Revealing the Mysteries of Population Mobility Amid the COVID-19 Pandemic in Canada: Comparative Analysis With Internet of Things-Based Thermostat Data and Google Mobility Insights

BACKGROUND

The COVID-19 pandemic necessitated public health policies to limit human mobility and curb infection spread. Human mobility, which is often underestimated, plays a pivotal role in health outcomes, impacting both infectious and chronic diseases. Collecting precise mobility data is vital for understanding human behavior and informing public health strategies. Google's GPS-based location tracking, which is compiled in Google Mobility Reports, became the gold standard for monitoring outdoor mobility during the pandemic. However, indoor mobility remains underexplored.

OBJECTIVE

This study investigates in-home mobility data from ecobee's smart thermostats in Canada (February 2020 to February 2021) and compares it directly with Google's residential mobility data. By assessing the suitability of smart thermostat data, we aim to shed light on indoor mobility patterns, contributing valuable insights to public health research and strategies.

METHODS

Motion sensor data were acquired from the ecobee "Donate Your Data" initiative via Google's BigQuery cloud platform. Concurrently, residential mobility data were sourced from the Google Mobility Report. This study centered on 4 Canadian provinces-Ontario, Quebec, Alberta, and British Columbia-during the period from February 15, 2020, to February 14, 2021. Data processing, analysis, and visualization were conducted on the Microsoft Azure platform using Python (Python Software Foundation) and R programming languages (R Foundation for Statistical Computing). Our investigation involved assessing changes in mobility relative to the baseline in both data sets, with the strength of this relationship assessed using Pearson and Spearman correlation coefficients. We scrutinized daily, weekly, and monthly variations in mobility patterns across the data sets and performed anomaly detection for further insights.

RESULTS

The results revealed noteworthy week-to-week and month-to-month shifts in population mobility within the chosen provinces, aligning with pandemic-driven policy adjustments. Notably, the ecobee data exhibited a robust correlation with Google's data set. Examination of Google's daily patterns detected more pronounced mobility fluctuations during weekdays, a trend not mirrored in the ecobee data. Anomaly detection successfully identified substantial mobility deviations coinciding with policy modifications and cultural events.

CONCLUSIONS

This study's findings illustrate the substantial influence of the Canadian stay-at-home and work-from-home policies on population mobility. This impact was discernible through both Google's out-of-house residential mobility data and ecobee's in-house smart thermostat data. As such, we deduce that smart thermostats represent a valid tool for facilitating intelligent monitoring of population mobility in response to policy-driven shifts.

Breast cancer, alcohol, and phosphate toxicity

Alcohol consumption is associated with an increased risk of breast cancer, even at low alcohol intake levels, but public awareness of the breast cancer risk associated with alcohol intake is low. Furthermore, the causative mechanisms underlying alcohol's association with breast cancer are unknown. The present theoretical paper uses a modified grounded theory method to review the research literature and propose that alcohol's association with breast cancer is mediated by phosphate toxicity, the accumulation of excess inorganic phosphate in body tissue. Serum levels of inorganic phosphate are regulated through a network of hormones released from the bone, kidneys, parathyroid glands, and intestines. Alcohol burdens renal function, which may disturb the regulation of inorganic phosphate, impair phosphate excretion, and increase phosphate toxicity. In addition to causing cellular dehydration, alcohol is an etiologic factor in nontraumatic rhabdomyolysis, which ruptures cell membranes and releases inorganic phosphate into the serum, leading to hyperphosphatemia. Phosphate toxicity is also associated with tumorigenesis, as high levels of inorganic phosphate within the tumor microenvironment activate cell signaling pathways and promote cancer cell growth. Furthermore, phosphate toxicity potentially links cancer and kidney disease in onco-nephrology. Insights into the mediating role of phosphate toxicity may lead to future research and interventions that raise public health awareness of breast cancer risk and alcohol consumption.

Forecasting daily COVID-19 cases with gradient boosted regression trees and other methods: evidence from U.S. cities

Introduction

There is a vast literature on the performance of different short-term forecasting models for country specific COVID-19 cases, but much less research with respect to city level cases. This paper employs daily case counts for 25 Metropolitan Statistical Areas (MSAs) in the U.S. to evaluate the efficacy of a variety of statistical forecasting models with respect to 7 and 28-day ahead predictions.

Methods

This study employed Gradient Boosted Regression Trees (GBRT), Linear Mixed Effects (LME), Susceptible, Infectious, or Recovered (SIR), and Seasonal Autoregressive Integrated Moving Average (SARIMA) models to generate daily forecasts of COVID-19 cases from November 2020 to March 2021.

Results

Consistent with other research that have employed Machine Learning (ML) based methods, we find that Median Absolute Percentage Error (MAPE) values for both 7-day ahead and 28-day ahead predictions from GBRTs are lower than corresponding values from SIR, Linear Mixed Effects (LME), and Seasonal Autoregressive Integrated Moving Average (SARIMA) specifications for the majority of MSAs during November-December 2020 and January 2021. GBRT and SARIMA models do not offer high-quality predictions for February 2021. However, SARIMA generated MAPE values for 28-day ahead predictions are slightly lower than corresponding GBRT estimates for March 2021.

Discussion

The results of this research demonstrate that basic ML models can lead to relatively accurate forecasts at the local level, which is important for resource allocation decisions and epidemiological surveillance by policymakers.

Breast Cancer and Bone Mineral Density in a U.S. Cohort of Middle-Aged Women: Associations with Phosphate Toxicity

Breast cancer is associated with phosphate toxicity, the toxic effect from dysregulated phosphate metabolism that can stimulate tumorigenesis. Phosphate toxicity and dysregulated phosphate metabolism are also associated with bone mineral abnormalities, including excessive bone mineral loss and deposition. Based on shared associations with dysregulated phosphate metabolism and phosphate toxicity, a hypothesis proposed in the present mixed methods-grounded theory study posits that middle-aged women with incidence of breast cancer had a greater magnitude of changes in bone mineral density over time compared with women who remained cancer-free. To test this hypothesis, a mixed-effects model was used to analyze the associations of breast cancer incidence with spinal bone mineral density changes in the U.S. Study of Women's Health Across the Nation. Compared with women in the cohort who remained cancer-free, women who self-reported breast cancer had higher bone mineral density at baseline, but had more rapid losses in bone mineral density during follow-up visits. These findings agree with the hypothesis that a greater magnitude of changes in bone mineral density over time is associated with breast cancer in a cohort of middle-aged women. The findings also have implications for studies investigating dysregulated phosphate metabolism and phosphate toxicity as causative factors of bone metastasis in metastatic breast cancer. Additionally, the authors previously found increased breast cancer risk associated with high dietary phosphate intake in the same cohort of middle-aged women, and more studies should investigate a low-phosphorus diet to reduce bone mineral abnormalities and tumorigenesis in breast cancer patients.

Examining the Association Between Household Enrollment in the Pantawid Pamilyang Pilipino Program (4Ps) and Wasting and Stunting Status Among Children Experiencing Poverty in the Philippines: A Cross-Sectional Study

This study assessed whether enrollment in a national conditional cash transfer program was associated with wasting and stunting among children experiencing extreme poverty in the Philippines. Data were drawn from cross-sectional surveys collected from 10 regional areas in the Philippines between April 2018 and May 2019. A total of 2945 children aged between six months and 12 years comprised the analytical sample. Multilevel logistic regression was conducted to estimate the association between enrollment in Pantawid Pamilyang Pilipino Program (4Ps) and stunting and wasting, controlling for sociodemographic factors and clustering by region. There was no meaningful association between household enrollment in 4Ps and the wasting status of children, but enrollment in 4Ps was associated with lower odds of stunting and differed by geography type. Findings suggest that the current design of 4Ps may not address sudden shocks that contribute to wasting, but may address the underlying socioeconomic risk factors associated with stunting.

