Skip to main content
. Author manuscript; available in PMC: 2012 Jul 30.
Published in final edited form as: Obesity (Silver Spring). 2010 Jun 17;19(1):83–93. doi: 10.1038/oby.2010.120

Table 2. Adjusted means of change in weight and WOMAC scores generated from mediation analyses, path β.

Adjusted mean changea (s.e.) P for difference Raw mean change (s.d.) Effect size for change between ILI and DSEb
Weight (kg) changea N = 1,755
 DSE −0.78 (0.49) 0.001 −0.90 (5.14) −1.18
 ILI −9.02 (0.48) −9.08 (8.31)
WOMAC pain changec N = 1,758
 DSE (n = 856) −0.11 (0.23) 0.04 −0.28 (3.57) 0.06
 ILI (n = 902) −0.41 (0.22) −0.51 (3.45)
WOMAC physical function changec N = 1,759
 DSE (n = 857) −1.28 (0.69) 0.001 −0.73 (11.32) −0.15
 ILI (n = 902) −2.73 (0.67) −2.30 (9.92)
WOMAC stiffness subscale changec N = 1,759
 DSE (n = 857) −0.34 (0.11) 0.16 −0.24 (1.72) −0.07
 ILI (n = 902) −0.44 (0.11) −0.36 (1.69)
WOMAC Summary Score changec N = 1,759
 DSE (n = 857) −1.73 (0.96) 0.004 −1.22 (15.34) −0.13
 ILI (n = 902) −3.54 (0.94) −3.10 (13.57)

DSE, Diabetes Support and Education; ILI, Intensive Lifestyle Intervention; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index.

a

Adjusted for age, gender, race/ethnicity, study site, Beck Depression Inventory Score, use of nonsteroidal anti-inflammatory drugs and baseline weight.

b

Effect size calculated as (ILImean − DSEmean)/pooled standard deviation.

c

Adjusted for baseline values of outcome variable, age, gender, race/ethnicity, study site, Beck Depression Inventory Score, use of nonsteroidal anti-inflammatory drugs and baseline weight.