Validation of digital photographs, as a tool in 24-h recall, for the improvement of dietary assessment among rural populations in developing countries

Table 2 Amount of food estimated by FP 24-hR and compared with amount weighed in WFR

FOOD CATEGORY	(n)	FP 24-hR	WR	Sperman r	Median difference FP24hR – WFR^a		Bland Altman Analysis FP24hR – WFR (antilog)^b
FOOD CATEGORY	(n)	Median (P25, P75)	Median (P25, P75)	Sperman r	Median (P25, P75)	%	Geometric mean ratio	95% Limits of agreement
Bread [g]	26	55 (50, 60)	55 (47, 65)	0.81	-1.5 (-4.3,3.0)	-2.43	0.98	0.79	1.22
Rice [g]	43	165 (105, 200)	165 (108, 237)	0.95	-13.0 (-30.0, 5.0)	-6.76	0.93	0.71	1.20
Noodles [g]	43	175 (154, 256)	187 (150, 263)	0.93	-12.0 (-20.0, 9.0)	-5.41	0.97	0.73	1.28
Potatoes [g]	80	114 (61, 160)	115 (71, 168)	0.98	-5.0 (-10.0, 7.8)	-5.80	0.96	0.79	1.18
Cassava [g]	19	117 (64, 156)	108 (66, 143)	0.98	-1.0 (-8.0, 8.0)	-2.33	0.99	0.80	1.22
Meat [g]	48	36 (25, 51)	34 (26,49)	0.96	-2.0 (-4.0, 2.0)	-4.88	0.95	0.71	1.28
Egg [g]	15	50(50, 50)	54 (46, 57)	0.75	-3.0 (-7.0, -1.0)	-6.54	0.94	0.76	1.17
Vegetables [g]	198	25 (13, 43)	25 (14, 43)	0.96	-1.0 (-4.0, 2.0)	-5.44	0.98	0.65	1.43
Leafy vegetables [g]	17	25 (25, 50)	27 (20, 46)	0.90	2.0 (-5.0, 7.0)	8.70	1.09	0.70	1.69
Beverages [g]	19	250 (200, 325)	260 (210, 310)	0.95	4.0 (-10.0, 10.0)	1.63	1.01	0.93	1.10

^a Median difference between FP 24-hR and WFR, in grams with 25^th, 75^th percentiles for the difference of each food category, and median of percentage percentage of the difference in parenthesis (calculated as: % of the difference = ((amount from FP 24hR- amount from WFR) / mean amount from WFR)*100).
^b Bland Altman analysis, show the antilog values of the mean difference between FP 24-hR and WFR, represented as the geometric mean ratio of amounts estimated by the FP 24-hR and weighed by WFR, and the 95% limits of agreement, represent the range of proportional agreement between both methods.

ISSN: 1475-2891