Multivariate analysis

Abstract

Multivariate analyses are conducted in this chapter that assess the independent contribution of each of the risk factor variables simultaneously. Within each league, stepwise backwards elimination survival analyses will remove non-significant variables until a final model is derived that contains significant items only. Between-league multivariate survival analyses will also be conducted with all risk factor variables entered. Life expectancy will be analyzed in this chapter. Within the NBA, African-American race and overweight BMI are significant independent risk factors for mortality. In addition, 10 or more years of NBA playing experience significantly reduces the risk of mortality. Controlling for the other significant variables, African-American race and overweight BMI within NBA players increases mortality risk by 58% and 57% respectively. Playing 10 or more years in the NBA reduces mortality risk by 39%. Within NFL players, all the risk factor variables are significant independent predictors of mortality. African-American race and obese BMI increase mortality risk by 30% and 47% respectively. Also within NFL players, birth in the South and 3–5 years of career playing experience raise mortality risk by 42% and 23% respectively. Between-league multivariate results indicate that NBA players have a 26% elevated risk of mortality compared to NFL players which is significant. Within- and between-league life expectancy results resemble survival findings in terms of which variables are significant. NFL players have a life expectancy that is 0.8 years longer than NBA players and this difference is statistically significant.