High Dietary Phosphorus Is Associated with Increased Breast Cancer Risk in a U.S. Cohort of Middle-Aged Women

Research has shown that high amounts of dietary phosphorus that are twice the amount of the U.S. dietary reference intake of 700 mg for adults are associated with all-cause mortality, phosphate toxicity, and tumorigenesis. The present nested case-control study measured the relative risk of self-reported breast cancer associated with dietary phosphate intake over 10 annual visits in a cohort of middle-aged U.S. women from the Study of Women's Health Across the Nation. Analyzing data from food frequency questionnaires, the highest level of daily dietary phosphorus intake, >1800 mg of phosphorus, was approximately equivalent to the dietary phosphorus levels in menus promoted by the United States Department of Agriculture. After adjusting for participants' energy intake, this level of dietary phosphorus was associated with a 2.3-fold increased risk of breast cancer incidence compared to the reference dietary phosphorus level of 800 to 1000 mg, which is based on recommendations from the U.S. National Kidney Foundation, (RR: 2.30, 95% CI: 0.94-5.61, p = 0.07). Despite the lack of statistical significance, likely due to the small sample size of the cohort, the present nested case-control study's clinically significant effect size, dose-response, temporality, specificity, biological plausibility, consistency, coherence, and analogy with other research findings meet the criteria for inferred causality in observational studies, warranting further investigations. Furthermore, these findings suggest that a low-phosphate diet should be tested on patients with breast cancer.

Examining the use of decision trees in population health surveillance research: an application to youth mental health survey data in the COMPASS study

INTRODUCTION

In population health surveillance research, survey data are commonly analyzed using regression methods; however, these methods have limited ability to examine complex relationships. In contrast, decision tree models are ideally suited for segmenting populations and examining complex interactions among factors, and their use within health research is growing. This article provides a methodological overview of decision trees and their application to youth mental health survey data.

METHODS

The performance of two popular decision tree techniques, the classification and regression tree (CART) and conditional inference tree (CTREE) techniques, is compared to traditional linear and logistic regression models through an application to youth mental health outcomes in the COMPASS study. Data were collected from 74 501 students across 136 schools in Canada. Anxiety, depression and psychosocial well-being outcomes were measured along with 23 sociodemographic and health behaviour predictors. Model performance was assessed using measures of prediction accuracy, parsimony and relative variable importance.

RESULTS

Decision tree and regression models consistently identified the same sets of most important predictors for each outcome, indicating a general level of agreement between methods. Tree models had lower prediction accuracy but were more parsimonious and placed greater relative importance on key differentiating factors.

CONCLUSION

Decision trees provide a means of identifying high-risk subgroups to whom prevention and intervention efforts can be targeted, making them a useful tool to address research questions that cannot be answered by traditional regression methods.

Do school characteristics, based on the Comprehensive School Health Framework, contribute to youth meeting national physical activity recommendations over time?

INTRODUCTION

Canadian youth are insufficiently active, and schools may play a role in promoting student physical activity (PA). Based on the Comprehensive School Health (CSH) framework, this study examined whether school characteristics are associated with secondary school students meeting national PA recommendations over time.

METHODS

We used COMPASS survey data from 78 schools in Ontario and Alberta and 9870 Grade 9 and 10 students attending those schools. Students who provided two years of linked PA data (2013/14 and 2015/16) and gender were included. Multilevel analysis was conducted by gender, evaluating the relationship of school-level characteristics (guided by CSH) with students achieving all three PA recommendations after two years (≥ 60 min/day of moderate-to-vigorous PA, vigorous PA ≥ 3 days/week, strengthening activities ≥ 3 days/week).

RESULTS

More than half (56.9%) of students achieving the PA recommendations at baseline were no longer achieving them after two years, and just a quarter (25.6%) of students not achieving the recommendations at baseline achieved them after two years. School-level factors were significantly associated with students achieving the recommendations, but these factors differed by student strata (i.e. by gender and baseline PA status). Generally, student access to equipment, public health partnerships and staff time for health were associated with increased odds of achieving the PA recommendations for certain students.

CONCLUSION

Modifications to school characteristics within CSH may play a role in supporting students in achieving or continuing to achieve the PA recommendations after two years. Further research is needed to better understand the underlying dynamics of the observed relationships.

Using Decision Trees to Examine Environmental and Behavioural Factors Associated with Youth Anxiety, Depression, and Flourishing

Modifiable environmental and behavioural factors influence youth mental health; however, past studies have primarily used regression models that quantify population average effects. Decision trees are an analytic technique that examine complex relationships between factors and identify high-risk subgroups to whom intervention measures can be targeted. This study used decision trees to examine associations of various risk factors with youth anxiety, depression, and flourishing. Data were collected from 74,501 students across Canadian high schools participating in the 2018-2019 COMPASS Study. Students completed a questionnaire including validated mental health scales and 23 covariates. Decision trees were grown to identify key factors and subgroups for anxiety, depression, and flourishing outcomes. Females lacking both happy home life and sense of connection to school were at greatest risk for higher anxiety and depression levels. In contrast with previous literature, behavioural factors such as diet, movement and substance use did not emerge as differentiators. This study highlights the influence of home and school environments on youth mental health using a novel decision tree analysis. While having a happy home life is most important in protecting against youth anxiety and depression, a sense of connection to school may mitigate the negative influence of a poor home environment.

A Comprehensive School Health Approach to Student Physical Activity: A Multilevel Analysis Examining the Association between School-Level Factors and Student Physical Activity Behaviors

BACKGROUND

The comprehensive school health (CSH) framework has four components: social and physical environment; partnerships and services; teaching and learning; and policy. This study examines associations between CSH and student physical activity (PA).

METHODS

Using 2015/2016 COMPASS study survey data of 37,397 students (grades 9-12) from 80 secondary schools in Ontario and Alberta, Canada, associations between school-level factors within CSH and student PA outcomes (weekly moderate-to-vigorous PA [MVPA] minutes and achieving the national PA recommendations of ≥60 min of MVPA daily, vigorous PA ≥3 days/week, strengthening activities ≥3 days/week) were analyzed using multilevel regression models stratified by gender and grade.

RESULTS

Factors within all four CSH components were associated with student PA. Four student subgroups were more likely to achieve the recommendations if their school had youth organization partnerships (Range of AORs:1.15-1.33, p <.05) and female students were less likely if their school had low prioritization of PA (AOR = 0.77, 95% CI: [0.65-0.92]). Grade 9 students had higher MVPA when provided non-competitive PA opportunities (β = 100.4, 95%CI: [30.0-170.9]). All student subgroups had better PA outcomes when schools provided access to equipment during non-instructional time.

CONCLUSION

There is opportunity to improve student PA through CSH-guided interventions, but different strategies may be more effective for each gender/grade.

NextGen Public Health Surveillance and the Internet of Things (IoT)

Recent advances in technology have led to the rise of new-age data sources (e.g., Internet of Things (IoT), wearables, social media, and mobile health). IoT is becoming ubiquitous, and data generation is accelerating globally. Other health research domains have used IoT as a data source, but its potential has not been thoroughly explored and utilized systematically in public health surveillance. This article summarizes the existing literature on the use of IoT as a data source for surveillance. It presents the shortcomings of current data sources and how NextGen data sources, including the large-scale applications of IoT, can meet the needs of surveillance. The opportunities and challenges of using these modern data sources in public health surveillance are also explored. These IoT data ecosystems are being generated with minimal effort by the device users and benefit from high granularity, objectivity, and validity. Advances in computing are now bringing IoT-based surveillance into the realm of possibility. The potential advantages of IoT data include high-frequency, high volume, zero effort data collection methods, with a potential to have syndromic surveillance. In contrast, the critical challenges to mainstream this data source within surveillance systems are the huge volume and variety of data, fusing data from multiple devices to produce a unified result, and the lack of multidisciplinary professionals to understand the domain and analyze the domain data accordingly.

Low-frequency exercise training improves cardiovascular fitness and strength during treatment for breast cancer: a single-arm intervention study

Aerobic and resistance exercise during and after cancer treatment are important for health-related outcomes, however treatment-specific barriers may inhibit adherence. We explored the effect of lower-frequency exercise training on fitness, body composition, and metabolic markers (i.e. glucose and lipids) in a group of recently diagnosed breast cancer patients. Fifty-two females ≥ 18 years with stage I-IIIB breast cancer were instructed to attend 2 cardiovascular and strength training sessions/week over 12 weeks, but program length was expanded as needed to accommodate missed sessions. Pre- and post-intervention, we measured: (1) cardiovascular fitness, (2) isometric strength, (3) body composition (dual-energy X-ray absorptiometry), and (4) fasting glucose, insulin, c-peptide, and lipids. Pre-intervention, participants were 53 ± 10 years old (mean ± SD) and overweight (BMI: 27.5 ± 5.4 kg m^-2, 40.1 ± 6.5% body fat). Forty participants completed the program over a median 20 weeks (range: 13-32 weeks, median frequency: 1.2 sessions/week), over which predicted VO₂peak improved by 7% (2.2[0.1-4.4] mL/kg/min) (delta[95% CI]), and strength increased by 7-9% (right arm: 2.3[0.1-4.5] N m; right leg: 7.9[2.1-13.7] N m; left leg: 7.8[1.9-13.7] N m). Body composition and metabolic markers were unchanged. An exercise frequency of 1.2 sessions/week stimulated significant improvements in fitness, and may represent a practical target for patients during active treatment.

