I am a Physical Activity Epidemiologist and Lecturer in Physical Activity in Adults at the University of Bristol. My research focuses on the associations between physical activity and various health outcomes such as vascular diseases and fractures in adults.
Background Quantifying movement behaviours over 24-hours enables the combined effects of and inte... more Background Quantifying movement behaviours over 24-hours enables the combined effects of and interrelations between sleep, sedentary time and physical activity (PA) to be understood. This is the first study describing 24-hour movement behaviours in school-aged children and adolescents in SouthEast Asia. Further aims were to investigate between-participant differences in movement behaviours by demographic characteristics and timing of data collection during Ramadan and COVID-19 restrictions.
BACKGROUND Risk factors for obesity prevalence can help target treatment, but less is known about... more BACKGROUND Risk factors for obesity prevalence can help target treatment, but less is known about risk factors for obesity incidence which may be distinct and more pertinent for preventing excessive weight gain. OBJECTIVE To explore factors associated with child/adolescent BMI and childhood obesity incidence over 5 years in Malaysia including parental cardiometabolic risk, socio-demographic and baseline child BMI. METHODS Longitudinal data on 1768 children (aged 6-14 years) and parents were drawn from a Malaysian cohort and followed up after 5 years. Multilevel models estimated associations between parent cardiometabolic risk factors (overweight/obesity, central obesity, hypertension, hyperglycaemia), socio-demographics (ethnicity, employment, education) and child BMI z-score at follow-up, adjusting for baseline and potential confounders. Associations with five-year child obesity incidence were explored using logistic multilevel models. RESULTS Obesity prevalence was stable (16-17%) and average annual obesity incidence ranged from 2.2% for 6-10 years (childhood) to 1.2% for 11-14 years (adolescence). Higher baseline BMI z-score was associated with a higher follow-up BMI z score in childhood (0.60; 95% CI: 0.55, 0.65) and adolescence (0.76; 95% CI: 0.70, 0.82). Associations were smaller with maternal overweight/obesity and childhood BMI (overweight: 0.16; 95% CI: -0.03, 0.36, obesity: 0.41; 95% CI: 0.20, 0.61), and paternal overweight/obesity and adolescent BMI (overweight: 0.22; 95% CI: 0.01, 0.43, obesity: 0.16; 95% CI: -0.10, 0.41). Five-year obesity incidence in adolescence was associated with parental obesity (paternal OR= 4.37; maternal OR= 3.38). There were no marked associations with parental hypertension/hyperglycaemia or socio-demographics. CONCLUSIONS Parental overweight/obesity was prospectively associated with higher child BMI and obesity incidence, but the largest follow-up BMIs were among children with a higher baseline BMI. Early childhood, rather than parent cardiometabolic risk, maybe a better target for obesity prevention.
BACKGROUND Traditionally, self-reported physical activity questionnaires have been used for physi... more BACKGROUND Traditionally, self-reported physical activity questionnaires have been used for physical activity (PA) surveillance in children and adolescents, especially in free-living conditions. Objective measures are more accurate at measuring physical activity, but high cost often creates a barrier to their use in low- and middle-income settings. The advent of smartphone technology has greatly influenced mobile health (mHealth) and offers new opportunities in health research, including PA surveillance. OBJECTIVE This scoping review aimed to systematically explore the use of smartphone technology for PA surveillance in children and adolescents, specifically focusing on the use of smartphone applications. METHODS A literature search was conducted involving 5 databases (PubMed, Scopus, CINAHL, Medline, Web of Science), and including Google Scholar to identify articles relevant to the topic that were published from 2008 to 2021. Articles are included if they 1) include children and adolescents within the age range of 5–18 years old; 2) use smartphone technology as PA surveillance; 3) have PA behavioural outcomes such as energy expenditure, steps count, PA levels; 4) are written in the English language; and 5) are published between 2008 and 2021. RESULTS In total, 7 studies (3 cross-sectional, 3 cohort and 1 exploratory) were identified and analysed. All participants involved were 12-18 years old and all studies were conducted in high-income countries. Participants were recruited from school, primary care and voluntarily. Five studies used mobile applications specifically and purposely developed for the study, while 2 studies used mobile applications downloadable from website, Apple Appstore and Android Play Store. PA surveillance using these apps were conducted from 24-hours to 4 weeks. Pros of smartphone technology in measuring PA included quick feedback, accurate measurement in comparison with self-report methods, easy self-monitoring, and being free and downloadable from websites and apps stores. Meanwhile, cons included some apps had not been empirically validated, being unable to measure water activities, some apps required data translation (data are not readily available to use directly) and some are only compatible for Android phones. CONCLUSIONS Evidence of PA surveillance using smartphone technology in children and adolescents was insufficient, which demonstrated the knowledge gap. Additional research is needed to further study the feasibility and validity of smartphone applications in PA surveillance among children and adolescents, especially in low- and middle- income countries.
