Does machine learning have a role in the prediction of asthma in children?

Asthma is the most common chronic lung disease in childhood. There has been a significant worldwide effort to develop tools/methods to identify children's risk for asthma as early as possible for preventative and early management strategies. Unfortunately, most childhood asthma prediction tools using conventional statistical models have modest accuracy, sensitivity, and positive predictive value. Machine learning is an approach that may improve on conventional models by finding patterns and trends from large and complex datasets. Thus far, few studies have utilized machine learning to predict asthma in children. This review aims to critically assess these studies, describe their limitations, and discuss future directions to move from proof-of-concept to clinical application.

Joint effects of heatwaves and air quality on ambulance services for vulnerable populations in Perth, western Australia.

As the frequency, intensity, and duration of heatwaves increases, emergency health serviceutilization, including ambulance service, has correspondingly increased across the world. The negative effects of air pollution on health complicate these adverse health effects. This research work is the first known study to analyze the joint effects of heatwaves and air quality on the ambulance service in Western Australia (WA). The main objective is to investigate the potential joint effects of heatwaves and air quality on the ambulance service for vulnerable populations in the Perth metropolitan area. A time series design was used. Daily data on ambulance callouts, temperature and air pollutants (CO, SO2, NO2, O3, PM10 and PM2.5) were collected for the Perth metropolitan area, WA from 2006 to 2015. Poisson regression modeling was used to assess the association between heatwaves, air quality, and ambulance callouts. Risk assessments on age, gender, socio-economic status (SES), and joint effects between heatwaves and air quality on ambulance callouts were conducted. The ambulance callout rate was higher during heatwave days (14.20/100,000/day) compared to non-heatwave days (13.95/100,000/day) with a rate ratio of 1.017 (95% confidence interval 1.012, 1.023). The ambulance callout rate was higher in males, people over 60 years old, people with low SES, and those living in coastal areas during period of heatwaves. Exposure to CO, SO2, O3 and PM2.5 increased risk on ambulance callouts and exposure to NO2 showed joint effect with heatwave and increased risk of ambulance callouts by 3% after adjustment of all other risk factors. Ambulance callouts are an important indicator for evaluating heatwave-related emergency morbidity in WA. As the median concentrations of air pollutants in WA were lower than the Australian National Standards, the interactive effects of heatwaves and air quality on ambulance service need to be further examined, especially when air pollutants exceed the standards.

Joint effect of heatwaves and air quality on emergency department attendances for vulnerable population in Perth, Western Australia, 2006 to 2015.

BACKGROUND: As global warming and the frequency and intensity of heatwaves increases, health service utilization, including emergency department attendances (EDA) have correspondingly increased across the world. The impact of air quality on health adds to the complexity of the effects. Potential joint effects between heatwaves and air quality on EDA have been rarely reported in the literature, prompting this study. OBJECTIVES: To investigate the potential joint effect of heatwaves and air quality on the EDA for vulnerable populations in the Perth metropolitan area, Western Australia. METHODS: A time series design was used. Daily data on EDA, heatwaves (excess heat factor>0) and air pollutants (CO, SO2, NO2, O3, PM10 and PM2.5) were collected for Perth, Western Australia from 2006 to 2015. Poisson regression modelling was used to assess the associations between heatwaves, air quality, and EDA. Risk assessments on age, gender, Aboriginality, socio-economic status (SES), and joint effect between heatwaves and air quality on EDA were conducted. RESULTS: The EDA rate was higher in heatwave days (77.86/100,000/day) compared with non-heatwave days (73.90/100,000/day) with rate ratio of 1.053 (95% confidence interval 1.048, 1.058). The EDA rate was higher in males, people older than 60 years or younger than 15 years, Aboriginal people, and people with low SES. Exposure to CO, SO2, O3 and PM2.5 increased risk on EDA and exposure to PM2.5 showed joint effect with heatwave and increased risk of EDA by 6.6% after adjustment of all other risk factors. CONCLUSIONS: EDA is an important indicator to evaluate heatwave related morbidity for emergency medical service as EDA rate increased during heatwaves with relative high concentrations of air pollutants. As all air pollutants measured in the study were lower than the Australian National Standards, the joint effect of heatwaves and air quality needs to be further examined when it exceeds the standards.