Review Of: Merlin M, Baldino K, Lehrfeld D, et al. Use of a limited lights and sirens protocol in the prehospital setting vs. standard usage. Am J of Emerg Med. 2011;May 11(Epub ahead of print)
This prospective cohort study attempts to find whether the implementation of a lights & sirens (L&S) protocol decreases the number of L&S transports and effects on patient outcomes. The study took place in eight New Jersey cities—EMS systems in the four intervention groups and four in the control groups. The study’s results showed that the EMS systems that implemented the protocol were 5.6 times less likely to transport with L&S compared to the control groups. The authors conclude that the “judicious” use of L&S may have a significant effect on transport safety, and they reported no impact on patient outcomes.
Dr. Wesley: When I discovered this study, I hoped it would provide answers to this extremely important subject. However, I must admit that I was terribly disappointed after reading it. The appropriate use of lights and sirens (Code 3) during patient transport is both a public safety and patient care issue.
The authors attempt to determine whether the implementation of a Code 3 transport protocol reduced L&S use, and more importantly, they investigate whether an association between patient disposition and patient outcome exists.
Unfortunately, they failed, in my opinion. Although the authors report that the Code 3 transports were significantly less (29% vs. 49.7%) between the four towns that used the protocol (implementation group) and the four towns that did not (controls), I think this difference is misleading for a few reasons.
First, the authors failed to determine the baseline incidence of L&S use in the four towns in the implementation group prior to the study. They assumed their rate was the same as the control group. But significant differences exist between the implementation towns and control towns. For example, the control towns were on average 8–12 miles farther away from the receiving hospital than the implementation towns. Previous studies have already proven that distance to the ED increases the use of L&S.
Secondly, although the authors found a significant difference in the use of L&S between the two groups, this difference didn’t add up to a difference in rates of admission overall between the two groups. This causes me to question the protocol itself, which leads me to the third and most important point.
The protocol supported the use of L&S for the following conditions:
• Airway not secure, failed intubations
• Sats < 93% despite NRB
• Flail chest, pneumothorax
• SBP <90 despite IV bolus
• Symptomatic bradycardia and tachycardia
• Acute Stroke
• Spinal injury
• Status seizures
• Burn >20% BSA
• Penetrating injury to head/neck, torso, and proximal extremities
• Amputations above the wrist/ankle
The authors didn’t present specific data on which of these criteria were used to justify the use of L&S. My opinion is that much of these criteria are not time sensitive and would increase the use of L&S for an all-ALS service.
The bottom line? This study leaves the main question unanswered: For what conditions do L&S make a difference, and does the implementation of an L&S protocol reduce their use?
Medic Marshall: When I see these “use of lights and sirens” papers, I get a little nervous about what they’re going to say. Don’t get me wrong, I’m all for improving the safety and the well being of my fellow paramedics and EMTs, and I encourage continued research on making it safer for us to be out on the roads. But when I read this paper I couldn’t help but be thoroughly disappointed as well.
First off, I don’t think the study design fits here. By using a “pairwise” design, the authors tried to pair similar cities and then compare intervention (the protocol) to the control group. The authors then report a decrease in L&S use, but I just don’t see it. First off, were the cities similar? Well, it’s hard to say. If you look at Table 2 in the article, the authors look at protocol vs. nonprotocol, approximate distance (in miles), population, median income in thousands, and whether the residents had a bachelor’s degree or higher. I don’t see how the authors could come up with a pairwise design. To me, too much variability seems to exist in all the cities.
This would be a better study (albeit, perhaps just a better design) if it were a simple experiment that prospectively evaluated six months of data, then implemented the protocol and evaluated the results after another six months. I think the authors made this study far more complicated than it needed to be.
Secondly, the clinical criteria the authors used for their protocol seemed relatively specific: “more than two failed intubation attempts” or “oxygen saturation <93%,” for example. How about impending airway/ventilator compromise? This seems like a better guideline than trying to make this a clear-cut protocol. I’ve seen plenty of chronic chronic obstructive pulmonary disorder (COPD) patients who float between 89–91% SaO2. So by this protocol, do I have to transport L&S?
In closing, too much is wrong with this study to give much credence to it, in my opinion, starting with the design. Like I said, I’m all about being an advocate for improved safety in EMS, but this study does nothing to say that L&S does not have any effect on patient outcomes.