Challenges with alarm notification and fatigue have plagued the health care industry for decades. Long before alarm notification systems like Emergin (now Philips IntelliSpace Event Management) and GlobeStar Systems (ConnexAll) appeared, some hospitals addressed alarm issues with the original alarm notification system, monitoring techs. Monitoring techs remain an accepted and effective tool in the constant battle to reduce alarm fatigue and avoid failure-to-rescue events.
With the growing adoption of electronic alarm notification systems, is there still a role for monitoring techs? Are electronic alarm notification systems superior to flesh and bone monitoring techs? This blog post will explore monitoring techs as a solution and consider whether they might be a compliment to an alarm notification system, or whether an alarm notification system should take the place of monitor techs.
Role of Monitoring Techs
The role of monitoring techs is to observe waveforms and numeric data from monitored patients. Monitoring techs are also often aware of the patient's clinical context, such as medications, that can influence resulting waveforms or readings. The key goals are two fold: to screen out nuisance alarms so that caregivers only have to respond to "real" alarms, and to ensure that alarms receive a timely and adequate response from caregivers.
Good alarms are those that are clinically significant and elicit an intervention from caregivers. These alarms are often referred to as "actionable" alarms. Nuisance alarms are "unactionable" alarms because the device and/or the patient's condition does not warrant an intervention by the caregiver. Unactionable alarms result from poorly set default alarm limits, false positive alarms (e.g., arrhythmia or other types of alarms generated from motion artifact), and transient alarms where parameters exceed alarm thresholds for brief periods and then fall back to normal ranges.
In addition to providing vigilance for monitored patients, monitoring techs often are responsible for changing batteries on telemetry monitors, and can even be responsible for applying and removing patient monitors, and managing patient attached sensors (they are certainly the first ones to see artifacts attributable to poorly applied or "bad" sensors).
Depending on the technology used to notify caregivers of alarms, monitoring techs may have to keep track of the nurse to patient assignments on each shift they are supporting. Even when notification via multiple "bat phones" or a call to the nursing station is used, it is helpful for monitoring techs to know which caregiver is responsible for the alarming patient. This knowledge can help ensure timely alarm responses.
Monitoring Tech Solutions
The components of a monitoring tech solution starts with a workspace for the monitoring techs and their central station displays. These central stations display the key waveforms and numerics for each monitored patient. A communications system is also required so monitoring techs can notify the caregiver for a timely alarm response. Communications systems can range from a series "bat phones" located throughout the unit and dedicated to calls from monitoring techs, to Vocera badges, wireless voice over IP handsets or smartphones using text or voice communications.
Monitoring techs can either be located in the unit with monitored patients, or centrally located where they monitor patients remotely. Sometimes called "war rooms," these central monitoring departments can observe patients throughout a single hospital and scale up to provide surveillance for monitored patients in multiple hospitals throughout a metropolitan area.
Monitoring Tech Advantages
Properly conceived and managed monitoring tech initiatives can be very good at reducing or eliminating alarm fatigue by screening out alarms that are not actionable. Such initiatives can also ensure timely and adequate response to alarms, minimizing or eliminating failure-to-rescue events with monitored patients. While this list of advantages is short, they are among the most important in the overall mission of a hospital. Secondary benefits from the foregoing advantages include increased staff and patient satisfaction due to reduced alarm fatigue and a more quiet care environment.
Monitoring Tech Challenges
There really aren't any inherent disadvantages to monitoring techs (except perhaps cost), but this approach is not without its challenges.
The use of monitoring techs can mask shortcomings in the clinical practice of alarms. This clinical practice includes:
- The purchase and use of sensors of sufficient quality as to minimize or eliminate sensor associated noise and artifact (that often results in false/positive alarms;
- Proper skin prep and application of sensors to minimize or eliminate sensor associated noise and artifact;
- The timely replacement of sensors on the patient to ensure continued sensor performance;
- The designation of default alarm limits that minimize nuisance (unactionable) alarms resulting in a high percentage of actionable alarms; and
- The policy and procedures to ensure the proper adjustment of alarm limits on a patient specific basis to minimize nuisance alarms.
Nuisance alarms can never be completely eliminated, even with top notch clinical practice of alarms. The alarm items above have a substantial impact on the ratio of nuisance to actionable alarms. Poor alarm clinical practice can be masked by good monitoring techs who will screen out all those nuisance alarms. But it should be kept in mind that alarm fatigue can impact monitoring techs as much as direct caregivers. For maximum effectiveness and reliability, adequate alarm clinical practice is a requirement.
Hospitals who have deployed monitoring techs must turn down alarm volumes in the unit and rely on their monitoring techs, otherwise the key advantages of monitoring techs will not be realized. The "belt and suspenders" approach - using monitoring techs and keeping alarm volumes loud enough to broadcast across the unit, just in case - will not reduce alarm fatigue or improve patient satisfaction.
