Algorithms are used in many fields, such as finance, engineering, and aviation. They are used in healthcare, but the usage is only a tiny fraction of its potential. So why don’t clinicians use healthcare algorithms more often in the care of their patients?
Traditional Clinical Decision Making Still Prevails
Despite technological advances, many aspects of healthcare decision-making are still done by hand by an individual. Using an algorithm requires that several steps be taken, any one of which can become a potential barrier.
A healthcare algorithm may not be used for four main reasons:
(1) failure to initiate
(2) failure to access a relevant algorithm
(3) failure to execute the algorithm
(4) failure to use the result appropriately
The clinician may not use an algorithm for a problem because he or she does not know that an algorithm exists. Alternatively, the clinician may not believe that an algorithm is important, helpful or reliable. In many clinical situations it is assumed that an algorithm is not required and so can be ignored or put off to a later time that never comes.
If the clinician does want to use an algorithm then the algorithm has to be located. The clinician may have to do a literature search and track down one or more references, which may takes a lot of time searching online or actually visiting a library. More often than not the clinician does not know where to find it or does not have the time or patience to track it down.
If the clinician does find a complete and reliable description of the algorithm then it must be executed. This may require unit conversion of available data, or the ordering of additional tests. Some algorithms use mathematical functions that can be challenging to do by hand, necessitating the use of a calculator or computer. Some clinicians are not very good at math, so they give up in frustration.
Finally, the clinician needs to use the result once it is available. Even the most valuable knowledge is worthless if it is not used properly. The clinician may not know how to use the information or may simply stare at it. Finally, after all of these steps the clinician must document the findings in the medical record.
Integrate Algorithms into EHRs for Seamless Access
Integrating medical algorithms into the electronic health record can overcome these barriers. Most of the data retrieval, calculation and documentation in the medical record can be done automatically. The results can be linked to resources that explain how to use the information, reducing the cognitive load on the clinician. An important result can be flagged for action and kept flagged until it has been acted upon.
The implementation needs to take into consideration several important design features such as using appropriate triggers, selecting the optimal algorithm for a situation, and presenting the information in a helpful but nonintrusive way. With medical algorithms accessible through the electronic health record and proper design of the user interface, medical algorithms can be seamlessly integrated into medical practice.
