top of page
Lifespan and Development
Abstracts & Posters
Perceived smartphone addiction and ADHD symptoms in Canadian middle school students: a longitudinal study
Emmett Sihoe, Dr. Sam Liu & Dr. Ulrich Mueller
Concerns have been raised about negative health outcomes associated with high levels of screen time in adolescent populations. Previous research provides evidence of an association between digital media usage and symptoms of Attention-Deficit/Hyperactivity Disorder (ADHD) in school-aged children. These findings have been criticized for a lack of longitudinal evidence, leaving the directionality of the relation between media usage and ADHD symptoms unclear. Using longitudinal data from a sample of Canadian middle school students in grade 6-8 (n=111, ages 11-14, 55% female), this study investigates the direction of the relationship between perceived smartphone addiction and symptoms of ADHD. We hypothesized that baseline smartphone addiction would predict future ADHD symptomatology, and that baseline ADHD symptomatology would predict future smartphone addiction. Data was collected at two time points over a one-year period, using self-report questionnaires. Smartphone addiction was measured using the Smartphone Addiction Scale-Short Form (SAS-SV). ADHD symptomatology was measured using a 17-item rating scale of DSM-5 symptoms. Using hierarchical regression analysis, results show that higher baseline SAS-SV scores predicted future ADHD symptomatology, after controlling for gender and prior ADHD symptoms. However, baseline ADHD symptomatology did not predict future SAS-SV scores, after controlling for gender and prior SAS-SV scores. This study provides important longitudinal evidence that perceived smartphone addiction may be a risk factor for future ADHD symptoms. Results of this study should be interpreted with caution due to the high SES of our sample, and the use of self-report data.
Concerns have been raised about negative health outcomes associated with high levels of screen time in adolescent populations. Previous research provides evidence of an association between digital media usage and symptoms of Attention-Deficit/Hyperactivity Disorder (ADHD) in school-aged children. These findings have been criticized for a lack of longitudinal evidence, leaving the directionality of the relation between media usage and ADHD symptoms unclear. Using longitudinal data from a sample of Canadian middle school students in grade 6-8 (n=111, ages 11-14, 55% female), this study investigates the direction of the relationship between perceived smartphone addiction and symptoms of ADHD. We hypothesized that baseline smartphone addiction would predict future ADHD symptomatology, and that baseline ADHD symptomatology would predict future smartphone addiction. Data was collected at two time points over a one-year period, using self-report questionnaires. Smartphone addiction was measured using the Smartphone Addiction Scale-Short Form (SAS-SV). ADHD symptomatology was measured using a 17-item rating scale of DSM-5 symptoms. Using hierarchical regression analysis, results show that higher baseline SAS-SV scores predicted future ADHD symptomatology, after controlling for gender and prior ADHD symptoms. However, baseline ADHD symptomatology did not predict future SAS-SV scores, after controlling for gender and prior SAS-SV scores. This study provides important longitudinal evidence that perceived smartphone addiction may be a risk factor for future ADHD symptoms. Results of this study should be interpreted with caution due to the high SES of our sample, and the use of self-report data.
Relationships Between Language and Executive Function Development in Early Childhood
Meryssa Waite and Dr. Ulrich Mueller
The current study examines the relationships between language and executive function (EF) development in early childhood through comparing participants’ (n = 32) measures of language and EF across two assessment points in preschool. Through utilizing a variety of measures and a longitudinal design, this study aims to better understand language and EF development. Participants were first assessed at approximately 36 to 42 months old, and then again one year later, at approximately 48 to 54 months old. Language measures included an aggregate language score, Kaufman Assessment Battery for Children (KABC-II; Kaufman & Kaufman, 2004) expressive vocabulary subtest, alongside measures of mean length of utterance (MLU) and number of different words (NDW) calculated from transcripts of free interactions between participants and their parents. Aggregate EF scores were comprised of participants’ z-scores from the Go-No, Shape School, Head Toes Knees Shoulders, Self-Ordered Pointing, and Snow-Grass EF tasks. It was hypothesized that there would be a concurrent and bidirectional relation between language and EF. Language and EF scores correlated concurrently at both assessment points. Hierarchical regression analyses found that early aggregate language predicted later EF, and early EF predicted later NDW. Overall, the current study indicates significant and complex relations between language and EF in early childhood. The practical significance of this study include its potential to inform future assessment and interventions and identify early indicators of language impairment and executive dysfunction. Future research should further examine the relations between language and EF in early childhood across additional assessment points, cultures, and clinical populations.
