Let Families Activate Rapid Response Teams, Says SCCM
SCCM recruits families to call rapid response teams directly
The Society of Critical Care Medicine (SCCM) issued a new guideline on how hospitals and care teams should respond to emergencies outside the ICU or ED (i.e., on the medical and surgical wards). Read it here.
Most of the advice is uncontroversial (e.g., collect timely and accurate vital signs, train team members on how to recognize deteriorating patients, etc).
Then there’s this:
The panel unanimously agreed that patients and care partners should be empowered to escalate their patient concerns to the healthcare team and, when deemed necessary, directly to the response team.
Whoa! Where did that one come from?
At a few hospitals, patients and families are given the phone number to the medical emergency team (MET)/rapid response team (RRT), and advised on situations that should prompt an emergency call (poorly responsive compared to baseline, difficulty breathing, etc).
SCCM now suggests this should be a standard practice at most hospitals.
I know what you’re thinking: this change could have long-term impacts on family/care team dynamics and health professionals’ job satisfaction, so it was probably extensively tested over years in carefully designed studies, with populations representative of the hundreds of thousands of complex, multiply-acutely-and-chronically-ill adults who exist in a liminal state between skilled nursing facilities, acute care hospitals, and long term acute care hospitals, who experience clinical deteriorations most frequently.
Well … no. The studies cited were:
A 2014 report on a program at Cincinnati Children’s hospital that permitted family activation of RRTs starting in 2007;
A 2010 report after two years of experience at the level 1 trauma center at Shands in Jacksonville, FL;
A 2013 “how-to” article in Nursing Management coaching hospitals on how to set up such a program, based on experience with adult patients at Duke;
A pilot at a Duke children’s hospital (during which 2 calls were made by families);
A 2012 poster abstract at an SCCM meeting, by some of the same team from the Cincinnati children’s hospital;
A 2009 report on a few years’ experience at the 140-bed North Carolina Children’s Hospital, during which it seems every child was transferred to a higher level of care upon family request/concern.
Panelists called this “low certainty evidence from five before-after studies that this intervention may be associated with lower mortality and fewer unsuccessful resuscitation events.”
But even that watered-down statement is unjustified. Let’s take a closer look.
Family-Activated METs in Adults Halves Mortality?
At Shands (the level 1 trauma center admitting adults), the panel relates the impressive results that mortality fell by 58%, and unplanned codes by 47% “after the widespread implementation of patient/ family activated RRT/MET.” This data is the primary basis for the panel’s claim that family MET calls “may be associated with lower mortality and fewer unsuccessful resuscitation events.”
But the reported mortality decline at Shands was clearly associated with a multi-pronged quality improvement initiative, of which family-and patient-initiated MET calls were an insignificant component:
Rapid response activations in total increased from 47 to 193 per month after the intervention, but families and patients only initiated 25 calls over two years.
This implies 0.5% of all RRT calls were initiated by families or patients during the two-year study period, during which there was a 310% increase in RRT calls by medical staff generally.
Another component of the program lowered the threshold to transfer patients to the ICU. Almost four times as many patients were transferred to higher levels of care during the study period (increasing from 12.8 to 45.4 transfers per month).
Despite all this, there was no statistically significant decline in cardiac arrests outside the ICU (which did fall nominally from 4.5 to 3 per 1000 discharges).
Intentionally or not, the panel misleads by failing to situate the family/patient-initiated calls in proper context with the system-wide interventions, implying instead that the observed mortality drop was directly attributable to permitting family/patient RRT calls. That could be because without this study, the evidence-based argument for this proposal in adults falls apart.
(And you did read that right: patients activated their own rapid response evaluations in over half of the 25 RRT calls at Shands, authors reported.)
Family-Initiated RRT Calls: Evidence in Children
Data from the Cincinnati children’s hospital provides another view. Over six years of study (2007-2013):
There were 83 MET calls activated by parents (~1.2 per month, about 3% of all MET calls).
Children were transferred to the ICU 24% of the time after families called, vs 60% for staff MET calls.
There was no decline in the rate of cardiac arrests outside the ICU after family-initiated calls were permitted.
Families reported they called a rapid response when they felt their concerns were being ignored or dismissed by the care team.
What’s Really Going On Here?
Why this proposal, and why now? These studies were done 15 years ago, mostly in kids, and the two in adults tell us almost nothing. Why put this out at all, much less a decade later?
One, like all of us, guideline panels want to make an impact. Rapid response teams have been around for over twenty years. There’s nothing really new to report, but this update still took a lot of hard work by many people. Wouldn’t it feel unsatisfying to be an expert panelist and after all that effort, leave it at “collect timely vital signs”? Wouldn’t you want to say something stronger, even if it required exaggerating the data a bit?
