Kamolrat Sangkharat1, Marliyyah A Mahmood1, John E Thornes2, Paul A Fisher3, Francis D Pope4. 1. School of Geography, Earth and Environmental Sciences, University of Birmingham, Birmingham, UK. 2. Chemicals and Environmental Effects, Public Health England, Oxfordshire, UK. 3. Institute of Applied Health Research, University of Birmingham, Birmingham, UK. 4. School of Geography, Earth and Environmental Sciences, University of Birmingham, Birmingham, UK. Electronic address: f.pope@bham.ac.uk.
Abstract
BACKGROUND: Associations between extreme temperatures and health outcomes, such as mortality and morbidity, are often observed. However, relatively little research has investigated the role of extreme temperatures upon ambulance dispatches. METHODS: A time series analysis using London Ambulance Service (LAS) incident data (2010-2014), consisting of 5,252,375 dispatches was conducted. A generalized linear model (GLM) with a quasi-likelihood Poisson regression was applied to analyse the associations between ambulance dispatches and temperature. The 99th (22.8°C) and 1st (0.0°C) percentiles of temperature were defined as extreme high and low temperature. Fourteen categories of ambulance dispatches were investigated, grouped into 'respiratory' (asthma, dyspnoea, respiratory chest infection, respiratory arrest and chronic obstructive pulmonary disease), 'cardiovascular' (cardiac arrest, chest pain, cardiac chest pain RCI, cardiac arrhythmia and other cardiac problems) and 'other' non-cardiorespiratory (dizzy, alcohol related, vomiting and 'generally unwell') categories. The effects of long-term trends, seasonality, day of the week, public holidays and air pollution were controlled for in the GLM. The lag effect of temperature was also investigated. The threshold temperatures for each category were identified and a distributed lag non-linear model (DLNM) was reported using relative risk (RR) values at 95% confidence intervals. RESULTS: Many dispatch categories show significant associations with extreme temperature. Total calls from 999 dispatches and 'generally unwell' dispatch category show significant RRs at both low and high temperatures. Most respiratory categories (asthma, dyspnoea and RCI) have significant RRs at low temperatures represented by with estimated RRs ranging from 1.392 (95%CI: 1.161-1.699) for asthma to 2.075 (95%CI: 1.673-2.574) for RCI. The RRs for all other non-cardiorespiratory dispatches were often significant for high temperatures ranging from 1.280 (95% CI: 1.128-1.454) for 'generally unwell' to 1.985 (95%CI: 1.422-2.773) for alcohol-related. For the cardiovascular group, only chest pain dispatches reported a significant RR at high temperatures. CONCLUSIONS: Ambulance dispatches can be associated with extreme temperatures, dependent on the dispatch category. It is recommended that meteorological factors are factored into ambulance forecast models and warning systems, allowing for improvements in ambulance and general health service efficiency.
BACKGROUND: Associations between extreme temperatures and health outcomes, such as mortality and morbidity, are often observed. However, relatively little research has investigated the role of extreme temperatures upon ambulance dispatches. METHODS: A time series analysis using London Ambulance Service (LAS) incident data (2010-2014), consisting of 5,252,375 dispatches was conducted. A generalized linear model (GLM) with a quasi-likelihood Poisson regression was applied to analyse the associations between ambulance dispatches and temperature. The 99th (22.8°C) and 1st (0.0°C) percentiles of temperature were defined as extreme high and low temperature. Fourteen categories of ambulance dispatches were investigated, grouped into 'respiratory' (asthma, dyspnoea, respiratory chest infection, respiratory arrest and chronic obstructive pulmonary disease), 'cardiovascular' (cardiac arrest, chest pain, cardiac chest pain RCI, cardiac arrhythmia and other cardiac problems) and 'other' non-cardiorespiratory (dizzy, alcohol related, vomiting and 'generally unwell') categories. The effects of long-term trends, seasonality, day of the week, public holidays and air pollution were controlled for in the GLM. The lag effect of temperature was also investigated. The threshold temperatures for each category were identified and a distributed lag non-linear model (DLNM) was reported using relative risk (RR) values at 95% confidence intervals. RESULTS: Many dispatch categories show significant associations with extreme temperature. Total calls from 999 dispatches and 'generally unwell' dispatch category show significant RRs at both low and high temperatures. Most respiratory categories (asthma, dyspnoea and RCI) have significant RRs at low temperatures represented by with estimated RRs ranging from 1.392 (95%CI: 1.161-1.699) for asthma to 2.075 (95%CI: 1.673-2.574) for RCI. The RRs for all other non-cardiorespiratory dispatches were often significant for high temperatures ranging from 1.280 (95% CI: 1.128-1.454) for 'generally unwell' to 1.985 (95%CI: 1.422-2.773) for alcohol-related. For the cardiovascular group, only chest pain dispatches reported a significant RR at high temperatures. CONCLUSIONS: Ambulance dispatches can be associated with extreme temperatures, dependent on the dispatch category. It is recommended that meteorological factors are factored into ambulance forecast models and warning systems, allowing for improvements in ambulance and general health service efficiency.
Authors: Santiago Gestal Romani; Dominic Royé; Luis Sánchez Santos; Adolfo Figueiras Journal: Int J Environ Res Public Health Date: 2020-12-03 Impact factor: 3.390
Authors: Osnat Wine; Alvaro Osornio Vargas; Sandra M Campbell; Vahid Hosseini; Charles Robert Koch; Mahdi Shahbakhti Journal: Int J Environ Res Public Health Date: 2022-01-28 Impact factor: 3.390