The current study examines the relationships between language and executive function (EF) development in early childhood through comparing participants’ (n = 32) measures of language and EF across two assessment points in preschool. Through utilizing a variety of measures and a longitudinal design, this study aims to better understand language and EF development. Participants were first assessed at approximately 36 to 42 months old, and then again one year later, at approximately 48 to 54 months old. Language measures included an aggregate language score, Kaufman Assessment Battery for Children (KABC-II; Kaufman & Kaufman, 2004) expressive vocabulary subtest, alongside measures of mean length of utterance (MLU) and number of different words (NDW) calculated from transcripts of free interactions between participants and their parents. Aggregate EF scores were comprised of participants’ z-scores from the Go-No, Shape School, Head Toes Knees Shoulders, Self-Ordered Pointing, and Snow-Grass EF tasks. It was hypothesized that there would be a concurrent and bidirectional relation between language and EF. Language and EF scores correlated concurrently at both assessment points. Hierarchical regression analyses found that early aggregate language predicted later EF, and early EF predicted later NDW. Overall, the current study indicates significant and complex relations between language and EF in early childhood. The practical significance of this study include its potential to inform future assessment and interventions and identify early indicators of language impairment and executive dysfunction. Future research should further examine the relations between language and EF in early childhood across additional assessment points, cultures, and clinical populations.
Caring for Our Caregivers: Predicting the Likelihood of Older Adults Transitioning from Home Care to Residential Care Using the MAPLe Algorithm.
Jannell Walton, Timothy Lukyn, & Stuart MacDonald
Currently, there is a severe lack of home care resources for the older population, which elucidates the need for care providers to develop a method of adequately allocating the limited home care resources to the individuals at the highest risk of transitioning into residential care (Hirdes et al., 2008). In our current study, we aimed to effectively identify individual differences in rates of change across participants’ score on the Resident Assessment Instrument- Home Care (RAI-HC) using the Method for Assigning Priority Levels (MAPLe) algorithm to effectively predict which individuals are most likely to transition from home care to residential care. We predicted that here will be significant increases in MAPLe scores over time and in individual differences in rates of change and that individuals who show significant increases in MAPLe scores will likely be at an increased risk of transitioning to residential as compared to those with stable scores. Our study utilized a sample of 93,928 participants from secondary data from a Canadian Institute of Health Information (CIHI) study conducted by Sheets et al. (2015). The RAI-HC was used to assess each participant between 2006 to mid-2016, and multilevel modelling (MLM) was used to estimate both the initial intercepts and the slopes of each participant’s RAI scores across 6-month retest assessments. Following the MLM analysis, individual slopes of change will be used as a predictor in a logistic regression analysis to predict the likelihood of transitioning to subsequent levels of care by computing the odds ratio. The odds ratio for MAPLe intercepts (i.e., the mean/average scores at most recent time of assessment) was 2.0420, which indicates that a 1-unit increase in a MAPLe score at the most recent time of assessment is associated with a 2.04 times increased likelihood (odds) of transitioning to residential care. In contrast, the odds ratio for MAPLe slopes (i.e., the average change per each additional year of assessment) was 8.14, which denotes that any individual with a 1-unit higher rate of MAPLe score change is 8.14 times more likely to transition to residential care. These findings indicate that using the data from all RAI-HC assessments (i.e., the change slopes) rather than merely using the most recent RAI-HC assessment data is much more accurate and effective for determining a given older individual’s need to transition from receiving home care to living in residential care. These results could have promising clinical implications for health care providers in effectively allocating the limited care resources available for the senior population.
Currently, there is a severe lack of home care resources for the older population, which elucidates the need for care providers to develop a method of adequately allocating the limited home care resources to the individuals at the highest risk of transitioning into residential care (Hirdes et al., 2008). In our current study, we aimed to effectively identify individual differences in rates of change across participants’ score on the Resident Assessment Instrument- Home Care (RAI-HC) using the Method for Assigning Priority Levels (MAPLe) algorithm to effectively predict which individuals are most likely to transition from home care to residential care. We predicted that here will be significant increases in MAPLe scores over time and in individual differences in rates of change and that individuals who show significant increases in MAPLe scores will likely be at an increased risk of transitioning to residential as compared to those with stable scores. Our study utilized a sample of 93,928 participants from secondary data from a Canadian Institute of Health Information (CIHI) study conducted by Sheets et al. (2015). The RAI-HC was used to assess each participant between 2006 to mid-2016, and multilevel modelling (MLM) was used to estimate both the initial intercepts and the slopes of each participant’s RAI scores across 6-month retest assessments. Following the MLM analysis, individual slopes of change will be used as a predictor in a logistic regression analysis to predict the likelihood of transitioning to subsequent levels of care by computing the odds ratio. The odds ratio for MAPLe intercepts (i.e., the mean/average scores at most recent time of assessment) was 2.0420, which indicates that a 1-unit increase in a MAPLe score at the most recent time of assessment is associated with a 2.04 times increased likelihood (odds) of transitioning to residential care. In contrast, the odds ratio for MAPLe slopes (i.e., the average change per each additional year of assessment) was 8.14, which denotes that any individual with a 1-unit higher rate of MAPLe score change is 8.14 times more likely to transition to residential care. These findings indicate that using the data from all RAI-HC assessments (i.e., the change slopes) rather than merely using the most recent RAI-HC assessment data is much more accurate and effective for determining a given older individual’s need to transition from receiving home care to living in residential care. These results could have promising clinical implications for health care providers in effectively allocating the limited care resources available for the senior population.
