Accurate forecasting of patient arrivals at emergency departments (EDs) is vital for efficient resource allocation and high-quality patient care. In this study we investigate the relevance of exogenous variables, namely tourism, weather, calendar and demographic variables, in forecasting ED visits in the reference hospital in Palma de Mallorca, a city with significant seasonal population fluctuations due to tourism. Using a machine learning approach, we develop a model that predicts ED visits based solely on these exogenous variables. We test different machine learning algorithms (random forests, support vector machines, and feedforward neural networks) with different combinations of input variables and compare their symmetric mean average percentage errors (SMAPEs). Our findings reveal that calendar information, resident, and tourist population data are statistically significant for the accuracy of the predictions, while the addition of weather data does not provide any further improvement. Comparison of non-time-series with time-series prediction models reveals that the latter provide better accuracy for short prediction horizons (e.g., shorter than a week). Furthermore, time-series models become less or equally accurate to models relying only on exogenous variables for long prediction horizons (e.g., fortnight or month). Our study highlights the importance of carefully selecting predictive variables to ensure short- and long-term, robust and reliable forecasts. This demonstrates that, despite their lower complexity, non-time-series models with well-chosen input variables can be as effective as time-series models when predicting for long time horizons.

Introduction: Knee osteoarthritis (KOA) is characterized by pain, stiffness, and functional limitation. Aerobic exercise (AE) is a key treatment with proven benefits. Aims: To determine the most effective AE intensity for KOA. Methods: Searches were performed in seven databases, including randomized controlled trials with AE-only groups. Outcomes assessed were pain, walking and sit-to-stand performance, stiffness, and disability. Methodological quality and bias were evaluated. Bayesian random-effects network meta-analyses compared AE intensities, reporting standardized mean differences (Hedges’ g) with 95% credible intervals. Certainty of evidence was rated using GRADE. Results: Fifteen studies were included (mean PEDro score: 5.93), showing “some concerns” or “high risk” of bias. Two meta-analyses (pain and walking performance) were conducted. Comparisons between AE intensities showed non-significant, imprecise results with very low certainty of evidence. Conclusions: Evidence is insufficient to identify the optimal AE intensity for KOA. Limitations include high risk of bias, wide credible intervals, and reliance on indirect comparisons. In this context, clinicians should apply a structured, patient-centered approach to AE prescription, considering gradual progression and monitoring of tolerance, and combine AE with other recommended interventions. High-quality trials with direct comparisons of AE intensity should be conducted for this inconclusive gap.

Accurate forecasting of patient arrivals at emergency departments (EDs) is vital for efficient resource allocation and high-quality patient care. In this study we investigate the relevance of exogenous variables, namely tourism, weather, calendar and demographic variables, in forecasting ED visits in the reference hospital in Palma de Mallorca, a city with significant seasonal population fluctuations due to tourism. Using a machine learning approach, we develop a model that predicts ED visits based solely on these exogenous variables. We test different machine learning algorithms (random forests, support vector machines, and feedforward neural networks) with different combinations of input variables and compare their symmetric mean average percentage errors (SMAPEs). Our findings reveal that calendar information, resident, and tourist population data are statistically significant for the accuracy of the predictions, while the addition of weather data does not provide any further improvement. Comparison of non-time-series with time-series prediction models reveals that the latter provide better accuracy for short prediction horizons (e.g. shorter than a week). Furthermore, time-series models become less or equally accurate to models relying only on exogenous variables for long prediction horizons (e.g. fortnight or month). Our study highlights the importance of carefully selecting predictive variables to ensure short- and long-term, robust and reliable forecasts. This demonstrates that, despite their lower complexity, non-time-series models with well-chosen input variables can be as effective as time-series models when predicting for long time horizons.