Background: The aging process can be described as the change in health-related variables over time. Unfortunately, simple graphs of available data may be misleading if some people die, since they may confuse patterns of mortality with patterns of change in health. Methods have been proposed to incorporate death into self-rated health (excellent to poor) and the SF-36 profile scores, but not for other variables.

Objectives: (1) To incorporate death into the following variables: ADLs, IADLs, mini-mental state examination, depressive symptoms, body mass index (BMI), blocks walked per week, bed days, hospitalization, systolic blood pressure, and the timed walk. (2) To discuss variables and settings for which this approach is helpful. (3) To use the approach to illustrate the effect of stroke on these variables.

Setting: The Cardiovascular Health Study of 5,888 older adults, studied up to nine years. Mean age was 73 at baseline, and 658 had an incident stroke during follow-up.

Methods: We categorized each variable, added a category for death, and examined stacked bar graphs over time. We dichotomized each variable into healthy/not healthy, assigning dead to the “not healthy” category, and calculated the probability of being healthy one year later, with the deaths set to zero. Trajectories for the 11 variables in the three years before and after an incident stroke were tabled and plotted. Other transformations were derived and discussed.

Results: Graphs that did not account for death were too optimistic. Stroke had an adverse effect on all variables but systolic blood pressure (which improved) and BMI (which showed little change). The short-term effect of stroke was greatest on hospitalization, self-rated health, and IADLs. Alternative coding provided slightly different results.

Conclusions: Categories or values for death can be added to a variety of longitudinal variables, permitting a description of an entire cohort’s trajectory over time. These transformations provide an additional approach for longitudinal studies of the health of older adults where some people die.


Categorical Data Analysis | Health Services Research | Longitudinal Data Analysis and Time Series | Numerical Analysis and Computation | Vital and Health Statistics

Previous Versions

January 29, 2004