Tracking Gait Fluctuations among Persons with Dementia and their Caregivers
Board, M., Palmer, J., Wilden, M., McDowell, C., & MacDonald, S.W.S.
Introduction:
It has recently been found that a multivariate approach to tracking intraindividual variability (IIV) across gait indicators (i.e., dispersion) was able to accurately predict dementia classifications, with higher levels of this dispersion associated with a higher likelihood to being classified as having dementia (Halliday, 2020). This dispersion tracks inconsistencies in performance across gait domains, with higher inconsistencies indicating more cognitive impairment. However, as this research is brand new, not a lot is known about how these dispersion levels change over time, or if an intervention focused on ameliorating the comorbidities that are associated with dementia symptoms can reduce these levels.
In the current study, we examine if this multivariate operationalization of variability (dispersion) yields significant change across multiple time points, and if there are any systematic associations between gait dispersion levels with levels from several cognitive assessments taken over the course of the intervention (Halliday, 2002).
Methods:
Community-dwelling older adults with dementia (n=33, M-age=77.4 years; SD=10.5) participated in Voices in Motion (VIM), an intensive repeated-measures choir intervention spanning up to 10 assessments over 18 months. Participants completed single-task (walk only) and dual-task (walk while subtracting 7s) conditions on a GAITRite Computerized Walkway.
Analysis:
Dispersion was operationalized using a regression technique, which computes intraindividual standard deviation (ISD) scores from standardized scores, which are used to track inconsistencies in performance across trials or within-person assessments (Hultsch, et al., 2002). The 10 gait indicators will be converted into a common t-score, and then within-person standard deviation calculated across the 10 indicators for each individual.
Multilevel modelling (MLM) will be used to analyze the data as it allows for the assessment of nested data, in the present case, multiple within-person gait assessments (level 1) are nested within individuals (between-person differences, or level 2; Singer & Willett, 2003).
Introduction:
It has recently been found that a multivariate approach to tracking intraindividual variability (IIV) across gait indicators (i.e., dispersion) was able to accurately predict dementia classifications, with higher levels of this dispersion associated with a higher likelihood to being classified as having dementia (Halliday, 2020). This dispersion tracks inconsistencies in performance across gait domains, with higher inconsistencies indicating more cognitive impairment. However, as this research is brand new, not a lot is known about how these dispersion levels change over time, or if an intervention focused on ameliorating the comorbidities that are associated with dementia symptoms can reduce these levels.
In the current study, we examine if this multivariate operationalization of variability (dispersion) yields significant change across multiple time points, and if there are any systematic associations between gait dispersion levels with levels from several cognitive assessments taken over the course of the intervention (Halliday, 2002).
Methods:
Community-dwelling older adults with dementia (n=33, M-age=77.4 years; SD=10.5) participated in Voices in Motion (VIM), an intensive repeated-measures choir intervention spanning up to 10 assessments over 18 months. Participants completed single-task (walk only) and dual-task (walk while subtracting 7s) conditions on a GAITRite Computerized Walkway.
Analysis:
Dispersion was operationalized using a regression technique, which computes intraindividual standard deviation (ISD) scores from standardized scores, which are used to track inconsistencies in performance across trials or within-person assessments (Hultsch, et al., 2002). The 10 gait indicators will be converted into a common t-score, and then within-person standard deviation calculated across the 10 indicators for each individual.
Multilevel modelling (MLM) will be used to analyze the data as it allows for the assessment of nested data, in the present case, multiple within-person gait assessments (level 1) are nested within individuals (between-person differences, or level 2; Singer & Willett, 2003).
Lifespan
Lifespan
Share
Facebook
Twitter
Pinterest
Tumblr
Copy Link
Link Copied
Search video...
Now Playing
Perceived smartphone addiction predicts ADHD symptoms in middle school children:a longitudinal study
14:15
Now Playing
Relationships Between Language and Executive Function in Early Childhood, Making Waves Conference
05:23
bottom of page