Table 3.

Adjusted odds of ED and/or LUTS at year 17 by quantitative CAN measures at DCCT closeout and EDIC year 16/17

Characteristic	ED only (N=193) vs. No LUTS or ED (N=284)	LUTS only (N=61) vs. No LUTS or ED (N=284)	ED & LUTS (N=97) vs. No LUTS or ED (N=284)
CAN Measures*	Odds ratio (95% CI)
At DCCT Closeout
R-R Variation	0·96 (0·91–1·02)	0·98 (0·90–1·07)	1·11 (1·01–1·22)
Valsalva Ratio	1·47 (1·10–1·96)	1·31 (0·90–1·92)	1·50 (1·01–2·21)
At EDIC Year 16/17
R-R Variation	1·01 (0·94–1·08)	0·98 (0·89–1·08)	1·12 (1·01–1·25)
Valsalva Ratio	1·43 (0·99–2·08)	0·90 (0·56–1·45)	2·11 (1·17–3·78)

Data are odds ratios (95% confidence intervals) from separate multivariable logistic regression models with ED/LUTS status as the dependent variable and CAN measures as the independent variables. The odds ratios were evaluated according to a 5-unit decrease in R-R Variation or a ½-unit decrease in Valsalva ratio. Adjustments were made for DCCT cohort assignment, DCCT/EDIC time-weighted HbA1c, DCCT/EDIC time-weighted systolic blood pressure, and the following EDIC year 17 characteristics: age, BMI, smoking status, drinking status, and any antihypertensive medication use.

^*Sample sizes vary based on availability of CAN data.