Perspectives

I recently attended a conference hosted by the Association for the Advancement of Medical Instrumentation (AAMI).  It was a follow-up to the October 2011 Infusion Device Summit hosted by the FDA and AAMI which led to the formation of 10 working groups, each focused on  a different aspect of infusion safety.  The primary purpose of this meeting was to provide a forum for each working group to report on progress and to develop agendas for the next 12-18 months.

The working group I am involved with is focused on alarms management.  In an earlier blog entry, I discussed the growing national focus on medical device alarms and alerts. For this meeting, I was asked to share my perspective on alarms and alerts for IV infusion pumps.   For hospitals with the Alaris® System and wireless connectivity, CareFusion is able to capture alarm and alert data from a large number of hospitals who have implemented wireless connectivity capabilities.  In preparation for this meeting, I reviewed data on 7.6 million alarms and 5 million alerts from 131 hospitals, covering 7.5 million infusions and 2.5 million infusion pump days.

Prior to the meeting, I received an email from the chairman of the alarms management working group describing his experience with IV infusions during a series of ten recent hospitalizations. Before I go on, there are three things you should know about him:

1)  He is an engineer.

2)  His career has been focused on monitoring devices.

 3) Prior to his involvement in this working group, he had limited knowledge about infusion pumps and pump alarms and alerts.

Because of his interest in device alarms, and knowing he would be on an infusion pump for each treatment, he asked the hospital to also attach him to a multi-parameter monitor.  This monitoring was not required for his treatment, but he wanted to gain firsthand experience with the pump and the monitor in simultaneous use.

As a refresher, we define “alarms” as events that were not anticipated, compared to “alerts” which are programmed to occur at specific points, such as the end of a programmed infusion. During his treatment, he witnessed a series of four occlusion alarms (one after the other during the same hospitalization), one alarm for an air bubble detected below the pump, and ten alerts that signaled the infusions were complete. The series of four pump occlusions also detected the clotting off of his peripheral IV catheter, ultimately resulting in replacement of the IV access. There were also two visual alerts that provided safety “guardrails” to reprogram dosage.   In stark contrast, the multi-parameter monitoring device alarmed repeatedly, nearly 25 times more than the pump, often interrupting his sleep.  The multi-parameter monitor tracked oxygen saturation, heart rate, and blood pressure measurements.  Again, it is important to restate that his infusion therapy did not require this level of monitoring and was not expected to cause changes in the monitored parameters.  For our chairman, who had spent much of his career focusing on monitoring devices, this experience served as an epiphany of sorts as it relates to understanding the significance of infusion pump alarms.

So, how does this data from ten infusions compare to the 7.6 million alarms and 5,000,000 alerts (audible only) captured from 131 hospitals over several months. The results were surprisingly consistent.  Some highlights:   

• We reviewed 7,596,345 infusions with total of 13,169,210 alerts and alarms from July 1 to Dec. 31, 2011. 
• Each infusion averaged 1.73 alerts and alarms.
• 30 percent of alarms occurred while the caregiver was at the bedside.
• For each infusion pump, there was an average of 5.21 alarms and alerts per day.
• 71 percent of infusions had no alarms, and 38 percent of infusions had no alerts.
• 30 percent of the alarms occurred while the caregiver was at the bedside (some related to potentially unsafe actions such as a free flow condition, door open, etc.).
• More than 40 percent of alerts did not require a trip to the bedside.
• Infusions that had alarms typically had more than one.
• The most common alarm was occlusion below the pump, and the second most common was detection of air.
• The most common alert was to signal the end of an infusion.

So, one patient experienced four alarms in one hospitalization that were true occlusion alarms, one true  air alarm in a second hospitalization, and ten alerts – one per infusion – signaling that the infusion was completed.  This data is pretty consistent with the millions of other infusions we reviewed.

Now that comprehensive infusion data can be collected wirelessly, we are beginning to paint a picture of the true incidence of pump alarms and alerts.  Looking to the future, it may be possible to use this retrospective alarm data to begin to reduce pump alarms and some of the alerts through better matching of pump performance settings to clinical care workflow, rethinking certain long-standing practices and copying some of the monitoring innovation with self-cancelling alarms.  Infusion pump dose error reduction systems (DERS) include data capture on alerts and subsequent actions by the caregiver.  This treasure trove of new information has led to many new discoveries and safety enhancements.  I believe that what the DERS alert logs did for making infusion pumps smarter and safer will be repeated for alarms and alerts.

