**1. Introduction**

Dietary assessment studies are important for the development of public nutrition policies and interventions because they can identify population groups at risk of nutritional health problems and describe their dietary habits [1]. In Europe, some immigrant groups have a higher risk of developing nutrition-related diseases than the host population, in particular overweight/obesity and diabetes mellitus type 2 [2–4]. This may be due to changes in dietary habits and physical activity patterns influenced by a process of acculturation, urbanization and westernization [2,5]. Only relatively few studies describe dietary habits in immigrant groups, but these indicate diets with increased consumption of processed food after migration that replace healthy dietary components, such as fruits, vegetables, nuts, and whole grains [2,3].

Dietary studies among immigrant groups are hampered by a lack of suitable cultural-sensitive assessment methods and data collection procedures. Missing food composition data on ethnic foods and the use of dietary assessment methods, which are not critically assessed for suitability in these groups, may limit the reliability of dietary intake data among immigrants [2]. Furthermore, various methodological aspects, such as sampling and recruitment, tools and method of administration, among others, often require special attention [6]. In addition, dietary assessment among infants and children in particular has several inherent challenges, and these might be amplified among immigrant groups [1,7]. It is, therefore, important to exercise considerable caution when conducting dietary studies in this study group in order to reduce possible errors and increase validity.

Previous Norwegian national dietary surveys among infants ("Spedkost", aged six months and 12 months) and toddlers ("Småbarnskost", aged 24 months) excluded children of mothers born outside Scandinavia [8–10]. There are two main reasons for this exclusion: (1) the dietary assessment method used in the study was a food frequency questionnaire (FFQ) and the food list was not adapted to non-Scandinavian food habits and (2) the FFQ was only available in Norwegian. Thus, the method was less suitable for population groups with atypical food habits, poor Norwegian skills or lower literacy levels. The researchers expressed a need for separate studies among children of immigrant parents using more appropriate methods [11].

The "InnBaKost" study was initiated to address the limited knowledge about dietary habits and health among children in Norway with immigrant backgrounds. Children of Somali- and Iraqi-born mothers were chosen because they are the two non-Western immigrant groups currently with the highest number of births in Norway [12]. The aim of this research project was to collect information about breastfeeding practices and feeding patterns among infants aged six months with follow-up at 12 and 24 months to supplement the Spedkost and Småbarnskost surveys [8–10]. A modified FFQ was used for the data collection at six months of age. However, a structured 24-h recall method was considered more appropriate at ages 12 and 24 months. This because the FFQ requires that variations in food habits must be known and included in the food list to develop suitable FFQs [13,14]. The aim of the present study was to develop, pilot test and evaluate a protocol for a 24-h recall procedure, with registration forms and visual tools, to assess the dietary intake of toddlers of Somali- and Iraqi-born mothers living in Norway.
