Physical activity, screen time and self-rated health and mental health in Canadian adolescents.

OBJECTIVE: Physical activity (PA) and screen time (ST) are associated with self-rated health (SRH) in adults; however, SRH has been less studied among youth, and information about self-rated mental health (SRMH) is lacking. This study examined the associations of PA and ST with SRH and SRMH among adolescents. METHODS: Cross-sectional data from the 2011-2012 Canadian Community Health Survey included 7725 participants aged 12-17years, representing 1,820,560 Canadian adolescents. Associations of self-reported PA and ST to SRH and SRMH were assessed, controlling for age, race/ethnicity, smoking, highest household education and weight status. RESULTS: Excellent/very good SRH was reported by 78% of active vs. 62% of inactive adolescents, and 77% of those meeting vs. 70% of those exceeding ST guidelines (both p<0.001). Excellent/very good SRMH was reported by 81% of active vs. 76% of inactive adolescents, and 84% of those meeting vs. 78% of those exceeding ST guidelines (both p<0.001). Inactive adolescents had twice higher odds of sub-optimal SRH, and inactive girls had 30% greater odds of sub-optimal SRMH. Adolescents exceeding 2h/day ST had 30% greater odds of sub-optimal SRH, and 30-50% greater odds of sub-optimal SRMH. CONCLUSION: PA and ST are independently associated with health perceptions among Canadian adolescents. Interventions should consider health perceptions in addition to biomedical outcomes.

Sedentary behavior in a cohort of 8- to 10-year-old children at elevated risk of obesity.

OBJECTIVES: Sedentary behavior (SB) is distinct from physical inactivity. Children's guidelines recommend ≤2h/day screen time and ≥60min/day moderate-to-vigorous physical activity (MVPA). This study describes SB in children at elevated risk of obesity, including the possibility of high SB in otherwise active children. METHODS: Participants were 534 children from Quebec, Canada, aged 8-10years with ≥1 obese parent in 2005-2008. SB and MVPA were measured by accelerometer and specific SBs by self-report, and height and weight were directly measured. RESULTS: Overweight/obese children were significantly more sedentary overall and reported higher screen time than normal weight children. About 19% of boys and 46% of girls met screen time but not PA guidelines; 28% of boys and 5% of girls met PA but not screen time guidelines. About 46% of overweight/obese children met neither guideline (32% normal weight); only 5% overweight/obese children met both (21% normal weight). Reported behaviors contributed 60%-80% of total SB time; the most sedentary children had the most unaccounted for SB time. CONCLUSIONS: Overweight/obese children are more sedentary overall and report higher screen time than normal weight children. Public health efforts targeting PA and SB in children must consider sex and weight status while being cognizant that being sufficiently active is not exclusive of high levels of SB.

Association between accelerometer-measured physical activity intensities and sedentary time in 8- to 10-year-old children.

This study examines the association between objectively-measured physical activity (PA) intensities and sedentary behavior (SED) in a cohort of 532 children aged 8-10 y. PA and SED were assessed by accelerometer over 7-days. Television and computer/video-game use were self-reported. Associations between PA intensities and SED variables were assessed by Spearman correlations and adjusted multiple linear regression. Higher mean daily moderate-to-vigorous and vigorous PA (MVPA, VPA) were negatively associated with mean daily SED (r = -0.47 and -0.37; p < .001), and positively associated with mean daily total PA (r = .58 and 0.46; p < .001). MVPA was also positively associated with light PA (LPA; r = .26, p < .001). MVPA and VPA were not significantly associated with TV, computer/video or total screen time; accelerometer SED was only weakly associated with specific SED behaviors. On average, for each additional 10 min daily MVPA, children accumulated >14 min less SED, and for each additional 5 min VPA, 11 min less SED. Thus, over the course of a week, higher mean daily MVPA may displace SED time and is associated with higher total PA over and above the additional MVPA, due to concomitant higher levels of LPA. Public health strategies should target both MVPA and SED to improve overall PA and health in children.

Self-rated health and life satisfaction among Canadian adults: associations of perceived weight status versus BMI.