Objectives To evaluate the association between adherence to the traditional Mexican diet (TMexD) ... more Objectives To evaluate the association between adherence to the traditional Mexican diet (TMexD) and cardiovascular disease (CVD), obesity, and diabetes-related outcomes. Methods Secondary data analysis of the 2018–19 Mexican National Health and Nutrition Survey, including cross-sectional data from 10,180 adults collected by trained personnel via visits to randomly selected households. Adherence to the TMexD was measured using a validated food frequency questionnaire and an adapted TMexD index, developed by systematically reviewing the literature and consulting expert opinion. Outcomes included CVD biomarkers and diagnosis, hypertension values and diagnosis, anthropometric measurements, and diabetes biomarkers and diagnosis. The percentage differences and odds ratios for presenting non-communicable disease (NCD)-related outcomes (with 95% confidence intervals [CI]), were calculated using multiple linear and logistic regression, adjusting for relevant variables. Sensitivity analyses were conducted using multiple imputation, according to sex, and excluding people with an NCD diagnosis. Results High, compared to low, TMexD adherence was associated with lower total cholesterol (−3.5%; 95% CI: −5.2, −1.8), low-density lipoprotein-cholesterol (−4.3%; 95% CI −6.9, −1.5), non-high-density lipoprotein-cholesterol (−3.9%; 95% CI: −6.1, −1.7), and insulin (−9.8%; 95% CI: −16.0, −3.3) concentrations (p < 0.004) in fully adjusted models. Weaker associations were observed in multiple imputation models, whereas men and adults with no NCD diagnosis had stronger associations. No other associations were observed. Conclusions TMexD adherence was associated with a favorable profile for some NCD-related outcomes, highlighting the potential of the TMexD in reducing the risk of some NCD-related outcomes in Mexican adults. Funding Sources This research was funded by Consejo Nacional de Ciencia y Tecnología (CONACYT). CONACYT had no role in the design, analysis or writing of this abstract.
Background Quantifying movement behaviours over 24-hours enables the combined effects of and inte... more Background Quantifying movement behaviours over 24-hours enables the combined effects of and interrelations between sleep, sedentary time and physical activity (PA) to be understood. This is the first study describing 24-hour movement behaviours in school-aged children and adolescents in SouthEast Asia. Further aims were to investigate between-participant differences in movement behaviours by demographic characteristics and timing of data collection during Ramadan and COVID-19 restrictions.
BACKGROUND Risk factors for obesity prevalence can help target treatment, but less is known about... more BACKGROUND Risk factors for obesity prevalence can help target treatment, but less is known about risk factors for obesity incidence which may be distinct and more pertinent for preventing excessive weight gain. OBJECTIVE To explore factors associated with child/adolescent BMI and childhood obesity incidence over 5 years in Malaysia including parental cardiometabolic risk, socio-demographic and baseline child BMI. METHODS Longitudinal data on 1768 children (aged 6-14 years) and parents were drawn from a Malaysian cohort and followed up after 5 years. Multilevel models estimated associations between parent cardiometabolic risk factors (overweight/obesity, central obesity, hypertension, hyperglycaemia), socio-demographics (ethnicity, employment, education) and child BMI z-score at follow-up, adjusting for baseline and potential confounders. Associations with five-year child obesity incidence were explored using logistic multilevel models. RESULTS Obesity prevalence was stable (16-17%) and average annual obesity incidence ranged from 2.2% for 6-10 years (childhood) to 1.2% for 11-14 years (adolescence). Higher baseline BMI z-score was associated with a higher follow-up BMI z score in childhood (0.60; 95% CI: 0.55, 0.65) and adolescence (0.76; 95% CI: 0.70, 0.82). Associations were smaller with maternal overweight/obesity and childhood BMI (overweight: 0.16; 95% CI: -0.03, 0.36, obesity: 0.41; 95% CI: 0.20, 0.61), and paternal overweight/obesity and adolescent BMI (overweight: 0.22; 95% CI: 0.01, 0.43, obesity: 0.16; 95% CI: -0.10, 0.41). Five-year obesity incidence in adolescence was associated with parental obesity (paternal OR= 4.37; maternal OR= 3.38). There were no marked associations with parental hypertension/hyperglycaemia or socio-demographics. CONCLUSIONS Parental overweight/obesity was prospectively associated with higher child BMI and obesity incidence, but the largest follow-up BMIs were among children with a higher baseline BMI. Early childhood, rather than parent cardiometabolic risk, maybe a better target for obesity prevention.
