A prospective study of health and risk of mortality after spinal cord injury

Abstract

Objective: To test hypothesized relationships between multiple health parameters and mortality among persons with spinal cord injury (SCI) while controlling for variations in biographical and injury characteristics.

Design: Prospective cohort study with health data collected in late 1997 and early 1998 and mortality status ascertained in December 2005.

Setting: A large rehabilitation hospital in the Southeastern United States.

Participants: A total of 1389 adults with traumatic SCI, at least 1 year postinjury.

Interventions: Not applicable.

Main outcome measures: The primary outcome was time from survey to mortality (or time of censoring). Mortality status was determined using the National Death Index and the Social Security Death Index. There were 225 deaths (16.2%) by December 31, 2005.

Results: Cox proportional hazards modeling identified several significant health predictors of mortality status, while controlling for biographic and injury factors. Two sets of analyses were conducted--the first identifying the significance of a single variable of interest and the second analysis building a comprehensive model based on an optimal group of variables. Multiple types of health conditions were associated with mortality. The best set of health predictors included probable major depression, surgeries to repair pressure ulcers, fractures and/or amputations, symptoms of infections, and days hospitalized. Inclusion of these variables, along with a general health rating, improved prediction of survival compared with biographic and injury variables alone, because the pseudo R(2) increased from .12 to .18 and the concordance from .730 to .776.

Conclusions: In addition to secondary conditions that have been the traditional focus of prevention efforts (eg, pressure ulcers, urinary tract infections), amputations, fractures, and depressive symptoms were associated with higher risk for mortality; however, further research is needed to identify the association of specific conditions with causes of death and to determine whether interventions can modify these conditions and ultimately improve survival.

Fig 1

Estimated survival curves for the 5 functional injury classifications at the mean level of all other variables included in the final model.

Fig 2

Estimated survival curves for persons free of health complications (Good Health) relative to persons with numerous complications (Poor Health). Good Health is defined as nondepressed, zero days hospitalized over the past year, free of symptoms of infections, no broken bones or amputations, and no surgical repairs of ulcers. Poor Health was defined as probable major depression, with 13.1+1 SD days hospitalized, 20.0+1 SD occurrences of infection-like symptoms, a broken bone or amputation, and 2.1+1 SD surgeries to repair pressure ulcers. All other variables included in the model were assumed to be at the mean level.