A look at how AI-powered algorithms can help curb the threats posed by increased incidents of workplace violence in hospitals and clinicsÂ
By Matt Powell
For as long as video surveillance systems have been leveraged by public and private sector organizations, there have been a multitude of regulations issued by federal, state and local authorities. These regulations govern not only where and how the technology can be used, but also the circumstances under which it is a requirement of doing business.
From cities mandating convenience stores to install cameras and retain footage for a predetermined length of time to aid in crime prevention and law enforcement efforts to casinos monitoring gaming machines and tables round-the-clock to detect cheating, video surveillance has become a part of business backbone infrastructure solutions as much as IT networks and business continuity systems for companies today.
However, unlike the typical analog deployments of 30 years ago, contemporary video solutions deliver a wealth of data to end users that not only satisfy basic security requirements, but also enable businesses to operate more efficiently. The introduction of artificial intelligence (AI) and machine learning technologies into the video space are powering a new generation of video analytics. These evolving technologies are helping organizations become more proactive when it comes to mitigating against various threats as well as meeting regulatory requirements specific to certain industries with healthcare being chief among them.
An unfortunate but well-known fact about the healthcare industry nationwide today is that doctors, nurses and other clinical staff face an outsized threat of violence relative to other professions. In fact, according to data published by the U.S. Bureau of Labor Statistics, healthcare workers accounted for 73% of all nonfatal workplace injuries and illnesses due to violence in 2018.
Many states have enacted legislation that stiffens penalties for those found guilty of perpetrating acts of violence against healthcare staff; however, the problem has become so pervasive that federal lawmakers have also been spurred to action.
Federal Legislation Introduced
Last year, the U.S. House of Representatives passed a bill dubbed, the “Workplace Violence Prevention for Health Care and Social Service Workers Act,” that would require healthcare providers to formally adopt and implement a workplace violence prevention plan.
Aside from developing policies and procedures for responding to and reporting acts of violence in hospitals and clinics, the legislation also outlines various mitigation measures that could be implemented by these facilities, which includes video surveillance among a slew of other physical security controls. The bill has yet to be voted on in the senate, but it underscores the importance of having solutions in place that are more preventative rather than reactive in nature.
The search and archiving capabilities of traditional video management systems are generally adequate for addressing issues related to liability claims, theft and other types of standard security events. However, unless a camera network is being actively monitored 24 hours a day by highly trained operators, it is only by having an analytics layer on top of these systems that a healthcare provider could have advanced warning of a potentially violent incident on their premises.
With algorithms that are trained on thousands of hours of human behavior, today’s video analytics can detect a wide range of violent actions – punching, kicking, etc. – and alert security personnel, who can subsequently respond and contain the perpetrator before they can harm additional staff.
The Role of Facial Recognition
Some facial analytic modules available today allow hospitals to place photos of known offenders into a database so that their security teams can be alerted when their presence is detected on campus. Like many school districts, some children’s hospitals have implemented facial recognition technology to be able to block convicted sex offenders from gaining access to their facilities.
Of course, facial recognition technology can be used for much more than detecting the presence of previously violent patients/visitors or sexual predators. The technology is also capable of being leveraged to provide more robust access control at points of ingress and egress as well as other sensitive areas, such as where prescription narcotics are stored and dispensed. In this scenario, the facial analytics are utilized as a credential as opposed to cards or other technologies which can be easily misused. This comprehensive method is highly secure and creates a video record of the access event, which cards or other hardware-based solutions are not able to provide.
Though thefts of opioids and other controlled substances by outsiders is rare, the costs of failing to adequately protect prescription drugs – be it from a patient or visitor with a stolen credential or a malicious insider- can be quite substantial. In 2015, for example, Massachusetts General Hospital agreed to pay a $2.3 million settlement for what the federal government alleged were lax control that allowed employees to steal drugs from the facility.
Because facial recognition links authorized users with biometric data, the chances of an outsider or a malicious insider abusing their access privileges to divert narcotics becomes much more remote. Additionally, if a breach occurs, there is video evidence of the misuse of access privileges, allowing for rapid investigation and policy adjustment if required.
COVID-19Leads to New Analytic Needs
With the onset of the COVID-19 pandemic in early 2020, many healthcare providers were confronted with the need to enforce a variety of mandates to ensure the spread of the virus was mitigated to the greatest extent possible. Among these was the requirement that patients, visitors and staff wear masks, which largely fell on employees to enforce on a person-by-person basis.
This unsurprisingly led to several violent confrontations across the country given the controversy surrounding the issue. However, video analytics have also found a critical role to play here as mask detection became a clear need in the early days of the pandemic as well as throughout the various subvariant surges.
With an analytic solution like mask detection deployed, healthcare providers can immediately determine those who are not wearing them and respond appropriately, even restricting access to the facility if necessary.
As we learn the lessons of the COVID-19 pandemic and pull these lessons learned forward to be better prepared for another pandemic, analytics can continue to play an important role. For example, utilizing advanced tracking analytics, exposure risks can be quickly identified. Once an exposure is identified, analytics in a healthcare facility can be used to determine who came into contact with the infected person so that notifications of exposure can be made rapidly.
Whether it is addressing longstanding concerns like workplace violence or meeting new challenges, such as those presented by COVID-19, video analytics provide healthcare organizations with a proven solution to mitigate risks and improve their overall security posture. And just like all other technologies, analytics will only continue to improve over time, giving security teams the tools they need to adapt to an ever-evolving threat environment.
About the Author:
Matt Powell is Managing Director for North America at ISS (Intelligent Security Systems), a pioneer and leader in the development of video intelligence and data awareness solutions. He has over two decades of experience in security and transportation technologies having formerly served as Principal-Infrastructure Markets at systems integrator Convergint Technologies and as a developer of ITS/DoT market strategies for Videolarm and Moog prior to that. He can be reached at matt@issivs.com.