Establishment of a seafood index to assess the seafood consumption in pregnant women

Figures & data

Fig. 1.  Flow chart of the seafood index.

Table 1. The seafood index – conversion of frequencies to numerical values

Reported frequency	Numerical interval per week^†	Seafood index (summary question)^‡	Seafood index (detailed item question)^§
Never	0	0	0
<1 time/month	>0–0.25	0.15	0.1^∣∣
1–3 times/month	0.25–0.75	0.50	0.25
1–2 times/week	1–2	1.5	1
≥3 times/week	≥3	3	3

^†Numerical interval based on the average weekly frequency of seafood intake for dinner and seafood intake as spread.

^‡Seafood index assigned the average weekly frequency of seafood intake for dinner and seafood intake as spread.

^§Seafood index based on the lowest possible weekly intake of items eaten as dinner and items eaten as spread.

^∣∣Numerical value set to 0.1 to enable distinction between the two lowest frequencies.

Table 2. The supplement index – conversion of frequencies to numerical values

Reported frequency	Numerical interval per week^†	Supplement index^‡
Never	0	0
1–3 times/month	0.25–0.75	0.5
1–3 times/week	1–3	2
4–6 times/week	4–6	5
Daily	7	7^§

^†Numerical interval based on the average weekly intake of supplements.

^‡Supplement index based on the mean weekly intake of supplement.

^§When calculating the total seafood index, the highest frequency is set to the same as for the second highest frequency to correspond to the frequency range in the two main questions (seafood as dinner and seafood as spread).

Table 3. Frequency estimates from the seafood index

Interval^†	Frequency estimate
0–0.14	Never
0.15–0.45	<1 time/month
0.5–0.95	1–3 times/month
1.0–2.9	1–2 times/week
3^‡→	≥3 times/week

^†Sum of type of seafood or seafood product.

^‡Maximum sum is set to 3.

Table 4. Socio-economic and behavioural characteristics of study participants (n=55)

Characteristics	Count	Percent
Education
Lower secondary school	0	0
Higher secondary school	16	29.1
< 4 years of university education^†	24	43.6
≥4 years of university education^†	15	27.3
Employment
Full-time (80–100%)	48	87.3
Part-time (50–79%)	2	3.6
Part-time (<50%)	1	1.8
Homemaker	1	1.8
Other	3	5.5
Marital status^‡
Married	23	43.4
Cohabiting	28	52.8
Single	2	3.8
Own income in NOK^§
< 1,50,000	2	3.6
1,50,000–1,99,999	1	1.8
2,00,000–2,99,999	12	21.8
3,00,000–3,99,999	27	49.1
4,00,000–4,99,999	10	18.2
> 5,00,000 NOK	3	5.5
Self-reported smoking during pregnancy^‡
Non-smoker	52	98.1
Current smoker	1	1.9
Self-reported use of snuff during pregnancy^∣∣
Non-user	54	100
Current user	0	0

^†University or University College.

^‡ n=53.

^§1,00,000 NOK 14,000 EUR.

^∣∣ n=54.

Table 5. Summary statistics for biomarkers in all participants (n=55) and non-supplement users (n=19)

	All participants		Non-supplement users only
Biomarker	Mean±SD	Min, max	Mean±SD	Min, max
25OH vitamin D* (nmol/L)^†	71±30	23, 148	54±22^¶	23, 98
EPA (mg/g RBC)^‡	0.025±0.02	0.0, 0.07	0.017±0.01^¶	0.0, 0.06
DPA (mg/g RBC)^§	0.052±0.01	0.02, 0.08	0.050±0.01	0.02, 0.08
DHA (mg/g RBC)^‡	0.17±0.04	0.0, 0.25	0.16±0.04	0.09, 0.25
Omega-3 index^§	6.9±1.7	3.5, 11.0	6.4±1.6	3.7, 10
RBC HUFA score^?	37±6	23, 52	34±5^¶	28, 49

*Vitamin D refers to serum 25OHD3 concentration.

^†All participants: n=51, non-supplement users only: n=18.

^‡All participants: n=54.

^§The content of EPA+DHA in red blood cells membranes expressed as a percent of total fatty acid 13.

^∣∣Total HUFA is the sum of the omega-3 and the omega-6 HUFAs, and the red blood cells omega-3 HUFA score equals 100% – omega-6 HUFA 14.

^¶Differences between supplement users and non-supplement users are statistically significant at the level 0.05. Results from supplement users are not shown.

Fig. 2.  The figure illustrates the association between the reported seafood for dinner frequency (summary question) and the estimated seafood for dinner frequency (based on seafood index from all dinner items). The sum of points equals 100. The line represents matching frequencies. Spearman's rho between the reported seafood dinner frequency and the estimated seafood dinner frequency was 0.59, with a significance level of P>0.001.

