Tag Archives: trauma systems

Secondary Overtriage: What Is It, And Why Is It Bad?

Simply put, secondary overtriage (SO) is the unnecessary transfer of a patient to another hospital. How can you, as the referring trauma professional, know that it is unnecessary? Almost by definition, you can’t, unless you have some kind of precognition. If you knew it wasn’t necessary, you wouldn’t do it in the first place, right?

But using the retrospectoscope, it’s much easier. The classic definition describes a patient who is discharged from the hospital shortly after arrival there. What is “shortly?” Typically, it occurs within 48 hours in a patient with low injury severity (ISS < 16) and without operative intervention. Definitions may vary slightly.

And why is it bad?

Several states with rural trauma systems have scrutinized this issue. The first study is from West Virginia, where six years of state registry data were analyzed. Over 19,000 adults were discharged home from a non-Level I center within 48 hours after an injury. Of those, about 1,900 (10%) had been transferred to a “higher level of care” and discharged from that center (secondary overtriage, could be any higher-level trauma center).

The factoids:

  • Patients with ISS > 15 and requiring blood transfusion were more likely to be SO. (I would argue that this is appropriate triage in most cases!)
  • Neurosurgical, spine and facial injuries were also associated with SO. (This one is a little more interesting, see below).
  • SO was more likely for transfers during the night shift, when resources are often more scarce

The problem is that this study is descriptive only. It doesn’t really help us figure out which patients could/should be kept based on any of the variables they collected.

The next study is from Dartmouth in New Hampshire and examines transfers into that single Level I center from 72 other hospitals. Registry data were examined over 5 years, identifying transfer patients with ISS < 15 who were discharged within 48 hours without an operation.

Yet more factoids:

  • 62% of the nearly 8,000 patients received by this center were transfers
  • Overall SO rate was 26%
  • A quarter of adult patients and one half of pediatric patients were considered SO, and about 15% of them were actually discharged from the ED (!)
  • Head and neck, and soft tissue injuries were most common among SO patients

The real bottom line: Here are my thoughts on what you can do to try to decrease the number of your patients with SO and optimize the transfer process:

  • Work with your upstream trauma center to determine how much imaging you really need to perform
  • Develop a reliable method of getting those images to them
  • Ask them to help you develop practice guidelines and educate your hospital/ED staff to help manage common diagnoses that often result in SO from your center
  • If you are located in a rural area, inquire about RTTD courses you might attend


  • Secondary overtriage in a statewide rural trauma system. J Surg Research 198:462-467, 2015.
  • Secondary overtriage: the burden of unnecessary interfacility transfers in a rural trauma system. JAMA Surg 48(8):763-768, 2013.

Are State Trauma Systems Cost-Effective?

Every state in the US now has a formal trauma system. Several studies are available that document the advantages of these systems in terms of outcomes and survival. Trauma professionals get this. But the governmental agencies and legislators who help create, fund, and maintain them tend to focus on cost as well.

Arkansas was the last state in the union to implement a trauma system. A study in press from the University of Arkansas details their experience. They examined the impact of the new system on mortality, patient care, and attempted to calculate a return on investment from the taxpayers in an effort to show the added value.

The study was commissioned by the Arkansas Department of Health and carried out by the state Trauma Advisory Council. It was led by out of state investigators in an effort to maintain impartiality. A comprehensive review of records was performed by a panel of 5 surgeons, 1 emergency physician, 2 trauma program managers, 1 ground and 1 flight paramedic. Preventable and potentially preventable deaths were identified and analyzed in depth. Value of life lost was calculated by using a conservative $100,000 per year lost. A total of 290 charts were reviewed pre-system, and 382 post-trauma system implementation using proportional sampling of about 2500 trauma deaths in one year.

Here are the factoids:

  • A significantly higher percentage of patients were triaged to Level I trauma centers after the system was implemented
  • Preventable mortality was decreased from 30% to 14% (!!)
  • This means that 79 extra lives were saved due to implementation of the trauma system
  • Non-preventable deaths with opportunity for improvement remained constant at about 55%
  • Non-preventable deaths without opportunity for improvement increased from 16% to 38% (!)
  • Using the most conservative VLL calculation, this equates to $2.4M in savings per life saved
  • This adds up to $186M in savings to the taxpayers of Arkansas, a 9-fold return on their investment of $20M in the trauma system. 

Bottom line: Wow! This nicely done studies gives us excellent insight into the hows and whys of the value of an organized state trauma system. It is likely that the triage system directed more patients to the most appropriate level of care, leading to fewer preventable deaths. And it enticed hospitals to up their game and make the move toward formal trauma center designation. This improved education and training at those centers, leading to better patient care.

There is a wealth of information in this study, and I recommend that everyone with an interest in or are already participating in their state trauma system read it in its entirety. Hospitals that are reluctant to join or are lagging in meeting criteria need to recognize that they are not serving their communities as well as they think. And legislators must realize that the financial impact of even a small investment has real and significant consequences to their constituents.

Related posts:

Reference: Does the Institution of a Statewide Trauma
System Reduce Preventable Mortality and Yield
a Positive Return on Investment for Taxpayers? JACS in press, 2017.

How Many Trauma Centers Should There Be?

