Validation of a photographic seafood portion guide to assess fish and shrimp intakes

Abstract

Objective: To validate a novel photographic portion guide as a tool to estimate consumption of fish and shrimp. Application of such a validated tool can facilitate accurate individual and community seafood intake assessments and provide meaningful data relative to health benefits and hazard assessment, particularly in response to environmental contamination and disasters.

Design: A photographic fish and shrimp portion guide presenting a stepped range of cooked portion sizes was used by participants to estimate their typical portion sizes. Participants selected their typical portion size from the photographic guide and also from a selection of freshly cooked reference meals. Photographic portions selections were compared with plated reference portions for each participant.

Setting: Academic sensory testing laboratory in the USA.

Subjects: Separate groups of adults (25-64 years) contributed to fish (n 54) and shrimp (n 53) portion size comparison studies.

Results: In the fish study, there was no difference between photographic portion selections (6·59 (sd 2·65) oz (186·8 (sd 75·1) g)) and reference plate selections (7·04 (sd 2·63) oz (199·6 (sd 74·6) g); P=0·384). Similarly in the shrimp study, there was no difference between photographic portion selections (6·88 (sd 3·40) oz (195·0 (sd 96·4) g)) and reference plate selections (6·06 (sd 2·65) oz (171·8 (sd 75·1) g); P=0·159). Photographic portions predicted plated reference portions for both fish and shrimp based on linear regression (P<0·001). Bland-Altman plot analyses showed good agreement between the two methods, <1 oz (<28·3 g) bias, in both fish and shrimp studies.

Conclusions: This validated photographic seafood portion guide provides a utilitarian tool for accurately assessing fish and shrimp intake in a community setting.

Keywords: Fish; Food intake; Photographic seafood portion guide; Portion estimation; Shrimp.

Fig. 1

Shrimp and fish portion images from the photographic seafood portion guide. Participants selected the image that best represented their typical portion. Letters for shrimp (top row) and fish (bottom row) correspond to portion sizes of: A=2oz, B=4oz, D=8oz and F=12oz. A high-resolution, complete version of this seafood portion guide is available for referenced downloading in the online supplementary material to this paper

Fig. 2

Top: Bland–Altman plots comparing the photographic portion guide and the plated reference portion responses to estimate ‘typical’ participant intake mass (oz) for fish and shrimp. The 95 % CI of the limits of agreement (dashed lines indicating ±2sd of the mean difference) estimates the magnitude of possible sampling error. The mean difference (solid lines) for the two methods indicates sampling bias; the 95 % CI of the mean difference includes the line of equality (zero) for fish data, but not for shrimp data. Bottom: Distribution plots of differences between participant selections from the photographic portion guide and their plated reference portion. Histograms, supported by χ ² goodness-of-fit, indicated fish and shrimp data had a normal distribution