Skip to main content
. Author manuscript; available in PMC: 2011 Feb 9.
Published in final edited form as: Am J Psychiatry. 2005 Dec;162(12):2249–2255. doi: 10.1176/appi.ajp.162.12.2249

TABLE 4.

Cox Regression and Discriminant Function Analyses of the Relationship Between Items on the Eating Disorder Examination Questionnaire and the Development of Eating Disorders

Cox Regression Analysesa
Sample Estimates
Jackknifed Estimates
Discriminant Function
Analysis
Item and Optimal Cut Point Hazard
Ratio
95% CI Hazard
Ratio
95% CI Variable
Loading
Standardized
Coefficient
Objective bulimia (≥3 episodes/28 days) 2.29** 1.40–3.75 2.24** 1.26–3.97 0.532 0.416
Purging (≥2 episodes/28 days) 3.60*** 1.63–5.43 3.23 0.96–10.86 0.534 0.498
Low body mass index (<19 kg/m2) 2.59** 1.50–4.47 2.59** 1.35–4.96 0.391 0.394
Secret eating (≥13 instances/28 days) 2.97*** 1.63–5.43 2.92** 1.43–5.99 0.543 0.410
Fear of losing control over eating (≥23 instances/28 days) 1.77* 1.05–2.98 1.70 0.88–3.26 0.447 0.305
Desire to have an empty stomach (≥16 instances/28 days) 1.63* 1.02–2.61 1.67 0.98–2.83
a

Data are presented for dichotomous independent variables; the results were similar for undichotomized variables.

*

p<0.05

**

p<0.01

***

p<0.001.