Risk Factors for Mortality Among Patients With Diabetes

The Translating Research Into Action for Diabetes (TRIAD) Study

  1. Laura N. McEwen, PHD12,
  2. Catherine Kim, MD, MPH13,
  3. Andrew J. Karter, PHD4,
  4. Mary N. Haan, MPH, DRPH2,
  5. Debashis Ghosh, PHD5,
  6. Paula M. Lantz, PHD, MS6,
  7. Carol M. Mangione, MD, MSPH7,
  8. Theodore J. Thompson, MS8 and
  9. William H. Herman, MD, MPH12
  1. 1Department of Internal Medicine, University of Michigan, Ann Arbor, Michigan
  2. 2Department of Epidemiology, University of Michigan, Ann Arbor, Michigan
  3. 3Department of Obstetrics and Gynecology, University of Michigan, Ann Arbor, Michigan
  4. 4Division of Research, Kaiser Permanente, Oakland, California
  5. 5Department of Biostatistics, University of Michigan, Ann Arbor, Michigan
  6. 6Department of Health Management and Policy, University of Michigan, Ann Arbor, Michigan
  7. 7Department of Medicine, University of California, Los Angeles, California
  8. 8Diabetes Translation, Centers for Disease Control and Prevention, Atlanta, Georgia
  1. Address correspondence and reprint requests to Laura N. McEwen, PhD, Internal Medicine/Metabolism, Endocrinology, and Diabetes, 1500 East Medical Center Dr., 3920 Taubman Center, Ann Arbor, MI 48109. E-mail: lmattei{at}med.umich.edu


OBJECTIVE— We sought to examine demographic, socioeconomic, and biological predictors of all-cause, cardiovascular, and noncardiovascular mortality in patients with diabetes.

RESEARCH DESIGN AND METHODS— Survey, medical record, and administrative data were obtained from 8,733 participants in the Translating Research Into Action for Diabetes Study, a multicenter, prospective, observational study of diabetes care in managed care. Data on deaths (n = 791) and cause of death were obtained from the National Death Index after 4 years. Predictors examined included age, sex, race, education, income, duration, and treatment of diabetes, BMI, smoking, microvascular and macrovascular complications, and comorbidities.

RESULTS— Predictors of adjusted all-cause mortality included older age (hazard ratio [HR] 1.04 [95% CI 1.03–1.05]), male sex (1.57 [1.35–1.83]), lower income (<$15,000 vs. >$75,000, HR 1.82 [1.30–2.54]; $15,000–$40,000 vs. >$75,000, HR 1.58 [1.15–2.17]), longer duration of diabetes (≥9 years vs. <9 years, HR 1.20 [1.02–1.41]), lower BMI (<26 vs. 26–30 kg/m2, HR 1.43 [1.13–1.69]), smoking (1.44 [1.20–1.74]), nephropathy (1.46 [1.23–2.73]), macrovascular disease (1.46 [1.23–1.74]), and greater Charlson index (≥2–3 vs. <1, HR 2.01 [1.04–3.90]; ≥3 vs. <1, HR 4.38 [2.26–8.47]). The predictors of cardiovascular and noncardiovascular mortality were different. Macrovascular disease predicted cardiovascular but not noncardiovascular mortality.

CONCLUSIONS— Among people with diabetes and access to medical care, older age, male sex, smoking, and renal disease are important predictors of mortality. Even within an insured population, socioeconomic circumstance is an important independent predictor of health.


  • Published ahead of print at http://care.diabetesjournals.org on 27 April 2007. DOI: 10.2337/dc07-0305.

    A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.

    The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C Section 1734 solely to indicate this fact.

    • Accepted April 15, 2007.
    • Received February 13, 2007.
| Table of Contents

This Article

  1. Diabetes Care vol. 30 no. 7 1736-1741
  1. All Versions of this Article:
    1. dc07-0305v1
    2. 30/7/1736 most recent