To predict the risk of hospital-acquired pressure injury using machine learning compared with standard care.

We obtained electronic health records (EHRs) to structure a multilevel cohort of hospitalised patients at risk for pressure injury and then calibrate a machine learning model to predict future pressure injury risk. Optimisation methods combined with multilevel logistic regression were used to develop a predictive algorithm of patient-specific shifts in risk over time. Machine learning methods were tested, including random forests, to identify predictive features for the algorithm. We reported the results of the regression approach as well as the area under the receiver operating characteristics (ROC) curve for predictive models.

Hospitalised inpatients.

EHRs of 35 001 hospitalisations over 5 years across 2 academic hospitals.

Longitudinal shifts in pressure injury risk.

The predictive algorithm with features generated by machine learning achieved significantly improved prediction of pressure injury risk (p

These data could help hospitals conserve resources within a critical period of patient vulnerability of hospital-acquired pressure injury which is not reimbursed by US Medicare; thus, conserving between 30 000 and 90 000 labour-hours per year in an average 500-bed hospital. Hospitals can use this predictive algorithm to initiate a quality improvement programme for pressure injury prevention and further customise the algorithm to patient-specific variation by facility.

The pandemic and public health response to contain the virus had impacts on many aspects of young people’s lives including disruptions to daily routines, opportunities for social, academic, recreational engagement and early employment. Consequently, children, adolescents and young adults may have experienced mental health challenges that required use of mental health services. This study compared rates of use for inpatient and outpatient mental health services during the pandemic to pre-pandemic rates.

Population-based repeated cross-sectional study.

Publicly delivered mental healthcare in primary and secondary settings within the province of Ontario, Canada.

All children 6–12 years of age (n=2 043 977), adolescents 13–17 years (n=1 708 754) and young adults 18–24 years (n=2 286 544), living in Ontario and eligible for provincial health insurance between March 2016 and November 2021.

Outpatient mental health visits to family physicians and psychiatrists for: mood and anxiety disorders, alcohol and substance abuse disorders, other non-psychotic mental health disorders and social problems. Inpatient mental health visits to emergency departments and hospitalisations for: substance-related and addictive disorders, anxiety disorders, assault-related injuries, deliberate self-harm and eating disorders. All outcomes were analysed by cohort and sex.

During the pandemic, observed outpatient visit rates were higher among young adults by 19.01% (95% CI: 15.56% to 22.37%; 209 vs 175 per 1000) and adolescent women 24.17% (95% CI: 18.93% to 29.15%; 131 vs 105 per 1000) for mood and anxiety disorders and remained higher than expected. Female adolescents had higher than expected usage of inpatient care for deliberate self-harm, eating disorders and assault-related injuries.

Study results raise concerns over prolonged high rates of mental health use during the pandemic, particularly in female adolescents and young women, and highlights the need to better monitor and identify mental health outcomes associated with COVID-19 containment measures and to develop policies to address these concerns.