Identifying subpopulations of septic patients: A temporal data-driven approach

Sepsis is one of the deadliest diseases in North America and in spite of the vast amount of research on this topic there is still uncertainty in the outcome of sepsis treatments. This study aimed at investigating the informativeness of temporal electronic health records (EHR) in stratifying septic patients and identifying subpopulations of septic patients with similar trajectories and clinical needs. We performed hierarchical clustering and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) analyses using data from septic patients in the MIMIC III intensive care unit database. The t-Distributed Stochastic Neighbor Embedding (t-SNE) method was utilized to map patients to a two-dimensional space. We utilized silhouette index and cluster-wise stability assessment by resampling to investigate the validity of the clusters. The hierarchical clustering with Euclidean metric identified twelve clinically recognizable subgroups that demonstrated different characteristics in spite of sharing common conditions. Our results demonstrated that data-driven approaches can help in customizing care platforms for septic patients by identifying similar clinically relevant groups.

Multilingualism and Dementia Risk: Longitudinal Analysis of the Nun Study

BACKGROUND

Multilingualism is associated with enhanced executive function and may thus prevent cognitive decline and reduce the risk of dementia.

OBJECTIVE

To determine whether multilingualism is associated with delayed onset or reduced risk of dementia.

METHODS

Dementia was diagnosed in the Nun Study, a longitudinal study of religious sisters aged 75+ years. Multilingualism was self-reported. Dementia likelihood was determined in 325 participants using discrete-time survival analysis; sensitivity analyses (n = 106) incorporated additional linguistic measures (idea density and grammatical complexity).

RESULTS

Multilingualism did not delay the onset of dementia. However, participants speaking four or more languages (but not two or three) were significantly less likely to develop dementia than monolinguals (OR = 0.13; 95% CI = 0.01, 0.65, adjusted for age, apolipoprotein E, and transition period). This significant protective effect of speaking four or more languages weakened (OR = 0.53; 95% CI = 0.06, 4.91) in the presence of idea density in models adjusted for education and apolipoprotein E.

CONCLUSION

Linguistic ability broadly was a significant predictor of dementia, although it was written linguistic ability (specifically idea density) rather than multilingualism that was the strongest predictor. The impact of language on dementia may extend beyond number of languages spoken to encompass other indicators of linguistic ability. Further research to identify the characteristics of multilingualism most salient for risk of dementia could clarify the value, target audience, and design of interventions to promote multilingualism and other linguistic training as a strategy to reduce the risk of dementia and its individual and societal impacts.

More-2-Eat implementation demonstrates that screening, assessment and treatment of malnourished patients can be spread and sustained in acute care; a multi-site, pretest post-test time series study

BACKGROUND

Malnutrition in medical and surgical inpatients is an on-going problem. More-2-Eat (M2E) Phase 1 demonstrated that improved detection and treatment of hospital malnutrition could be embedded into routine practice using an intensive researcher-facilitated implementation process. Yet, spreading and sustaining new practices in diverse hospital cultures with minimal researcher support is unknown.

AIMS

To demonstrate that a scalable model of implementation can increase three key nutrition practices (admission screening; Subjective Global Assessment (SGA); and medication pass (MedPass) of oral nutritional supplement) in diverse acute care hospitals to detect and treat malnutrition in medical and surgical patients.

METHODS

Ten hospitals participated in this pretest post-test time series implementation study from across Canada, including 21 medical or surgical units (Phase 1 original units (n = 4), Phase 1 hospital new units (n = 9), Phase 2 new hospitals and units (n = 8)). The scalable implementation model included: training champions on implementation strategies and providing them with education resources for teams; creating a self-directed audit and feedback process; and providing mentorship. Standardized audits of all patients on the study unit on an audit day were completed bi-monthly to track nutrition care activities since admission. Bivariate comparisons were performed by time period (initial, mid-term and final audits). Run-charts depicted the trajectory of change and qualitatively compared to Phase 1.

RESULTS

5158 patient charts were audited over the course of 18-months. Admission nutrition screening rates increased from 50% to 84% (p < 0.0001). New Phase 1 units more readily implemented screening than Phase 2 sites, and the original Phase 1 units generally sustained screening practices from Phase 1. SGA was a sustained practice at Phase 1 hospitals including in new Phase 1 units. The new Phase 2 units improved completion of SGA but did not reach the levels of Phase 1 units (original or new). MedPass almost doubled over the time periods (7%-13% of all patients p < 0.007). Other care practices significantly increased (e.g. volunteer mealtime assistance).

CONCLUSION

Nutrition-care activities significantly increased in diverse hospital units with this scalable model. This heralds the transition from implementation research to sustained changes in routine practice. Screening, SGA, and MedPass can all be implemented, improve nutrition care for all patients, spread within an organization, and for the most part, sustained (and in the case of original Phase 1 units, for over 3 years) with champion leadership.

The impact of an alcohol policy change on developmental trajectories of youth alcohol use: examination of a natural experiment in Canada

OBJECTIVES

In 2015, the Liquor Control Board of Ontario (LCBO) authorized sale of alcohol in some Ontario grocery stores. This research evaluates the impact of the new policy on alcohol use patterns of youth in a quasi-experimental setting with two control groups.

METHODS

The sample consists of 2267 grade 9 students attending 60 secondary schools across Ontario (n = 56) and Alberta (n = 4), who provided 4-year linked longitudinal data (2013-2014 to 2016-2017) in the COMPASS study. The study used the frequency of drinking and the frequency of binge drinking to characterize alcohol use behaviours.

RESULTS

Latent transition analysis found four statuses of alcohol use: abstainer, periodic drinker, low-risk drinker, and high-risk regular drinker. The new policy had no negative impact among periodic and low-risk drinkers, but the risk of transitioning from the abstainer (lowest risk status) to high-risk regular drinker (highest risk status) among the exposed cohort was 1.71 times greater post-policy than pre-policy change, compared with those of Ontario-unexposed (0.50) and Alberta-unexposed cohorts (1.00). The probability of sustaining high-risk drinking among the exposed cohort increased by a factor of 1.76, compared with 1.13-fold and 0.89-fold among the Ontario-unexposed and Alberta-unexposed cohorts, respectively.

CONCLUSION

Youth are more likely to transition from abstinence to high-risk regular drinking, and high-risk regular drinkers are more likely to maintain their behaviours in the jurisdictions exposed to the latest change in LCBO policy authorizing grocery stores to sell alcohol. When formulating policy interventions, youth access to alcohol should be considered in order to reduce their harmful alcohol consumption.

Mid-level constraint may correct coronal plane imbalance without compromising patient function in patients with severe osteoarthritis

Introduction

Mid-level constraint (MLC) in Total Knee Arthroplasty (TKA) offer surgeons the opportunity to obtain a well-balanced varus-valgus balance in the setting of slight ligament imbalance. As such, we sought to evaluate patient reported outcome measures (PROMs), alignment correction, and rate of revision between the MLC group and a cohort of posterior stabilized (PS) knees in a cohort of patients with preoperative degenerative arthritis.

Materials/methods

We performed a retrospective review of 57 MLC knees and 96 PS knees from a single manufacturer that were implanted by a single surgeon. We found the average age (68.91 vs. 68.40, p = 0.72), average BMI (30.88 vs. 29.14, p = 0.10), and gender breakdown (25:32 vs. 28:69, p = 0.08) to be comparable between the two cohorts. The latest follow-up was 4.0 years in the MLC group and 3.8 years in the PS group, p = 0.26.

Results

The two cohorts inherently resulted in significantly different preoperative deformities (MCL knees: average varus deformity 13.75°, average valgus deformity 12.37°; PS knees average varus deformity 15.14°, average valgus deformity 10.8°). There were more valgus knees in the MLC group (36 vs. 22 (p < 0.001), respectively), but the postoperative alignment was the same. MLC cohort: preoperative varus group had 4.74° of valgus postoperatively, preoperative valgus group had 5.43° of valgus postoperatively. PS cohort: preoperative varus group had 5.40° of valgus postoperatively, preoperative valgus group had 4.80° of valgus postoperatively. We found a significant difference in Knee Injury and Osteoarthritis Outcome Score (KSCRS-Total) between the two groups (MLC 163.9 vs. PS 132.8, p = 0.003). There was no significant difference in terms of Range of Motion (ROM) (MLC 121° vs. PS 122°, p = 0.58), anterior knee pain (MLC 1.75 vs. PS 1.81, p = 0.39), or Pain VAS (MLC 25.1 vs. PS 28.6, p = 0.46). There was similar rate of revision between the cohorts (3.5% MLC vs. 2.10% PS, p = 0.13). There was no significant difference in manipulation rate (8.78% MLC vs. 9.40% PS, p = 0.38).

