ChatGPT And Your Research Paper

Generative artificial intelligence (AI) is the newest shiny toy. The best-known example, ChatGPT, burst onto the scene in November 2022 and caught most of us off guard. The earliest versions were interesting and showed great promise for a variety of applications.

The easiest way to think about this technology is to compare it to the auto-complete feature in your search engine. When you start to type a query, the search engine will show a list of commonly entered queries that begin the same way.

Generative AI does the same thing, just on a vastly expanded level. It looks at the question the user posed and displays its best guess on how to complete it based on the huge amount of data it has trained on. As the algorithms get better and the training data more extensive, the AI continues to improve.

However, there are drawbacks. Early iterations of ChatGPT demonstrated that the AI could very convincingly generate text that didn’t really match reality.  There are several very public cases of this. One online sports reporting service used AI extensively and found that nearly 50% of the articles it produced were not accurate. A lawyer submitted a legal brief prepared by AI without checking it. The judge determined that the cases cited in the brief were totally fabricated by the algorithms.

One of the most significant controversies facing academics and the research community is how to harness this valuable tool reliably. College students have been known to submit papers partially or fully prepared by AI. There is also a concern that it could be used to write parts or all of research papers for publication. Some enterprising startups have developed tools for spotting AI-generated text. And most journals have adopted language in their guidelines to force authors to disclose their use of this tool.

Here’s some sample language from the Elsevier Guide for Authors on their website:

“Where authors use generative artificial intelligence (AI) and AI-assisted technologies in the writing process, authors should only use these technologies to improve readability and language. Applying the technology should be done with human oversight and control, and authors should carefully review and edit the result, as AI can generate authoritative-sounding output that can be incorrect, incomplete, or biased. AI and AI-assisted technologies should not be listed as an author or co-author, or be cited as an author. Authorship implies responsibilities and tasks that can only be attributed to and performed by humans, as outlined in Elsevier’s AI policy for authors.

Authors should disclose in their manuscript the use of AI and AI-assisted technologies in the writing process by following the instructions below. A statement will appear in the published work. Please note that authors are ultimately responsible and accountable for the contents of the work.

Disclosure instructions
Authors must disclose the use of generative AI and AI-assisted technologies in the writing process by adding a statement at the end of their manuscript in the core manuscript file, before the References list. The statement should be placed in a new section entitled ‘Declaration of Generative AI and AI-assisted technologies in the writing process’.

Statement: During the preparation of this work the author(s) used [NAME TOOL / SERVICE] in order to [REASON]. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.”

But apparently, some researchers don’t heed the guidelines. Here is a snippet of text from a case report published in an Elsevier journal earlier this year. I’ve highlighted the suspicious text at the end of the discussion section. Click the image to make it readable in full size:

It appears the author tried to use AI to generate this text, but its disclaimer logic was triggered. This represents a failure on many levels:

  • The authors apparently used AI to generate one or more parts of their manuscript.
  • The authors did not thoroughly proofread the final copy before submission to the journal.
  • The authors did not disclose their use of AI as required.
  • The editors and reviewers for the journal did not appear to read it very closely either.
  • And it’s now in the wild for everyone to see.

The big question is, how reliable is the rest of the text? How much is actually just a bunch of stuff pulled together by generative AI, and how accurate is that?

Bottom line: First, use generative AI responsibly. Some excellent tools are available that allow researchers to gather a large quantity of background information and quickly analyze it. However, it ultimately needs to be fact-checked before being used in a paper.

Second, refrain from copying and pasting AI material into your paper. After fact-checking the material, write your own thoughts independent of the AI text.

Finally, if you do use AI, be sure to follow the editorial guidelines for your journal. Failure to do so may result in your being banned from future submissions!

Reference: Successful management of an Iatrogenic portal vein and hepatic artery injury in a 4-month-old female patient: A case report and literature review. Radiology Case Reports 19(2024):2106-2111.

How To Design Your Trauma Bay

In the last two posts, I discussed the size of your trauma bay and how to measure it. This can obviously be helpful if you are updating or building new resuscitation rooms. But what about all the stuff that goes into it? Where is the best place to put it? If you are in the enviable position of being able to stock a brand-new room, here are some tips.

Figure out what you really need in the trauma bay. You don’t have to put everything and the kitchen stove in there. It’s fine to have less commonly used equipment somewhere else, but it must be close! You don’t want someone to have to walk 50 yards to look for something you need right now.