Monitoring techs' jobs are challenging. Besides the alarm fatigue risk mentioned above, the job offers a mix of chaotic stress-filled periods combined with extended periods where nothing happens and it become difficult to maintain adequate vigilance. Due to the stress, staff turn over can be an issue. Monitoring techs are required to complete an 8 to 12 week EKG recognition course. Typical positions require a high school diploma and it is not unusual for positions to be filled by paramedics and those with other technical medical experience. In some hospital markets it can be difficult to fill monitoring tech positions.
The number of patients that can be monitored or observed by a single monitoring tech can vary from 24 to 60. There is no standard or widely accepted "best practice" for the ratio of patient to monitor techs. The number of patients watched by a monitoring tech is influenced by the skill and experience of the individual techs, along with unit census. Patient acuity or their unique physiological condition can also drive the ratio of patients to monitoring tech. Due to an unknown perverse confluence of patient condition and arrhythmia analysis algorithms, an occasional patient will generate almost continual false/positive arrhythmia alarms. Finally, much of this variability depends on the patient monitoring system in use and how many patient's can be displayed and viewed at a workstation. Workstations typically have from 4 to 6 displays.
Another challenge of monitoring tech solutions is management data. To manage something, it must be measured, and there is plenty to be measured and recorded when it comes to alarm generation, notification and response. Patient monitoring log files can provide some data on alarms generated (the ease of access and user friendliness of the log files in patient monitoring systems can range from impossible to barely understandable with a decoder ring). Any data on screened alarms (nuisance or false positive vs "good" alarms) and caregiver response times for alarm calls, must be manually logged during the shift to compile a historical record. Manual record keeping like this is time intensive, prone to human error, and requires considerable effort for a consistent long term execution.
The costs of a monitoring tech solution are mostly ongoing operating costs. For a 400 bed hospital costs will be in the low seven figures per year. Capital costs include remote patient monitoring central stations needed to cover the maximum number of monitored patient and for use by each monitoring tech. Based on monitoring tech costs, the return on investment for most alarm notification systems is one to two years.
Monitoring Techs and Alarm Notification Systems
It's hard to beat the nuisance alarm screening capabilities of monitoring techs. I suppose that some day alarm notification systems will have artificial intelligence or clinical decision support sufficient to be able to screen nuisance alarms. Maybe. The present day best practice solution is to implement robust alarm clinical practice (whether for monitoring techs or an alarm notification system) to minimize nuisance alarms. For an alarm notification system to fully meet this screening requirement, the caregiver must be presented with the clinical context of an alarm, often a sample waveform, so that they can do the screening themselves. Without this contextual data associated with the alarm, the caregiver is forced to run to the bedside to screen those nuisance alarms, potentially contributing to alarm fatigue.
When it comes to the routing of alarms to the proper caregiver and ensuring an adequate alarm response, it's hard to beat an alarm notification system. Both routing and escalation are tasks easily automated by today's software applications. Sure, monitoring techs can route and escalate alarms, ensuring a proper response, but at a substantial operating cost.
And when it comes to capturing quantitative data on alarm generation, screening and response the clear winner is the alarm notification system. Alarm notification systems capture alarm events, messaging events, escalation and response events - basically everything from the medical device to the resolution of the alarm is captured. How accessible that information is and whether it's presented in a meaningful way for reporting purposes varies from system to system.
A Final Wrinkle
Patient flow and the need to monitor a wider variety of patients in growing numbers have been impacting hospitals for many years. In response, a lot of hospitals have increased the number of telemetry monitors to cover many patients in addition to those found in their telemetry unit. Yes, there are some patients on other services (say oncology or orthopedics) with cardiac or arrhythmia conditions that need telemetry outside of the telemetry unit. However many of the patients outside the telemetry unit are being monitored with telemetry packs because they are at risk of a deteriorating clinical condition and the telemetry pack is relatively inexpensive and patient worn. I would guess half the telemetry packs sold are used on patients of this type, rather than cardiology patients.
Using telemetry packs to monitor non cardiac patients is a poor match to the application. Patients with a deteriorating clinical condition are usually awfully close to cardiac or respiratory arrest before they start throwing arrhythmias. The good news is that there are a growing number of low acuity monitors better suited for this application. For those many telemetry packs still being "misused," this creates a false demand for monitoring techs in their most powerful role of screening out nuisance alarms.
Hospitals in this situation may find that good clinical alarm practice will reduce nuisance alarms sufficiently to obviate the need for monitoring techs. Some may not. The role of monitor techs in these hospital should be determined on a case by case basis.
Tim - as usual, a very well thought out, comprehensive post on a topic that is growing in importance.
The use of “monitoring techs” has a sort of built-in limitation in the sense that monitoring techs are all about just one class of devices and systems that produces alarms and contributes to alarm fatigue - which is of course patient monitors. And the heart of the issue is nursing interruptions that come not only from monitors, but from many different sources including other medical devices (especially ventilators and IV pumps), but also nurse call, EMR’s and ancillary lab systems.
