Table 1.
Review of studies that have reported agreement between weighed and recalled portion estimations using a photographic atlas
| Time between | Bland–Altman | ||||||
|---|---|---|---|---|---|---|---|
| Study | Country | Sample (n) | Food types (n) | consumption and recall | Mean difference (weight – recall estimate) | 95 % limits of agreement | Cohen’s κ |
| Huybregts et al.( 34 ) | Burkina Faso | 137 women for all food types | 8 | 24 h | Not applicable | Not applicable | Range 0·52 to 0·92 |
| Korkalo et al.( 35 ) | Mozambique | 99 girls (aged 13–18 years) 25–52 recalls per food type | 5 | 30 min | Range −2 to −61 g between food types | Approx. 120, −225 g for rice and stiff maize porridge Exact values not reported; only shown on plots | Not applicable |
| Lazarte et al. ( 36 ) | Bolivia | 34 women 15–198 recalls per food type | 10 | 24 h | Range −13 to 4 g (median) between food types 237 kJ (56·7 kcal; mean) for all types | Proportional and log limits reported† 49·0, 162·6 | Not reported |
| Steyn et al. ( 37 ) | South Africa | 92 adolescents n per food type not reported | 11 | 30 s* | Range −66·0 to 29·6 g between food types 2·5 kJ (0·6 kcal) for all types | Not reported Only reported 2·7 % and 0·2 % fell above or below the limits | Not reported |
| Tueni et al.( 38 ) | Lebanon | 50 adults 11–67 recalls per food type | 9 (212 items) | 24 h | Range −36·8 to 17·1 g between food types | Reported that they showed ‘good agreement for all dishes’† | Not reported |
| Turconi et al. ( 39 ) | Italy | 448 adults and children 45–3513 per food type | 6 (434 items) | 5–10 min | Range −1·3 to 23·2 g for different food types Overall 13·6 g | Widest limits −153·9, 107·5 g† Overall limits −114·9, 87·8 g† | Not reported |
*
Respondents were only shown the portion and did not consume it.
†
Multiple measurements per individual in these analyses mean that data assumption of non-independence of scores for Bland–Altman limits of agreement does not hold.