Table 2.

Impact of pre-existing cardiovascular diseases on adequate glycemic control

Logistic regressions on percentage of patients with HbA1c < 6.5%
Variable	Odds Ratio	95% confidence intervals	P-value
With CVD vs. without CVD	0.62	0.46, 0.82	0.001
Time indicators
<1 year from index	6.21	3.61, 10.70	<0.001
1-2 years from index	6.39	3.72, 10.98	<0.001
2-3 years from index	6.69	3.81, 11.75	<0.001
3-4 years from index	7.66	4.24, 13.87	<0.001
>= 4 years from index	5.62	3.02, 10.46	<0.001
Variable	Odds Ratio	95% confidence intervals	P-value
With CVD vs. without CVD	0.62	0.46, 0.82	0.001
Study period vs. baseline	6.47	3.71, 11.29	<0.001
Years after the index	1.00	0.92, 1.10	0.96
Linear regressions on HbA1c (%)
Variable	Coefficient	95% confidence intervals	P-value
With CVD vs. without CVD	0.02	-0.10, 0.14	0.75
Time indicators
<1 year from index	-0.80	-0.97, -0.62	<0.001
1-2 years from index	-0.78	-0.96, -0.60	<0.001
2-3 years from index	-0.76	-0.96, -0.57	<0.001
3-4 years from index	-0.88	-1.07, -0.68	<0.001
>= 4 years from index	-0.84	-1.07, -0.60	<0.001
Variable	Coefficient	95% confidence intervals	P-value
With CVD vs. without CVD	0.02	-0.10, 0.14	0.75
Study period vs. baseline	-0.78	-0.97, -0.59	<0.001
Years after the index	-0.01	-0.05, 0.03	0.65

Note: CVD: cardiovascular diseases. Other controlled covariates in the regression analyses: concurrent treatment of the HbA_1c measurement (metformin+SU combination therapy, metformin+glitazones combination therapy, SU monotherapy, glitazones monotherapy, metformin monotherapy, any therapies with insulin), index SU vs. glitazones, age, gender, Caucasian, duration of diabetes, never used alcohol, physical activity, country, BMI, physician specialty.