Bmj Usa

Better systems, not guidelines, for glucose monitoring

BMJ 2004; 329 doi: (Published 09 December 2004) Cite this as: BMJ 2004;329:E332
  1. Paris Roach (paroach{at}, associate professor of clinical medicine1
  1. 1 Division of Endocrinology and Metabolism Indiana University School of Medicine 545 Barnhill Drive, Emerson Hall Room 421 Indianapolis, IN 46202

    Reynolds and Strachan propose that routine home blood glucose monitoring (HBGM) not be used to monitor non-insulin treated patients with type 2 diabetes due to lack of evidence of its benefit. Their underlying concerns are that unnecessary HBGM will result in unjustified costs and that it may adversely affect patients' quality of life. They also suggest that large randomized trials be performed to provide guidance on the optimal use of HBGM in this population.

    The effectiveness of any glucose monitoring program is highly dependent on the ability of patients and providers to integrate HBGM into an overall program of self-care and therapeutic decision making. Current guidelines reflect this reality and seem appropriately non-specific given the wide variation observed in clinical circumstances.1,2 Instead of undertaking the often conflicting task of developing more specific yet more generalizable guidelines for HBGM, providing more guidance regarding the effective incorporation of HBGM into the care plan may be of more practical use to clinicians and should serve to curtail unnecessary testing. Responsibility should be placed on the care system to actively promote patient understanding of the monitoring plan and its purpose.3 Confirmation of patients' understanding of medical information is critical but is frequently overlooked, especially in individuals with low health literacy who are most in need of this counseling.4

    Quality of care and the optimal use of HBGM may be advanced more rapidly if clinicians and trialists focus on identifying patient, provider, and system factors that facilitate the reasoned use or nonuse of HBGM. This is preferable to testing specific monitoring plans in randomized trials in which protocol constraints may limit translatability of the results into clinical practice. Additional study of these issues is warranted, but observational trials may provide a better approach to exploring these issues, as suggested by others.5

    The continuation of routine HBGM in patients with type 2 diabetes who are not meeting their A1c goals on maximal non-insulin therapy is clearly of no benefit, and these patients should be advised to begin insulin therapy in a timely fashion. The fact that the initiation of insulin therapy is often delayed in patients who are poorly controlled on oral agents may partially explain the poor correlation between HBGM frequency and A1c values in cross-sectional studies. Addressing the root causes of this “clinical inertia”6 should both decrease the prevalence of unnecessary HBGM and improve glycemic control in the large population of patients with type 2 diabetes.


    • Competing interests PR has received consulting fees and research grants from Roche Diagnostics, Micron…, and Eli Lilly; worked for Lilly from 1995-2001; and owns stock in Lilly.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.