PURPOSE: Obesity is associated with impaired health-related quality of life (HRQL). As perceived constructs, self-rated health (SRH) and general life satisfaction (LS) might be more strongly related to perceived weight status than actual weight status. The aim was to assess agreement between perceived weight status and self-reported body mass index (BMI), and to investigate their associations with SRH and LS as indicators of HRQL. METHODS: Cross-sectional data included 87,545 adults aged 18-65 years from the 2005 Canadian Community Health Survey. Agreement between perceived weight status and self-reported BMI was assessed. Prevalence of suboptimal SRH and LS was estimated by perceived weight status and BMI, and adjusted logistic regression used to assess the odds of suboptimal outcomes. RESULTS: Overall agreement between perceived weight status and self-reported BMI was only moderate (females: κ = 0.58; males: κ = 0.42). The lowest prevalences of suboptimal SRH and LS were in those who reported both a healthy weight BMI and "about right" weight perception. Discordance between perceived weight status and BMI status, and congruence (i.e. perceived weight status = BMI) around underweight or overweight/obese were associated with poorer SRH and LS. For weight perceptions of "about right", BMI status had minimal influence on suboptimal SRH or LS, while perceptions of underweight or overweight were associated with higher odds of suboptimal SRH and LS, independent of BMI. CONCLUSIONS: Adults' weight status perceptions often do not agree with their actual weight status, even when self-reported. While both perceived and actual weight status influence self-rated health and life satisfaction, perceptions are more strongly associated with these HRQL indicators and should be considered when informing obesity-targeted policies and programmes.

Sex differences in the association of youth body mass index to adult health-related quality of life: the physical activity longitudinal study.

OBJECTIVE: The long-term consequences of youth overweight on adult health-related quality of life (HRQL) have not been documented. This study examines sex differences in the association between youth body mass index (BMI) and adult HRQL. METHODS: Subjects included 139 male and 142 female participants aged 7-18 in the 1981 Canada Fitness Survey, followed up in 2002-04. The associations of youth BMI to adult HRQL (SF-36) were examined with bivariate correlations, differences in means and multivariate linear regression analyses. RESULTS: Bivariate analyses revealed positive associations between youth overweight and mental aspects of adult HRQL in females, and weak negative associations with physical aspects, but no significant associations in males. All overweight male and female youth scored the maximum (100) on Role Emotional (RE). In females, compared to healthy weight youth, overweight youth scored 16.0, 13.4, 12.7, and 10.9 points higher on general health (GH), vitality (VT), mental health (MH), and the mental component score (MCS) in adulthood, respectively; a 1 unit increase in youth BMI led to 1.7, 1.5, and 1.4 point increases in adult VT, MH and MCS scores, respectively. Associations were attenuated with the removal of adult BMI from the models, but remained strong for MH and MCS. CONCLUSIONS: Overweight in youth did not have a significant negative impact on HRQL two decades later; rather, a positive association was found with mental aspects of adult HRQL in females.

Are youth BMI and physical activity associated with better or worse than expected health-related quality of life in adulthood? The Physical Activity Longitudinal Study.

PURPOSE: Body mass index (BMI) and physical activity (PA) affect health-related quality of life (HRQL); however, the long-term impact of youth BMI and PA on adult HRQL is unknown. We investigated the relationship of youth BMI and PA to adult HRQL 22 years later. METHODS: Subjects included 310 participants aged 7 to 18 in the 1981 Canada Fitness Survey, followed up in 2002-2004. The associations of youth BMI and leisure time PA to adult HRQL were examined, comparing to age- and sex-adjusted Canadian SF-36 norms. RESULTS: Bivariate analyses revealed positive associations between youth overweight and mental aspects of adult HRQL, but little association with physical aspects. In logistic regression adjusting for adult BMI and other covariates, overweight youth were 7 times more likely than healthy weight youth to score at/above the norm on both mental health (MH) and bodily pain, and almost 18 times more likely on the mental component score (MCS). Youth BMI was also positively associated with general health (GH), social functioning, and role emotional. Removing adult BMI from the models led to attenuated associations with mental HRQL and no association with GH. Longitudinal BMI status change was explored, and findings supported the main regression results. Youth PA was not associated with adult HRQL. CONCLUSIONS: Youth overweight conveyed a long-term positive impact on several aspects of adult HRQL, and this impact may be both direct and indirect through BMI change and the effect on adult BMI. Youth PA had no long-term impact on adult HRQL.

Active living in Saskatchewan: A review of official community plans.

