The Effect of Adherence to Dietary Tracking on Weight Loss: Using HLM to Model Weight Loss over Time

Table 2

Comparison of model parameters (standard error) across the null, control, and final models on final weight outcome in pounds.

	Null model	Control model	Final model
Intercept
Intercept2	−4.69 (1.75)^a	−4.58 (1.88)^b	−2.40 (2.29)
Wk-1 loss		−0.37 (0.65)	0.03 (0.62)
BMI_0		−0.39 (0.25)	−0.45 (0.23)
HbA1c_0		1.00 (1.28)	0.49 (1.20)
Gender		0.33 (3.83)	1.49 (3.56)
Inconsistent			4.97 (5.15)
Consistent			−7.59 (3.60)^b
Linear time slope
Intercept2	−0.53 (0.07)^a	−0.46 (0.08)^a	−0.61 (0.12)^a
Wk-1 loss		0.01 (0.03)	0.003 (0.03)
BMI_0		−0.01 (0.01)	−0.01 (0.01)
HbA1c_0		−0.01 (0.05)	−0.002 (0.05)
Gender		0.30 (0.16)	0.28 (0.23)
Inconsistent			0.42 (0.23)
Consistent			0.22 (0.16)
Quadratic time slope
Intercept2	−0.02 (0.003)^a	−0.02 (0.004)^a	−0.03 (0.01)^a
Wk-1 loss		0.0004 (0.001)	−0.0007 (0.001)
BMI_0		0.0002 (0.0005)	0.0004 (0.0005)
HbA1c_0		−0.004 (0.002)	−0.003 (0.002)
Gender		0.007 (0.007)	0.004 (0.008)
Inconsistent			0.016 (0.01)
Consistent			0.024 (0.007)^a
Cubic time slope
Intercept2	−0.0002 (0.00005)^a	−0.0002 (0.00005)^a	−0.0003 (0.00007)^a
Wk-1 loss		−0.0000 (0.00002)	−0.00002 (0.00002)
BMI_0		0.00001 (0.00001)	0.00001 (0.00001)
HbA1c_0		−0.00006 (0.00003)	−0.00005 (0.00003)
Gender		0.00002 (0.0001)	−0.00002 (0.0001)
Inconsistent			0.0002 (0.0002)
Consistent			0.0003 (0.0001)^a

, , (standard error).