Increasing the Breadth and Quantity of Injury Data in AWISS

AWISS continuously strive to increase the breadth and quantity of injury related data in the AWISS system.  By integrating additional datasets into the AWISS, we can build a more complete picture about the causes of injury, at risk groups, and effectiveness of injury prevention interventions.

Listed below are several datasets which AWISS staff have either recently incorporated into AWISS system or are in the process of doing so.

School Travel and Child Safety Survey (STCSS) dataset

AWISS staff are working in collaboration with Neath Port Talbot Council to develop an online ‘School Travel and Child Safety Survey’. This survey was initially developed and piloted as part of the European Eurosafe TACTICS project (http://www.childsafetyeurope.org/tactics/). The survey has subsequently been implemented on NPT’s systems, and was piloted in schools in NPT in June 2016. NPT Council plan to roll the survey out to all schools in NPT at the start of the 2017 academic year. The survey has been designed to enable the results to be both analysed spatially by the council (e.g. identify areas with high and low active travel rates) and also to be incorporated anonymously into the SAIL database. The implementation of this survey will enable the medium and long term effectiveness of child safety/active travel interventions to be evaluated, and it’s incorporation into the SAIL database will allow associations between child behaviour, health and educational outcomes to be explored. Whilst the AWISS system has access to lots of health outcome data, this will be one of the first exposure based datasets to be incorporated in to the AWISS system. AWISS staff are also in discussions with the National School Travel Advisory group to develop a national school travel survey to be rolled out to all schools in Wales using an online system called ‘Hwb. We are currently exploring the potential to bring the results of this national survey into SAIL.

Unintentional House Fire dataset

Samantha Turner, as a part of her PhD project, acquired anonymised unintentional house fire data from all three Welsh Fire and Rescue Services (North, Mid & West, South). Samantha is linking house fire data to other datasets in SAIL to establish the types of households in Wales at increased risk of experiencing unintentional house fire incidents, injuries or deaths. 6943 case households from across Wales have been matched to 347,150 control households (case control ratio 1:50). Conditional Logistic regression will be performed to identify household level factors associated with an increased risk of experiencing a FRS attended unintentional fire. Samantha has also undertaken a systematic review of the risk factors associated with unintentional house fire incidents, injuries and deaths. A report has been provided to PHW summarising the findings, and a scientific paper will be published shortly. An abstract based on this work has been accepted at the Safety 2016 conference.

Welsh Burns Centre Unit (WBCU) dataset

AWISS recently acquired anonymised data from the Welsh Burns Centre Unit (WBCU). AWISS staff have conducted initial analyses on the dataset, exploring the trends in treated burns between 2003 and 2012. Further work will be carried out in 2016/17 to explore the impact of changing the type of dressings used on healthcare utilisation. An updated data extract from the Welsh Burns Centre will be included in SAIL in 2016/17.

Critical Care Minimum Data Set (CCMDS)

The Critical Care Minimum Data Set (CCMDS) was recently incorporated into the SAIL database in 2015. AWISS staff have begun exploring the utility of the data set. Analysis of survival following discharge from critical care is currently being carried out. The critical care dataset was linked with PEDW inpatients data to determine if the hospital spell was related to an injury.

AWISS analysts have begun Cox Regression analysis to determine the factors that are associated with increased risk of mortality with follow-up to 30 days and 365 days (for all patients, not only injury patients). Initial analysis suggests the estimated risk of mortality 30 days and 365 days following discharge from critical care is 8.1% and 20.1% respectively. Preliminary Cox regression results show that those who were admitted to hospital due to injuries have better mortality outcomes compared with those admitted for non-injuries (30 day mortality HR (CI): 0.85 (0.74,0.98) (baseline:non-injury), 365 day mortality HR (CI): 0.89 (0.81,0.97) (baseline:non-injury)). Further work will be carried out.

Trauma Audit Research Network (TARN) and the Intensive Care National Audit Research Centre (ICNARC)

Data from two national audits, TARN and ICNARC are in the process of being incorporated into the SAIL database. Both contain data on injury diagnoses and severity which are missing from other datasets. TARN and ICNARC will specifically be used in the EMRTS evaluation (page??). They will allow the follow up of critically ill and injured patients who received specialist care at English and Welsh hospitals.

Ambulance Data

There are plans to incorporate ambulance service clinical data when a new digital pen electronic patient clinical record data collection system becomes operational from September 2015. Together with colleagues from the Health and Care Research Wales NCPHWR and PRIME centres we will explore the utility of using dispatch data from Welsh Ambulance Service Trust to support injury surveillance