The Validity of Physician and Model-Predicted Life Expectancy for Older Type 2 Diabetes Patients

This is a pilot study that will compare the validity of physician-predicted and model-predicted five-year survival of older patients with type 2 diabetes.  While nearly half of patients with type 2 diabetes are over the age of 65, it remains unclear whether diabetes care goals (e.g., glycosylated hemoglobin (HbA1C) <7%) developed for the general population of diabetes patients are appropriate for all older patients.  Older diabetes patients are highly heterogeneous in terms of functional status, comorbidities and life expectancy, and this heterogeneity may alter the risks, benefits, and importance of achieving general population diabetes care goals.  In 2003, geriatric diabetes care guidelines were published that encouraged older patients and their providers to consider less intensive glucose control goals (HbA1C <8%) among older patients with limited life expectancy (<5 years) or significant functional impairment, while continuing to pursue intensive glucose control (HbA1C <7%) among relatively healthy older patients.  While these recommendations are clinically sensible, there are no validated tools for stratifying older diabetes patients by life expectancy.  In current clinical practice, physicians are left to use clinical intuition to predict if a patient has limited life expectancy.  In hopes to better inform clinical decision making we have developed a decision analytic model, the Chicago Type 2 Geriatric Diabetes Model that predicts life expectancy and complications related to diabetes.  Merging data from existing survey data regarding 473 older diabetes patients and the National Death Index, we will create a retrospective cohort study that we will use to compare the validity of physician and computer simulation model predicted five-year survival.  Establishing the validity of the simulation model will be crucial to the future design of prognostic tool interventions for clinical practice.  Identifying the optimal tools for prognostic stratification may have broader use for diabetes quality improvement policies.

Principal Investigator: Elbert S. Huang, MD, MPH
Research Team:
Claire O’Hanlon, MPP, PhDc
Marshall H. Chin, MD MPH
Priya John, MPH
Sang Mee Lee, PhD

Funding Source:
Center on Demography and Economics of Aging Pilot Award