OBJECTIVE: Municipal policies may have a significant impact on the development of environments that provide sustainable opportunities for individuals to engage in healthy, active lifestyles. Little is known about how explicitly community planning in Canada integrates strategies to promote physical activity. In the context of Active Saskatchewan 2020 (AS2020), the strategic plan of Saskatchewan in motion, such an analysis would create a basis for identifying policy gaps and ongoing monitoring. The objective of this study was to review the official community plan (OCP) of each city in the province of Saskatchewan, Canada and identify policies supportive of physical activity. METHODS: A conventional content analysis was completed of the OCPs of the 17 cities in Saskatchewan. Each OCP was reviewed and text extracted that related to supporting physical activity. Extracted text was thematically organized within and across cities, creating a set of indicators for ongoing monitoring. RESULTS: Overall, 17 indicators were identified. The frequency of inclusion of these indicators within the 17 OCPs varied from 17.6% to 82.4%. The mean frequency of indicators identified per OCP was 7.4. The most commonly included indicators included residential neighbourhood plans associated with active living, downtown cycling and pedestrian plans, and joint-use agreements between communities and schools. CONCLUSIONS: Most Saskatchewan OCPs make little direct reference to policies supportive of physical activity. Impacting community-level policy is an expected outcome of AS2020/Saskatchewan in motion. This study identifies a range of indicators for monitoring this process and highlights potential areas for policy development within OCPs.

Smoking and multiple sclerosis: A systematic review and meta-analysis using the Bradford Hill criteria for causation.

BACKGROUND: Despite being one of the most common neurological disorders globally, the cause(s) of multiple sclerosis (MS) remain unknown. Cigarette smoking has been studied with regards to both the development and progression of MS. The Bradford Hill criteria for causation can contribute to a more comprehensive evaluation of a potentially causal risk factor-disease outcome relationship. The objective of this systematic review and meta-analysis was to assess the relationship between smoking and both MS risk and MS progression, subsequently applying Hill's criteria to further evaluate the likelihood of causal associations. METHODS: The Medline, EMBASE, CINAHL, PsycInfo, and Cochrane Library databases were searched for relevant studies up until July 28, 2015. A random-effects meta-analysis was conducted for three outcomes: MS risk, conversion from clinically isolated syndrome (CIS) to clinically definite multiple sclerosis (CDMS), and progression from relapsing-remitting multiple sclerosis (RRMS) to secondary-progressive multiple sclerosis (SPMS). Dose-response relationships and risk factor interactions, and discussions of mechanisms and analogous associations were noted. Hill's criteria were applied to assess causality of the relationships between smoking and each outcome. The effect of second-hand smoke exposure was also briefly reviewed. RESULTS: Smoking had a statistically significant association with both MS risk (conservative: OR/RR 1.54, 95% CI [1.46-1.63]) and SPMS risk (HR 1.80, 95% CI [1.04-3.10]), but the association with progression from CIS to CDMS was non-significant (HR 1.13, 95% CI [0.73-1.76]). Using Hill's criteria, there was strong evidence of a causal role of smoking in MS risk, but only moderate evidence of a causal association between smoking and MS progression. Heterogeneity in study designs and target populations, inconsistent results, and an overall scarcity of studies point to the need for more research on second-hand smoke exposure in relation to MS prior to conducting a detailed meta-analysis. CONCLUSION: This first review to supplement systematic review and meta-analytic methods with Hill's criteria to analyze the smoking-MS association provides evidence supporting the causal involvement of smoking in the development and progression of MS. Smoking prevention and cessation programs and policies should consider MS as an additional health risk when aiming to reduce smoking prevalence in the population.

Sedentary behaviours among adults across Canada.

OBJECTIVES: While cross-Canada variations in physical activity and weight status have been illustrated, less is known about sedentary behaviour (SB). The aim of this study was to describe various SBs and their correlates among Canadian adults. METHODS: Cross-sectional data from the 2011-2012 Canadian Community Health Survey included 92,918 respondents aged 20-75+ years, representative of >22 million Canadian adults. TV/video viewing, computer, video game playing and reading time were self-reported. Associations with socio-demographic, health and health behaviour variables were examined. RESULTS: About 31% of adults reported >2 hours/day TV viewing, while 47% of men and 41% of women reported >5 hours/week computer use, 24% of men and 12% of women reported ≥1 hour/week video game playing, and 33% of men and 46% of women reported >5 hours/week reading; 28% of respondents reported ≥5 hours/day total SB time. Age was the strongest correlate: adults 75+ had 5 and 6 times greater odds respectively of reporting >2 hours/day TV viewing and >5 hours/week reading, but far lesser odds of reporting high computer or video game time, compared to adults 20-24. Other variables associated with specific SBs included gender, marital status, education, occupation, income and immigrant status, as well as BMI, weight perceptions, smoking, diet and physical activity. CONCLUSION: Common sedentary behaviours were associated with numerous socio-demographic, health and health behaviour characteristics in a large representative sample of Canadians. These correlates differed according to the type of SB. Public health interventions targeting SB should be behavior-specific and tailored to the population segment of interest.