Here’s a list of the critical stuff:

  • Temperature and light controls.  These must be inside the room for easy and rapid access. And the doors should close to contain the heat. Resus rooms that are used frequently should be kept warm, doors closed, at all times.
  • Personal protective equipment. This should be located just inside the room (if space is available) or just outside. It absolutely must be near the entrance and easily accessible or no one will wear it.
  • Airway cart and video laryngoscope. These items must always be located near the head of the bed for immediate availability.
  • Difficult airway cart. These are not used frequently, so need not be placed inside the room. But make sure it is close by.
  • Travel ventilator. This can be stored outside unless you have lots of space available.
  • IV start/blood draw carts. One of these should be stationed on either side of the patient.
  • Rapid infuser. This may be located inside or outside of your trauma room based on the number of times it is typically needed.
  • Procedure packs. These should be located inside the trauma bay, and clearly organized inside cabinets.
  • Medication dispenser. This must be inside the room. Period.
  • Other commonly used equipment/supplies.  These should be placed intuitively in the bay and/or cabinets depending on frequency of usage of each item. Clear marking is essential.
  • Scribe stand. Don’t forget the scribe. They obviously have to be in the room, and need some space for the (preferably) paper trauma flow sheet.
  • Pediatric cart. This can be stored inside or outside the resus room, but should be nearby. Make sure that the measuring card that translates child size into equipment size is easily located.
  • Blood refrigerator. This item is optional, but is becoming more common. It can be located inside or close outside the trauma bay depending on space available.
  • Blanket and sheet warmer. These are nice to have, wherever you have room to put one. The patients will appreciate it.
  • Procedure lights. Ceiling mounted are best because they don’t take up floor space. However, these are notorious for developing a mind of their own as they age. After a while, they never seem to stay focused on your field.
  • Forced air warming blanket unit. This is important here in Minnesota, but also anywhere your patients can get cold. Which is pretty much everywhere. The airflow unit itself is relatively small and can usually be tucked under a counter somewhere. Otherwise, keep it nearby.
  • Linens hamper. You need to get rid of that gown / those sheets and blankets / or whatever. There’s no reason to take up space in the room for this. Park it outside.
  • Laundry basket. This is a valuable item that is generally overlooked. What do you usually do with all that stuff you cut off the patient? Drop it on the floor, right? This is setting you up to lose your patient’s stuff. Get a cheap plastic laundry basket from Target and put it under one of the counters. Toss clothing, shoes, etc in it as they are removed.
  • Cast cart. These are typically huge. They can be anywhere else but inside the trauma bay. Roll it outside the door when needed.

Now where do you put all this stuff? Most trauma centers already have an established layout and flow in their existing trauma bays. When you are moving to a new one, plan ahead! Hopefully you will have more room, so you’ll have some additional flexibility as to where to place everything.

But designing the placement and flow on paper alone is of limited use. You must try it out in advance! How do you do this? Have your contractor mock up a space exactly the size of your new resuscitation room. Move actual carts, cabinets, and equipment into it. If it’s not possible to cart the exact stuff into it, have the contractor build mock-ups of them and place them in the bay.

Now have actual trauma team members practice simulations of common types of resuscitation: basic no frills, basic with intubation, basic with splinting/casting, advanced with all of the above plus multiple procedures. Take careful notes of flow and any glitches that arise. Then move your stuff around to fix any problems, and try again!

How To Measure Your Trauma Bay

In my last post, I detailed some standard info on trauma bay size. Today, I’ll describe what I found when I brought in my trusty tape measure a few years ago to check out the old trauma bays at Regions Hospital. I came up with several helpful measurements to help gauge the relative utility of the rooms.

Here are the indices that I came up with:

  • TBTA: Trauma Bay Total Area. This is the total square footage (meterage?) measured wall to wall.
  • TBWA: Trauma Bay Working Area. This is the area that excludes equipment carts next to a wall, and areas under countertops that extend away from the wall.
  • TBAA: Trauma Bay Available Area. This is the TBWA less any other unusable areas in the room. We have an equipment post near one corner that eats up 16.5  sq ft of space. Also remember to subtract the area taken up by the patient bed, as this area is not available to the trauma team, either.
  • TBSI: Trauma Bay Space Index. This value is derived by dividing the TBAA by the number of team members in the room. It gives an indication of how much space is available for each trauma team member to work in.

Values in the old trauma bays at my trauma center:

  • TBTA: 291 sq ft
  • TBWA: 220.5 sq ft
  • TBAA: 186.5 sq ft
  • TBSI: 15.5

What does it all mean? You’ll have to work out the details using measurements from your own resuscitation room. For my old rooms, it meant we each had a 4×4 foot square to move around in, on average. This was fairly tight, I would say. Fortunately, we’ve moved to new rooms with much, much more space.

Tune in to my next post this week on my thoughts on outfitting your resuscitation room.

How Big Should Your Trauma Bay Be?

Trauma professionals are never satisfied with the size of their trauma bay. Today, I’ll write about optimal trauma bay size. Next week, I’ll describe my system for quantifying the space in your trauma bay and address the equipment layout in your resuscitation room.

Trauma resuscitation rooms vary tremendously. They can range from very spacious…

to very tight…

Most trauma bays that I have visited were somewhere between 225 and 300 square feet (21-28 sq meters), although some were quite large (Rashid Hospital in Dubai at nearly 50 sq meters!).

