The Diets of On-reserve First Nations Youth: An Exploration of the Factors Associated with Healthy Food Choices and the Impact of School and Community Programs

Abstract

Background. In Canada, First Nations (FN) people experience poorer health outcomes and shorter life expectancies than non-Aboriginal people. Of particular concern are the high and rising prevalence rates of overweight and obesity. This is especially true of on-reserve populations, who may be lacking in the opportunities and resources necessary to facilitate healthy lifestyles (e.g., reasonably priced healthy foods, access to organized sports or recreation facilities). The development of obesity for FN people is exceptionally complex and not well understood, although energy intake is a known contributor. The contributors to the food choices of young people are also numerous, interacting and complex. This is compounded by the severe and lasting effects of colonialism and marginalization, which have left much of the Canadian FN population disadvantaged with respect to many of the social determinants of health and distanced them from their culture. First Nations Canadians have, as a result, undergone a relatively rapid and recent transition towards a more ‘Western’ way of life that is increasingly reliant on store-bought foods and more sedentary in nature. The reduction in traditional, land-based food procurement and consumption among this population, given the known nutritional benefits of these foods and their importance to the holistic health of FN people, is of particular interest. Because of the high prevalence of food insecurity within on-reserve FN populations and the relative lack of economic opportunities in isolated and remote locations, these nutrient-rich traditional foods may sometimes be replaced with energy dense, nutrient poor store-bought alternatives. The food environment in many FN communities does not support or facilitate healthy choices. Healthy food may cost in excess of two times what it does in larger cities, and in remote and isolated locations, such foods may be inconsistently available and of unacceptable quality when they are. Access to traditional foods may be limited by high economic costs associated with their procurement, changes in the migratory patterns of traditional species, losses of traditional knowledge and fears of environmental contamination. If obesity and chronic disease are to be prevented in remote FN communities, children and adolescents are a logical target for population health interventions. In order to plan, initiate and evaluate program or policy initiatives, knowledge of the factors influencing the food intake of youth is necessary. Objectives. This thesis aimed to (a) review the current literature on the diets of Aboriginal youth in Canada, (b) explore the socioeconomic and sociocultural predictors of traditional food intake among a large nationwide sample of Canadian on-reserve FN youth, (c) explore the individual, behavioural and socioeconomic predictors of healthy store-bought food choices among a large nationwide sample of Canadian on-reserve FN youth, (d) assess the nutritional significance of traditional food consumption for youth in two remote, isolated FN communities in northern Ontario before and after a harvest sharing program, and (e) investigate the barriers and facilitators to the sustainability of a school-based healthy snack program for FN youth in a remote, isolated northern Ontario community and the impact of the program on diet. Methods. This thesis is presented as five individual studies. Study 1 presents a systematic literature review of refereed publications on the diets of school-aged Aboriginal youth residing in Canada, published between January 2004 and 2014. Using a predetermined search strategy, Medline (PubMed), Scopus, Education Resources Information Center, Web of Science and Google Scholar databases were scanned to identify the articles included. Each of the included articles was summarized for purpose, design, year, sample population, setting, dietary assessment method and main findings. Studies 2 and 3 utilized First Nations Regional Health Survey 2008/10 data, collected by the First Nations Information Governance Centre, to explore the individual, behavioural, socioeconomic and sociocultural predictors of traditional and store-bought food intake among 12-17 year-old youth living on reserves in Canada. Traditional and store-bought food intakes were elucidated via two separate food frequency questionnaires. The individual (i.e., age, sex, body mass index), behavioural (i.e., school attendance, physical activity, smoking, alcohol use), socioeconomic (i.e., number of children in the household, living with ones biological parents, parent education level) and sociocultural (i.e., use of a FN language, attending community cultural events, number of people helping to understand culture, food sharing) predictors of traditional food, milk and milk products, and vegetable intake were assessed via binary logistic regression modelling for four different age and sex subgroups. Studies 4 and 5 originated from longstanding research collaborations with the remote and isolated subarctic FN communities of the Mushkegowuk Territory on the western coast of James Bay in northern Ontario, Canada. Study 4 focused on two separate communities in which a harvest sharing program was initiated in May 2011 to support local access to lesser snow geese. Dietary data from grades 6-8 youth in both communities were collected via a validated web-based 24-hour dietary recall and food frequency questionnaire. The significance of snow goose consumption to food group and nutrient intake and the likelihood of achieving current dietary standards were assessed via Analysis of Variance and Pearson chi-square tests, where appropriate. Potential changes in diet from baseline to the post-pilot phase of the program were assessed via independent samples t-tests and Pearson chi-square tests. Study 5 focused on a