The Effect of Antivirus Software on the Laboratory Information System
Liron Pantanowitz ; Baystate Medical Center/Tufts University School of Medicine;
Content:
Security is a key feature of any laboratory information system (LIS). The Clinical and Laboratory Standards Institute recommend that laboratory computer systems be monitored and protected against malicious intrusions (e.g., spyware, Trojan horses, cookies, worms, viruses, and root kits) and attacks. With computers connected to the Internet forming an integral component of the modern LIS, the use of virus protection software is recommended. Antivirus programs perform audits (scans) of applications and data files on hard disks for viruses (e.g. looking for viral signatures) and remove any that are found. It is also suggested that antivirus software be installed and run daily on servers and all end-user devices that access laboratory data. The impact of antivirus software on the LIS has, to the best of our knowledge, not been investigated. Therefore, the aim of this study was to determine the effect antivirus software has on LIS performance.
Technology:
Anatomical Pathology-LIS (CoPath Plus version 2.5, Cerner), Servers (database server Compaq ProLiant ML570; interface server Compaq ProLiant DL360), Enterprise antivirus and antispyware software (McAffee version 8.5.0i), Application performance monitor (Citrix EdgeSight).
Design:
Our Anatomical Pathology-LIS was installed and operating on an array of networked database and interface servers. Each server had antivirus and antispyware software installed that was continuously running scans involving all folders. Edgesight software was installed to collect data (errors, central processing unit usage, network activity) for trend analysis in order to monitor performance as experienced by the end-user.
Results:
Increased network system activity spikes on our servers were detected that corresponded to antivirus and antispyware scanning activity. These activity spikes resulted in decreased LIS bidirectional data transmission over our network that accounted for LIS end-user feedback of delays (so-called episodes of hanging ). Excluding LIS application folders (i.e. .exe folders) in the directory from antivirus scans alleviated this problem.
Conclusion:
Antivirus and antispyware software programs may affect LIS activity resulting in diminished performance. In order to avoid this problem we recommend excluding application folders during antivirus scans. Utilizing performance monitoring software to view LIS utilization and performance will allow such issues to be proactively addressed.
