Format

Send to

Choose Destination
See comment in PubMed Commons below
Diabetes Care. 2010 Jun;33(6):1347-52. doi: 10.2337/dc09-1444. Epub 2010 Mar 18.

Derivation and validation of a new cardiovascular risk score for people with type 2 diabetes: the new zealand diabetes cohort study.

Author information

1
Department of General Practice and Primary Health Care, School of Population Health, University of Auckland, Auckland, New Zealand. c.elley@auckland.ac.nz

Abstract

OBJECTIVE:

To derive a 5-year cardiovascular disease (CVD) risk equation from usual-care data that is appropriate for people with type 2 diabetes from a wide range of ethnic groups, variable glycemic control, and high rates of albuminuria in New Zealand.

RESEARCH DESIGN AND METHODS:

This prospective open-cohort study used primary-care data from 36,127 people with type 2 diabetes without previous CVD to derive a CVD equation using Cox proportional hazards regression models. Data from 12,626 people from a geographically different area were used for validation. Outcome measure was time to first fatal or nonfatal cardiovascular event, derived from national hospitalization and mortality records. Risk factors were age at diagnosis, diabetes duration, sex, systolic blood pressure, smoking status, total cholesterol-to-HDL ratio, ethnicity, glycated hemoglobin (A1C), and urine albumin-to-creatinine ratio.

RESULTS:

Baseline median age was 59 years, 51% were women, 55% were of non-European ethnicity, and 33% had micro- or macroalbuminuria. Median follow-up was 3.9 years (141,169 person-years), including 10,030 individuals followed for at least 5 years. At total of 6,479 first cardiovascular events occurred during follow-up. The 5-year observed risk was 20.8% (95% CI 20.3-21.3). Risk increased with each 1% A1C (adjusted hazard ratio 1.06 [95% CI 1.05-1.08]), when macroalbuminuria was present (2.04 [1.89-2.21]), and in Indo-Asians (1.29 [1.14-1.46]) and Maori (1.23 [1.14-1.32]) compared with Europeans. The derived risk equations performed well on the validation cohort compared with other risk equations.

CONCLUSIONS:

Renal function, ethnicity, and glycemic control contribute significantly to cardiovascular risk prediction. Population-appropriate risk equations can be derived from routinely collected data.

PMID:
20299482
PMCID:
PMC2875452
DOI:
10.2337/dc09-1444
[Indexed for MEDLINE]
Free PMC Article
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for HighWire Icon for PubMed Central
    Loading ...
    Support Center