Science & Technology, UK (Commonwealth Union) – The amalgamation of police and healthcare data has unveiled the ability to identify highly vulnerable individuals even before police intervention becomes necessary. A recent investigation conducted by researchers from Swansea University underscores the potential of data sharing and linkage in curbing police callouts and future instances of emergency medical admissions.
This comprehensive study, featured in The Lancet Public Health and spearheaded by the National Centre for Population Health & Wellbeing Research, sought to showcase the viability and advantages of connecting police and healthcare data.
The Crime and Disorder Act of 1998, designed to preempt criminal activities, mandates collaboration between the police, local government, and the NHS to formulate joint strategies for crime reduction, incorporating data sharing to inform targeted responses. While sharing individual data is a practiced approach, the exploration of the full potential inherent in sharing an entire agency’s dataset and linking it with data from other organizations is seldom undertaken.
The research team delved into historical data to uncover the variables linked with predicting emergency medical admissions subsequent to the submission of a public protection referral involving the instigator of domestic violence.
Encompassing 8,709 residents in South Wales who were subjected to a public protection referral between 12/08/2015 and 31/03/2020, the study categorized participants into two groups: victims who encountered an emergency medical admission or fatality within a year of the public protection referral, and those who did not.
Employing the Secure Anonymised Information Linkage (SAIL) Databank, situated at the University, the study tapped into nationwide electronic health records spanning primary and secondary care. SAIL also houses administrative and demographic data, along with mortality records compiled by the Office of National Statistics. Notably, the study incorporated police-generated data to construct public protection referral records within the SAIL framework.
As a component of the research, the team additionally employed decision tree analysis to pinpoint the factors linked to the likelihood of a future emergency admission for the victim.
One year following the initial submission of the public protection referral, within the pool of 8,709 participants, a total of 3,544 victims underwent admissions to the Accident & Emergency (A&E) department, with a cumulative count of 48 fatalities.
Within a span of 12 months from the public protection referral, several pivotal elements were identified as linked to the likelihood of emergency medical admissions for the victim. These encompassed factors such as having visited emergency healthcare services over three times within three years, being either under 19 or above 70 years of age, engaging in cigarette smoking or having received smoking cessation advice (indicating addictive behavior), sustaining injuries at the scene, being prescribed central nervous system drugs or medications for treating infections, and the presence of household pregnancy.
Examining healthcare data spanning one to three years prior to the public protection referral revealed that the most significant risk indicator for victims prone to future emergency admissions was their past emergency admissions history. The victims were stratified into distinct risk groups: the highest-risk category comprised individuals familiar to emergency healthcare services, followed by those less recognized by emergency healthcare services but known to the police, and finally, individual’s unknown to both healthcare and police but identifiable by their General Practitioners (GPs).
Dr. Tash Kennedy, a lead researcher, remarked that the linkage between police and healthcare data has illuminated the detectability of highly vulnerable individuals across multiple healthcare datasets even before police intervention. The study has successfully identified and compiled a roster of pivotal risk factors associated with subsequent emergency medical admissions following the initial public protection referral response.
“This research shows that data sharing and linkage could help reduce callouts to the police and future emergency medical admissions.” Data analysis with the aid of AI has demonstrated remarkable capabilities, however data privacy concerns in relation the misuse of data has also been a serious concern among the public and civil liberties groups.
