The pareto-analysis in medical education
In 1887, the Italian economist Vilfredo Pareto observed an exponential relation between the amount of wealth an inhabitant owned and the rank-order of the inhabitant. He discovered that 80% of the property is owned by merely 20% of the inhabitants, a pattern which later was popularized in the 1950s by management consultant Joseph M. Juran as the Pareto-principle or ‘80–20 rule’. The Pareto-principle is best known for its use in increasing business returns by identifying the vital-few causes responsible for the bulk of income within a company and consequently increasing its efficiency by focusing investments on these company facets. But the Pareto-principle has also been observed in many other fields such as the frequency of words in any human language, the force of earthquakes and the number of hits on web pages. Does the Pareto-principle exist in medical education, in specific, the surgical training of a basic surgical procedure?
Previous studies that have described content criteria for surgical training based their findings mainly on cognitive task analysis, human reliability analysis or expert opinion. While these methods all provide valuable information for surgical training curriculum development, they do not provide us with a description of the aspects of surgical expertise that requires the most time and energy during training in the OR. Meanwhile, the Pareto-analysis might provide a valuable tool in the reduction of training duration in the OR by identifying those aspects of surgical skills that require the most resources to instil in trainees. Applying the Pareto-principle might be a solution.
Together with the laparoscopic surgery research team of the Medical Center Leeuwarden in the Netherlands, I conducted an analysis of operating room videos to assess the verbal guidance provided by 9 supervising surgeons to 12 trainees performing 64 laparoscopic cholecystectomies in the OR. The verbal corrections were documented, tallied and clustered according to the aimed change in novice behaviour as shown in the figure below.
The corrections were rank-ordered and a cumulative distribution curve was used to calculate which corrections accounted for 80% of the total number of verbal corrections. In total, 253 different verbal corrections were uttered 1587 times and were categorized into 40 different clusters of aimed changes in novice behaviours. The 35 highest-ranking verbal corrections (14%) and the 11 highest-ranking clusters (28%) accounted for 80% of the total number of verbal corrections. Below is the cumulative distribution curve of the latter and a table overview of clusters of verbal corrections contributing 80% of the total number of corrections given by clinical supervisors.
In summary, the majority of verbal corrections in the OR were directed towards a relatively small number of novice behaviours. By using the Pareto-analysis we can identify which behaviours are the most difficult and therefore take the most time and energy to teach trainees in the OR.
I expect that the same principle is true for other kinds of medical training and I would therefore hope to see more medical training and education centres do a similar analysis. We can use the results to try to reduce the necessary resources for training. The outcomes can be used to tackle the relatively small amount of stumbling points that are responsible for the bulk of training. It’s time we make use of this more than 100 years old principle to improve the efficiency and efficacy of our educational system.
This analysis was part of an article published by the journal of surgical education in 2016: “Kramp, K. H., van Det, M. J., Veeger, N. J., & Pierie, J. P. E. (2016). The Pareto analysis for establishing content criteria in surgical training.