Proper Temperature Monitoring Practices in the Clinical Lab

January-February 2015 - Vol.4 No. 1 - Page #14
Category: Temperature Monitoring

Q&A with
Franklyn Garland, MT(ASCP)BB

Medical Lab Management: From an information systems perspective, what value does an automated temperature monitoring system provide to hospital operations?
Franklyn Garland: Temperature monitoring systems are used by numerous departments in a hospital outside the common-use areas of laboratory and pharmacy. Practice areas, including nursing, surgical services, food services, and patient care services, all have a need to monitor the state (eg, temperature, status, CO2 level, etc) of temperature-dependent storage devices. Therefore, perhaps the greatest single consideration is that by adopting an enterprise-wide solution, the hospital can unify all these disparate areas under a single application. Such enterprise-wide solutions tend to reduce the scope of maintenance and support necessary for regulatory compliance and patient safety, as well as render the processes of temperature and environment monitoring more cost-effective. Most notably, proactive monitoring of temperature-dependent storage devices can help prevent the potentially significant financial costs associated with lost inventory when a storage device fails. 

MLM: What are some of the most important factors relative to laboratory operations when researching automated temperature monitoring systems?
Garland: Fortunately, there are many different systems available on the market now, and, with few exceptions, they are all sufficiently capable of meeting regulatory requirements (eg, Joint Commission, CAP, AABB, etc). Most of these systems also can be calibrated, with some systems using proprietary sensors and others employing more generic hardware. The main advantage of generic hardware is that the cost of new sensors often is greatly reduced. That being said, the following characteristics are hallmarks of quality systems:

  • The system must permit the definition of selected groups such that different areas of the lab can be segregated with alarm notifications limited to the areas that directly rely upon the devices being monitored. This should include available configurations for both local alerts (beeping directly at the source) and distributable alerts via email, pager, or cell phone notification. These alerts, in turn, should be scalable in the event a primary contact is unavailable (as is often the case during off hours). 
  • The system should be end-user serviceable so that the costs of maintenance or expansion can be mitigated. 
  • The system must be accurate (+/- 1°C or less, depending upon applicable standards).
  • Data should be easy to query and export for reporting, monitoring, and compliance purposes.
  • System redundancy is necessary to prevent loss of data during power outages and/or network failures. 
  • Remote system access also is desirable in the event primary contacts are unavailable.

All systems being roughly equivalent in their base functionality, the factors that should influence acquisition boil down to the system’s ability to be expanded at an affordable cost, its overall reliability including built-in redundancies, and the relative responsiveness of the vendor. A responsive vendor with expeditious and comprehensive service often is more important than the core hardware of the chosen system.

MLM: What type of monitoring system is used at your facility? 
Garland: At Northwestern Memorial Hospital (NMH), we use a hard-wired, centralized system that employs a Web-based application. It has been in place here about a decade and is one of several systems we continue to use. We also utilize a variety of system-compatible probes throughout the facility that serve units ranging from -196°C freezers to +37°C incubators, as well as humidity and door-ajar sensors.

MLM: How should probes be positioned within temperature control devices to maximize their utility? 
Garland: Whenever possible, remote probes should be placed in close physical proximity to the sensor that runs the digital display to harmonize the two readings. Furthermore, all probes should be calibrated at least annually so they read +/- 1°C between the remote probe and the local display. This helps to reduce confusion at the end-user level. 

At NMH, we place monitoring sensors inside of aluminum blocks, which respond to temperature changes much like commonly used 10% glycerol solutions; however, aluminum blocks do not spill or dry out, and the lack of liquid eliminates the chance of a liquid leak shorting out a sensor. Regardless, without an aluminum block or glycerol bottle to hold and secure the sensor within refrigerators and freezers, the sensor will respond to rapid changes in air temperature (eg, whenever the compressor turns on) and not the true storage temperature.

MLM: Who operates and maintains the temperature monitoring system at NMH? 
Garland: Each end-user group is responsible for their devices and can, if they so desire, bring in the vendor for service on a time and materials basis. At present, most areas coordinate for an annual recalibration of sensors. Some areas, such as the blood bank, do this themselves and do not contract out such routine maintenance to the vendor. Our system is largely end-user maintained, although our information services department pays for an annual support contract that covers the system’s server maintenance and telephone technical support.

MLM: What types of alerts are issued in the lab and how are they resolved?
Garland: In our clinical laboratory, temperature-control-related alerts are common (eg, door left ajar too long while doing inventory), so processes are in place to ensure that each alarm is properly managed and documented within the system. For standard refrigerators, the normal operating range is typically 1°C to 10°C, and in the blood bank, the normal range is 1.5°C to 5.5°C. This range refers to the programmed alarm range for the device, which should align with the normal operating specifications for the device. However, this may not always be the case: Consider outpatient clinics that store vaccines in a monitored refrigerator that is not medical-grade; disparities between the alarm range and the operating range may be common.

