Obesity, Physical Activity and Sedentary Behaviour Among Canadian First Nations Youth: An Exploration of Associated Factors and Evaluation of a School Sports Program

Abstract

Background: Secondary to inequities in several determinants of health, Canadian Aboriginal youth are affected by a higher prevalence rate of obesity and chronic disease than the general population. While the factors contributing to obesity have been extensively studied among the general population, relatively little data are available for First Nations [FN] youth living on reserve. This dissertation will focus on two risk factors, physical inactivity and sedentary behaviour, amenable to change via community-level health promotion initiatives. Purpose: To (a) review quantitative evidence on physical activity [PA], fitness and sedentary behaviour among Canadian Aboriginal youth, (b) explore the potential relationships of socioeconomic, cultural and health behavioural factors with overweight/obesity among FN youth living on reserve, (c) explore the potential relationships of socioeconomic, cultural and health-related factors with high (>1.5 hours/day) levels of television viewing among FN youth living on reserve, (d) conduct a needs assessment for PA programming in one subarctic Ontario FN, and (e) evaluate the implementation and outcomes of a school sport program in this community. Methods: In Study 1, a systematic review of peer-reviewed quantitative research on the PA and fitness of Canadian Aboriginal youth was conducted. The studies, which also described sedentary behaviours, were summarized and the results were presented according to adherence to Canadian guidelines, age and sex differences, and associations with health outcomes. Studies 2 and 3 used data collected for the 12-17 year old subset of the 2008-2010 First Nations Regional Health Survey [RHS]. The relationships between a number of independent variables and (a) overweight/obesity or (b) high television viewing were assessed using logistic regression, stratified by age and sex. Analyses were weighted to account for the sampling strategy. Studies 4 and 5 responded to a subarctic Ontario FN community’s desire for school-based sport programming. A participatory, utilization-focused approach to program evaluation was employed in first conducting a needs assessment for PA and sport in the community (Study 4). Anthropometrics (body mass index [BMI], waist circumference, body fat percentage), PA (3 days of accelerometry) and physical fitness (informed by the Canadian Physical Activity, Fitness and Lifestyles Approach) were measured. Descriptive statistics were compared to reference data and group-level differences (by BMI, waist circumference and body fat status) were tested using Mann-Whitney U and Chi-square tests. Barriers and supports were assessed qualitatively. In Study 5, implementation and outcome evaluations for a school sports program were conducted by repeating the quantitative measures used in Study 4. Changes in these variables were tested using bootstrapped paired samples t-tests. Information on participation, satisfaction and program implementation was collected qualitatively. The qualitative data from Studies 4 and 5 were analyzed inductively for themes by hand, and verified by a second researcher. Results: The systematic review in Study 1 revealed 23 records, 35% (n=8) of which reported on national survey data and 65% (n=15) that reported regional data from several FN sub-populations. National reports showed that about half of youth are ‘active’ and up to two-thirds spend more than 15 hours per week in screen-based sedentary time. Results from regional reports were highly variable. Cardiorespiratory endurance was the only measure of fitness reported in regional studies, and generally appeared low. Girls were almost universally less active than boys. Physical activity was associated with a number of health outcomes; most commonly, an inverse association with BMI or reduced odds of obesity at higher PA levels were reported. Study 2 showed that among 12-17 year old FN youth living on reserve (n=2 888, representing 29 988 individuals in the weighted sample, 56.4% male), overweight/obesity (45.2% prevalence) was associated with younger age (p=0.024), knowledge of a FN language (p=0.015), lower PA level (p=0.001) and lower levels of video game use (p=0.042). Numerous age- and sex-based interactions prompted stratifying the sample for further analysis. No significant relationships were detected for boys aged 12-14 years. Among girls aged 12-14 years, reduced odds of overweight/obesity were observed among those who were moderately active (OR=0.48, 95% CI=0.26, 0.88) or active (OR=0.56, 95% CI=0.32, 0.96) as compared to inactive (p=0.048), and among those who spent more than 1.5 hours using a computer per day (OR=0.66, 95% CI=0.45, 0.96) as compared to those with lesser use (p=0.028). Among boys aged 15-17 years the odds of overweight/obesity were increased among those with parents in the middle education level (high school diploma; OR=1.78, 95% CI=1.15, 2.76) as compared to the lowest level (p=0.032), and those who always or almost always participated in their community’s cultural events (OR=2.06, 95% CI=1.23, 3.44) as compared to less frequent participation (p=0.005). The odds of overweight/obesity were reduced among those playing more than 1.5 hours of video games per day (OR=0.69, 95% CI=0.47, 0.99) as compared to those with lesser use (p=0.044). Among girls aged 15-17 years, the odds of overweight/obesity were increased among those with knowledge of a FN language (OR=1.67, 95% CI=1.15, 2.42) as compared to those without (p=0.007). The odds of overweight/obesity were reduced for those playing more than 1.5 hours of video games per day (OR=0.43, 95% CI=0.27, 0.67) as compared to those with lesser use (p=0.009), and those who reported sometimes (OR=0.43, 95% CI=0.27, 0.67) or always (OR=0.51, 95% CI=0.30, 0.87) eating a nutritious diet as compared to those reporting rarely or never (p=0.001). In all cases, the bivariate relationships remained significant in the multivariate adjusted analyses. Study 3 showed that among on-reserve FN youth (n=3 658, representing 40 462 individuals in the weighted sample, 53.1% male), 39.9% watched more than 1.5 hours of television daily. For the group, only attendance at community cultural events was associated with high television viewing (p=0.005). No significant effects of the tested independent variables were detected for younger (12-14 year old) youth. Among boys aged 15-17 years the odds of high television watching were increased among those with parents having a higher education level (OR=1.73, 95% CI=1.11, 2.69 for high school diploma; OR=1.74, 95% CI=1.16, 2.60 for post-secondary education) as compared to the lowest level (p=0.010), those who always or almost always participated in their community’s cultural events (OR=1.84, 95% CI=1.23, 2.75) as compared to less frequent participation (p=0.003), and those who were overweight or obese (OR=1.44, 95% CI=1.04, 2.00) as compared to normal weight (p=0.028). Among 15-17 year old girls the odds of high television viewing were reduced among those in the largest household size (OR=0.64, 95% CI=0.46, 0.89) as compared to the smallest size (p=0.030), and those reporting very good or excellent mental health (OR=0.63, 95% CI=0.45, 0.86) as compared to poorer mental health (p=0.004). The community needs assessment in Study 4 revealed that in a subarctic Ontario FN, 63% of participating youth in grades 6-7 (n=72, 61.1% male) were overweight or obese, 51% were abdominally obese and 21% had excess body fat. Meanwhile, 86% met Canadian PA guidelines. Cardiorespiratory endurance appeared low in comparison to Canadian Health Measures Survey findings for the general population. Those who were overweight, obese, and/or abdominally obese had lower cardiorespiratory endurance than other youth (p<0.001). Barriers and supports to youth PA fell under the themes of motivation, role modelling, personnel and facilities, and environment and programs. The findings informed a locally implemented school sports program. In Study 5, after a 9-months (one school year) of the program, 57 youth (12.8±1.0 years, 59.6% male) provided at least one of the process and outcome evaluation measures. Over the course of the program, youth increased their mean participation in moderate-to-vigorous PA by 47.9 minutes/day (p=0.016). Boys completed an average additional 10.5 shuttle run stages (p=0.006), and showed a non-significant trend toward improved cardiorespiratory endurance (p=0.057). Girls did not experience significant improvement in cardiorespiratory endurance. Both boys (+8.2 kg, p=0.002) and girls (+5.6 kg, p=0.012) improved their muscular strength. Flexibility improved for the group (+2.2 cm, p=0.015). Boys appeared to participate in the program more often than girls (not statistically significant); 28% of boys reported daily participation as compared to 15% of girls. Qualitative findings revealed barriers to implementation that were related to resource and infrastructure limitations. These were somewhat overcome by mobilizing of the few available resources and volunteer personnel. Conclusions: Many Canadian Aboriginal youth fail to meet established PA and sedentary behaviour guidelines, and physical fitness appears low among a few FN populations. Available data support an inverse association for both higher amounts of PA and lesser time in sedentary behaviour relative to overweight and obesity, providing impetus for PA programming in vulnerable communities. Data from the RHS showed that among FN youth living on reserve, overweight/obesity and high levels of television viewing are prevalent across all age and sex groups. The identification of several factors affecting the odds of overweight/obesity and television viewing has provided some evidence that may inform health promotion initiatives, however more research is needed to fully understand the observed relationships. The findings suggest that programming will need to be tailored to various and and sex categories. In a remote, northern Ontario FN community, although the majority of youth met Canadian PA guidelines, low physical fitness (particularly cardiorespiratory endurance) and a high prevalence of obesity and abdominal obesity supported the community’s desire for school-based PA programming. In the same community, many pre-existing barriers were overcome to implement a school sports program, which was well received. The program was associated with positive outcomes, more so among boys. Thus, future programming should target the participation of girls specifically. Nonetheless, this work demonstrates the many beneficial outcomes that can be achieved when a community with relatively few pre-existing resources is empowered to implement local, sustainable programming supporting the health of their youth.