Conclusion/discussion

This study demonstrated that the use of MLC in TKA allows surgeons to correct preoperative deformities with equal or improved functional outcomes compared to PS knees. In general, we recommend that surgeons try to balance the knee and use the least amount of constraint possible but should consider MLC when needed and use such implants if they are unable to balance the varus-valgus gap.

Identifying patterns of alcohol use among secondary school students in Canada: A multilevel latent class analysis

INTRODUCTION

Harm from alcohol use depend not only on the volume of consumption but also on drinking patterns. This study identifies patterns of alcohol consumption in youth and investigates how these patterns vary across schools and whether individual- and school-level factors are associated with engagement in patterns of alcohol consumption.

METHODS

The sample consists of 45,298 grade 9 to 12 students attending 89 secondary schools across Ontario and Alberta (Canada), who participated in the COMPASS study during the school year 2013-14. The frequency of drinking, the frequency of binge drinking, and age of alcohol-use initiation were used to characterize alcohol use patterns.

RESULTS

The multilevel latent class analysis identified 4 student-level latent groups and 2 school-level latent groups. Student-level groups of youth were characterized as non-drinkers (44.2%), light drinkers (41.8%), regular drinkers (11.1%), and heavy drinkers (2.9%). Two groups of schools were characterized as either low-use (44.9%) or high-use (55.1%) schools, with significantly different probability of membership in each student-level group. Male students (OR 1.30) and upper grades (OR 1.93) were significantly associated with membership in higher use groups of individuals. The median household income and the number of off-premise alcohol outlets had no significant association with patterns of alcohol consumption within schools.

CONCLUSIONS

A large proportion of students reported a level of drinking, suggesting that, in addition to delaying the onset of alcohol use, interventions need to encourage drinker students to quit drinking or lower their consumption. Schools may need to select and/or alter external interventions according to the dominant patterns of alcohol use among their students.

Anatomic dual mobility compared to modular dual mobility in primary total hip arthroplasty: a matched cohort study

Background

Dual mobility (DM) has been used in primary total hip arthroplasty recently for their low dislocation rates, low revision rates, and improved patient functional outcomes. We compared 2 DM systems, anatomic dual mobility (ADM; Stryker, Mahwah, NJ) and modular dual mobility (MDM; Stryker, Mahwah, NJ), to determine differences in dislocation rates, revision rates, and patient outcome scores.

Methods

The study was a single-center matched retrospective review of prospectively collected data of patients who underwent primary total hip arthroplasty surgery with an ADM or MDM system by a single surgeon from 2012 to 2017. Demographics, operative details, postoperative patient-reported outcomes, and clinical outcomes were recorded. A Kaplan-Meier survivorship curve to compare survival time between groups was collected as well.

Results

Five hundred seventy-four patients were included in the study with 287 patients matched in each group with mean 2.86 years of follow-up. The dislocation rate in each cohort was 0%, the acetabular-specific revision rate was 0%, and in each cohort, overall revision rate in each cohort was 1.7%. In general, patient-reported outcomes were similar for each group (Harris Hip Score Pain (P = .919), Harris Hip Score Function (P = .736), Western Ontario and McMaster Universities Osteoarthritis Index (P = .139), Pain Visual Analog Scale (P = .146), Veterans RAND 12-Item Health Survey (P = .99), University of California, Los Angeles (P = .417), and Harris Hip Score Total (P = .136). There was a slight clinically insignificant increase in hip flexion between the cohorts favoring the ADM group (98.6 ± 9.8 vs 94.0 ± 9.7, P < .001).

Conclusions

Both DM systems had similar patient-reported outcomes that were quite favorable. At 2.86 years of follow-up, neither the ADM nor MDM systems demonstrated dislocation, and both had low acetabular-specific and overall revision rates in this matched cohort study.

Handgrip strength predicts length of stay and quality of life in and out of hospital

BACKGROUND

Handgrip strength (HGS) is a practical measure of strength and physical function that can be used to identify frailty among hospitalized patients, but its utility in this setting is unclear. To be considered useful, any functional measure needs to provide pertinent information on the patient and predict relevant outcomes such as health-care utilization (e.g., length of stay (LOS)) and patient-reported quality of life (QOL). The purpose of this study was to determine if HGS predicted LOS and QOL. A second aim was to examine the best sensitivity (SE) and specificity (SP) for predicting length of stay (>7 or >13 days) using previously published cut-points for HGS.

METHODOLOGY

HGS was measured on 1136 medical patients shortly after admission with a Lafayette dynamometer. QOL was assessed with the self-reported SF-12 completed with an interviewer during hospitalization and 30- days after discharge via telephone. Physical (PCS) and mental (MCS) component scores of SF-12 were calculated. A variety of covariates were assessed (e.g., nutritional status). Multivariate analyses stratified by sex were completed.

RESULTS

The mean LOS was 12.71 days (median = 8.00; SD = 13.20), 12.88 days (SD = 13.82) for males, and 12.58 days (SD = 12.68) for females. Lower admission HGS scores were associated with longer LOS (male X² = 7.85, p < 0.05; female X² = 14.9, p < 0.0001). The average quality of life scores were as follows: in hospital PCS: 34.66, MCS: 46.49; post discharge PCS: 36.17; MCS: 51.22. HGS predicted PCS during hospitalization (male X² = 36.22, p < 0.0001; female X² = 19.87, p < 0.0001) and post hospitalization (male X² = 6.98, p < 0.01; female X² = 10.99, p < 0.01). Various reference cut-points for HGS were tested against LOS, with none being considered appropriate (e.g., SE and SP both < 70) when adjusting for age and sex.

CONCLUSION

Admission HGS adds predictive value for both LOS and physical components of QOL and is worth pursuing in practice to identify potential frailty and the need for proactive steps to mitigate further functional decline during hospitalization. However, HGS cut-points for LOS specific to acute care patients need to be defined and tested.

Noncomprehensive and Intermittent Obesity-Related School Programs and Policies May Not Work: Evidence from the COMPASS Study

BACKGROUND

The school environment has the potential to influence student body mass index (BMI) through programs and policies. Our objective was to examine the effect of modifying obesity-related school policies and programs on youths' BMI trajectories.

METHODS

Obesity-related school policies and programs related to physical activity and healthy eating were collected from 41 schools across Ontario at baseline (2012-2013) and year 2 (2013-2014) of the COMPASS study. Self-reported height and weight were collected from the 4951 grades 9 and 10 students who attended those schools for 3 years. Linear mixed effects regression models examined the effect of modifying obesity-related school policies and programs on youths' BMI trajectories.

RESULTS

Between Y₁ and Y₂ , 26 of the 41 schools implemented distinct new obesity-related programs or policies related to physical activity or healthy eating. Five of the interventions were associated with BMI trajectories of students attending those schools compared to students attending a pooled sample of control schools, predicting a higher BMI trajectory.

CONCLUSIONS

Isolated programs and policies may not successfully improve youths' BMI trajectories. Further research is required to test the implementation of policies and programs that follow a comprehensive school health (CSH) approach, targeting BMI and associated behaviors.

Identifying trajectories of alcohol use in a sample of secondary school students in Ontario and Alberta: longitudinal evidence from the COMPASS study

INTRODUCTION

Despite evidence indicating a rapid progression in use of alcohol during adolescence, little is known about the ways patterns of drinking develop over time. This study investigated patterns of alcohol use within a cohort of youth in Ontario and Alberta and the probability of changes between patterns.

METHODS

The sample consists of two-year linked longitudinal data (school year 2013/14 to 2014/15) from 19 492 students in Grades 9 to 12 in 89 secondary schools across Ontario and Alberta, Canada, who participated in the COMPASS study. The latent class analysis used two self-reported items about the frequency of drinking (measured as none, monthly, weekly, or daily use) and the frequency of binge drinking (measured as none, less than or once a month, 2-4 times a month, or more than once week) to characterize patterns of alcohol use. The effects of gender, ethnicity and cannabis and cigarette use on alcohol use patterns were examined.

