Weather patterns can help predict demand for emergency CT scans
Weather patterns may help providers predict demand for emergency CT scans, according to new research published Tuesday in the European Journal of Radiology [1].
Resource planning can be a crucial component in hospitals and especially radiology departments, experts noted. And since weather conditions often correlate with emergency department visits, Swedish researchers theorized that this information could aid in forecasting imaging demand.
To back their hypothesis, radiologist Tobias Heye, MD, and colleagues retrospectively analyzed all polytrauma CTs performed at their institution between 2011 and 2022. University Hospital Basel, Switzerland, tallied 6,683 scans for instances when patients sustained multiple injuries during the study period. This included a high number of CT exams on warm days with more sunshine and ultraviolet light, less wind and fewer clouds.
“The amount of daily polytrauma CTs seems to correlate with warmer and sunnier weather, allowing to partially forecast above median daily number of polytrauma CTs using weather data,” Heye and co-authors concluded. “Prediction of polytrauma CT volumes may help to improve resource planning in the hospital, as especially unforeseen emergency CT examinations pose a considerable impact on workload, which was historically difficult to account for.”
Heye et al. pulled their weather information from meteoblue.com, normalizing the data to fit their analysis. They applied logistic regression and machine learning algorithms as prediction models, using data from 2012-2020 for training and 2021-2022 for validation.
University Hospital Basel acquired more polytrauma CTs in summer compared to winter months (at a median of 2.35 vs. 2.08), “demonstrating a seasonal change.” They also noted that temperature, sunshine duration and ultraviolet light amount all correlated positively, while wind velocity and cloudiness correlated negatively with CT occurrence. Their regression model achieved an accuracy of 87% on training data from 2011-2020. But when forecasting for 2021-2022, accuracy dipped to 65%. AI upped accuracy to 72%, with wind velocity and ultraviolet light amount deemed as the most important parameters.
Read more about the results, including potential study limitations, at the link below.
Marty Stempniak has covered healthcare since 2012, with his byline appearing in the American Hospital Association's member magazine, Modern Healthcare and McKnight's. Prior to that, he wrote about village government and local business for his hometown newspaper in Oak Park, Illinois. He won a Peter Lisagor and Gold EXCEL awards in 2017 for his coverage of the opioid epidemic.