Table 3

Disease-specific RRs for PFOA from Cox regression for outcomes with 20 or more cases

DiseaseNumber of cases*RR quartile 1RR quartile 2†RR quartile 3RR quartile 4p Value trend log cumulative exposurep Value trend via categories‡
No lag ulcerative colitis281.001.57 (0.52 to 4.76)0.57 (0.10 to 3.11)2.74 (0.78 to 9.65)0.050.26
10-year lag ulcerative colitis281.003.00 (0.82 to 11.0)3.26 (0.70 to 15.10)6.57 (1.47 to 29.40)0.050.05
No lag rheumatoid arthritis231.002.11 (0.49 to 8.98)4.08 (0.96 to 17.2)4.45 (0.99 to 19.9)0.540.04
10-year lag rheumatoid arthritis231.001.74 (0.45 to 6.77)2.12 (0.4 to 11.1)2.62 (0.47 to 14.7)0.750.06
No lag bladder cancer291.000.32 (0.08 to 1.330.95 (0.28 to 3.14)0.23 (0.05 to 0.93)0.040.19
10-year lag bladder cancer291.000.55(0.12 to 2.61)0.47 (0.10 to 2.21)0.31 (0.06 to 1.54)0.060.03
No lag colorectal cancer411.000.58 (0.18 to 1.87)1.43 (0.49 to 4.19)1.20 (0.39 to 3.62)0.910.68
10-year lag colorectal cancer411.000.31 (0.90 to 1.11)0.99 (0.33 to 2.94)1.06 (0.34 to 3.31)0.860.89
No lag prostate cancer1291.001.81 (0.69 to 4.78)2.45 (0.96 to 6.25)1.88 (0.72 to 4.88)0.830.11
10-year lag prostate cancer1291.001.92 (0.56 to 6.58)1.89 (0.57 to 6.34)2.15 (0.64 to 7.26)0.910.10
No lag melanoma411.001.16 (0.38 to 3.54)1.45 (0.47 to 4.45)0.88 (0.26 to 2.95)0.160.72
10-year lag melanoma411.000.85 (0.27 to 2.71)1.10 (0.34 to 3.58)0.75 (0.21 to 2.67)0.550.33
No lag non-hepatitis liver351.000.55 (0.17 to 1.77)1.27 (0.44 to 3.65)0.88 (0.27 to 2.85)0.860.83
10-year lag non-hepatitis liver351.001.46 (0.42 to 5.04)2.13 (0.59 to 7.71)2.02 (0.50 to 8.10)0.400.08
No lag male thyroid disease821.001.64 (0.82 to 3.29)1.13 (0.50 to 2.54)2.16 (0.98 to 4.77)0.980.10
10-year lag male thyroid disease821.001.23 (0.57 to 2.66)1.70 (0.74 to 3.91)1.71 (0.68 to 4.25)0.550.06
No lag female thyroid disease771.001.00 (0.54 to 1.87)1.02 (0.48 to 2.17)0.33 (0.08 to 1.26)0.970.35
10-year lag female thyroid disease771.000.79 (0.42 to 1.50)0.87 (0.37 to 2.02)0.23 (0.05 to 1.01)0.270.13
No lag coronary heart disease3801.001.03 (0.71 to 1.49)1.23 (0.84 to 1.78)1.03 (0.70 to 1.52)0.780.39
10-year lag coronary heart disease3801.001.20 (0.82 to 1.75)1.06(0.71 to 1.58)0.93 (0.61 to 1.41)0.750.99
No lag medicated hypertension14301.000.95 (0.81 to 1.11)0.97 (0.82 to 115)1.04 (0.87 to 1.25)0.950.99
10-year lag medicated hypertension14301.000.95 (0.81 to 0.98)0.91 (0.75 to 1.09)0.95 (0.77 to 1.16)0.540.12
No lag medicated
High cholesterol
12981.001.11 (0.94 to 1.30)1.06 (0.89 to 1.27)1.05 (0.87 to 1.27)0.560.29
10-year lag medicated
High cholesterol
12981.000.93 (0.79 to 1.10)1.01 (0.84 to 1.22)0.96 (0.78 to 1.18)0.620.50
No lag osteoarthritis1961.000.88 (0.58 to 1.34)0.97 (0.71 to 1.54)0.97 (0.59 to 1.59)0.920.48
10-year lag osteoarthritis1961.000.74 (0.49 to 1.10)0.56 (0.34 to 0.93)0.67 (0.39 to 1.14)0.130.15
No lag stroke1081.002.63 (1.06 to 6.56)2.13 (0.83 to 6.44)2.07 (0.81 to 5.29)0.350.17
10-year lag stroke1081.001.48 (0.56 to 3.69)1.53 (0.60 to 3.89)1.33 (0.51 to 3.43)0.640.22
No lag COPD1461.001.20 (0.64 to 2.27)1.25 (0.65 to 2.37)1.13 (0.59 to 2.17)0.860.20
10-year lag COPD1461.000.75 (0.38 to 1.48)1.16 (0.60 to 2.26)0.77 (0.38 to 1.57)0.970.51
No lag medicated asthma821.000.76 (0.28 to 2.04)0.50 (0.17 to 1.49)0.53 (0.16 to 1.69)0.270.05
10-year lag medicated asthma821.000.48 (0.10 to 2.29)0.57 (0.11 to 2.93)0.52 (0.09 to 2.83)0.530.17
No lag chronic kidney disease431.000.50 (0.13 to 1.81)0.69 (0.21 to 2.12)0.67 (0.15 to 3.05)0.920.22
10-year lag chronic
Kidney disease
431.001.32 (0.32 to 5.43)0.50 (0.11 to 2.34)0.67 (0.15 to 3.05)0.990.30
No lag type 2 diabetes4081.000.94 (0.67 to 1.32)1.08 (0.76 to 1.52)1.10 (0.77 to 1.57)0.660.27
10-year lag type 2 diabetes4081.001.06 (0.75 to 1.49)1.10 (0.76 to 1.61)1.12 (0.76 to 1.66)0.940.05
  • *Some cases were lost to analysis due to missing covariates; hence these numbers are slightly lower than validated cases in table 2.

  • †Quartile cut-offs for the unlagged analyses were 3.03, 6.16, and 11.42 μg/mL-year and in the 10 year lagged analysis were 0.8, 3.44, 7.04 μg/mL-year (cumulative exposure, PFOA in serum). These were common across all diseases analysed, and were chosen by combining all diagnosed chronic disease cases and choosing cut points such that diagnosed cases were evenly divided across quartiles.

  • ‡Trend test conducted using a weighted linear regression of log RRs on the midpoints of quartile categories, with the line constrained to pass through the origin, and weights equal to the inverse of the variance of the log RRs.

  • COPD, chronic obstructive pulmonary disease; PFOA, perfluorooctanoic acid; RR, rate ratio.