RESULTS

The study identified four drinking patterns: non-drinker, periodic drinker (reported monthly drinking and no binge drinking), low-risk drinker (reported monthly drinking and limited binge drinking) and high-risk regular drinker (reported drinking 1-3 times a week and binge drinking 2-4 times a month). Non-drinker was the most prevalent pattern at baseline (55.1%) and follow-up (39.1%). Periodic drinkers had the highest likelihood of an increase in alcohol consumption, with 40% moving to the low-risk pattern. A notable proportion of participants returned to a lower severity pattern or transitioning out of drinking.

CONCLUSION

There are four distinct youth alcohol-use patterns. The high probability of transitioning to drinking during the secondary school years suggests the need for preventive interventions in earlier stages of use, before drinking becomes habitual.

Fall Risk Classification in Community-Dwelling Older Adults Using a Smart Wrist-Worn Device and the Resident Assessment Instrument-Home Care: Prospective Observational Study

Background

 Little is known about whether off-the-shelf wearable sensor data can contribute to fall risk classification or complement clinical assessment tools such as the Resident Assessment Instrument-Home Care (RAI-HC).

Objective

 This study aimed to (1) investigate the similarities and differences in physical activity (PA), heart rate, and night sleep in a sample of community-dwelling older adults with varying fall histories using a smart wrist-worn device and (2) create and evaluate fall risk classification models based on (i) wearable data, (ii) the RAI-HC, and (iii) the combination of wearable and RAI-HC data.

Methods

 A prospective, observational study was conducted among 3 faller groups (G₀, G₁, G₂₊) based on the number of previous falls (0, 1, ≥2 falls) in a sample of older community-dwelling adults. Each participant was requested to wear a smart wristband for 7 consecutive days while carrying out day-to-day activities in their normal lives. The wearable and RAI-HC assessment data were analyzed and utilized to create fall risk classification models, with 3 supervised machine learning algorithms: logistic regression, decision tree, and random forest (RF).

Results

 Of 40 participants aged 65 to 93 years, 16 (40%) had no previous falls, whereas 8 (20%) and 16 (40%) had experienced 1 and multiple (≥2) falls, respectively. Level of PA as measured by average daily steps was significantly different between groups (P=.04). In the 3 faller group classification, RF achieved the best accuracy of 83.8% using both wearable and RAI-HC data, which is 13.5% higher than that of using the RAI-HC data only and 18.9% higher than that of using wearable data exclusively. In discriminating between {G₀+G₁} and G₂₊, RF achieved the best area under the receiver operating characteristic curve of 0.894 (overall accuracy of 89.2%) based on wearable and RAI-HC data. Discrimination between G₀ and {G₁+G₂₊} did not result in better classification performance than that between {G₀+G₁} and G₂₊.

Conclusions

 Both wearable data and the RAI-HC assessment can contribute to fall risk classification. All the classification models revealed that RAI-HC outperforms wearable data, and the best performance was achieved with the combination of 2 datasets. Future studies in fall risk assessment should consider using wearable technologies to supplement resident assessment instruments.

A New Insight Into Missing Data in Intensive Care Unit Patient Profiles: Observational Study

Background

The data missing from patient profiles in intensive care units (ICUs) are substantial and unavoidable. However, this incompleteness is not always random or because of imperfections in the data collection process.

Objective

This study aimed to investigate the potential hidden information in data missing from electronic health records (EHRs) in an ICU and examine whether the presence or missingness of a variable itself can convey information about the patient health status.

Methods

Daily retrieval of laboratory test (LT) measurements from the Medical Information Mart for Intensive Care III database was set as our reference for defining complete patient profiles. Missingness indicators were introduced as a way of representing presence or absence of the LTs in a patient profile. Thereafter, various feature selection methods (filter and embedded feature selection methods) were used to examine the predictive power of missingness indicators. Finally, a set of well-known prediction models (logistic regression [LR], decision tree, and random forest) were used to evaluate whether the absence status itself of a variable recording can provide predictive power. We also examined the utility of missingness indicators in improving predictive performance when used with observed laboratory measurements as model input. The outcome of interest was in-hospital mortality and mortality at 30 days after ICU discharge.

Results

Regardless of mortality type or ICU day, more than 40% of the predictors selected by feature selection methods were missingness indicators. Notably, employing missingness indicators as the only predictors achieved reasonable mortality prediction on all days and for all mortality types (for instance, in 30-day mortality prediction with LR, we achieved area under the curve of the receiver operating characteristic [AUROC] of 0.6836±0.012). Including indicators with observed measurements in the prediction models also improved the AUROC; the maximum improvement was 0.0426. Indicators also improved the AUROC for Simplified Acute Physiology Score II model—a well-known ICU severity of illness score—confirming the additive information of the indicators (AUROC of 0.8045±0.0109 for 30-day mortality prediction for LR).

Conclusions

Our study demonstrated that the presence or absence of LT measurements is informative and can be considered a potential predictor of in-hospital and 30-day mortality. The comparative analysis of prediction models also showed statistically significant prediction improvement when indicators were included. Moreover, missing data might reflect the opinions of examining clinicians. Therefore, the absence of measurements can be informative in ICUs and has predictive power beyond the measured data themselves. This initial case study shows promise for more in-depth analysis of missing data and its informativeness in ICUs. Future studies are needed to generalize these results.

Improving the standard of nutrition care in hospital: Mealtime barriers reduced with implementation of the Integrated Nutrition Pathway for Acute Care

BACKGROUND

Poor food intake is common in hospital patients and is associated with adverse patient and healthcare outcomes; diverse mealtime barriers to intake often undermine clinical nutrition care.

AIM

This study determines whether implementation of locally adaptable nutrition care activities as part of uptake of the Integrated Nutrition Pathway for Acute Care (INPAC) reduced mealtime barriers and improved other patient outcomes (e.g. length of stay; LOS) when considering other covariates.

METHODS

1250 medical patients from 5 Canadian hospitals were recruited for this before-after time series design. Mealtime barriers were tallied with the Mealtime Audit Tool after a meal, while proportion of the meal consumed was assessed with the My Meal Intake Tool. Implementation of new standard care activities occurred over 12 months and three periods (pre-, early, and late) of implementation were compared. Regression analyses determined the effect of time period while adjusting for key covariates.

RESULTS

Mealtime barriers were reduced over time periods (Period 1 = 2.5 S.D. 2.1; Period 3 = 1.8 S.D. 1.7) and site differences were noted. This decrease was statistically significant in regression analyses (-0.28 per time period; 95% CI -0.44, -0.11). Within and across site changes were also observed over time in meal intake and LOS; however, after adjusting for covariates, time period of implementation was not significantly associated with these outcomes.

DISCUSSION

Mealtime barriers can be reduced and sustained by implementing improved standard care procedures for patients. The More-2-Eat study provides an example of how to implement changes in practice to support the prevention and treatment of malnutrition.

TRIAL REGISTRATION

Retrospectively registered ClinicalTrials.gov Identifier: NCT02800304, June 7, 2016.

Handgrip Strength, but Not 5-Meter Walk, Adds Value to a Clinical Nutrition Assessment

BACKGROUND

Decreased physical functioning is associated with malnutrition and common in acute care patients; determining loss of function is often considered part of a comprehensive nutrition assessment. Handgrip strength (HGS) and 5-meter timed walk (5m) are functional measures used in a variety of settings. This analysis sought to determine which functional measure could be added to a hospital nutrition assessment, based on its feasibility and capacity to discriminate patient subgroups.

METHODS

Eligible medical patients (no delirium/dementia, admitted from community; n = 1250), recruited from 5 hospitals that participated in a previous multisite action research study, provided data on demographics, HGS, 5m, nutrition status, perceived disability, and other characteristics.

RESULTS

Significantly more patients (z = 17.39, P < .00001) were able to complete HGS than 5m (92% versus 43%, respectively). Median HGS was 28.0 kg for men and 14.7 kg for women. Of patients who completed the 5m, mean completion time was 8.98 seconds (median, 6.79 seconds, SD = 6.59). 5m and HGS scores were significantly worse with patient-perceived disability (z = -9.56, t = 10.69, respectively; P < .0001; 95% confidence interval [CI], [7.33, 10.63]; [1.76, 3.18]). HGS was associated with nutrition status (t = 4.13, P < .001; 95% CI [2.02, 5.67]), although it showed poor validity as a single nutrition indicator.

