Table 2.
Exposure quartile | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Serum n-6 PUFA | 1 (n 421) | 95 % CI | 2 (n 421) | 95 % CI | 3 (n 422) | 95 % CI | 4 (n 421) | 95 % CI | P for trend § | Mean difference between the highest and the lowest quartile | 95 % CI |
Total n-6 PUFA (%) | 13·99–30·13 | 30·14–33·29 | 33·30–36·17 | 36·18–47·25 | |||||||
Model 1† | 11·98 | 11·71, 12·26 | 12·35 | 12·07, 12·62 | 12·59 | 12·32, 12·86 | 12·88 | 12·61, 13·16 | <0·001 | 0·90 | 0·51, 1·29 |
Model 2‡ | 12·05 | 11·78, 12·32 | 12·36 | 12·10, 12·63 | 12·58 | 12·31, 12·84 | 12·82 | 12·55, 13·09 | <0·001 | 0·77 | 0·38, 1·16 |
LA (%) | 10·30–23·82 | 23·83–26·70 | 26·71–29·63 | 29·64–41·30 | |||||||
Model 1 | 12·04 | 11·77, 12·32 | 12·32 | 12·05, 12·59 | 12·58 | 12·30, 12·85 | 12·86 | 12·59, 13·13 | <0·001 | 0·82 | 0·43, 1·20 |
Model 2 | 12·08 | 11·80, 12·35 | 12·31 | 12·04, 12·57 | 12·60 | 12·34, 12·86 | 12·82 | 12·55, 13·09 | <0·001 | 0·74 | 0·34, 1·14 |
GLA (%) | 0·06–0·20 | 0·21–0·27 | 0·28–0·35 | 0·36–0·87 | |||||||
Model 1 | 12·64 | 12·36, 12·91 | 12·35 | 12·08, 12·63 | 12·50 | 12·22, 12·77 | 12·31 | 12·04, 12·59 | 0·19 | –0·32 | –0·71, 0·07 |
Model 2 | 12·67 | 12·41, 12·93 | 12·32 | 12·06, 12·58 | 12·45 | 12·19, 12·71 | 12·36 | 12·09, 12·63 | 0·21 | –0·31 | –0·69, 0·07 |
DGLA (%) | 0·57–1·15 | 1·16–1·33 | 1·34–1·50 | 1·51–3·02 | |||||||
Model 1 | 12·37 | 12·09, 12·64 | 12·15 | 11·88, 12·42 | 12·66 | 12·38, 12·93 | 12·63 | 12·35, 12·90 | 0·06 | 0·24 | –0·13, 0·65 |
Model 2 | 12·29 | 12·02, 12·56 | 12·23 | 11·97, 12·49 | 12·63 | 12·37, 12·90 | 12·65 | 12·38, 12·92 | 0·03 | 0·35 | –0·03, 0·75 |
AA (%) | 1·68–4·14 | 4·15–4·76 | 4·77–5·44 | 5·45–9·21 | |||||||
Model 1 | 12·16 | 11·88, 12·43 | 12·26 | 11·99, 12·53 | 12·69 | 12·42, 12·97 | 12·69 | 12·41, 12·96 | <0·001 | 0·53 | 0·14, 0·91 |
Model 2 | 12·27 | 11·99, 12·45 | 12·31 | 12·05, 12·58 | 12·66 | 12·40, 12·92 | 12·76 | 12·48, 12·94 | 0·01 | 0·50 | 0·29, 0·80 |
LA, linoleic acid; GLA, γ-linolenic acid; DGLA, dihomo-γ-linolenic acid and AA, arachidonic acid.
Values are means (95 % CI). Data were analysed with ANCOVA.
Model 1: adjusted for age and examination years.
Model 2: adjusted for model 1 plus BMI, smoking status, leisure-time physical activity, alcohol intake, use of antihypertensive medication, education, income and serum long-chain n-3 PUFA concentrations.
P for trend refers to the linear trend across the fatty acid quartiles.