Immigrant status and having a regular medical doctor among Canadian adults.

OBJECTIVE: New immigrants generally arrive in Canada with a health advantage over their Canadian counterparts, but lose that advantage over time. Difficulties in acquiring a physician may contribute. Past studies relied on older data, and lacked control for many confounders and assessment of gender differences. We assessed the relationship between immigrant status and having a regular doctor among Canadian adults. METHODS: Cross-sectional data from the 2011-2012 Canadian Community Health Survey were self-reported by 73,958 respondents aged 18-64, representing >20 million Canadian adults. The relationship between immigrant status and having a doctor was analyzed using χ2 and logistic regression analyses stratified by gender. Age, body mass index, race, education, province of residence, physical activity, chronic conditions, self-perceived health, and number of children in household were controlled. RESULTS: Approximately 77% of males and 87% of females reported having a doctor. About 7% of respondents reported being new immigrants in Canada (0-9 years), while 16% were established immigrants (≥10 years). For males (M) and females (F) respectively, 78% and 88% of non-immigrants, 55% and 68% of new immigrants, and 84% and 91% of established immigrants reported having a doctor (p < 0.001). Compared to non-immigrants, new immigrants were significantly less likely to have a doctor (OR (95% CI) M: 0.43 (0.38-0.47); F: 0.36 (0.32-0.41)), while established immigrants were significantly more likely to have a doctor (M: 1.13 (1.03-1.24); F: 1.16 (1.03-1.30)). CONCLUSION: New Canadian immigrants are less likely to have a regular doctor compared to non-immigrants, and should be targeted by policies and programs facilitating finding a doctor.

Correlates of sedentary behaviour in 8- to 10-year-old children at elevated risk for obesity.

The purpose of this study was to describe correlates of sedentary behaviour (SED) in children at elevated risk of obesity because of parental obesity. Participants were 534 children aged 8-10 years with ≥ 1 obese parent. SED and physical activity (PA) were measured by accelerometer, screen time by self-report, and height, weight, waist circumference, and cardiovascular fitness objectively measured. Data describing the child, parents, friends, and home and neighbourhood environments were from child self-report. Higher total SED time was significantly positively associated with child's age, mother's age, Tanner stage, weight status or waist circumference, less self-reported PA, choosing screen time over PA/sport, mother saying PA/sport good for them, and fewer weekly physical education (PE) classes. Exceeding 2 h/day screen time was significantly associated with child's age, male sex, weight status or waist circumference, choosing screen time over PA/sport, and dinnertime TV viewing. Children regularly watching TV with dinner had 2.3 times greater odds (95% confidence interval (CI) 1.52, 3.58) of exceeding screen time guidelines compared with children rarely watching TV with dinner; children reporting ≤ 2 PE classes/week had 2.4 times greater odds (95% CI 1.41, 4.10) of being in the highest SED tertile compared with children reporting >2 PE classes/week. Hence, the most sedentary children are older, more biologically mature, less active, more overweight/obese, have fewer PE classes, and are more likely to choose screen time over PA and watch TV with dinner compared with less sedentary children. PE opportunities and mealtime TV viewing are potentially modifiable targets for reducing total SED and screen time in children.

Combined physical activity/sedentary behaviour associations with indices of adiposity in 8- to 10-year-old children.

