Table 3. Multinomial regression analysis using CKD as predictor variable with outcome of “new” or “worsening” monofilament insensitivity.

Monofilament sensitivity	Unadjusted analysis OR and 95% CI	Adjusted analysis OR and 95% CI
Between years 2000–01 and 2007–08	CKD vs non-CKD
Participants who “maintained normal function” as reference category for outcome of “new” monofilament insensitivity #
Maintained normal function (N = 388)	Reference category
New light touch insensitivity (N = 183)	1.41 (0.85–2.35)	1.51 (0.89–2.54)
New standard touch insensitivity (N = 64)	1.62 (0.79–3.39)	1.47 (0.70–3.08)
Participants who “maintained function” as reference category for outcome of “worsening sensitivity” and “improving sensitivity” $
Maintained function (light touch insensitivity) (N = 136)	Reference category
Worsened to standard touch insensitivity (N = 89)	1.79 (0.92–3.51)	2.09 (1.03–4.29)
Improved to normal sensitivity (N = 175)	1.42 (0.78–2.56)	1.64 (0.86–3.13)

# Forward stepwise model. Only participants with normal sensitivity at initial visit included

$ Forward stepwise model. Final model adjusted for age, gender, race, smoking status at year 3 and peripheral arterial disease. Reverse confounding variable for outcome of “worsened to standard touch insensitivity”: smoking and peripheral arterial disease