Table 5.
Literature review on the validation of food‐frequency questionnaire (FFQ) among pregnant women: characteristics of the published studies along with those of the new Portuguese study
| First author Year of publication* | Country of study | Reference method | n | Age | Characteristics of the FFQ | Validation studies Range of correlations (energy adjusted) | Reproducibility assessment | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Type of FFQ | Number of food items | Application method | Timing of application | Reference time period | |||||||
| Suitor et al. (1989) | USA | 3 × 24 h recall | 95 | 14–43 years | Sqt. | 111 | Self | Different gestational ages | 1 month | 0.12 † vitamin A; 0.60 † calcium | Yes |
| Greeley et al. (1992) | USA | 2 × 24 h recall | 50 | 20–35 years | Sqt. | 116 | Self | 21–35 weeks | 2 months | 0.12 [Link] , [Link] protein; 0.46 [Link] , [Link] calcium | No |
| Robinson et al. (1996) | UK | 4 days food diary Biomarker (serum vitamin C) | 569 | >16 years | Sqt. | 100 | Interviewer | 15 weeks | 3 months | 0.27 § starch; 0.55 § riboflavin | No |
| Brown et al. (1996) | USA | 4 days weighed food diary | 56 | Mean = 31.9 years (SD = 3.4) | Sqt. | 64 | Self | Mid‐pregnancy | N/A | 0.21 § saturated fat; 0.67 § iron | No |
| Wei et al. (1999) | USA | 3 × 24 h‐recall | 101 | N/A | Ql. | N/A | Interviewer | Different gestational ages | 1 month | 0.03 † vitamin B12; 0.46 † folate | No |
| Erkkola et al. (2001) | Finland | 2 × 5 days food diary | 113 | Mean = 29.6 years (SD = 5.1) | Sqt. | 181 | Self | 8th month | 1 month | 0.19 † vitamin E; 0.68 † total retinol | Yes |
| De Vriese et al. (2001) | Belgium | 2 × 7 days food diary | 26 | N/A | Sqt. | 180 | Interviewer | 6–22 weeks; 32–40 weeks | 1 month | 0.62 † monounsaturated FA; 0.68 † polyunsaturated FA | No |
| Parra et al. (2002) | Mexico | Biomarkers (erythrocyte FA) | 146 | 18–42 years | Sqt. | 104 | Interviewer | N/A | Recent intake | 0.30 † docosahexaenoic acid; 0.52 † alpha‐linolenic acid | No |
| Saldana et al. (2004) | USA | 3 × 24 h recall | 99 | Mean = 26 years (SD = 6.2) | Sqt. | N/A | Self | Between 24 and 29 weeks | 3 months | 0.26 † carbohydrates; 0.44 † protein | No |
| Fawzi et al. (2004) | USA | Biomarkers (alfa‐carotene, lycopene, lutein and zeaxanthin, gama‐tocoferol, long chain n‐3 FA, alpha‐ linolenic FA, trans‐FA) | 204 | Mean = 31.6 years (SD = N/A) | Sqt. | 179 | Self | <14 weeks | First trimester | N/A | No |
| Baer et al. (2005) | USA | 6 × 24 h recall | 521 | 16–40 years | Ql. | 100 | Self | 12 weeks, 16–17 weeks and 28 weeks | 1 month | 0.03 † Polyunsaturated FA; 0.51 † riboflavin | Yes |
| Venter et al. (2006) | UK | 4 × 7 days food diary | 57 | 20–44 years | Ql. | N/A | Interviewer | 12–13 weeks | N/A | N/A | Yes |
| Mouratidou et al. (2006) | UK | 2 × 24 h recall | 123 | Mean = 29.0 years (SD = 6.4) | Sqt. | 62 | Interviewer | 14–18 weeks | 1 month | −0.16 † niacin; 0.47 † englyst fibre | No |
| Mikkelsen et al. (2006) | Denmark | 7 days weighed food diary Biomarkers (urinary nitrogen excretion, retinol in plasma, erythrocyte folic acid, erythrocyte EPA) | 88 | Median = 28 years (IQR = 25–31.5) | Sqt. | 360 | Self | 30–31 weeks | 1 month | 0.30 § urinary flavenoids; 0.32 § plasma carotene | No |
| Brantsaeter et al. (2008) | Norway | 4 days weighed food diary Biomarkers (urinary nitrogen and iodine excretion, 25‐hydroxi‐vitamin D in plasma, serum folic acid) | 119 | Mean = 31.2 years (SD = 4.1) | Sqt. | 255 | Self | 16–18 weeks | 4 months | 0.25 § retinol total; iron total 0.54 § fibre | No |
| Present study (2008) | Portugal | 3 × 3 days food diary | 101 | Mean = 29.8 years (SD = 4.9) | Sqt. | 86 | Interviewer | Puerperium | Whole pregnancy | 0.16 † protein; 0.47 † riboflavin | Yes |
FA, fatty acids; EPA, eicosapentaenoic acid; IQR, interquartile range; Sqt., semi‐quantitative; Ql., qualitative; N/A, not given in publication; *Search terms used to identify relevant articles published in English, French, Spanish and Portuguese and indexed in PubMed: (‘pregnant’ or ‘pregnancy’) and (‘food’ or ‘diet’) and ‘food frequency’ and (‘validation’ or ‘validity’) until December, 2007. A total of 21 papers were identified through electronic search and four through additional manual searches. Ten were excluded after review of the abstract or full article because they were reviews, did not assess diet during pregnancy or were duplicate publications (only the most complete evaluation was included). Thus, a total of 15 articles were included in the review; †Pearson correlation coefficients; ‡Only unadjusted correlation coefficients available; §Spearman correlation coefficients.