Table 4.

Cox regression analyses of Sjögren's syndrome associated with different treatment duration and cumulative doses of metformin in type 2 diabetic patients.

Variable	Event	Person-year	IR	Crude		Adjusted*
	N = 36		100,000 person-years	HR (95% CI)	P-value	HR (95% CI)	P-value
Treatment duration (day)
None	24	58,779	40.83	1 (Reference)		1 (Reference)
1–29	3	7,603	39.46	0.99 (0.30, 3.28)	0.9809	0.85 (0.25, 2.85)	0.7937
30–89	4	9,141	43.76	1.09 (0.38, 3.15)	0.8683	1.00 (0.35, 2.91)	0.9963
≥ 90	5	54,618	9.15	0.24 (0.09, 0.62)	0.0033	0.27 (0.10, 0.71)	0.0077
Cumulative doses (mg)
None	24	58,779	40.83	1 (Reference)		1 (Reference)
1–14,999	3	6,332	47.38	1.17 (0.35, 3.89)	0.7980	0.99 (0.30, 3.32)	0.9914
15,000–44,999	3	7,089	42.32	1.06 (0.32, 3.51)	0.9281	0.96 (0.29, 3.21)	0.9490
≥45,000	6	57,941	10.36	0.27 (0.11, 0.65)	0.0037	0.30 (0.12, 0.74)	0.0087

CI, Confidence Interval; HR, Hazard Ratio; IR, Incidence Rate.

^*Adjusted for demographics, baseline comorbidities, and medications listed above.