Elsevier

Annals of Epidemiology

Volume 23, Issue 2, February 2013, Pages 74-79
Annals of Epidemiology

Impact of diabetes control on mortality by race in a national cohort of veterans

https://doi.org/10.1016/j.annepidem.2012.11.002Get rights and content

Abstract

Purpose

The association between glycated hemoglobin (HbA1c), medication use/adherence, and mortality stratified by race/ethnicity was examined in a national cohort of veterans with type 2 diabetes.

Methods

A total of 892,223 veterans with diabetes in 2002 were followed through 2006. HbA1c category was the main exposure (i.e., HbA1c <7%, HbA1c 7%–8% [reference], HbA1c 8%–9%, and HbA1c >9%). Covariates included age, sex, marital status, rural/urban residence, geographic region, number of comorbidities, and diabetes medication use/adherence (i.e., adherent, medication possession ratio ≥80%; nonadherent; and nonusers). HbA1c and medication use/adherence varied over time, and Cox regression models accounting for time-varying variables were used.

Results

In nonmedication users, HbA1c greater than 9% predicted higher mortality risk relative to HbA1c of 7%–8% in non-Hispanic whites (hazard ratio [HR], 1.55; 95% confidence interval [95% CI], 1.43–1.69), non-Hispanic blacks (NHB) (HR, 1.58; 95% CI, 1.34–1.87), and Hispanics (HR, 2.22; 95% CI, 1.75–2.82). In contrast, in nonadherent medication users, HbA1c less than 7% predicted higher mortality risk in NHB (HR, 1.12; 95% CI, 1.05–1.20), whereas HbA1c greater than 9% only predicted mortality in non-Hispanic whites (HR, 1.11; 95% CI, 1.06–1.16). In adherent medication users, HbA1c less than 7% predicted higher mortality in NHB (HR, 1.18; 95% CI, 1.07–1.31), whereas HbA1c greater than 9.0% predicted higher mortality risk across all race/ethnic groups.

Conclusion

We found evidence for racial/ethnic differences in the association between glycemic control and mortality, which varied by medication use/adherence.

Introduction

Veterans Affairs/Department of Defense Clinical Practice Guidelines for the Management of Diabetes Mellitus in Primary Care consider HbA1c levels 9.0% or greater to be uncontrolled and recommend targeted HbA1c levels less than 7.0% only “for patients with very mild or no microvascular complications of diabetes and those free of major concurrent illnesses and with a reasonable life expectancy [1].” In contrast, the American Diabetes Association Clinical Practice Recommendations consider HbA1c levels 7.0% or greater to be uncontrolled but acknowledge that “less stringent HbA1c goals may be appropriate for patients with a history of hypoglycemia, limited life expectancy, advanced microvascular or macrovascular complications, and extensive comorbid conditions [2].” Recent clinical trials demonstrate that tight control of hyperglycemia in individuals with diabetes does not prevent the associated complications and mortality [3], [4]. In light of these findings, it is important to consider what the best HbA1c cut point is for poor control and whether these cut points should vary by race/ethnicity.

Although prior studies have focused on racial/ethnic differences in HbA1c levels and racial/ethnic differences in mortality, few studies have examined whether diabetes control (i.e., tight or poor) differentially impacts outcomes based on race/ethnic group. Recent meta-analyses of the effect of glycemic control on outcomes comprise randomized controlled trials, which included racial/ethnic minorities; however, presented analyses do not examine whether race/ethnic group modified the association between HbA1c levels and outcomes [5], [6], [7], [8], [9]. Although glucose is not the only factor that impacts HbA1c levels, HbA1c levels are highly correlated with average glucose levels over the proceeding 3-month period [10]. The social determinants of poor medication adherence need to be better elucidated to understand why Hispanics and non-Hispanic black (NHB) have high rates of poorly controlled diabetes. Moreover, it is critical to know whether the relationship between HbA1c and outcomes differs by racial/ethnic group. In other words, does the documented racial/ethnic difference in mean HbA1c levels translate to poorer outcomes in the minority populations, or is the relationship between HbA1c level and outcome dependent on the racial/ethnic group. If diabetes control differentially impacts outcomes based on race/ethnic group, the information is critical to setting targeted individualized levels for good control. The present study fills an important gap by examining the association between HbA1c and mortality stratified by racial ethnic group in a national cohort of veterans with type 2 diabetes over a 5-year follow-up period. Our hypothesis is that diabetes control (i.e., tight and poor) differentially impacts mortality based on race/ethnic group. In addressing our hypothesis, we felt that it was important to consider medication use and adherence as potential modifiers of the association between HbA1c and mortality.

Section snippets

Study population

A national cohort of veterans with type 2 diabetes was created by linking multiple patient and administrative files from the Veterans Health Administration (VHA) National Patient Care and Pharmacy Benefits Management (PBM) databases. Veterans were included in the cohort if they had type 2 diabetes defined by two or more International Classification of Diseases, Ninth Revision codes for diabetes (250, 357.2, 362.0, and 366.41) in the previous 24 months (2000 and 2001) and during 2002 from

Results

The study population consisted of 892,223 veterans with diabetes in 2002 who were followed until death, loss to follow-up, or through December 2006. Mean HbA1c was 7.3% in NHW, 7.8% in NHB, 7.6% in Hispanics, and 7.4% in the other race/ethnic group category. During the follow-up period, 20.83% of individuals in the cohort died. Table 1 shows the characteristics of the sample stratified by race/ethnic group.

In 2002, unadjusted mean HbA1c levels were 6.5% in individuals not using medication for

Discussion

After adjustment for demographic factors, comorbidity burden, and medication status (i.e., medication use and adherence), HbA1c levels greater than 9% predicted higher mortality across all ethnic groups; however, mortality varied by medication use and adherence status. HbA1c levels of 8.0%–9.0% (compared with levels of 7.0%–8.0%) were associated with an increased mortality risk of 10% in NHW, 9% in Hispanics, and 25% in the other racial/ethnic group but were not associated with increased

Acknowledgments

This study was supported by grant IIR-06-219 funded by the VHA Health Services Research and Development program. The funding agency did not participate in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.

The manuscript represents the views of the authors and not those of the VHA or Health Services Research and Development Service.

Drs. Kelly J Hunt, Mulugeta Gebregziabher, and Leonard E.

References (39)

  • E. Selvin et al.

    Meta-analysis: glycosylated hemoglobin and cardiovascular disease in diabetes mellitus

    Ann Intern Med

    (2004)
  • D.M. Nathan et al.

    Translating the A1C assay into estimated average glucose values

    Diabetes Care

    (2008)
  • D.R. Miller et al.

    Who has diabetes? Best estimates of diabetes prevalence in the Department of Veterans Affairs based on computerized patient data

    Diabetes Care

    (2004)
  • A.N. West et al.

    Defining “rural” for veterans' healthcare planning

    J Rural Health

    (2010)
  • Department of Veterans Affairs Field Research Advisory Committee. Department of Veterans Affairs, Veterans Health...
  • H. Quan et al.

    Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data

    Med Care

    (2005)
  • D.R. Cox

    Regression models and life tables

    J R Stat Soc Ser B

    (1972)
  • D.Y. Lin et al.

    Checking the Cox model with cumulative sums of martingale-based residuals

    Biometrika

    (1993)
  • L.D. Fisher et al.

    Time-dependent covariates in the Cox proportional-hazards regression model

    Annu Rev Public Health

    (1999)
  • Cited by (0)

    View full text