Data-driven order set generation and evaluation in the pediatric environment.

From Clinfowiki
Jump to: navigation, search

Article Review Zhang Y, Levin JE, Padman R. Data-driven order set generation and evaluation in the pediatric environment. AMIA Annu Symp Proc. 2012[1]


CPOE has the potential to improve care through improved order legibility, completeness, and advanced clinical decision support features to guide care and catch errors before they reach the patient. Typically, orders are entered in to CPOE system one item at a time. However, even the earliest implementations included items grouped by clinical purpose into "order sets".[1]

Order sets may improve care delivery by making it faster and easier for the physician to enter orders and by guiding care according to known best practices. The creation of order sets may be labor intensive, not only in their creation but in their maintenance. While order sets derived from scientific evidence-based guidelines are the industry standard, the increasing amounts of data related to orders that are compiled via CPOE and networked across enterprises provide the capacity to automate order set creation from practice-based evidence derived from historical data, including the ability to derive new types of order sets.[1] In addition, K-means clustering was applied to orders to generate evidence-based order sets that are learned from historical hospital data.

Scope of Study

This study examines current utilization patterns of order sets and "a la carte" (individual order entry) orders in a pediatric environment with a preliminary investigation of methods to automate the creation and modification of order sets using historical ordering data. We examine the current usage of order sets associated with Asthma Minor and Appendectomy Minor patients to understand how physicians are utilizing order sets, and how order sets usage is associated with the time of ordering and characteristics of order sets.[1]


Order sets usage increase with order activities. However, low coverage rate of coverage rate of order set items reflect the fact that physicians often have to go through repetitive mouse-clicks to un-select items out of order sets when prescribing, resulting in inefficiency and non-use. Data-driven order set creation and testing was attempted utilizing historical data. It was demonstrated that order set composition was improved from evidence-based only order sets by using K-means clustering (a simple and widely used machine learning technique). Performance comparisons suggest that order sets generated using clustering can capture more order items than current ones, and with higher coverage rate.[1]


Order sets are intended to increase efficiency, enter a group of orders faster and easier. However, when the sets are not set up according to physician's historical practice, order sets may not be used as frequently due to the constant revision, checking, un-checking of individual components. It is necessary to take in consideration evidence and data base information to make sure that order sets are appropriately created. Before I read this study, I thought that order sets should not be generated primarily using physicians' preferences as the only criteria since if each physician request his own order set, it will be a difficult task to maintain all of them. This study did not evaluate the maintenance of the sets which would help out to determine the amount of labor it will be needed.


  1. 1.0 1.1 1.2 1.3 1.4 Zhang Y, Levin JE, Padman R. Data-driven order set generation and evaluation in the pediatric environment