To reduce nuisance alarms, such as those that may occur during inventory, some areas of the lab have opted for a 15-minute delay on monitored inputs; an alarm triggers only if the temperature is out of range for more than 15 minutes. The blood bank, for example, has no alarm delays on refrigerators, but it does on freezers. In fact, all freezers at NMH incorporate at least a 15-minute delay. Otherwise, every time a freezer door is opened, the cold air rushing out and the disparity between freezer and room temperature would send the device immediately into alarm mode. Refrigerators do not suffer from this issue given the narrower differential between room and storage temperature. 

In addition, in the blood bank, our system permits warning limits and alarm limits. The system will emit a warning at 1.5°C and 5.5°C and an actual alarm at 1.0°C and 6.0°C. The FDA and AABB require blood to be stored between 1°C and 6°C, but the standard requires an alarm before the device moves outside of acceptable range (eg, 0.9°C or 6.1°C). By setting a warning range of 1.5°C and 5.5°C, blood bank staff is alerted when the storage temperature is within a half of a degree of the limits. This alarm protocol allows for a proactive response and helps maintain a compliant storage range. Once triggered, these alarms will repeat every 20 minutes until the input is either disabled or the alarm condition passes.

MLM: What is the process for responding to and resolving alerts?
Garland: Responding to alerts is a two-phase task. First, the recipient must listen to the telephone call that indicates which device is in alarm and why (this is a robotic text-to-speech voice call). The recipient must then acknowledge the call by pressing the pound (#) key at the end of the message. If the call is not acknowledged in this manner, the system will continue calling. The recipient then logs in to the system, acknowledges the alarm in the application, and documents the problem. There are standard responses to choose from, as well as an area for free text entry.

While our system can be programmed to notify different areas based on time of day, unfortunately, it is not currently weekend or holiday aware. However, this is an improvement we plan to adopt soon. Serving as a forcing function for proper documentation, the alarm items remain on the alert log until they are acknowledged within the system. The alarms also are color-coded, so active alarms can be determined at a glance. This process, combined with daily reports, helps to ensure all alarms are responded to and documented.

MLM: Have there been questions regarding temperature monitoring during regulatory inspections? 
Garland: Absolutely. I have spoken with inspectors about details of the system during virtually every inspection.
MLM: How does your temperature monitoring system impact operations? Garland: In general, automated monitoring has eliminated most of the daily, manual temperature logs that were easy targets for citation during inspection. As we have an extensive system of devices to monitor in addition to refrigerators and freezers—including dry blocks, water baths, room temperature, etc—all relevant information is located and retained in one place and is easy to query upon demand. This has made a previously tedious process much more manageable.

MLM: Are there any major maintenance issues or considerations for dedicated power during power outages? 
Garland: Dedicated emergency power, if available, is beneficial. Many systems, including ours, have built-in backup batteries and will continue to record temperatures during a power outage. Remote sensors for blood products and critical items are connected to emergency power.

MLM: Do you have any recommendations for other labs looking to automate their temperature monitoring? 
Garland: Automated temperature and environment monitoring systems can provide significant time savings and improve efficiency and compliance, but the lab must own the system. There is an unfortunate, yet common, belief that once system automation is implemented, monitoring and maintenance are no longer required; this is simply not the case. For example, CAP standards require daily documentation of proper system functionality, so the system cannot be ignored. With a manual system, there is a risk that—rather than investigate, correct, and document issues—a staff member will simply disable the alarm citing annoyance and then ignore it. Such actions can easily result in significant material losses should the monitored device fail and no one is alerted. 

Furthermore, a centralized system makes it easier to comprehensively monitor numerous devices in multiple areas, but this requires calculated management by relevant and invested personnel. It is never a good idea to have alarms forwarded to staff members that have no access to, or knowledge of, the contents of a monitored device. Likewise, in the event of a device failure or breakdown, the lab specialty responsible for the contents of the device should always be designated to assess the alarm, relocate the contents if necessary, and then call facility engineering or the vendor to fix the device.

Franklyn D. Garland, MT(ASCP)BB, is a graduate of Benedictine University and has been employed at Northwestern Memorial Hospital in Chicago for more than 25 years in a variety of roles, including more than fifteen years in the blood bank with nearly ten of those years as the blood bank compliance coordinator. At present, Franklyn serves as Northwestern Memorial Information Services’ senior application technical analyst, where he is responsible for the centralized monitoring system used by the laboratory.


Like what you've read? Please log in or create a free account to enjoy more of what has to offer.

Current Issue