CONCLUSIONS

These data indicate that HGS is a more useful functional measure than 5m when added to a hospital nutrition assessment. Determination of HGS cutpoints to identify low strength in acute care patients will promote its use.

Towards Consensus on Essential Components of Physical Examination in Primary Care-based Memory Clinics

Background

Primary care-based memory clinics were established to meet the needs of persons with memory concerns. We aimed to identify: 1) physical examination maneuvers required to assess persons with possible dementia in specialist-supported primary care-based memory clinics, and 2) the best-suited clinicians to perform these maneuvers in this setting.

Methods

We distributed in-person and online surveys of clinicians in a network of 67 primary care-based memory clinics in Ontario, Canada.

Results

90 surveys were completed for an overall response rate of 66.7%. Assessments of vital signs, gait, and for features of Parkinsonism were identified as essential by most respondents. There was little consensus on which clinician should be responsible for specific physical examination maneuvers.

Conclusions

While we identified specific physical examination maneuvers deemed by providers to be both necessary and feasible to perform in the context of primary care-based memory clinics, further research is needed to clarify interprofessional roles related to the examination.

Observation of High School Students' Food Handling Behaviors: Do They Improve following a Food Safety Education Intervention?

Youth are a key audience for food safety education. They often engage in risky food handling behaviors, prepare food for others, and have limited experience and knowledge of safe food handling practices. Our goal was to investigate the effectiveness of an existing food handler training program for improving safe food handling behaviors among high school students in Ontario, Canada. However, because no schools agreed to provide control groups, we evaluated whether behaviors changed following delivery of the intervention program and whether changes were sustained over the school term. We measured 32 food safety behaviors, before the intervention and at 2-week and 3-month follow-up evaluations by in-person observations of students ( n = 119) enrolled in grade 10 and 12 Food and Nutrition classes ( n = 8) and who individually prepared recipes. We examined within-student changes in behaviors across the three time points, using mixed effects regression models to model trends in the total food handling score (of a possible 32 behaviors) and subscores for "clean" (17 behaviors), "separate" (14 behaviors), and "cook" (1 behavior), adjusting for student characteristics. At baseline, students ( n = 108) averaged 49.1% (15.7 of 32 behaviors; standard deviation = 5.8) correct food handling behaviors, and only 5.5% (6) of the 108 students used a food thermometer to check the doneness of the chicken (the "cook" behavior). All four behavior score types increased significantly ∼2 weeks postintervention and remained unchanged ∼3 months later. Student characteristics (e.g., having taken a prior food handling course) were not significant predictors of the total number of correctly performed food handling behaviors or of the "clean" or "separate" behaviors, working or volunteering in a food service establishment was the only characteristic significantly associated with food thermometer use (i.e., "cook"). Despite the significant increase in correct behaviors, students continued to use risky practices postintervention, suggesting that the risk of foodborne disease remained.

Multi-site implementation of nutrition screening and diagnosis in medical care units: Success of the More-2-Eat project

BACKGROUND

Improving the detection and treatment of malnourished patients in hospital is needed to promote recovery.

AIM

To describe the change in rates of detection and triaging of care for malnourished patients in 5 hospitals that were implementing an evidence-based nutrition care algorithm. To demonstrate that following this algorithm leads to increased detection of malnutrition and increased treatment to mitigate this condition.

METHODS

Sites worked towards implementing the Integrated Nutrition Pathway for Acute Care (INPAC), including screening (Canadian Nutrition Screening Tool) and triage (Subjective Global Assessment; SGA) to detect and diagnose malnourished patients. Implementation occurred over a 24-month period, including developmental (Period 1), implementation (Periods 2-5), and sustainability (Period 6) phases. Audits (n = 36) of patient health records (n = 5030) were conducted to identify nutrition care practices implemented with a variety of strategies and behaviour change techniques.

RESULTS

All sites increased nutrition screening from Period 1, with three achieving the goal of 75% of admitted patients being screened by Period 3, and the remainder achieving a rate of 70% by end of implementation. No sites were conducting SGA at Period 1, and sites reached the goal of a 75% completion rate or referral for those identified to be at nutrition risk, by Period 3 or 4. By Period 2, 100% of patients identified as SGA C (severely malnourished) were receiving a comprehensive nutritional assessment. In Period 1, the nutrition diagnosis and documentation by the dietitian of 'malnutrition' was a modest 0.37%, increasing to over 5% of all audited health records. The overall use of any Advanced Nutrition Care practices increased from 31% during Period 1 to 63% during Period 6.

CONCLUSION

The success of this multi-site study demonstrated that implementation of nutrition screening and diagnosis is feasible and leads to appropriate care. INPAC promotes efficiency in nutrition care while minimizing the risk of missing malnourished patients.

TRIAL REGISTRATION

Retrospectively registered ClinicalTrials.gov Identifier: NCT02800304, June 7, 2016.

Development of a bedside viable ultrasound protocol to quantify appendicular lean tissue mass

BACKGROUND

Ultrasound is a non-invasive and readily available tool that can be prospectively applied at the bedside to assess muscle mass in clinical settings. The four-site protocol, which images two anatomical sites on each quadriceps, may be a viable bedside method, but its ability to predict musculature has not been compared against whole-body reference methods. Our primary objectives were to (i) compare the four-site protocol's ability to predict appendicular lean tissue mass from dual-energy X-ray absorptiometry; (ii) optimize the predictability of the four-site protocol with additional anatomical muscle thicknesses and easily obtained covariates; and (iii) assess the ability of the optimized protocol to identify individuals with low lean tissue mass.

METHODS

This observational cross-sectional study recruited 96 university and community dwelling adults. Participants underwent ultrasound scans for assessment of muscle thickness and whole-body dual-energy X-ray absorptiometry scans for assessment of appendicular lean tissue. Ultrasound protocols included (i) the nine-site protocol, which images nine anterior and posterior muscle groups in supine and prone positions, and (ii) the four-site protocol, which images two anterior sites on each quadriceps muscle group in a supine position.

RESULTS

The four-site protocol was strongly associated (R2  = 0.72) with appendicular lean tissue mass, but Bland-Altman analysis displayed wide limits of agreement (-5.67, 5.67 kg). Incorporating the anterior upper arm muscle thickness, and covariates age and sex, alongside the four-site protocol, improved the association (R2  = 0.91) with appendicular lean tissue and displayed narrower limits of agreement (-3.18, 3.18 kg). The optimized protocol demonstrated a strong ability to identify low lean tissue mass (area under the curve = 0.89).

CONCLUSIONS

The four-site protocol can be improved with the addition of the anterior upper arm muscle thickness, sex, and age when predicting appendicular lean tissue mass. This optimized protocol can accurately identify low lean tissue mass, while still being easily applied at the bedside.

Patient Similarity in Prediction Models Based on Health Data: A Scoping Review

Background

Physicians and health policy makers are required to make predictions during their decision making in various medical problems. Many advances have been made in predictive modeling toward outcome prediction, but these innovations target an average patient and are insufficiently adjustable for individual patients. One developing idea in this field is individualized predictive analytics based on patient similarity. The goal of this approach is to identify patients who are similar to an index patient and derive insights from the records of similar patients to provide personalized predictions..

Objective

The aim is to summarize and review published studies describing computer-based approaches for predicting patients’ future health status based on health data and patient similarity, identify gaps, and provide a starting point for related future research.

Methods

The method involved (1) conducting the review by performing automated searches in Scopus, PubMed, and ISI Web of Science, selecting relevant studies by first screening titles and abstracts then analyzing full-texts, and (2) documenting by extracting publication details and information on context, predictors, missing data, modeling algorithm, outcome, and evaluation methods into a matrix table, synthesizing data, and reporting results.

Results

After duplicate removal, 1339 articles were screened in abstracts and titles and 67 were selected for full-text review. In total, 22 articles met the inclusion criteria. Within included articles, hospitals were the main source of data (n=10). Cardiovascular disease (n=7) and diabetes (n=4) were the dominant patient diseases. Most studies (n=18) used neighborhood-based approaches in devising prediction models. Two studies showed that patient similarity-based modeling outperformed population-based predictive methods.

Conclusions

Interest in patient similarity-based predictive modeling for diagnosis and prognosis has been growing. In addition to raw/coded health data, wavelet transform and term frequency-inverse document frequency methods were employed to extract predictors. Selecting predictors with potential to highlight special cases and defining new patient similarity metrics were among the gaps identified in the existing literature that provide starting points for future work. Patient status prediction models based on patient similarity and health data offer exciting potential for personalizing and ultimately improving health care, leading to better patient outcomes.

