Talk to the Veterans Crisis Line now
An official website of the United States government
Search | Search by Center | Search by Source | Keywords in Title
Pentakota S, Tseng C, Rajan M, Pogach LM. Impact of comorbid conditions on long-term glycemic control in Veterans with newly identified diabetes mellitus. Poster session presented at: AcademyHealth Annual Research Meeting; 2012 Jun 25; Orlando, FL.
Research Objective: There is increasing awareness among clinicians and policy makers regarding the complexity of care in patients with comorbidities, who tend to be often excluded from clinical trials. Lack of taxonomy for classification of comorbidities, is one gap in our knowledge on how comorbidities impact care. Our objective was to evaluate long term glycemic control among veterans with newly identified diabetes mellitus (DM), stratified by age-groups and categories of comorbidity based on degree of concordance with glycemic management. We hypothesized that individuals with concordant conditions would have improved control, but not discordant conditions when compared to those without comorbid conditions. Study Design: Retrospective cohort study Population Studied: Veterans (n = 27,654) with newly identified DM in FY2002 based upon administrative ICD 9-CM codes or DM-related medication records in a 2 year look-back period (FY00-01) were followed until end of FY09. Veterans were classified into 5 chronic comorbid illness groups (CCIG) using a competing demands framework: none, only concordant (DM related), only discordant (non-DM related), both, or dominant (short life expectancy). Approximately 60 comorbidities were identified using a minimum of two codes during the look-back (FY00-01) and baseline year (FY02). All HbA1c values in FY02-09, at least 30 days apart, were included in the analysis. Annual means of the HbA1c for the 5 CCIGs were analyzed. General linear regressions were used to evaluate annual trends of HbA1c by CCIG and age-groups. Principal Findings: We identified 27,564 veterans: 96.6% male, 30% < 55 years, 25% 55-64 years, 45% > 65 years, and 25% had no comorbidity. Number of HbA1c tests decreased from 50,272 in FY02 to 32,618 in FY09. The overall mean (sd) HbA1c and range by CCIG in FY02 was [7.06 (1.83); range: 6.88(dominant)- 7.17(only concordant)]; it decreased to [6.63(1.26); 6.47- 6.67] in FY03 and gradually increased in the following years to [7.18(1.46); 6.62-7.23] in FY09. Glycemic control was worse in veterans < 55years in all years, with FY02 (7.43(2.20), range: 7.14-7.92) and FY09 (7.56(1.72); 7.25-7.93). In all age groups, veterans with only concordant or no comorbidities had higher HbA1c levels, followed by only discordant, both (concordant/discordant), and dominant groups. Age-and-sex adjusted annual slopes were similar across all CCIGs [FY02-FY03: -0.54%; FY03-FY09: 0.08%]. Among veterans < 55 years, the annual slope for FY03- FY09 was 0.14% (no comorbidities); patients in only discordant (0.11%), both (0.10%), and dominant (0.10%) groups were significantly different from those without comorbidities. Among older age groups, annual slopes did not differ by CCIG. Conclusions: HbA1c trends were not linear. They followed a downward trend for the first year, and then gradually increased in all CCIGs. Contrary to our hypothesis, younger patients with discordant conditions had better glycemic control over time after the initial follow-up year compared to those without comorbidities. On the other hand, glycemic control among those with only concordant conditions was similar to those without comorbidities. Implications for Policy, Delivery, or Practice: Since younger patients would benefit the most from improved glycemic control, our findings indicate the need to
