• quality measurement;
  • diabetes mellitus;
  • glycemia


To evaluate the effect of including of clinical actions within 6 months of a glycosylated hemoglobin (HbA1c) level greater than 8% upon measure adherence (pass rates) and to assess the association between patient factors and the likelihood of not passing.


Veterans Health Administration.


Retrospective cohort study for FY2002 to FY2004.


One hundred fifty-three thousand one hundred thirty-two veterans aged 65–74 with diabetes mellitus not taking insulin; 99% were male and 86% white.


The clinical action measure included three categories: (a) initial pass (index HbA1c < 8%); (b) modified pass (index HbA1c ≥ 8%), and the hierarchical occurrence of one of the following events within 6 months after date of index HbA1c: subsequent HbA1c < 8%, being started on insulin (100% weight), new oral medication (50% weight), care in a diabetes mellitus–related clinic (25% weight); and (c) failure (no category met or HbA1c > 9%). Multinomial logistic regression models were used to evaluate associations between participant factors and the likelihood of not passing initially.


Most (82.6%) or the participants had an index HbA1c of less than 8%, and 10.6% were in the modified pass group. The failure rate (17.4%) fell to 6.8% when actions were weighted equally and to 9.4% using different weights. Veterans who are African American (odds ratios (ORs) = 1.43 and 1.44), unmarried (ORs = 1.19 and 1.24), poor (ORs = 1.36 and 1.17), or taking two or more oral antihyperglycemic agents (ORs = 2.61 and 3.72) were significantly more likely to be in the modified pass and failure groups, respectively.


Most veterans with an initial HbA1c of 8% or greater had clinical actions within 6 months. A measure that incorporates multiple treatment options, including education and nutrition, could be of benefit by encouraging dialogue of such options between patients and clinicians.