Table 3.
Relative risk of prostate cancer-specific mortality among 926 men initially diagnosed with non-metastatic prostate cancer by post-diagnostic fat intake.
| Quartile of intake | Continuous model | ||||||
|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 |
P- trend1 |
HR (95% CI) | p-value | |
| Saturated fat | |||||||
| Median percent energy | 6.6 | 8.6 | 10.4 | 12.7 | |||
| Events/person-years | 12/2109 | 7/2093 | 16/1961 | 21/1930 | Unit of change = 5% energy | ||
| Model 1 HR (95% CI) 2 | 1.00 | 0.54 (0.21, 1.37) | 1.12 (0.52, 2.39) | 1.71 (0.84, 3.49) | 0.04 | 1.56 (0.98, 2.47) | 0.06 |
| Model 2 HR (95% CI) 3 | 1.00 | 0.51 (0.18, 1.40) | 1.22 (0.44, 3.42) | 2.13 (0.57, 7.90) | 0.11 | .2.78 (1.01, 7.64) | 0.05 |
| Monounsaturated fat | |||||||
| Median percent energy | 6.9 | 8.8 | 10.8 | 13.1 | |||
| Events/person-years | 13/2085 | 12/2019 | 15/2020 | 16/1969 | Unit of change = 5% energy | ||
| Model 1 HR (95% CI) 2 | 1.00 | 0.86 (0.39, 1.88) | 1.02 (0.48, 2.15) | 1.12 (0.53, 2.34) | 0.65 | 1.26 (0.77, 2.04) | 0.36 |
| Model 2 HR (95% CI) 3 | 1.00 | 0.54 (0.23, 1.28) | 0.47 (0.18, 1.29) | 0.39 (0.11, 1.40) | 0.19 | 0.52 (0.16, 1.63) | 0.26 |
|
Polyunsaturated fat n-6 Fatty Acids | |||||||
| Median percent energy | 2.9 | 3.6 | 4.3 | 5.5 | |||
| Events | 14/1960 | 20/2005 | 11/2069 | 11/2059 | Unit of change = 5% energy | ||
| Model 1 HR (95% CI) 2 | 1.00 | 1.47 (0.74, 2.92) | 0.72 (0.33, 1.59) | 0.77 (0.35, 1.71) | 0.23 | 0.62 (0.19, 1.99) | 0.42 |
| Model 2 HR (95% CI) 3 | 1.00 | 1.85 (0.89, 3.87) | 0.93 (0.39, 2.19) | 0.98 (0.35, 2.76) | 0.58 | 1.01 (0.20, 5.09) | 0.99 |
| n-3 Fatty Acids | |||||||
| Median percent energy | 0.3 | 0.4 | 0.5 | 0.6 | |||
| Events | 15/1948 | 17/1967 | 12/2094 | 12/2084 | Unit of change = 1% energy | ||
| Model 1 HR (95% CI) 2 | 1.00 | 1.07 (0.53, 2.14) | 0.82 (0.38, 1.76) | 0.81 (0.38, 1.74) | 0.47 | 0.35 (0.05, 2.61) | 0.31 |
| Model 2 HR (95% CI) 3 | 1.00 | 1.17 (0.56, 2.45) | 1.02 (0.44, 2.39) | 1.12 (0.42, 3.00) | 0.90 | 0.63 (0.06, 6.82) | 0.70 |
| Trans fat | |||||||
| Median percent energy | 0.5 | 0.8 | 1.0 | 1.3 | |||
| Events | 10/2097 | 12/2040 | 18/2034 | 16/1922 | Unit of change = 1% energy | ||
| Model 1 HR (95% CI) 2 | 1.00 | 1.15 (0.49, 2.66) | 1.65 (0.76, 3.58) | 1.50 (0.68, 3.34) | 0.23 | 1.72 (0.81, 3.68) | 0.16 |
| Model 2 HR (95% CI) 3 | 1.00 | 0.88 (0.35, 2.23) | 1.04 (0.39, 2.77) | 0.93 (0.27, 3.25) | 0.97 | 0.94 (0.24, 3.63) | 0.92 |
| Animal fat | |||||||
| Median percent energy | 9.5 | 13.9 | 17.6 | 23.1 | |||
| Events | 12/2098 | 10/2081 | 14/1981 | 20/1933 | Unit of change = 10% energy | ||
| Model 1 HR (95% CI) 2 | 1.00 | 0.77 (0.33, 1.78) | 1.08 (0.50, 2.34) | 1.46 (0.71, 3.02) | 0.17 | 1.48 (0.98, 2.23) | 0.06 |
| Model 2 HR (95% CI) 3 | 1.00 | 0.54 (0.21, 1.38) | 0.65 (0.24, 1.78) | 0.74 (0.21, 2.65) | 0.89 | 1.47 (0.62, 3.53) | 0.38 |
| Vegetable fat | |||||||
| Median percent energy | 6.7 | 9.2 | 11.5 | 15.3 | |||
| Events | 19/1937 | 11/2048 | 12/2060 | 14/2049 | Unit of change = 10% energy | ||
| Model 1 HR (95% CI) 2 | 1.00 | 0.58 (0.28, 1.22) | 0.62 (0.30, 1.29) | 0.73 (0.36, 1.47) | 0.46 | 0.69 (0.34, 1.42) | 0.32 |
| Model 2 HR (95% CI) 3 | 1.00 | 0.64 (0.29, 1.38) | 0.80 (0.36, 1.77) | 0.93 (0.41, 2.14) | 0.97 | 0.81 (0.35, 1.91) | 0.63 |
Abbreviations: HR, hazard ratio; CI, confidence interval.
P-trend calculated by modeling the median of each category as a continuous term.
Cox proportional hazards regression adjusted for age at diagnosis (continuous), energy (continuous), and time since diagnosis (continuous).
Cox proportional hazards regression adjusted for variables in Model 1 plus treatment (radical prostatectomy, radiation therapy, other, missing), modified D’amico risk category (high, intermediate, low), body mass index (kg/m2, continuous), smoking (ever, never), and intake of alcohol (continuous %energy) and protein (continuous %energy). Additionally, the fats were adjusted for one another; and animal and vegetable fat were adjusted for one another and trans fat. These multivariate nutrient density models are modeling the association between replacing daily calories from carbohydrate with the same amount of calories from the fat of interest on risk of all-cause mortality.