Table 2.
Associations of fish oil use with risk of dementia.
| Non-users | Fish oil-users | ||
|---|---|---|---|
|
| |||
| All-cause dementia | P-value | ||
| Cases, N (%) | 3,654 (1.2) | 2,141 (0.6) | |
| Model 1 a | 1(reference) | 0.90 (0.85–0.96) | 0.001 |
| Model 1+ APOE ε4 dosage b | 1(reference) | 0.91 (0.86–0.96) | 0.001 |
| Model 1+ PRS (without APOE) b | 1(reference) | 0.90 (0.85–0.96) | 0.001 |
|
| |||
| Alzheimer's disease | Non-users | Fish oil-users | P-value |
|
| |||
| Cases, N (%) | 1,413 (0.5) | 969 (0.7) | |
| Model 1 a | 1(reference) | 0.99(0.91–1.09) | 0.901 |
| Model 1+ APOE ε4 dosage b | 1(reference) | 1.00 (0.92–1.10) | 0.983 |
| Model 1+ PRS (without APOE)b | 1(reference) | 1.00 (0.91–1.09) | 0.948 |
|
| |||
| Vascular dementia | Non-users | Fish oil-users | P-value |
|
| |||
| Cases, N (%) | 834 (0.3) | 432 (0.3) | |
| Model 1 a | 1(reference) | 0.85 (0.75–0.97) | 0.013 |
| Model 1+ APOE ε4 dosage b | 1(reference) | 0.85 (0.75–0.97) | 0.016 |
a
Model 1: Cox proportional hazards regression adjusted for age, sex, assessment centers, Townsend index, education, BMI, physical activity, healthy diet, smoking, moderate drinking, frequency of depression, social isolation, hearing problems, other supplements use and history of diabetes, history of high cholesterol, history of hypertension and history of cardiovascular disease.
b
Further adjustment for the first 10 genetic principal components and relatedness.