Ambulance Calls for Substance Use and Alcohol in a Pandemic (ASAP)

Overview

The Primary objective is to explore ambulance service attendance at incidents involving alcohol and/or substance use over the period of the pandemic lockdown, and the following months. This will be to determine prevalence and explore factors such as patient gender, age, ethnicity or location. Analysis will examine the calls over the course of the year prior to the lockdown, and then compare this to the period of lockdown and following months.

The Primary objective is to explore ambulance service attendance at incidents involving alcohol and/or substance use over the period of the pandemic lockdown, and the following months. This will be to determine prevalence and explore factors such as patient gender, age, ethnicity or location. Analysis will examine the calls over the course of the year prior to the lockdown, and then compare this to the period of lockdown and following months. A time series analysis will be conducted to examine the calls over the course of the year prior to the lockdown, and then compare this to the period of lockdown and following months. This will use the 'Interrupted Time Series' (ITS) approach. To explore this regression models will be built that examine the causal models for attendance prior to the pandemic and compared to the lockdown time frame. A multivariable regression model will be built. Initially a Directed Acyclic Graph (DAG) will allow the identification of confounders and exposures relevant to the model. A logistic regression model will be used to calculate the relative risk of call during lockdown compared to the data prior to lockdown. The model will be fit using p\<0.05 as the definition of statistical significance. Descriptive statistics, trend analysis and predictive analysis will be conducted on the data set to determine trends across time, factors that predict patients requiring ambulance attendance, and factors that predict treatment outcomes. Missing data will be examined for systematic bias, and where found to be missing at random will be excluded from analysis. Where not missing at random, sensitivity analysis will be conducted. Analysis will examine covariates. Age will be defined as single year continuous variable and examined in categories such as 5-year age groups. Ethnicity will be categorised as groups, such as black, Asian, other minority and mixed ethnic groups will be explored. Census data such as the deprivation, rurality, income, employment, disability and education will look at the decile as defined.