As seen on SurgicalProducts.com

Anesthesia professionals in the operating room have a unique role and responsibility in that they are the only medical personnel who prescribe, secure, prepare, administer, and document medications—a process that can require more than 40 steps—usually within a very short time interval. While advances in safer medication therapy, such as bar code medication administration (BCMA), have been shown to help nurses administer IV medications, many of these systems do not fit the workflow for anesthesiologists.  The checks and balances that function well outside the OR may not  exist in anesthesia practice.  Although published evidence of medication errors in the OR is limited, there have been some studies that demonstrate error occurrence and risk. Consider the following:

• In 1984, medication errors were reported as a leading cause of adverse events during anesthesia.¹
• In 2001, one drug administration error was reported for every 133 anesthetics administered.²
• More recently, the United States Pharmacopeia’s MEDMARX database identified nearly 3,300 medication errors in U.S. operating rooms from 1998 to 2005.At a recent Anesthesiology Patient Safety Foundation (ASPF) board of directors workshop, a high percentage of anesthesiologists indicated they or a colleague had been involved in a serious adverse drug event. Sixty percent of respondents acknowledged that among a list of safety initiatives, safer medication practices was the most likely to increase safety in the OR.  In addition, there was close to unanimous support for drug standardization and premix/prefilled medications. 

Since its founding over 25 years ago, the Anesthesia Patient Safety Foundation (APSF) has focused on implementing safe anesthesia practices. In response to growing awareness of medication errors in the OR, the APSF has advocated a new campaign, “Medication Safety in the Operating Room – Time for a New Paradigm.”  In much the same way that improvements in aviation safety practices have advanced safe air travel, the APSF is advocating for a team approach that includes four pillars – standardization, technology, pharmacy, and culture. 

As an APSF board member and a pharmacist involved in medication safety, I am excited to announce the release of a new video that focuses on improving medication safety in the OR.  This video discusses the medication safety issue, discuses the four pillars and provides examples of model practices initiated by anesthesia departments.  Distribution of this video is made possible through an unrestricted grant from CareFusion. You can also download a similar video by APSF and the ECRI Institute focusing on preventing OR fires. 

1. Cooper JB, Newbower RS, Kitz RJ. An analysis of major errors and equipment failures in anesthesia management: considerations for prevention and detection. Anesthesiology. 1984;60:34-42.

2. Webster CS, Merry AF, Larson L, McGrath KA, Weller J. The frequency and nature of drug administration errors during anesthesia. Anaesth Intensive Care. 2001;29:494-500.

As a kid, I spent much of my time playing outside in a forest near my home. I recall my parents warning me that the stream running through the woods was off limits. Void of any living creatures, the stream was clearly contaminated – the result of untreated water and pollutants discharged by a number of nearby factories. Many of us have heard stories of rivers and lakes that were so polluted they caught fire. Over the years, as factories and corporations adopted safer and more responsible environmental practices – enforced by strict state and federal regulations – I came to believe that we could once more trust our water sources. That bubble burst last month when I attended a lecture by Dr. Fred Massoomi at the ASHP MidYear Annual Meeting.  Dr Massoomi is the Pharmacy Operations Coordinator at Nebraska Methodist Hospital and an expert on a different type of waste now polluting our waterways – pharmaceutical waste. His presentation included statistics that were sobering to say the least:

• Health facilities flush more than 250 million pounds of drugs each year.
• A 2002 U.S. geological survey found pharmaceutical waste, including antibiotics, mood stabilizers and reproductive hormones in 80 percent of sampled water streams.

As he spoke about the problems, I started thinking back to my days as a pharmacy resident and later as a satellite pharmacy supervisor, and I realized that pharmacists could be as guilty as the factory managers who polluted the stream in my home town. Every drug that we wasted went into the sink or into the trash. I’ve since read a 2008 Associated Press investigation that found a vast array of pharmaceutical trace compounds in drinking water for 41 million Americans in 24 major metropolitan areas.

This is not a new problem, and while waste management regulations have existed for hospitals for some time, only recently has the Environmental Protection Agency (EPA) and state health departments started to enforce the regulations. Still, it is estimated that as many as 60 percent of U.S. hospitals have not implemented comprehensive pharmaceutical waste disposal programs, putting them at risk of compliance infractions that carry significant financial penalties. 

Later this month, the CareFusion Center for Safety and Clinical Excellence will host a continuing education webcast to help educate pharmacists and pharmacy technicians on how to implement pharmaceutical waste management solutions.  In addition to Dr. Massoomi, the session faculty will include Lisa Lauer from the EPA who will provide background on medical waste management regulations. 

Please join us January 20 at Noon Eastern Time (9 a.m. PST).  You can RSVP online to participate or look for a replay here following the session.

I recently had an opportunity to hear a presentation by James Orlikoff who spoke on healthcare reform at one of our annual meetings. Mr. Orlikoff is a senior consultant to the Center for Healthcare Governance and the national advisor on governance and leadership to the American Hospital Association and Health Forum. During his talk, he shared a slide titled “What Fueled the Reform Fire,” on which he referenced the “Gawande/McAllen Effect.” For those who aren’t familiar with this concept, the name comes from an op-ed piece published in the June 1, 2009, issue of The New Yorker magazine, written by Atul Gawande, a general surgeon at the Brigham & Women’s Hospital.  Dr. Gawande is one of the primary movers behind the use of checklists to promote safety in operating rooms, which means this won’t be the last time you’ll see him referenced on this blog.

