Table 2. Relative RRs for PSA Testing and Prostate Biopsy Frequency and Odds of Undergoing a Biopsy Within 12 Months of an Elevated PSA Measure According to Exposure to Antidiabetic Medications.

Exposure	Frequency, RR (95% CI)a		Biopsy Following Elevated PSA, OR (95% CI)b,c
	PSA Testing	Prostate Biopsy	≥3.0 ng/mL (n = 53 357)	≥4.0 ng/mL (n = 38 719)
Metformin	1.07 (1.06-1.09)	0.76 (0.70-0.83)	0.87 (0.80-0.96)	0.87 (0.79-0.96)
Sulfonylurea	1.06 (1.03-1.08)	0.93 (0.83-1.04)	0.88 (0.78-1.00)	0.88 (0.76-1.01)
Insulin	0.79 (0.77-0.81)	0.67 (0.60-0.75)	0.83 (0.74-0.93)	0.81 (0.71-0.92)
Any antidiabetic medicationd	0.93 (0.92-0.94)	0.59 (0.55-0.62)	0.87 (0.80-0.96)	0.77 (0.71-0.82)

Abbreviations: OR: odds ratio; PSA, prostate-specific antigen; RR, rate ratio.

SI conversion factor: To convert PSA to μg/L, multiply by 1.0.

^aRate ratios derived from zero-inflated Poisson regression models adjusted for age, calendar year, education level, marital status, family history of prostate cancer, and specific medications (simultaneously). Follow-up time for exposure was split at first prescription for the specific class of antidiabetic drug.

^bOdds ratios derived from binary logistic regression models for biopsy procedure within 12 months of elevated PSA level, adjusted for age group (10-year bands), year of PSA test, education, marital status, family history of prostate cancer, and Charlson Comorbidity Index score (0, 1, 2, or ≥3). Antidiabetic medications were modeled simultaneously, with a separate model for any diabetic medication.

^cIndex PSA was the highest PSA value during follow-up if more than 1 test result was above the cutoff level.

^dSeparate model for ever vs with never user of any diabetic medications during follow-up period.