TABLE 10-5Logistic Regression Models of Mortality Adding Age, Sex, and One Health Measure

All Age <50 Age 50+
Pseudo-R2 Gain Odds Ratioa Pseudo-R2 Gain Odds Ratioa Pseudo-R2 Gain Odds Ratioa
Age 0.179b 1.09*** 0.001b 0.99 0.092b 1.11***
Female 0.001 0.80 0.001 0.99 0.001 0.09
Adding a single health indicator to model with age and sex
Self-Reported
SRH 0.012 2.28*** 0.028 5.57* 0.012 2.08***
ADL score 0.024 0.76*** 0.048 .51*** 0.023 0.79***
Major disease 0.006 1.82** 0.028 5.45* 0.006 1.64**
Observed
ADL score 0.014 0.78*** 0.001 1.17 0.020 0.78***
BMI < 16 0.010 2.28*** 0.017 3.63 0.010 2.00***
Observations 45,098 29,807 15,291

NOTE: Calculations treat all observations for an individual as a cluster and use individual sample weights. SRH = self-reported health; ADL = activities of daily living; BMI = body mass index.

a

The odds ratio for female is from the model with age and sex; those for the health indicators are from models with age and sex plus that indicator as predictors.

b

This figure is the pseudo-R2 for the model with age as the single predictor.

***

p<.001.

*

p<.05;

**

p<.01;

From: 10, Survey Measures of Health: How Well Do Self-Reported and Observed Indicators Measure Health and Predict Mortality?

Cover of Aging in Sub-Saharan Africa
Aging in Sub-Saharan Africa: Recommendation for Furthering Research.
National Research Council (US) Committee on Population; Cohen B, Menken J, editors.
Washington (DC): National Academies Press (US); 2006.
Copyright © 2006, National Academy of Sciences.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.