OBJECTIVE: Individuals may achieve high physical activity (PA) yet also be highly sedentary (SED). This study assessed adiposity in children classified by PA/SED groups. METHODS: Participants were 520 8- to 10-year-old children with ≥ 1 obese parent. Moderate-to-vigorous PA (MVPA) and SED were measured by accelerometer, and screen-time was measured by self-report. Height, weight, waist circumference (WC), body fat percentage (BF%), and VO2peak were objectively measured; energy intake was measured by dietary recall. Elevated adiposity was defined as BMI ≥ 85th percentile, WC ≥ 90th percentile, BF% ≥ 85th percentile, or waist-to-height ratio (WHR) ≥ 0.5. RESULTS: Up to 27% of boys and 15% of girls were active/SED. Adiposity was lowest for active/non-SED, highest for inactive/SED, and intermediate and similar for active/SED and inactive/non-SED. Using 60 min/d MVPA and 2 h/d screen-time cut-offs, prevalence ranges for elevated adiposity in the active/non-SED, active/SED, inactive/non-SED, and inactive/SED groups were 0% to 14%, 15% to 44%, 16% to 40%, and 32% to 51%, respectively. Corresponding odds and 95% confidence intervals of being overweight/obese for the latter groups were 3.8 (95% CI, 1.7-8.4), 3.8 (1.8-8.2), and 4.9 (2.3-10.3) versus active/non-SED. PA/SED-adiposity associations were mediated by fitness but not energy intake. CONCLUSIONS: Combined PA/SED levels are strongly associated with adiposity in children, but associations are mediated by fitness. Active children who accumulate >2 h/d of screen time and inactive children are equally likely to be overweight/obese.

Self-rated health in children at risk for obesity: associations of physical activity, sedentary behaviour, and BMI.

BACKGROUND: Self-rated health (SRH) is a common indicator of health-related quality of life; however, little is known about SRH in children. This study explored the associations of physical activity (PA), sedentary behavior (SED), and BMI with SRH in children at risk for obesity. METHODS: Participants were 527 children aged 8-10 years in the Quebec Adipose and Lifestyle Investigation in Youth study (inclusion criteria: ≥ 1 parent clinically obese). PA and SED were measured by accelerometer, specific SED behaviors by self-report, and height and weight measured. RESULTS: About 40% of children were overweight or obese; 48% reported less-than- excellent health. The odds of reporting less-than-excellent health were higher among obese girls (OR 3.0, 95% CI 1.4-5.2) and boys (OR 2.7, 95% CI 1.5-6.1) versus healthy weight children. Boys not meeting PA guidelines and boys in the lowest moderate-to-vigorous PA tertile were at 2 and 6 times higher odds of less-than-excellent SRH, respectively, versus more active boys. In girls, higher computer/video time and reading time were associated with higher and lower odds of less-than-excellent SRH, respectively. CONCLUSIONS: Obesity is inversely associated with SRH in boys and girls, PA is positively associated with SRH in boys, and SED behaviors are associated with SRH in girls.

Keeping the weight off: physical activity, sitting time, and weight loss maintenance in bariatric surgery patients 2 to 16 years postsurgery.

BACKGROUND: Bariatric surgery patients often exhibit low levels of physical activity (PA), despite the presumed importance of PA as an adjunct to surgery for successful weight loss. Little is known regarding the associations of PA and sedentary behaviors to weight loss outcomes in the long term following surgery. The objective of the study was to assess the associations of PA and sitting time with weight status, weight loss, and weight maintenance outcomes in bariatric patients 2-16 years postsurgery. METHODS: A total of 303 Roux-en-Y Gastric Bypass patients (73% female; mean age 47 ± 10 years, mean 7 ± 4 years since surgery) completed a telephone questionnaire. Patients reported moderate-to-vigorous PA (MVPA: # sessions/week ≥30 min) and average daily sitting time (h/day). Associations with various weight outcomes were assessed. RESULTS: Only 48% of patients reported ≥1 session/week MVPA, and mean reported sitting time was 7 ± 4 h/day. Neither MVPA nor sitting time was associated with weight loss outcomes at patients' lowest weight postsurgery. However, both MVPA and sitting time were independently positively and inversely, respectively, associated with total (kg) weight loss, % weight loss, and % excess weight loss at current weight, as well as weight loss maintained vs. regained, controlling for age, sex, surgery type, presurgery BMI, total initial weight loss, and time since surgery. CONCLUSIONS: Results demonstrate associations between MVPA and high sitting time and weight loss outcomes among bariatric patients in the long term. The implications for long-term weight management and concomitant health outcomes highlight the need for appropriate follow-up and interventions in this unique high-risk patient population.

Physical activity, body mass index, and health-related quality of life in Canadian adults.