Fig. 3.  The figure illustrates the association between the reported seafood as spread frequency (summary question) and the estimated seafood as spread frequency (sum of all seafood as spread items). The sum of points equals 100. The line represents matching frequencies. Spearman's rho between the reported seafood as spread frequency and the estimated seafood as spread frequency was 0.82, with a significance level of P>0.001.

∧The data points could be allocated with the data points below as the estimated seafood as spread frequency was maximum ≥3 times/week.

Table 6. Spearman's rho coefficients of the association between the different seafood indexes and biomarkers in all participants, n=54

	Seafood dinner indexes^†		Seafood spread indexes^‡
Biomarker	Oily fish dinner index	Lean fish dinner index	Spread index	Oily fish spread index	Lean fish spread index	Supplement index^§	Total Seafood index^∣
25OHD3 (nmol/L)^∣∣	0.37*	0.33	−0.09	0.27	−0.05	0.62**	0.44*
EPA (mg/g RBC)	0.09	0.05	0.27	0.28*	0.12	0.37**	0.53**
DPA (mg/g RBC)	0.14	0.0	0.22	0.21	0.14	0.19	0.34*
DHA (mg/g RBC)	0.07	−0.08	0.33*	0.39**	0.03	0.22	0.37**
Omega-3 index^#	0.15	0.01	0.28*	0.32*	0.17	0.21	0.36**
RBC omega-3 HUFA score^††	0.13	0.03	0.35**	0.41**	0.11	0.35**	0.49**

^†The seafood dinner indexes were defined based on the respondents reported average frequency of seafood eaten as dinner. Seafood as dinner was grouped into five categories oily, lean (results in table), processed, shellfish and freshwater fish (results not in table).

^‡The seafood spread indexes were defined based on the respondents reported average frequency of seafood eaten as spread. Seafood as spread was grouped into two categories (oily and lean).

^§The supplement index was defined based on the respondents reported average frequency of omega-3 supplement intake.

^¶The total seafood index was computed as a sum of the participants’ seafood dinner index, seafood spread index and supplement index.

^∣∣Participants who answered in the winter only n=30.

^#The content of EPA+DHA in red blood cells membranes expressed as a percent of total fatty acid 13.

^††Total HUFA is the sum of the omega-3 and the omega-6 HUFAs, and the red blood cells omega-3 HUFA score equals 100% – omega-6 HUFA 14.

^*/**Indicates statistical significance at the 0.05 or 0.001 level, respectively.

Fig. 4.  The association between the total seafood index (fish, other seafood, and omega-3 supplement intake) and the omega-3 index, in all participants (n=54).

Fig. 5.  The association between the total seafood index (fish, other seafood, and omega-3 supplement intake) and the omega-3 HUFA score, in all participants (n=54).

Fig. 6.  The association between the total seafood index (fish, other seafood, and omega-3 supplement intake) and serum 25OH vitamin D, in participants who answered in the winter months (October–April) (n=30).

Table 7. Concentrations of biomarkers in all participants according to total seafood index quartiles

		Total seafood index, quartile (range)
Biomarker	n	Q1 (0≤4)	Q2 (>4–6.7)	Q3 (>6.7–8)	Q4 (>8–9.5)
25OH vitamin D (nmol/L)^†	30	41±14^b	57±28^a,b	94±23^a	77±33^a
EPA (mg/g RBC)^‡	54	0.015±0.01^b	0.017±0.01^b	0.029±0.02^a	0.032±0.02^a
DPA (mg/g RBC)^‡	54	0.049±0.01^a,b	0.046±0.01^b	0.056±0.01^a	0.056±0.01^a
DHA (mg/g RBC)^‡	54	0.16±0.03^b,c	0.16±0.04^b,c	0.18±0.04^a	0.182±0.03^a,c
Omega-3 index^§	54	6.1±1.5^b	6.2±1.5^b	7.4±1.6^a,b	7.4±1.7^a
RBC HUFA score^?	54	32±4^b,c	35±3^b	38±7^a,b	39±7^a

^a,b,cDifferent letters indicate statistical significant difference, P<0.005 (ANOVA Fisher LSD post hoc test).

^†Participants who answered in the winter: n=30.

^‡ n=54.

^§The content of EPA+DHA in red blood cells membranes expressed as a percent of total fatty acid 13.

^∣∣Total HUFA is the sum of the omega-3 and the omega-6 HUFAs, and the red blood cells omega-3 HUFA score equals 100% – omega-6 HUFA 14.

Harris WS, von Schacky C. The Omega-3 Index: a new risk factor for death from coronary heart disease?. Prev Med. 2004; 39: 212–20. 10.3402/fnr.v57i0.19272.

 PubMed Web of Science ®Google Scholar

Lands B. Measuring blood fatty acids as a surrogate indicator for coronary heart disease risk in population studies. World Rev Nutr Diet. 2009; 100: 22–34.

 Google Scholar