Elevated C-Peptides, Abdominal Obesity, and Abnormal Adipokine Profile are Associated With Higher Gleason Scores in Prostate Cancer

BACKGROUND

Prostate cancer development is associated with numerous lifestyle factors (i.e., physical activity, nutrition intake) and metabolic perturbations. These factors have been studied independently; here, we used an integrative approach to characterize these lifestyle and metabolic parameters in men undergoing diagnostic prostate biopsies.

METHODS

We prospectively evaluated 51 consecutive men for body composition, metabolic factors including glucose- and lipid-related measures, as well as lifestyle factors prior to prostate biopsy. Evaluations were performed in a blinded manner and were subsequently related to biopsy outcomes for: (i) presence or absence of cancer; and (ii) where cancer was present, Gleason score.

RESULTS

Serum C-peptide concentrations were significantly greater in participants with Gleason scores ≥4 + 3 (2.8 ± 1.1 ng/ml) compared to those with Gleason 3 + 3 (1.4 ± 0.6 ng/ml) or Gleason 3 + 4 (1.3 ± 0.8 ng/ml, P = 0.002), suggesting greater insulin secretion despite lack of differences in fasting glucose concentrations. Central adiposity, measured by waist circumference, was significantly greater in participants with Gleason ≥4 + 3 (110.1 ± 7.4 cm) compared to those with Gleason 3 + 4 (102.0 ± 9.5 cm, P = 0.028). Men with Gleason ≥4 + 3 also had significantly greater leptin concentrations than those with lower Gleason scores (Gleason ≥4 + 3: 15.6 ± 3.3 ng/ml vs. Gleason 3 + 4: 8.1 ± 8.1 ng/ml, P < 0.05) and leptin:adiponectin ratio (Gleason ≥4 + 3: 9.7 ± 6.1 AU, Gleason 3 + 4: 2.9 ± 3.2, Gleason 3 + 3: 2.4 ± 2.1 AU, P = 0.013).

CONCLUSIONS

We profiled a cluster of obesity-related metabolic perturbations (C-peptide, central adiposity, leptin, and leptin:adiponectin ratios) which may associate with more aggressive prostate cancer histology. Prostate 77:211-221, 2017. © 2016 Wiley Periodicals, Inc.

Clustering of risk-related modifiable behaviours and their association with overweight and obesity among a large sample of youth in the COMPASS study

BACKGROUND

Canadian youth exhibit a number of risky behaviours, some of which are associated with overweight and obesity. The purpose of this study was to examine the prevalence of 15 modifiable risk behaviours in a large sample of Canadian youth, to identify underlying subgroups based on patterns of health behaviours, and to examine the association between identified subgroups and overweight/obesity.

METHODS

Data from 18,587 grades 9-12 students in Year 1 (2012-13) of the COMPASS study and latent class analysis were used to identify patterns and clustering among 15 health behaviours (e.g., physical inactivity, sedentary behaviour, unhealthy eating, substance use). A logistic regression model examined the associations between these clusters and overweight/obesity status.

RESULTS

Four distinct classes were identified: traditional school athletes, inactive screenagers, health conscious, and moderately active substance users. Each behavioural cluster demonstrated a distinct pattern of behaviours, some with a greater number of risk factors than others. Traditional school athletes (odds ratio (OR) 1.15, 95% CI 1.03-1.29), inactive screenagers (OR 1.33; 1.19-1.48), and moderately active substance users (OR 1.27; 1.14-1.43) were all significantly more likely to be overweight/obese compared to the health conscious group.

CONCLUSIONS

Four distinct subpopulations of youth were identified based on their patterns of health and risk behaviours. The three clusters demonstrating poorer health behaviour were all at an increased risk of being overweight/obese compared to their somewhat healthier peers. Obesity-related public health interventions and health promotion efforts might be more effective if consideration is given to population segments with certain behavioural patterns, targeting subgroups at greatest risk of overweight or obesity.

Glucose metabolism during the acute prostate cancer treatment trajectory: The influence of age and obesity

BACKGROUND & AIMS

Obesity and age, key risk factors for aggressive prostate cancer, are associated with insulin resistance. Glucose-related parameters in patients with aggressive prostate cancer were compared with 2 reference groups: men of similar age and body mass index (BMI) without cancer, and healthy young men. Acute changes in these parameters following radiation treatment were also evaluated.

METHODS

Nine patients with aggressive prostate cancer underwent metabolic assessments prior to treatment (baseline), 7 and 33 weeks post-baseline (post-treatment initiation). Baseline measures were compared with the 2 reference groups. Evaluations included: 1) fasting and oral glucose tolerance test (OGTT) blood samples for glucose, C-peptide, and insulin, 2) fasting blood samples for triglycerides, cholesterols, leptin, adiponectin, IL-6, and TNF-α, 3) body composition, 4) nutrition, and 5) physical activity.

RESULTS

At baseline, patients had normal fasting glucose concentrations (<5.6 mM; 4.9 ± 1.2 mM) but impaired 2-h OGTT glucose concentrations (>7.8 mM; 8.7 ± 2.9 mM). Both reference groups had normal fasting (matched males: 4.2 ± 0.5 mM; young males: 3.7 ± 0.4 mM) and 2-h OGTT glucose concentrations (matched males: 5.6 ± 1.8 mM; young males: 3.1 ± 0.1 mM) that were significantly lower than patient values. During the OGTT, patients had higher insulin (120 min) and C-peptide (45, 60, 90, 120 min) concentrations compared to the matched males. At 7 weeks, 2-h OGTT glucose concentrations in patients improved to healthy ranges without changes in insulin, C-peptide, IGF-1, IGFBP-3 or other metabolic parameters.

CONCLUSIONS

At baseline patients with aggressive prostate cancer demonstrated impaired glucose tolerance compared with men of similar age and body size. Following treatment, glucose tolerance improved in the absence of changes in expected modifiers of glucose metabolism. These improvements may be related to treatment.

Food safety knowledge of undergraduate students at a Canadian university: results of an online survey

Background

Foodborne diseases are an important public health issue, and young adults are an important demographic to target with food safety education. Our objective was to assess the food safety knowledge of undergraduate students at a Canadian university, to identify potential areas for such education.

Methods

In February 2015, we conducted an online survey of 485 undergraduate students at a university in Ontario, Canada. We assessed various food-related factors, including cooking frequency and prior food handling or preparation education. We then modeled the relationship between ‘overall knowledge score’ and the demographic and food skills/cooking experience predictors using multivariable log-binomial regression, to determine factors associated with relatively higher proportions of correct responses.

Results

Respondents were, on average, 20.5 years old, and the majority (64.8 %) lived off campus. Students cooked from basic ingredients infrequently, with 3 in 4 doing so a few times a year to never. Students averaged 6.2 correct answers to the 11 knowledge questions. Adjusting for other important covariates, older age and being a current food handler were associated with relatively higher knowledge, whereas working/volunteering in a hospital and infrequent cooking were associated with relatively lower knowledge. Males in the Faculty of Science had relatively higher knowledge than females in the Faculty of Science, both of whom had relatively higher knowledge than all students in other Faculties. Among students who had never taken a food preparation course, knowledge increased with self-reported cooking ability; however, among students who had taken such a course, knowledge was highest among those with low self-reported cooking ability.

Conclusions

Consistent with other similar studies, students in Faculties outside of the Faculty of Science, younger students, and those who cook infrequently could benefit from food safety education. Supporting improved hand hygiene, in particular clarifying hand washing versus hand sanitizing messages, may also be important. Universities can play a role in such education, including as part of preparing students for work or volunteer placements, or as general support for student health and success.

Electronic supplementary material

The online version of this article (doi:10.1186/s12889-016-3818-y) contains supplementary material, which is available to authorized users.

