Table 2.

Adjusted HRs for PC by diabetes and random plasma glucose at baseline in CKB

	No. events	No. people	Rate, per 100,000	Person‐time, PYAR	HR (95% CI) a
Diabetes status
No diabetes	499	480,307	103.89	3,894,806	Reference
Total diabetes	86	30,007	286.60	230,790	1.87 (1.48, 2.37)
Diabetes status
No diabetes	499	480,307	103.89	3,894,806	1.00 (0.90, 1.11)
Screen‐detected	30	14,007	214.18	108,552	1.54 (1.08, 2.20)
Previously diagnosed	56	16,000	350.00	122,238	2.08 (1.56, 2.76)
Duration of diabetes b
No diabetes	499	480,307	103.89	3,894,806	1.00 (0.90, 1.11)
Screen‐detected	30	14,007	214.18	108,552	1.54 (1.07, 2.20)
Previously diagnosed <5 years	21	8,341	311.71	64,714	2.11 (1.43, 3.09)
≥5 years	30	7,659	391.70	57,374	2.16 (1.50, 3.11)
p for trend					0.01
Random plasma glucose (mmol/L)c
≤5.5	207	242,737	85.28	1,893,962	1.00 (0.86, 1.16)
>5.5‐7.0	208	172,701	120.44	1,473,516	1.11 (0.97, 1.24)
≥7.0	97	70,751	137.10	555,171	1.22 (1.00, 1.50)
per 1 mmol/Ld	512	486,189	105.31	3,922,649	1.12 (1.04, 1.21)

^aEstimates were stratified by age at risk, sex and region, and adjusted for age at baseline, education, smoking and alcohol.

^bDiabetes duration data were missing or implausible for 20 participants.

^cThe analysis for random plasma glucose was restricted to participants without previously diagnosed diabetes and further adjusted for fasting time.

^dHR per 1 mmol/L in blood glucose was corrected for regression dilution. The regression dilution ratio is 0.45. 1 mmol/L= 18 mg/dL.

Abbreviations: PYAR = person years at risk.