Risk for Early Mortality After Spinal Cord Injury. Field Initiated Research Grant, National Institute on Disability and Rehabilitation Research, Department of Education; $449,944, October 1, 2003 to September 30, 2006.
The onset of spinal cord injury (SCI) is associated with diminished life expectancy. Although the life expectancy of people with SCI improved dramatically during the decades leading up to the mid-1990’s, recent research of former patients within the model spinal cord injury systems (MSCIS) has shown a reversal in this pattern (DeVivo, Krause, & Lammertse, 1999). Furthermore, there continues to be a substantial portion of deaths that may be prevented with appropriate health maintenance behaviors, underscoring the need for prospective studies that identify factors contributing to premature mortality after SCI. The purpose of the proposed project is to identify the relationships of life adjustment, quality of life (QOL), activities, fitness, and secondary conditions with length of survival and causes of death after SCI. Unlike previous prospective studies, the proposed study will have enhanced generalizability to women and minorities by virtue of oversampling these populations and will use the more sophisticated Cox proportional hazard models to analyze the relationships between the predictors and longevity.
Two prospective studies will be conducted, each utilizing distinct prospective data sets. The prospective data for study 1 was collected from 362 participants in 1990/1991; whereas the prospective data for study 2 was collected from 597 participants in 1995/1996. Similar, but not identical, measures were obtained from each of the two study samples. We will identify current mortality status using the National Death Index (NDI) and the Master Beneficiary Record and Summary Earnings files from the Social Security Administration (SSA). The NDI had a 99% sensitivity rate in a recent pilot study by the principal investigator and will be used for all years in which it is available (mortality status for the year 2001 becomes available about March of 2003). The SSA database (90% sensitivity) will be used for the short period the NDI is not available. Cox proportional hazard models will be used to identify the association of the key predictors with mortality. The most basic biographic and injury related variables will be entered as a block first as statistical controls (these have been the focus of most investigation). The general life adjustment and health variables (also previously investigated) and the more specific secondary conditions variables, none of which have been the focus of previous studies, will be investigated in more detail in relation to mortality, as they hold the greatest promise for intervention. Death certificates will be obtained and used to classify causes of death in accordance with the scheme traditionally used in the model spinal cord injury systems (MSCIS).
If the risk for early mortality is to be decreased after SCI, we must know more about how and why it occurs. The results of this study will help us to learn more about the specific behavioral and health factors leading to avoidable causes of death after SCI and will lay the foundation for prevention strategies to decrease preventable deaths after SCI and increase longevity. The proposed research is ideally poised to accomplish this goal as it expands on previous research by measuring behaviors and secondary health conditions that may directly relate to early mortality.