When you are resuscitating a baby, and you ask, how is the heart rate? What kind of answer do you get? “It’s good” “pulse is a bit slow” “I think its around 80”?

As Lou Halamek and his team recount (Yamada NK, et al. Impact of Standardized Communication Techniques on Errors during Simulated Neonatal Resuscitation. American journal of perinatology. 2015), such kinds of communication in an Air Traffic Control tower would see you booted out. In air traffic control, they realized that imprecise, or even just variable, ways of recounting what was going on, were leading to errors; so now, when there is a communication, it has to follow a strict format. Yamada et al have followed that lead and developed a standard lexicon of information transfer for neonatal resuscitation, including a closed-loop communication system for completing orders for medication or volume resuscitation.

In answer to the question above, the responses have to be one of the following “heart rate above 100″; ‘heart rate below 100”; “heart rate below 60”; or “we’re in deep shit’ (I made that last one up, it should be “heart rate zero”).

You can see how that would be better than “I can’t hear a heart rate” which might mean, “I have a middle ear infection”, or “my stethoscope just dropped on the floor”, or “the baby is asystolic”!

Similar structured phrases are presented for other issues during resuscitation.

For medications or volume, the order must include the name of medication, dose, concentration and route AND it must be repeated by the person taking the order and include all the same information.

They studied this in an RCT, with nurses trained in the lexicon. A cross-over design was used, so the trained nurses either used the lexicon or did not, and 13 people with some experience in NRP were enrolled as the study subjects.

This is a weakness in design of this study,  I think, the trained nurses were told for the control group to “follow the general pattern of imprecise communication that is typical of nonstandardized speech”. It is certainly possible that the nurse participants, who were probably invested in showing that this works, might have used even more imprecise phrases than usual. I don’t remember ever hearing “wow he’s crackly” (one of their examples) when asking about air entry, but I guess the point is that it could happen. I think it would have been better to use non-trained nurses as the controls, but I can see that would introduce other biases also.

The other problem is that although this seems like a major, and very obvious, improvement (although I didn’t think about it), the sample size was so small that most of the changes seen were not that significant. There were fewer errors of omission (failure to perform an intervention that was clinically indicated), cardiac compressions were started earlier, by about 8 seconds, PPV was started earlier by about 2 seconds. Communication techniques were used much more frequently; that comparison was statistically significant.

This is another place where I am not sure we need another trial; we should probably all start doing this, making sure communication is clear, by using standard phrases. I can’t see any down side, it wouldn’t cost anything, and the only thing a larger study is likely to find is that sometimes communication errors lead to screw ups.