healthy school snack program in one community, which was originally implemented in 2009 and previously evaluated for process and outcomes in 2010. The diets of grades 6-8 youth at baseline and at the four-year time point were assessed via a web-based 24-hour dietary recall, while qualitative information on the challenges and facilitators to the program’s sustainability were gathered via a focus group with the principal and coordinating committee. Possible changes in the food group and nutrient intakes of youth, as well as the proportion meeting current dietary standards were assessed via independent samples t-tests and Pearson chi-square tests. All quantitative analyses were conducted using Statistical Package for the Social Sciences (SPSS, v. 20 Complex Samples (Study 2 and Study 3) or v. 21 (Study 4 and Study 5), IBM Corporation, Armonk, New York) with a significance level of p≤0.05. Qualitative data were managed and analyzed inductively for themes by hand. The qualitative themes were substantiated by a second, independent researcher. Results. In Study 1, 24 studies were reviewed, all of which were cross sectional in design. The majority of studies (n=16, 67%) were from Ontario or Quebec, focused on FN youth (n=21, 88%) and took place in remote and/or isolated settings (n=18, 75%). Nearly all studies utilized the 24-hour recall approach to measure diet (n=19, 79%). On average, the diets of Aboriginal youth fall short of a number of important food groups and nutrients, with specific concerns for vegetables and fruit, milk and alternatives, fibre, folate, vitamin A, vitamin C, calcium and vitamin D across many studies, along with excess fast food and sugar sweetened beverage consumption. In Study 2, 31% of youth reported consuming traditional foods often (n=3587, representing a weighted population of 39 232, 50% 12-14 years old, 52% male). Frequent traditional food consumption was significantly associated with having a parent with a high school education (p=0.014), frequent participation in community cultural events (p<0.001), and often having traditional food shared with one’s household (p<0.001). For 12-14 year-old girls (n=834, weighted population of 9 451), the odds of consuming traditional foods often were increased for those who always or almost always participated in community cultural events as compared to those who rarely or never did (OR=2.43, CI=1.43-4.12, p<0.001), and those who often had traditional foods shared with their household as compared to those who had traditional foods shared with their household less often (OR=4.08, CI=2.26-7.34, p<0.001). For 15-17 year-old girls (n=960, weighted population of 9 376) there were increased odds of frequently consuming traditional foods for those who often had traditional foods shared with their household as compared to those who had foods shared with them less often (OR=4.61, CI=3.08-6.88, p<0.001). For 12-14 year-old boys (n=893, weighted population of 10 331), there were increased odds of frequently consuming traditional foods amongst those who spoke a FN language as compared to those who did not (OR=1.41, CI=1.01-1.96, p=0.024), those who always or almost always participated in community cultural events as compared to those who participated rarely or never (OR=1.76, CI=1.16-2.68, p=0.013), and those who often had traditional foods shared with their household as compared to those who had traditional foods shared with them less often (OR=5.13, CI=3.42-7.72, p<0.001). For boys aged 15-17 years (n=900, weighted population of 10 075), there were increased odds of often consuming traditional foods for those whose parents had a high school education as compared to those whose parents had not completed high school (OR=1.86, CI=1.17-2.95, p=0.024), those who always or almost always participated in community cultural events as compared to those who rarely or never did (OR=2.82, CI=1.63-4.88, p=0.001), and those who often had traditional food shared with their household as compared to those who had traditional food shared with them less often (OR=3.67, CI=2.54-5.30, p<0.001). With the exception of participation in community cultural events for 12-14 year-old boys, all of these relationships persisted following adjustment for other socioeconomic and sociocultural variables included in the multivariate models. In Study 3, more than two-thirds (68%) of participants reported eating vegetables at least daily (n=3721, representing a weighted population of 41 499, 52% male, 50% 12-14 years old) and over half (57%) reported consuming milk and milk products at least once per day (n=3717, representing a weighted population of 41 368, 52% male, 51% 12-14 years old). Daily vegetable consumption was associated with being younger (p<0.001), being physically active (p<0.001), being a non-smoker (p<0.001), abstaining from alcohol (p<0.001) and having at least one parent with a post-secondary education (p=0.026). For 12-14 year-old girls (n=891, weighted population of 9 968), the odds of eating vegetables at least daily were increased amongst those who were physically active as compared to those who were not (OR=1.80, CI=1.18-2.75, p=0.006) and those who did not use alcohol as compared to those who did (OR=1.70, CI=1.16-2.50, p=0.006). For 15-17 year-old girls (n=995, weighted population of 10 076), the odds of consuming vegetables daily were increased amongst those who did not smoke as compared to daily or occasional smokers (OR=1.61, CI=1.14-1.07, p=0.006). These relationships all persisted in the multivariate models. For 12-14 year-old boys (n=921, weighted population of 10 971), the odds of eating vegetables daily were increased amongst those whose parents had a post-secondary education as compared to those whose parents had not completed high school (OR=1.51, CI=1.00-2.28, p=0.050) and decreased amongst those who did not attend school as compared to those who did (OR=0.50, CI=0.26-0.95, p=0.031). For 15-17 year-old boys (n=914, weighted population of 10 484), the odds of eating vegetables