Last week’s New York Times front-page piece on colored wristbands highlighted an issue I’ve been fretting about for years. We can achieve consensus on a $700B bailout in 3 days (well, perhaps not), but can’t agree on what color a DNR wristband should be? Wow.

My interest in this subject began with a remarkable case I first described in Internal Bleeding and later in Understanding Patient Safety. The story went like this:

A young nurse in a teaching hospital goes in to visit her patient, an elderly man on the ortho floor, and finds him not breathing. She calls for a Code Blue, and the cavalry soon arrives and begins CPR. As the Code proceeds, the team leader repeatedly asks whether anybody knows the patient’s history, to no avail.

A young resident, trying to be helpful, dashes off the nurse’s station and grabs the patient’s chart. A moment later, he slides into the room and, to the horror of the Code Blue team, breathlessly announces that the patient has a DNR order in the chart. The Code is aborted, the patient left to die.

The patient’s young nurse was there for the Code and wondered about the DNR – after all, she had received sign-out an hour earlier, and was told the patient was a Full Code. But the young resident had spoken so forcefully, she was too intimidated to speak up. “Somebody must have changed the patient’s code status and not told me,” she thought sheepishly. Chalk it up to experience.

The nurse remained in the room, helping to clean up. She palpated her patient’s radial pulse, and felt his heartbeat slow and then extinguish, like the embers on a doused campfire. Her eyes wandered to the chart, where the page with the DNR form was open. “Oh my God,” she shrieked, “that’s not this patient.” The resident, you see, had inadvertently picked up the wrong chart out of the nurse’s station chart rack.

The Code Blue team was re-called to the room, but it was too late. The patient died.

I’ve recounted this story scores of times in talks, often ending my presentation by asking the audience how this tragedy might have been averted. “Color code the wristbands,” somebody usually shouts. Seems straightforward enough. So my colleague Niraj Sehgal and I set out to study whether hospitals had implemented this simple and elegant safety solution. We published our findings last year in the Journal of Hospital Medicine.

We surveyed 127 nursing executives at academic medical centers; 69 (54%) completed the survey. They reported that 17 of their hospitals were using color-coded wristbands to identify DNR status. Of these, 8 different color schemes were described. The colors (with the number of hospitals using each color for DNR) were:



• Green—5
• Yellow—3
• Blue—3
• White with blue stars versus green stars (full DNR versus limited DNR)—1
• Red—1
• Red and white—1
• Purple—1
• Gold—1
• Other (not listed)—1

  

According to the Times article, purple is now the color recommended for DNR in New York public hospitals. Hospitals have shied away from using blue, since it seems like a mixed message for a blue wristband to signify “No Code Blue.” I was pleased to see that no one was using black – that does seem a bit macabre.

Note that yellow was tied for the second most common color. This too is a bit problematic, since about 10 million people wear those cool Lance Armstrong Livestrong bracelets. A couple of years ago, a Florida newspaper reported a few near misses in which Livestrong bracelet wearers were nearly made “No Code” against their will.
 
We next asked the nurse leaders whether any other colors had meanings at their hospitals besides DNR. We found that there were 12 other color codes (I’ve since heard even more, including bladder training [yellow, obviously], diabetic, and seizure disorder). But the list from our paper (number in each category or color is in parentheses) is pretty impressive as is: 


• Drug/allergy (22): Red (16), Yellow (4), White (1), Orange (1)
• Fall risk (18): Orange (5), Green (3) (and lime green [1]), Blue (3), Purple (3), Yellow (2) (and fluorescent yellow [1])
• Same name alert (7): Blue (3), Orange (2), Yellow (2)
• Bleeding risk (3): Red (all)
  
• Patient identification (3): Green, Red, White
• Wandering risk (3): Pink (2) (and hot pink [1])
• Contact isolation (2): Green (all)
  
• Latex allergy (2): Purple (all)
  
• No blood draws on this arm (1): Orange
• MRSA infection (1): Green
• No blood products (1): Red
• Sleep apnea (1): Purple


Obviously, before long, the wrist of a patient unfortunate enough to have several problems begins to look like a Benetton ad.



More than 70% of the nurse executives recalled situations in which confusion over DNR status, sometimes involving the wristbands, placed patients at risk. Representative problems included:

The patient had a DNR order written in the chart but no other identifiers at bedside, so a consult service started CPR while trying to determine code status.

Prior to implementing the wristbands, there were delays in care. Once wristbands were implemented with stars only, there was confusion as to what a blue star meant and what a green star meant (limited versus no resuscitation efforts).

We used to place a sticker on the chart. A sticker was left on the chart of a discharged patient when a new patient was admitted. The mistake was caught before an incident occurred.

The situation depicted in our article and the NY Times piece might be amusing if it wasn’t so scary, and so emblematic of many of our safety problems.

I know that everybody is worried about too much standardization (“we’re different here at Acme Hospital”), and healthcare has a lot of bright people who have their own ideas about things like wristband colors. But the first principle of safety is to standardize what you can. The Times article describes a rather silly controversy over some of the colors and the language. Should a DNR wristband be purple or another color? Should it say “DNR” on it (along with being color-coded), or would that be too stigmatizing? Despite efforts in many states to standardize colors, these and similar small-bore disputes have blocked a uniform solution.

So what, you might ask? Problem is, with all these colors and indications, the chances that a nurse or doctor will forget what a given color means in the heat of battle are unacceptably high.

This struggle to "get to yes" on something as simple as wristband colors brings to mind my favorite saying about medical staffs: What do you call a 99-to-1 vote of the medical staff?  The answer: a tie. Start with this unrealistic need for complete consensus, stir in a bit too much sensitivity to patients’ privacy and to the newfangled goal of “patient-centeredness,” and you have a formula for perpetual inaction.

Folks – let’s just chose a set of colors and enforce them as national standards – stat. The Joint Commission should convene a group of safety experts, a few providers, a couple of human factors mavens, and some patient advocates, and keep the door locked and the vino flowing until they’ve hammered out an agreement. I remember when JCAHO released the list of “Do Not Use” abbreviations for prescriptions a few years ago. Sure, the list wasn’t exactly the one we would have developed ourselves, but having a national list probably saved us 10,000 hours of medication safety committee meetings – which is what it would have taken to hammer out our own list. More importantly, a uniform national list facilitated adherence: a resident who learned the list at UCSF didn't have to relearn it when she started her fellowship at Duke; ditto a registry nurse working at several different hospitals.

So let’s create – and enforce – a national standard for wristband colors ASAP. Lives will be saved. And the Customization Police will get over it eventually.