Interestingly, I did manage to find a set of published guidelines on this topic. The Facility Guidelines Institute (FGI) develops detailed recommendations for the design of a variety of healthcare facilities. Here are their guidelines for adult trauma bays:

  • Single patient room: The clear floor area should be 250 sq ft (23 sq m), with a minimum clearance of 5 feet on all sides of the patient stretcher.
  • Multiple patient room: The clear floor area should be 200 sq ft (18.5 sq m) with curtains separating patient areas. Minimum clearance of 5 feet on all sides of the patient stretcher should be maintained.

The FGI “clear floor area” corresponds to my “Trauma Bay Working Area”, which is the area that excludes all the carts, cabinets, and countertops scattered about the usual trauma room. California’s guideline of 280 sq feet seems pretty reasonable as the “Trauma Bay Total Area”, if you can keep your wasted space down to about 30 sq feet.

Bottom line: Once again, don’t try to figure out everything from scratch if you are designing new resuscitation rooms. Somebody has probably already done it (designed a trauma bay, developed a practice guideline, etc). But remember, a generic guideline or even one developed for a specific institution may not completely fit your situation. In this case, the FGI guidelines say nothing about the trauma team size, which is a critical factor in space planning. Use the work of others as a springboard to jump start your own efforts at solving the problem.

Related link:

EAST Practice Management Guideline: Handoffs And Transitions Of Care

Medicine, in general, and trauma care, specifically, require frequent communication. These communications may be between two providers to maintain continuity of care or between providers and patients to explain it. Unfortunately, the Joint Commission has identified breakdowns in the process as a root cause of preventable events and a significant factor in preventable death.

To address this problem, many centers have sought to standardize this process, which may include some of the principles in my previous post. However, until now, there have been no evidence-based recommendations for this practice.

The Eastern Association for the Surgery of Trauma performed a systematic review and meta-analysis of the literature to develop a practice guideline. They focused specifically on handoffs for acute care surgery during perioperative interactions, patients arriving in the trauma bay, and patients transitioning to or from the ICU and floor. The goal was to reduce complications, handoff errors, medical errors, and preventable events.

The literature on this topic was searched from 1960 to 2021, and only observational and randomized studies were included. This yielded only ten papers that met all search criteria. The reviewers then used these papers to answer three questions. These and their answers are outlined below.

Question 1.  Should perioperative interactions in the care of ACS patients (P) include a standardized handoff versus current process without a standardized handoff to help reduce clinical complications, handoff errors, medical errors, and preventable adverse events?

Patients who received a standardized handoff were significantly less likely to experience a handoff error.  However, the impact on medical errors and adverse events could not be gauged because only one paper covered these problems.

Question 2. Should EMS utilize a standardized handoff at the arrival of trauma patients versus the current process without a standardized handoff to help reduce clinical complications, handoff errors, medical errors, and preventable adverse events?

We instituted a trauma team EMS timeout process in 2012, which persists to this day. Please take a look at my post here. The prehospital providers like it because they feel like they are more a part of the team. The receiving team can listen to their report without distraction. But what does the literature say? Unfortunately, we don’t know yet. Only one published paper covered this topic, and it included only 18 patients.  Thus, no conclusions can be drawn.

Question 3. Should intra/inter floor and ICU interactions in the care of ACS patients include a standardized handoff versus currently process without a standardized handoff to help reduce clinical complications, handoff errors, medical errors, and preventable adverse events?

Significantly fewer preventable adverse events occurred when a standardized handoff was used. There was no difference in clinical complications. The impact on medical errors could not be evaluated because only one study assessed this.

Bottom line: The general belief is that using a standardized handoff is a good thing. But I think you see the theme here. As in most EAST systematic reviews, there is painfully little high-quality data available for us to prove it. Most of the mundane, day-to-day things we do and decisions we make as trauma professionals are too dull to perform a study about. 

From the few papers available for this guideline, standardized handoffs are a good thing. They decrease handoff errors and reduce preventable adverse events as well. The EMS to trauma team handoff is well-received and is subjectively valuable. Unfortunately, there is little real data to prove this.

Overall, the real data on this topic is weak, and much more work needs to be done. I would encourage all trauma professionals to develop and refine their handoff processes. I strongly recommend coupling that with your own study so you can teach the rest of us how good it really can be.

Reference: Handoffs and Transitions of Care: A Systematic Review, Meta-Analysis, and Practice Management Guideline from the Eastern Association for the Surgery of Trauma. J Trauma, Publish Ahead of Print
DOI: 10.1097/TA.0000000000004285

Home of the Trauma Professional's Blog

Do you want to get a daily email every time there’s a new post? See what I’m up to.

Click here to get details and subscribe!

[accua-form fid=”1″]

[mc4wp_form id=”2023″]