Skip to Main Content
Table 2—

Final multiple linear regression analysis predicting change in HbA1c for those (n = 677) with baseline HbA1c ≥7% and at least one HbA1c value available in the 12-month follow-up period: the variable coefficients, their standard error (SE), and the associated t test value and its statistical significance are shown

βSE (β)t
Sex (female)* −0.147 0.087 −1.69 
Age (years) −0.010 0.004 −2.61 
College graduate* −0.166 0.095 −1.75 
Duration of diabetes (years) 0.002 0.004 0.50 
Baseline HbA1c value −0.533 0.035 −15.08 
Compliant with medications* −0.492 0.112 −4.41 
RTC 0.764 0.329 2.32§ 
SF-12 physical score 0.011 0.005 2.05§ 
RTC × SF-12 physical score −0.018 0.007 −2.45§ 
βSE (β)t
Sex (female)* −0.147 0.087 −1.69 
Age (years) −0.010 0.004 −2.61 
College graduate* −0.166 0.095 −1.75 
Duration of diabetes (years) 0.002 0.004 0.50 
Baseline HbA1c value −0.533 0.035 −15.08 
Compliant with medications* −0.492 0.112 −4.41 
RTC 0.764 0.329 2.32§ 
SF-12 physical score 0.011 0.005 2.05§ 
RTC × SF-12 physical score −0.018 0.007 −2.45§ 

Details on model construction are provided in the text. Negative regression coefficients indicate a reduction in HbA1c.

*

Variable coding: 0 = no, 1 = yes.

P < 0.01.

Variable coding: 0 = precontemplation, 1 = contemplation or preparation, n = 617, F(9, 607) = 27.8, P < 0.0001, R2 = 0.29.

§

P < 0.05.

Close Modal

or Create an Account

Close Modal
Close Modal