Trauma centers seem to be popping up all over the US. Many metropolitan area have literally scores of centers at various levels. And yet there are swaths across this country where you won’t find a single Level I, and only a few Level IIs. In most states, there is little guidance from the designating authority regarding whether a new trauma center is feasible or even needed. The American College of Surgeons (ACS) has given little guidance over the years, except for a white paper in 2015 that essentially said that it is up to the designating authorities to determine this.

Last August, the ACS organized a consensus conference to try to develop an objective method for figuring out when enough is enough. There was unanimous support for developing a tool that would encourage designation to meet the needs of the trauma patient, not the financial needs of a hospital or hospital system. This Needs-Based Assessment of Trauma Systems (NBATS) tool looks at 6 factors, some of which take a little calculation to complete. A point score is arrived at that predicts the additional number of trauma centers that may be needed. Currently, this tool is in draft form and is in the process of being optimized.

Click here to download the draft document.

So far, this has been a theoretical exercise. But a group at Stanford decided to test the tool on the entire state of California. They used a variety of data sources to compile the needed numbers, and did some complicated spatial analyses of transport times to accurately calculate NBATS scores.

Here are the factoids:

  • 74 trauma centers were identified in the state – 15 Level I, 37 Level II, 14 Level III, and 8 Level IV
  • The state was broken down into 30 Local Emergency Medical Service Agency trauma service regions
  • Only 4 of the 30 regions had scores suggestion that they had enough trauma centers
  • The tool suggested that 9 regions needed 1 more trauma center, 13 would require 2 more, and 4 would require 3 more
  • The model also suggested that 3 regions already had more than needed

Bottom line: There is already literature showing that adding additional (too many?) trauma centers to a region can have a negative impact on patient volumes and resource availability at Level I and II centers. This tool may allow state trauma systems to more objectively determine exactly where more centers are justified, enabling them to rise above the usual political battles (maybe). Unfortunately, the tool does not take available surgical resources in the region (trauma surgeons, neurosurgeons,  orthopedic surgeons) into account, or provide guidance on which levels of new centers should be developed. But it’s a good start to help solve a sticky problem.

Reference: ACS needs based assessment of trauma systems (NBATS) tool: California example. AAST 2016, Paper #24.

AAST 2011: Benefit of Transport to a Trauma Center

Most trauma systems set certain prehospital criteria that, when met, direct that patient to a trauma center. It is now well-established that care of these patients results in improved survival if they are managed at those centers. Unfortunately, undertriage is still a problem, meaning that those patients may not always be taken to a hospital most appropriate to care for their injuries. What is the penalty that your patient pays if this happens?

The University of Toronto performed a nice, prospective study across a large region with both urban and rural areas. Database information was analyzed for all victims of motor vehicle crashes who had a severe injury (ISS>15) or who died. Over 6,000 crash victims’ data were analyzed. 

Just under half of the victims (45%) were triaged to a trauma center. Of those who were taken to other hospitals, slightly more than half (58%) were transferred to one within 24 hours, but nearly 5% died in the non-trauma center ED. The overall mortality for severely injured patients who were taken to a nontrauma center was 8.7%. This was a 30% increase in adjusted mortality compared to those taken to a trauma center directly.

Bottom line: Follow the rules! EMS authorities and trauma systems should make it a priority to adopt the CDC protocol (see below) or create trauma guidelines based on them that ensure patients with significant injuries are taken directly to a trauma center. Going to the nearest hospital (if it is not a trauma center) or bending to the patient’s preference is not in their best interest (and may kill them)!

Click here to download the Centers For Disease Control and Prevention (CDC) National Trauma Triage Protocol. This should be used as a standard!

Reference: The mortality benefit of direct trauma center transport in a regional trauma system: a population-based analysis. AAST 2011 Annual Meeting, Paper 50.

Q&A: Is Undertriage Bad?

After my discourses on under- and over-triage in the last week, I received an interesting question from a reader: although undertriage seems bad from a theoretical standpoint, are there any objective negative consequences?

As you might imagine, there is little literature on this topic. The incidence is low, so it’s tough to design a study with enough power to come to any solid conclusions. There are two studies that I can cite that shed as much light on the subject as possible.

The first looks at system undertriage at the EMS level. A Canadian study looked at patients with severe injuries (identified by ISS>15 after admission) who were taken to trauma centers (correct triage) vs non-trauma centers (undertriage). After solid statistical analysis of over 11,000 patients, they found that mortality in the undertriage group was 24% higher than the correctly triaged patients.

A second study looked at undertriage in one trauma center (1,424 patients) using their standard triage criteria, not ISS. The undertriage group had a significantly lower ISS (17 vs 25). The correctly triaged patients were more frequently intubated in the ED, more likely to be admitted to the ICU, and had longer ICU and hospital stays. Mortality was not significantly different. The problem with this study is that most of the undertriage group probably did not need a trauma activation, based on their lower ISS. The higher ISS patients (who met triage criteria) needed an airway earlier and required critical care more often. These data show that the institution probably needs to adjust its triage criteria!

Bottom line: The Canadian study shows the danger of undertriage prior to reaching definitive care. There is no good literature that illustrates its danger once the patient is at a trauma center. But there is support for the converse idea that appropriately triaged patients get definitive management sooner (airway, critical care). Any takers for designing the study to answer this question?


  1. Survival of the fittest: the hidden cost of undertriage of major trauma. J Amer Col Surgeons, 211:804-811, Dec 2010.
  2. Outcome assessment of blunt trauma patients who are undertriaged. Surgery 148(2): 239-245, Aug 2010.