The growing role of alarm notification systems (i.e. alarm middleware) takes us down the path where hospitals are starting to realize that they need to get their arms around all alarms and interruptions. Do you think an expanded concept of “alarm surveillance techs” would hold together when these “alarm techs” would have to get trained on a much broader set of medical devices, and the “alarm war room” would have to be enhanced to include alarm surveillance for all medical device alarms? I don’t know the answer, but it sure seems like it would be a challenge. What are your thoughts about this?
Brian, you are correct that there’s a lot more that alarm notification systems (i.e., messaging middleware) can do besides handle patient monitor alarms. In fact, it would not be unreasonable (if expensive) for a hospital to use and benefit from messaging middleware and still have monitoring techs.
Alarm notification systems, especially when used for more than just alarms, are more cost effective than monitoring techs. Neither solution is perfect and both require real management focus to ensure continued optimal operation. That said, I expect that over time, alarm notification systems will displace monitoring techs.
This morning I received the following comment via email from Maria Cvach, Asst Director of Nursing at Johns Hopkins.
A few thoughts regarding the key disadvantage to a monitor tech:
a. Cost; You need 4.8 FTEs to run a single monitor station which can observe anywhere from 24-60 monitors.
b. Finding qualified people to man these stations is very difficult in Baltimore. The money they expect to receive for this work is disproportionate to what we pay our clinical technicians, thus people don’t want to take these jobs. The graduates from ECG tech programs expect to receive much more than hospitals are willing to pay.
c. When you put monitor interpretation in the hands of a monitor tech, the nursing staff can become indifferent about alarms and may not act appropriately or even know exactly how to act when they are called by the monitor tech.
d. You are assuming that monitor techs are in the hospital and can change batteries and do lead placement; remote sites are often not on site or are not manned to do battery or lead placement replacement.
e. When a problem occurs, it is hard to get the nursing staff to act without a reliable notification system; hospitals need to have that in place.
f. Most importantly, there are no studies that indicate that monitor watch actually improves patient outcomes, other than what Marge Funk did in the 90s. Even then, she only found limited benefit related to VT alarms. For hospitals to invest in that much cost without well designed studies to prove its worth, is perhaps the biggest disadvantage. Cost avoidance is not good enough.
Item A; I’m watching 70 plus patient a Day. I’ve been @ this 30 years & make $18.65 hour if you figure in all my benefits per hour. New techs here @ AHS/Florida start @ $10.50 an hour. I’ve saved may lives per $$$.
Item B; of course it’s disproportionate… we’re patients guardian angels, if they get by us, it could be over.
Item C; That’s a fact, that’s why monitor techs should be on the floors with their patients and team members, not off in a bunker somewhere.
Item E; Another reason techs should be on their floors they are watching. Do you know how many sentinel events I’ve responded to myself… because no one would answer my calls or they were not even around or BUSY?
Very interesting post. We in the clinical engineering business call those telemetry ‘cockpits’ due to their similarity with the way a pilot would see their working environment. Overloading a person with a lot of information makes it difficult for them to be able to detect issues and/or quickly respond to them (in technical terms the display hinders cognitive processing). I am a private pilot and when any information overload occurs, it’s the checklist and basics (aviate, navigate, communicate). All else becomes secondary or tertiary.
With regard to the proliferation of telemetry, this is another example of the hope that technology will help to manage more patients with fewer personnel. In some cases that works and in some cases it doesn’t. I don’t know if the current trend of wanting to monitor everyone-all-the-time is going to improve the quality of healthcare. I do know that it means there is a *lot* more data available that will be amassing and needs to be either archived as interesting for research purposes or assessed quickly and acted upon. They key is what data is actionable and how to easily determine that.
One issue I have is in the display of the information. It is ‘engineering-centric’, i.e. we are forcing clinicians to be able to indirectly infer what is going on physiologically with the shape of the waveform and/or a series of numbers. The UI really needs to get better on this and the blog post I did on just that type of UI that Dr Moorman and the University of VA are doing is where the industry and healthcare needs to go. It does require good data and a robust model and based on some of my research and understanding we are nearly there.
I agree that cost avoidance is not the best reason to insert technology into a clinical workflow. Unfortunately, and especially in the USA, that tends to be one of the predominate solutions chosen. It has been interesting to me as I consult in both the EU and USA to see the different ways in which they are trying to manage the aging of the population, the concomitant increase in the use of healthcare resources and the constraints on those resources. Rarely have I seen any facility in the EU have wall-to-wall telemetry. Additionally, the overwhelming use of EHRs is not prevalent in the EU. This has driven them to use technology differently.
Regardless, your point about using middleware and the implied simple clinical decision support tools to help alleviate the information overload issue is a fair one. However, as @Douglas Spring noted, even if we get better at the tools, we still need to follow-through with the ‘action’ part to the ‘actionable’ information.