The use of a brief mental health screener to enhance the ability of police officers to identify persons with serious mental disorders

Police agencies in Canada and elsewhere have received much criticism over how they respond to persons with serious mental disorders. The adequacy of training provided to police officers on mental health issues and in particular on recognizing indicators of serious mental disorders has been a major concern. This paper describes the process that led to the development of a new brief mental health screener (interRAI Brief Mental Health Screener, BMHS) designed to assist police officers to better identify persons with serious mental disorders. The interRAI BMHS was developed in collaboration with interRAI, an international, not-for-profit consortium of researchers. The government of Ontario had previously partnered with interRAI to develop and implement the Resident Assessment Instrument for Mental Health (RAI-MH), the assessment system mandated for use on all persons admitted into inpatient psychiatric care in the province. Core items on the interRAI BMHS were obtained through analysis (N=41,019) of RAI-MH data together with input from representatives from health care, police services, and patient groups. Two police services in southwestern Ontario completed forms (N=235) on persons thought to have a mental disorder. Patient records were later accessed to determine patient disposition. The use of summary and inferential statistics revealed that the variables significantly associated with being taken to hospital by police included performing a self-injurious act in the past 30days, and others being concerned over the person's risk for self-injury. Variables significantly associated with being admitted included abnormal thought process, delusions, and hallucinations. The results of the study indicate that the 14-variable algorithm used to construct the interRAI BMHS is a good predictor of who was most likely to be taken to hospital by police officers and who was most likely to be admitted. The instrument is an effective means of capturing and standardizing police officer observations enabling them to provide more and better quality information to emergency department (ED) staff. Teaching police officers to use the form constitutes enhanced training on major indicators of serious mental disorders. Further, given that items on the interRAI BMHS are written in the language of the health system, language acts as common currency between police officers and ED staff laying the foundation for a more collaborative approach between the systems.

Food safety knowledge, attitudes and self-reported practices among Ontario high school students

OBJECTIVES

To measure the food safety knowledge, attitudes and self-reported practices of high school students in Ontario.

METHODS

We administered a school-wide paper survey to the student body (n = 2,860) of four Ontario high schools. We developed the survey by selecting questions from existing, validated questionnaires, prioritizing questions that aligned with the Canadian Partnership for Consumer Food Safety Education's educational messages and the food safety objectives from the 2013 Ontario High School Curriculum.

RESULTS

One in five students reported currently handling food in commercial or public-serving venues; of these, 45.1% had ever taken a course that taught them how to prepare food (e.g., food and nutrition classes, food handler certification). Food safety knowledge among respondents was low. For example, 17.3% knew that the best way to determine whether hamburgers were cooked enough to eat was to measure the temperature with a food thermometer. Despite low knowledge, most respondents (72.7%) reported being confident that they could cook safe, healthy meals for themselves and their families. Safe food handling practices were frequently self-reported. Most students (86.5%) agreed that being able to cook safe, healthy meals was an important life skill, although their interest in learning about safe food handling and concern about foodborne disease were less pronounced.

CONCLUSION

Our findings suggest that food safety knowledge is low, yet confidence in preparing safe, healthy meals is high, among high school students. Because work and volunteer opportunities put students in contact with both the public and food, this group is important to target for increased education about safe food handling.

Personalized mortality prediction driven by electronic medical data and a patient similarity metric

BACKGROUND

Clinical outcome prediction normally employs static, one-size-fits-all models that perform well for the average patient but are sub-optimal for individual patients with unique characteristics. In the era of digital healthcare, it is feasible to dynamically personalize decision support by identifying and analyzing similar past patients, in a way that is analogous to personalized product recommendation in e-commerce. Our objectives were: 1) to prove that analyzing only similar patients leads to better outcome prediction performance than analyzing all available patients, and 2) to characterize the trade-off between training data size and the degree of similarity between the training data and the index patient for whom prediction is to be made.

METHODS AND FINDINGS

We deployed a cosine-similarity-based patient similarity metric (PSM) to an intensive care unit (ICU) database to identify patients that are most similar to each patient and subsequently to custom-build 30-day mortality prediction models. Rich clinical and administrative data from the first day in the ICU from 17,152 adult ICU admissions were analyzed. The results confirmed that using data from only a small subset of most similar patients for training improves predictive performance in comparison with using data from all available patients. The results also showed that when too few similar patients are used for training, predictive performance degrades due to the effects of small sample sizes. Our PSM-based approach outperformed well-known ICU severity of illness scores. Although the improved prediction performance is achieved at the cost of increased computational burden, Big Data technologies can help realize personalized data-driven decision support at the point of care.

CONCLUSIONS

The present study provides crucial empirical evidence for the promising potential of personalized data-driven decision support systems. With the increasing adoption of electronic medical record (EMR) systems, our novel medical data analytics contributes to meaningful use of EMR data.

Use of caffeinated energy drinks among secondary school students in Ontario: Prevalence and correlates of using energy drinks and mixing with alcohol

OBJECTIVES

Caffeinated energy drinks have become increasingly popular among young people, raising concern about possible adverse effects, including increased alcohol consumption and related risk behaviours. The current study examined consumption of caffeinated energy drinks and use of energy drinks with alcohol, as well as associations with socio-demographic and behavioural characteristics, among a sample of secondary school students in Ontario.

METHODS

Survey data from 23,610 grade 9-12 students at 43 purposefully sampled Ontario secondary schools participating in the baseline wave (2012/13) of the COMPASS study were analyzed using generalized linear mixed-effects models. Outcomes were any energy drink use, frequency of use, and use of alcohol mixed with energy drinks; covariates were age, sex, race, spending money, bodymass index (BMI), weight-related efforts and alcohol use. Two-way interactions between sex and other covariates were tested.

RESULTS

Nearly one in five students (18.2%) reported consuming energy drinks in a usual week. Use of energy drinks was associated (p &lt; 0.01) with all socio-demographic variables examined and was more common among students who were male, off-reserve Aboriginal, had some spending money, had a BMI outside of "healthy" range, were trying to lose weight, and/or reported a higher intensity of alcohol use. Interactions with sex were observed for age, spending money and weight-related efforts. Use of energy drinks mixed with alcohol in the previous 12 months was reported by 17.3% of the sample, and was associated with race, spending money, and more frequent binge drinking.

CONCLUSION

Regular use of energy drinks was common among this sample of students and strongly linked to alcohol consumption.

Plasma amino acid profiles of breast cancer patients early in the trajectory of the disease differ from healthy comparison groups

This study describes and compares fasting plasma amino acid profiles of breast cancer patients near the initiation of chemotherapy with those of healthy age- and body mass index-matched females (HM), as well as young healthy females (HY). Breast cancer patients had significantly greater glutamate and histidine concentrations and significantly lower threonine concentrations compared with HM and HY females independent of protein or caloric intake. These differences may be related to metabolic perturbations associated with the disease.

The COMPASS study: a longitudinal hierarchical research platform for evaluating natural experiments related to changes in school-level programs, policies and built environment resources

Background

Few researchers have the data required to adequately understand how the school environment impacts youth health behaviour development over time.

Methods/Design

COMPASS is a prospective cohort study designed to annually collect hierarchical longitudinal data from a sample of 90 secondary schools and the 50,000+ grade 9 to 12 students attending those schools. COMPASS uses a rigorous quasi-experimental design to evaluate how changes in school programs, policies, and/or built environment (BE) characteristics are related to changes in multiple youth health behaviours and outcomes over time. These data will allow for the quasi-experimental evaluation of natural experiments that will occur within schools over the course of COMPASS, providing a means for generating “practice based evidence” in school-based prevention programming.

Discussion

COMPASS is the first study with the infrastructure to robustly evaluate the impact that changes in multiple school-level programs, policies, and BE characteristics within or surrounding a school might have on multiple youth health behaviours or outcomes over time. COMPASS will provide valuable new insight for planning, tailoring and targeting of school-based prevention initiatives where they are most likely to have impact.

Diet and lifestyle factors and risk of subtypes of esophageal and gastric cancers: classification tree analysis

PURPOSE

Although risk factors for squamous cell carcinoma of the esophagus and adenocarcinomas of the esophagus (EA), gastric cardia (GC), and other (noncardia) gastric (OG) sites have been identified, little is known about interactions among risk factors. We sought to examine interactions of diet, other lifestyle, and medical factors with risks of subtypes of esophageal and gastric cancers.

METHODS

We used classification tree analysis to analyze data from a population-based case-control study (1095 cases, 687 controls) conducted in Connecticut, New Jersey, and western Washington State.

RESULTS

Frequency of reported gastroesophageal reflux disease symptoms was the most important risk stratification factor for EA, GC, and OG, with dietary factors (EA, OG), smoking (EA, GC), wine intake (GC, OG), age (OG), and income (OG) appearing to modify the risk of these cancer sites. For esophageal squamous cell carcinoma, smoking was the most important risk stratification factor, with gastroesophageal reflux disease, income, race, noncitrus fruit, and energy intakes further modifying risk.

CONCLUSION

Various combinations of risk factors appear to interact to affect risk of each cancer subtype. Replication of these data mining analyses are required before suggesting causal pathways; however, the classification tree results are useful in partitioning risk and mapping multilevel interactions among risk variables.