%, P=0.012). The change in HbA(1c) levels between baseline and endpoint was classified into two classes: improvement if a change > 0.5% occurred, stability or worsening in case of a change< =0.5%; 57.1% of patients in the SMBG group vs 46.8% in the control group had an improvement in HbA(1c) level (P=0.007) after 3 months. A steady state was reached during the last 3 months. A multivariate logistic regression analysis was performed and identified factors predictive of improvement in HbA(1c) levels: HbA(1c) at baseline: odd ratio (OR)=1.749 (P<0.001), SMBG group (reference value: SMBG group): OR=0.665 (P=0.015), duration of diabetes: OR=0.953 (P=0.001) and BMI: OR=0.962 (P=0.039).

Conclusions: This study is the first multicenter, controlled, prospective trial conducted on a large number of patients demonstrating that SMBG was statistically associated with a better quality of metabolic control than usual traditional recommendations alone in type 2 diabetes.

Publication types

  • Clinical Trial
  • Multicenter Study
  • Randomized Controlled Trial

MeSH terms

  • Adult
  • Aged
  • Blood Glucose Self-Monitoring* / statistics & numerical data
  • Diabetes Mellitus, Type 2 / blood*
  • Diabetes Mellitus, Type 2 / complications
  • Diabetes Mellitus, Type 2 / therapy*
  • Diet
  • Exercise
  • Glycated Hemoglobin / analysis
  • Humans
  • Hypoglycemia / epidemiology
  • Hypoglycemic Agents / administration & dosage
  • Middle Aged
  • Odds Ratio
  • Patient Compliance
  • Prospective Studies


  • Glycated Hemoglobin A
  • Hypoglycemic Agents