Registro alimentar fotografado: desenvolvimento e validação de protocolo para pessoas com deficiência visual

Abstract

Research evaluating food and dietary intake of visually impaired people is still scarce nationally and internationally and does not employ validated methods for this population. The development of alternative strategies for assessing food and dietary intake adapted to the reality of the visually impaired person is necessary to support health promotion and disease prevention actions, as well as to provide more autonomy in health care and nutrition among this population. Thus, the objective of this study was to develop and validate a protocol for performing food records photographed by visually impaired people using a cell phone camera. This is a development and validation study, carried out in three stages. The first focused on the technique for photographic recording of food by visually impaired people using a mobile phone and the writing of protocols for its implementation. In the second, visually impaired people made the photographic record of three standardized meals (breakfast, lunch/dinner and snack), following the protocols developed in the previous step. In the third, the protocols were validated by evaluating the photographs by expert panels regarding the framing, focus and angle to identify the types and quantities of food / preparations from a five-point Likert scale. At this stage, the experts also performed an estimated food record based on the observation of the photos to compare with the heavy food record of their meals. Five protocols were developed and tested, remaining only two, considered the most adequate in relation to autonomy, safety, discretion and speed to obtain the photographs, called “Front Photo” (camera at 45º in relation to the meal) and “Aerial Photo” (90º camera in relation to the meal). Both were applied by 40 visually impaired people, mostly blind (77.5%), male (67.5%) aged ≥ 40 years (57.5%). The frequency of photos considered satisfactory in both protocols was high for all evaluated attributes. For the Frontal Photo, the agreement between the experts was substantial regarding the framework to identify the type of food (Fleiss Kappa = 0.70 and p <0.05), the amount of food (Fleiss Kappa = 0.65 and p). <0.05) and portion size (Fleiss Kappa = 0.60 and p <0.05). For the Aerial Photo, there was substantial agreement of the frame to identify the type and quantity of food (Fleiss Kappa = 0.60 and p <0.05, in both attributes). In both protocols the experts got 100% of the foods present in the meals, except for the coffee with milk and grilled chicken (90% correct). In the comparison between the heavy food register and the one estimated by the experts, in both protocols the estimate of the amount of food was better for French bread, coffee with milk and industrialized grape juice. It is concluded that the developed protocols are feasible for people with visual impairment and can be used for qualitative assessment of their food consumption. Finally, it is indicated the need use of adjacent strategies for quantitative assessment of dietary intake by the professional, to be forwarded along with the photographs.