PURPOSE: Physical activity (PA) positively influences health-related quality of life (HRQL), whereas obesity is associated with significant HRQL impairments. Active-obese persons often have similar or lower risk of cardiovascular outcomes and mortality than inactive-healthy weight persons; however, the combined PA-weight status effects on HRQL are unclear. The aim was to investigate the combined association of PA and body mass index (BMI) with HRQL in Canadian adults and older adults. METHODS: Cross-sectional data included 110,986 participants ≥ 18 yr from the 2005 Canadian Community Health Survey, representing an estimated 22,563,527 Canadians. HRQL indicators included: Self-Rated Health (SRH), Participation and Activity Limitation due to illness/injury (PAL), and Total Disability Days (physical + mental) during the past 14 d (TDD). Prevalence of adverse HRQL was estimated by BMI, PA, and combined BMI-PA categories. Adjusted logistic regression was used to assess the odds of adverse HRQL by BMI, PA, and BMI-PA. Analyses were stratified by sex and age (18-44, 45-64, ≥65 yr). RESULTS: In both men and women of all ages, inactive individuals had greater likelihood of fair/poor SRH, and sometimes/often PAL, at all BMI levels; conversely, in active individuals, being underweight, overweight, or obese had little effect on SRH and PAL. Associations were weaker for TDD, where the greatest influence was in older adults from inactivity combined with underweight. Overweight showed less association to HRQL in males and older adults, whereas underweight showed stronger association in males and older adults. CONCLUSIONS: When examining BMI-PA in combination, PA emerges as the more important correlate of HRQL, regardless of weight status. This reinforces the importance of PA to health outcomes over and above the benefits related to weight loss or maintenance.

Tracking of obesity and physical activity from childhood to adulthood: the Physical Activity Longitudinal Study.

OBJECTIVE: Body mass index (BMI) has shown moderate to strong stability through childhood into adulthood, while physical activity (PA) tracks less well. Tracking studies have often had limited follow-up lengths. The aim was to investigate BMI and PA tracking over 22 years from youth to adulthood. METHODS: Subjects included 374 participants aged 7 to 18 years in the 1981 Canada Fitness Survey, who were re-evaluated in 2002-04. The stability of BMI and leisure-time PA energy expenditure (AEE) was assessed by inter-age correlations, maintenance of extreme quintiles and BMI status, and the prediction of adult overweight from youth BMI. RESULTS: BMI tracking was moderate to strong (r=0.42-0.65) in females, and moderate (r=0.29-0.53) in males. Approximately 38% and 42% of youth in the highest and lowest BMI quintiles, respectively, remained in these quintiles as adults. About 83% of overweight youth remained overweight as adults, while 85% of overweight adults were not overweight youth. Almost all healthy weight adults had been healthy weight youth. The odds of being overweight in adulthood was 6.2 times greater (95% CI: 2.2-17.2) in overweight compared with healthy weight youth. PA tracking over 22 years was low and non-significant, but moderate over the final 15 years. Only 16% and 18% of youth in the highest and lowest PA quintiles, respectively, remained in these quintiles as adults. CONCLUSIONS: BMI, but not PA, tracked well over 22 years in this sample. The majority of overweight youth remained overweight as adults; however, the majority of overweight adults were not overweight youth.

Trends in physical activity research in Canada.

The purpose of this study was to assess Canada's physical activity research productivity by (i) assessing Canada's contribution to global physical activity research, (ii) determining trends in Canadian physical activity research over time, and (iii) identifying research areas of greater or lesser representation. A Medline search was performed to identify physical activity research published between 1990 and 2005 from Canadian institutions or using Canadian data. Trends over time were plotted and compared with global physical activity research patterns. Original and review articles were sorted based on subject characteristics, nature of physical activity contribution, province and institution of origin, and research area or "pillar". A total of 5302 Canadian physical activity articles were identified for the period 1990-2005, representing 4.9% of global physical activity research. After manual sorting of abstracts, 3829 relevant articles were included for further analysis. A majority of Canada's physical activity research has come from Ontario (49%), followed by Quebec (21%), Alberta (10%), and British Columbia (10%). Where physical activity was the primary research focus, the biomedical and clinical pillars each accounted for 39% of Canadian research, with lesser contributions from population health (14%) and health services (2%). Physical activity research productivity in Canada has paralleled global trends over the last 15 years. There is currently less physical activity research being conducted in population health and health services than in the biomedical and clinical areas; however, these areas play an important role in the development of public health policy and programs targeted at reducing the burden of physical inactivity and obesity in Canada.