Skip to main content

Table 3 Univariate and multivariate regression models of factors predicting adverse pathologic outcomes in men undergoing radical prostatectomy at University of Pennsylvania, 1990–2012

From: The impact of body mass index on treatment outcomes for patients with low-intermediate risk prostate cancer

Univariate analysis OR 95 % CI p value
Age 1 0.98 to 1.03 0.80
Racea    
White 1 Reference  
African-American/Black 1.21 0.83 to 1.77 0.32
 Serum PSA 1.08 1.05 to 1.11 <0.001
Clinical stagea    
T1c 1 Reference  
T2a 1.05 0.64 to 1.72 0.85
T2b 3.00 1.11 to 8.17 0.03
   > T2c 2.58 0.97 to 6.85 0.06
 Year of Prostatectomy 0.97 0.94 to 1.00 0.05
Clinical Gleason scorea    
   ≤ 6 1 Reference  
7 2.21 1.48 to 3.29 <0.001
   ≥ 8 5.06 3.03 to 8.43 <0.001
Body mass index, categoricala    
   < 25 kg/m2 1 Reference  
25 kg/m2 to <30 kg/m2 1.48 0.86 to 2.53 0.02
   ≥ 30 kg/m2 2.20 1.27 to 3.81 0.005
Multivariate analysis
 Age 1 0.97 to 1.03 0.74
Racea    
White 1 Reference  
African-American/Black 0.81 0.50 to 1.32 0.40
 Serum PSA 1.12 1.08 to 1.17 <0.001
Clinical stagea    
T1c 1 Reference  
T2a 1.24 0.72 to 2.14 0.43
T2b 2.28 0.70 to 7.38 0.17
   > T2c 1.21 0.37 to 4.03 0.75
 Year of Prostatectomy 0.99 0.95 to 1.04 0.69
Clinical Gleason scorea    
   ≤ 6 1 Reference  
7 2.01 1.24 to 3.25 0.005
   ≥ 8 5.97 3.02 to 11.78 <0.001
Body mass index, categoricala    
   < 25 kg/m 2 1 Reference  
25 kg/m 2 to <30 kg/m 2 1.58 0.81 to 3.07 0.18
   ≥ 30 kg/m 2 2.58 1.30 to 5.09 0.006
  1. Abbreviations: PSA prostate-specific antigen, ≥2 adverse pathologic features as endpoint
  2. aDenotes categorical variables. Body mass index, categorical uses BMI <25 kg/m2 as reference
  3. Race use White as reference category; “Other” race was dropped from model due to small numbers
  4. Clinical stage, categorical uses (T1; T2a, T2b; >T2c), with T1c as reference category
  5. Clinical Gleason score, categorical (6, 7, ≥8); with 6 as reference category
  6. P values derived from a logistic regression model
  7. Boldfaced values represent statistically significance