Immigrant parents of elementary school children from diverse backgrounds held positive beliefs about the benefits of walking to school and eating breakfast. Despite supportive individual beliefs, participants identified significant barriers to participation in walking to school and school breakfast programs across community/neighborhood, institutional and built/natural environment domains described in the socio-ecological model [22, 28]. These barriers demonstrate the numerous, potentially conflicting, influences on behavior, but also suggest multiple possible avenues for intervention.
For walking to school, predominant barriers among immigrant families emerged from within the community/neighborhood (kidnapping) and built environment domains (traffic safety) that elicited fear among parents for their children's safety. Findings from the present study are consistent overall with barriers to walking to school identified among non-immigrant families in the U.S. (traffic safety, distance, crime) , and in Australia (traffic safety, poor street crossings, distance) . Kerr et al. found that level of parental concern was the strongest explanatory factor for likelihood of children walking or bicycling to school  and Timperio et al. reported that concern about strangers was universal to parents in terms of child safety for walking and bicycling . Immigrant families seemed to place greater emphasis on the danger of child abductions relative to traffic injuries, which could be explained by distrust of neighbors due to language and other cultural barriers, which parents reported, and to living in low-income neighborhoods. Indeed, Weir et al. found associations between parent neighborhood fears with lower physical activity among children in poor urban neighborhoods . Focus group participants in this study frequently cited the media in highlighting abduction dangers, suggesting that immigrant parents depend on media for learning information about life in the U.S.  and may be more susceptible to alarming news reports.
The implications of the present findings for walking to school suggest several avenues for program development among immigrant families. As previously advocated by McMillan et al., , this study highlights the need to address both real and perceived barriers, such as objective built environment changes (crossing guards/signals, walking paths) and perceptions that engender fearfulness (media reports about abductions). First, programs within immigrant communities should highlight the cultural relevance and underlying value about the benefits of walking, as developed through experience walking in their home country. Second, while immigrant parents were receptive to the walking school bus model , language barriers and parent work schedules necessitate support from schools and community-based organizations to facilitate families meeting and building relationships to share supervision. Approaches might include walking in the afternoon when fewer parents work or using school staff as walking group leaders. Third, institutions could provide comprehensive access to school walking programs to working parents, such as offering walkable on-site before school care or activities, or physical activity programs once children arrive. Fourth, in addition to teaching safety to children, participants recommended improved safety supports and built environment design, such as the use of crossing guards, and structural improvements like pedestrian paths and traffic slowing around schools. There is evidence that such built environment changes can result in higher rates of walking to school .
In relation to breakfast, parents in all three ethnic groups described the importance of breakfast consumption for children's performance in school, which is consistent with findings reported in the literature from breakfast programs [38–40]. Although breakfast consumption has been associated with lower risks of child overweight [21, 41], none of the parents in this study mentioned weight control in relation to breakfast. For school breakfast participation, primary barriers emerged from within the community and institutional domains of the ecological model. Time scarcity, which has been identified as a barrier to eating nutritious meals among low-income families , was frequently described by parents in this study about their morning routines. Promoting school breakfast may therefore be especially helpful to these families to provide convenient, nutritious meals; however, parents' concerns need to be addressed. As for walking to school, the inability to supervise children was an important barrier among immigrant parents to support school breakfast. Similar to prior research among non-immigrant families, school bus arrival was a concern for immigrant parents about children not getting enough time for breakfast . Additional concerns not previously identified among non-immigrant populations included inadequate variety of culturally-appropriate breakfast foods (particularly hot meals), inadequate adult supervision during breakfast, and lack of reminders for children to eat breakfast.
The findings about school breakfast highlight the need for improved communication with immigrant families about school breakfast programs, including the hot and cold options available daily, efforts to ensure that school meals meet children's religious or other dietary restrictions, and the availability of adult supervision provided to ensure that children are reminded to get breakfast in the cafeteria and actually consume the food provided. The limited financial resources for school meals make providing appealing, nutritious foods in ethnically diverse schools a challenge. School meal programs may benefit from actively engaging children and families, such as through taste tests and parent surveys. Lengthening the time between bus arrival and the start of classes may also be warranted to provide children adequate time to arrive and consume breakfast at school.
Based on results of this study, preventing child obesity should not necessarily be considered a primary strategy to promote school programs among immigrant families. Parents tend to underestimate their children's weight [44–46], which may partly explain why few participants in this study expressed concern about their own children's weight. While participants acknowledged the overall problem of child obesity, child obesity prevention did not emerge as a personally motivating factor among immigrant parents. Instead, having energy and nourishment for school performance were cited as positive factors for walking to school and eating breakfast.
Limitations of this study include the small number of focus groups conducted in each language, which limits the extent of comparative analysis possible between the three ethnic groups studied. The design of this study, however, was to explore overall themes among immigrant parents rather than to draw specific conclusions about each ethnic group. A limitation of focus group research, in particular, is the challenge of discussing sensitive topics , such as cost of meals, which was not identified as a barrier to participation in school meals by parents in this study, although cost has been described as a factor in previous studies . Unlike quantitative methods, which uses probability sampling, sampling in qualitative methods, as used here, seeks to provide "information-rich" cases . Thus, results here are not intended to be statistically representative. Results of this study can, nonetheless, inform the perspective of those working with immigrant parents to enrich the understanding of cultural and personal factors relevant to promoting school-based physical activity and meals programs .