Williams AL, Lasky RE, Dannemiller JL, Andrei AM, Thomas EJ. Teamwork behaviours and errors during neonatal resuscitation. Qual Saf Health Care. 2010 Feb;19(1):60-4. (PubMed)

OBJECTIVE: To describe relationships between teamwork behaviours and errors during neonatal resuscitation.

METHODS: Trained observers viewed video recordings of neonatal resuscitations (n = 12) for the occurrence of teamwork behaviours and errors. Teamwork state behaviours (such as vigilance and workload management, which extend for some duration) were assessed as the percentage of each resuscitation that the behaviour was observed and correlated with the percentage of observed errors. Teamwork event behaviours (such as information sharing, inquiry and assertion, which occur at specific times) were counted in 20-s intervals before and after resuscitation steps, and a generalised linear mixed model was calculated to evaluate relationships between these behaviours and errors.

RESULTS: Resuscitation teams who were more vigilant committed fewer errors (Spearman's rho for vigilance and errors = -0.62, 95% CI -0.07 to -0.87, p = 0.031). Assertions were more likely to occur before errors than correct steps (OR = 1.44, 95% CI 1.10 to 1.89, p = 0.008) and teaching/advising occurred less frequently after errors (OR = 0.59, 95% CI 0.37 to 0.94, p = 0.028). Though not statistically significant, there was less information sharing before errors (OR = 0.90, 95% CI 0.77 to 1.05, p = 0.172).

CONCLUSIONS: Vigilance is an important behaviour for error management. Assertion may have caused errors, or perhaps was an indicator for some other factor that caused errors. Teams may have preferred to resolve errors directly, rather than using errors as opportunities to teach their teammates. These observations raise important questions about the appropriate use of some teamwork behaviours and how to include them in team training programmes.

---

Comments:
Teamwork is a vague concept, but in the context of task performance such as resuscitation specific behaviors are being identified that have the potential to reduce error or, as this study suggests, in some instances cause or contribute to error. Another study, also by this University of Texas group, has demonstrated the value of including teamwork training in NRP.  Further, from the general performance improvement literature, a recent meta-analysis confirms the value of training for teamwork.

(Jim Handyside)