Dr. Gawande’s article, titled “The Cost Conundrum” encapsulates what I feel is one of the most complex subjects in healthcare today – changing reimbursement mechanisms for U.S. hospitals. His story centers on McAllen, Texas, a border town near the southernmost tip of the state and one of the biggest national spenders on healthcare. Gawande notes:

“Only Miami—which has much higher labor and living costs—spends more per person on healthcare. In 2006, Medicare spent fifteen thousand dollars per enrollee [in McAllen], almost twice the national average. The income per capita is twelve thousand dollars. In other words, Medicare spends three thousand dollars more per person here than the average person earns.”    

Well, there’s an attention grabber!  Despite the inflated healthcare costs, McAllen is an economically depressed area. The area is plagued by high unemployment, high rates of alcohol abuse and high rates of obesity – all of which might inflate healthcare costs.  Gawande points out, however, that El Paso County has nearly the same patient demography as Hidalgo County, where McAllen is located, but Medicare payments are far less, roughly $7,500 per enrollee. Further, despite fewer specialists in Hidalgo County, data from 2001-2005 shows that patients were 66 percent more likely to see ten (yes, ten) subspecialists in a six-month period.  They were also more likely to undergo “big ticket” procedures, such as nerve conduction studies, urine-flow studies, gallbladder operations, knee replacements, breast biopsies, echocardiograms, implantable defibrillators, cardiac-bypass operations and coronary stents. 

No one suggests this data represents misuse, just dramatic overuse. Along these lines, Gawande in his study outlined some factors to consider:

1. While increased access can and often does lead to increased costs, high tech medicine is not synonymous with high Medicare costs. Case in point: Olmsted County, Minn., is home to the Mayo Clinic – one of the most advanced healthcare systems in the country. Olmsted County’s Medicare spend, however, ranks among the lowest 15 percent in the nation.  
2. Healthcare spending – and variability in spending – is not widely publicized or recognized by healthcare administrators. For example, hospital executives in McAllen were legitimately surprised by these findings in their own hospitals. This data is further complicated by widespread variation in spending and reimbursement models among private insurers.

Let me make a couple of my own observations, as well: 

1. First, I think many people still have a problem accepting that higher cost does not necessarily imply higher quality when it comes to healthcare. It’s just not how we’re used to seeing the world. Consider your experience buying cars, real estate, clothing or food. The “you get what you pay for” philosophy seems so pervasive but it may not be true when it comes to healthcare services. For those in doubt, have a look at the report by Baicker and Chandra that shows the inverse relationship between Medicare spending and quality measures.
2. While reading The New Yorker article, I couldn’t help but think that if the healthcare professionals in McAllen had instead been automotive manufactures, their story could have  landed on the  cover of Business Week, if not Time magazine.  Is there some kind of fundamental disconnect related to the economics of healthcare? 

The plans set forth to reform healthcare in the U.S. represent a monumental shift in how we purchase, pay for and consume healthcare – and aim to help right what has become a seriously misaligned system. With the emergence of Accountable Care Organizations (ACOs), bundled payments and performance-based reimbursement through Value Based Purchasing, we’re beginning to witness and test new models of care delivery.  While these initiatives have many desirable characteristics, there is trepidation among service providers. Who can get this right?  Will the new models really work?

As Gawande states, “We can turn to insurers (whether public or private), which have proved repeatedly that they can’t do it. Or we can turn to the local medical communities, which have proved that they can.”  

Systems like Mayo Clinic, Kaiser, Geisinger and Intermountain Health may be important models for the future, but can a culture of “patients first and economics will follow” truly become embraced across the vast plane of the U.S. healthcare system?  If spending rates continue at their current pace, we’ll need to find out and find out fast.

I recently wrote about the Alarms Management Summit that was convened by the Association for Advancement of Medical Instrumentation, FDA and the Joint Commission.  While the primary discussion and focus was on monitoring alarms, we saw how infusion pumps, ventilators, beds and other devices all contributed to the high frequency of alarms, many of which are categorized as “false” or “nuisance” alarms.

This week, a Boston Globe report showed ventilator alarms linked with more than 100 patient deaths. With the exception of two instances of ventilator malfunction, all of the cases involved inappropriate alarm settings, failure to hear the alarms, or failure to respond in a timely manner.  In 2010 alone, approximately 800 ventilator-associated adverse events were reported to the FDA. 
With the recent publication of the top 10 hospital safety issues from the ECRI Institute listing medical device  alarms  as the top safety concern,  the discussions and data coming out of the October 2011 Alarms Summit, and now this week’s Boston Globe report, there is little doubt that a major focus for the coming years will be on device alarm and alert safety, in particular:

• Strategies to reduce or eliminate false alarms;
• Methods to ensure immediate and appropriate response to critical alarms; and,
• Utilization of mobile wireless technology – including smart phones and central monitoring – to alert caregivers to critical patient needs and events.