It’s a politically safe move that fits with the overall (generally positive) trend toward empowerment of patients and their families as consumers of medical care: providing immediate access to labs, progress notes, formal ownership of medical data, on-demand apps, etc. If an American can call 911 anytime, why not ‘911 in the hospital’ (if you will)?
I suspect there’s an increasing unease among policymakers that soon, more patients may need to ‘call 911’ from inside hospitals—because almost anywhere you look, hospital-based acute care’s wheels are starting to wobble.
Patients have become more complex, hospitalists’ workload frequently exceeds what they consider safe, and nonphysician providers are seeing more (and sicker) patients with no physician involvement. Nursing shortages mean fewer staffed beds and more patients per nurse. More health care professionals are burned out or disengaged. Societal trust has declined. All this together strains the quality of care, communication and relationships between care teams and families.
Families have always been “empowered” to escalate their concerns to their care team, to ask for evaluation by the nurse, by the doctor, to repeatedly do so if they feel the need.
And they certainly do this (and should!) everywhere I’ve worked. I learned a long time ago to become highly vigilant if a family member says “something’s different,” “she’s not right,” “something’s wrong,” etc. In my anecdotal experience, these phrases have an impressively high positive predictive value for badness (early sepsis, often).
So SCCM smuggling in a pet policy prescription as “evidence-based,” with outdated, scant, inconclusive research mostly in children, reads to me like activism tinged with an acknowledgement that hospital stays are becoming increasingly dangerous.
One way to duct-tape these widening cracks is to deputize families as bedside backup. Also (imagining the panel crossing its fingers), maybe if nurses knew that families or patients could go around them to call RRTs directly, they might maintain better communication and vigilance to deterioration? Maybe at Shands, although patients and families almost never actually called, care teams became more attentive, knowing they might?
Patient-Centered, or Pandora’s Box?
SCCM authors could have advocated for their unanimously held opinion in opinion pieces, or pushed for new, quality research in hospitalized adults, rather than exaggerating scant evidence to support its agenda for allowing direct activation of RRTs by families and patients.
Leaving all that aside: is it still a good idea, anyway?
As it stands, if a patient is deteriorating and she or her family senses it but the nurse does not, or there’s a toxic relationship there, etc.—after asking for an evaluation, if no action is taken, the family and patient have no recourse. They just have to sit and wait until obvious critical illness or cardiac arrest occurs.
Shouldn’t there be some other option available to them besides “acting out” (or whatever it gets called where you work)?
At all the sites tested, patients and families directly activated rapid response teams infrequently—roughly one to four calls per month on average.
That would be manageable. But would that restraint persist as this became a routine, expected aspect of hospitalization? Think of the future TikTok influencers and memes in patient-advocacy subreddits and online forums (“You Make The Call!”). Why go through the doctor at all, if you don’t have to?
What Would You Want?
Now, imagine yourself or your loved one in the hospital bed.
The families of ill children in Cincinnati calling rapid responses were “right” 24% of the time (considering the need for ICU transfer as a true positive), compared to 60% for medical professionals. In one of the two studies in adults, families “scored” 34% right (vs. 47% for staff, in proportion of RRT calls resulting in transfers to higher levels of care).
Maybe some of those transfers were “defensive” and unjustified, but probably not all of them.
Some of those MET calls were made out of desperation and fear after care teams failed to take concerns seriously (from the family’s perspective, at least). One hopes this is uncommon, but it must occur, and might more frequently in the future. We don’t know, really.
Isn’t that uncertainty reason enough for families to be allowed to activate medical emergency teams directly, without further study?
What do you think?
I think the limited resources should be focused on appropriate nursing staffing and patient monitoring. After all it is the nurse's job to monitor the patient and call for help immediately if he/she deems it necessary. I think the chronic understaffing in hospitals since Covid contributes to the nurses not being able to do this job appropriately (this is not criticism, they are overworked and underpaid, generally). Certainly, in most hospitals, empowering all patients to call rapids would stress the mechanics of the RRT and perhaps skew their patient evaluations.
A different perspective on this: these things can and do already happen. We have seen families that are very bold, hit the code blue buttons on the wall when they felt they were in a dire situation. Nearly every time it was a valid reason albeit, we had questions just like the nurse did. Who hit the code button? Why?
This is not necessarily “bad”, however it did become concerning only one time in my experience when family indicated their rationale for doing so was the nurse call button wasn’t getting them what they wanted after multiple attempts. Water. They wanted a glass of water. 12 people rolling in hot, heavy and deep to deliver a glass of water (absolutely not). They received a sternly worded message, delivered by the entire team (no one was dismissed for emphasis to drive the point home for this family). Education and stern reinforcement of proper use of resources was implicitly stated.
We have also seen patients call out for help for their roommates because “they didn’t sound right” or “they stopped talking” or “were choking”. This action by the roommates started either an RRT or code and saved their lives.
Bottom line, they will find a way, so give it to them. Education and boundaries must always exist.