BACKGROUND Traditionally, self-reported physical activity questionnaires have been used for physi... more BACKGROUND Traditionally, self-reported physical activity questionnaires have been used for physical activity (PA) surveillance in children and adolescents, especially in free-living conditions. Objective measures are more accurate at measuring physical activity, but high cost often creates a barrier to their use in low- and middle-income settings. The advent of smartphone technology has greatly influenced mobile health (mHealth) and offers new opportunities in health research, including PA surveillance. OBJECTIVE This scoping review aimed to systematically explore the use of smartphone technology for PA surveillance in children and adolescents, specifically focusing on the use of smartphone applications. METHODS A literature search was conducted involving 5 databases (PubMed, Scopus, CINAHL, Medline, Web of Science), and including Google Scholar to identify articles relevant to the topic that were published from 2008 to 2021. Articles are included if they 1) include children and adolescents within the age range of 5–18 years old; 2) use smartphone technology as PA surveillance; 3) have PA behavioural outcomes such as energy expenditure, steps count, PA levels; 4) are written in the English language; and 5) are published between 2008 and 2021. RESULTS In total, 7 studies (3 cross-sectional, 3 cohort and 1 exploratory) were identified and analysed. All participants involved were 12-18 years old and all studies were conducted in high-income countries. Participants were recruited from school, primary care and voluntarily. Five studies used mobile applications specifically and purposely developed for the study, while 2 studies used mobile applications downloadable from website, Apple Appstore and Android Play Store. PA surveillance using these apps were conducted from 24-hours to 4 weeks. Pros of smartphone technology in measuring PA included quick feedback, accurate measurement in comparison with self-report methods, easy self-monitoring, and being free and downloadable from websites and apps stores. Meanwhile, cons included some apps had not been empirically validated, being unable to measure water activities, some apps required data translation (data are not readily available to use directly) and some are only compatible for Android phones. CONCLUSIONS Evidence of PA surveillance using smartphone technology in children and adolescents was insufficient, which demonstrated the knowledge gap. Additional research is needed to further study the feasibility and validity of smartphone applications in PA surveillance among children and adolescents, especially in low- and middle- income countries.
Objectives To evaluate the association between adherence to the traditional Mexican diet (TMexD) ... more Objectives To evaluate the association between adherence to the traditional Mexican diet (TMexD) and cardiovascular disease (CVD), obesity, and diabetes-related outcomes. Methods Secondary data analysis of the 2018–19 Mexican National Health and Nutrition Survey, including cross-sectional data from 10,180 adults collected by trained personnel via visits to randomly selected households. Adherence to the TMexD was measured using a validated food frequency questionnaire and an adapted TMexD index, developed by systematically reviewing the literature and consulting expert opinion. Outcomes included CVD biomarkers and diagnosis, hypertension values and diagnosis, anthropometric measurements, and diabetes biomarkers and diagnosis. The percentage differences and odds ratios for presenting non-communicable disease (NCD)-related outcomes (with 95% confidence intervals [CI]), were calculated using multiple linear and logistic regression, adjusting for relevant variables. Sensitivity analyses were conducted using multiple imputation, according to sex, and excluding people with an NCD diagnosis. Results High, compared to low, TMexD adherence was associated with lower total cholesterol (−3.5%; 95% CI: −5.2, −1.8), low-density lipoprotein-cholesterol (−4.3%; 95% CI −6.9, −1.5), non-high-density lipoprotein-cholesterol (−3.9%; 95% CI: −6.1, −1.7), and insulin (−9.8%; 95% CI: −16.0, −3.3) concentrations (p < 0.004) in fully adjusted models. Weaker associations were observed in multiple imputation models, whereas men and adults with no NCD diagnosis had stronger associations. No other associations were observed. Conclusions TMexD adherence was associated with a favorable profile for some NCD-related outcomes, highlighting the potential of the TMexD in reducing the risk of some NCD-related outcomes in Mexican adults. Funding Sources This research was funded by Consejo Nacional de Ciencia y Tecnología (CONACYT). CONACYT had no role in the design, analysis or writing of this abstract.
Uploads
Papers by Miranda Armstrong