daily were increased amongst those whose parents had a post-secondary education as compared to those who had not attended high school (OR=1.87, CI=1.16-3.01, p=0.033). These relationships did not persist when adjusted for other factors in the multivariate models. In this same study, milk and milk products consumption was associated with being female (p=0.025), attending school (p<0.001) and being a non-smoker (p<0.001). For 12-14 year-old girls (n=893, weighted population of 9 989), there were increased odds of consuming milk and milk products daily amongst those who did not smoke as compared to daily and occasional smokers (OR=1.93, CI=1.27-2.93, p=0.002) as well as amongst those who did not drink alcohol as compared to those who did (OR=1.67, CI=1.10-2.54, p=0.014). For 15-17 year-old girls (n=996, weighted population of 10 070), the odds of consuming milk and milk products daily were increased amongst those who were physically active as compared to those who were not (OR=1.64, CI=1.18-2.28, p=0.003), amongst those who were non-smokers as compared to those who smoked daily or occasionally (OR=1.56, CI=1.12-2.14, p=0.007) and amongst those who lived with one biological parent as compared to those who lived with neither biological parent (OR=1.91, CI=1.25-2.93, p=0.007). With the exception of the association between milk product intake and alcohol use amongst the younger sub-group of girls, all of these relationships persisted when controlled for the other variables included in the multivariate models. For 12-14 year-old boys (n=917, weighted population of 10 947), the odds of consuming milk and milk products daily were increased amongst those who were physically active as compared to those who were not (OR=1.51, CI=1.09-2.11, p=0.013) and decreased amongst those who lived with less than three children in their household as compared to those with greater than three children in their household (OR=0.68, CI=0.48-0.97, p=0.029) as well as those who did not attend school as compared to those who did (OR=0.20, CI=0.11-0.37, p<0.001). For 15-17 year-old boys (n=911, weighted population of 10 362), the odds of consuming milk and milk products daily were increased amongst non-smokers as compared to daily and occasional smokers (OR=1.81, CI=1.23-2.69, p=0.003) and decreased amongst those who did not attend school as compared to those who did (OR=0.40, CI=0.25-0.66, p=<0.001). All of these relationships persisted when adjusted for other variables included in the multivariate models. In Study 4 (n=84, 12.38±1.07 years old, 49% male, 67% overweight or obese), 39% of youth were found to consume snow goose weekly, and girls less so than boys (p=0.006). Although snow goose consumption was not associated with body mass index, the frequency of its consumption was associated with intake of meat and alternatives (p=0.019), protein (p=0.004), vitamin B12 (p=0.004), iron (p=0.011) and zinc (p=0.006), where nutrient intakes were highest among those who consumed snow goose moderately (2-6 times per week). Nutrient intake by goose consumption level interacted with sex for protein (p=0.018), thiamine (p=0.008), riboflavin (p=0.042) and iron (p=0.030). Although no changes in mean food group or nutrient intake intakes were observed following the harvest sharing program (n=73, 12.34±0.99 years old, 44% male), it appeared that a greater proportion of youth reported consuming snow goose at this time (63% vs. 48%), and girls were significantly more likely to report consuming the food (61% vs. 33%, p=0.015). In Study 5, youth had higher intakes of vegetables and fruit (p=0.048), milk and alternatives (p=0.017), ‘other’ foods (p=0.030), carbohydrates (p=0.025), fibre (p=0.019), thiamine (p=0.040), riboflavin (p=0.008), folate (p=0.006), calcium (p=0.015), iron (p=0.046), potassium (p=0.007), zinc (p=0.042) and kilocalories (p=0.021) following four years of the program (n=49, 12.7±1.0 years old, 59% male) as compared to baseline (n=43, 13.1±0.9 years old, 60% male). Although numerous challenges were encountered over the first four years of the program, many supporting strategies were noted by the principal and coordinating committee (n=3). Advocacy and community ownership of the program emerged as strong supports to the program’s expansion and sustainability. Acquiring reliable personnel and sustained funding, building new facilities and receiving donations of equipment, and moving away from local food purchasing were helpful approaches. Conclusions. The diets of Aboriginal Canadian youth of school age are typically energy dense and nutrient poor. Explorations of the individual, behavioural, socioeconomic and sociocultural predictors of traditional and healthy store-bought food intakes in a nationwide sample of on-reserve FN youth found a range of significant associations. Traditional food intake was strongly predicted by food sharing and closeness to one’s culture (i.e., participation in community cultural events and speaking a FN language), while the intake of healthy store-bought alternatives (i.e., vegetables, and milk and milk products) was predicted by school attendance, other health-related behaviours (i.e., not smoking, abstaining from alcohol, being physically active), and indicators of socioeconomic status (i.e., higher parent education level, fewer children residing in the household, living with both biological parents). These associations varied by age and sex and were not always in the expected direction. In three remote, isolated communities on the western coast of James Bay, school and community programs had a positive impact on the traditional food, healthy food and nutrient intakes of youth, but in all cases these programs were not sufficient to bring the micronutrient and food group intakes of these youth to recommended levels. Nevertheless, these investigations underscore the positive outcomes that can be achieved when the necessary resources are applied in communities that strongly value the good health of their young people.