Unintended Outcome: Issue Spotlight

Tina Dieckhaus, RN, MSN, CPHIMS, Director of PatieHIMSS Clinical Informatics Insights

Technology projects are implemented to address an issue.  This issue can be well supported by evidence or be perceived. When technology is implemented it is often accompanied by unintended outcomes, both positive and negative.  The literature often describes negative unintended consequences of CPOE. (Cowen, 2013)  CPOE was primarily implemented to address errors in transcription that can be associated with illegibility or simply human error.  It was also intended to shorten the timeframe between order conception and order entry.

Another intended consequence was the ability to present clinical decision support information to the provider at the time of order entry.  After implementing CPOE, many institutions saw an increase in a new category of errors including providers selecting “look-alike” orders or miss-clicking when presented with a list of order detail options and issues in workflow by providers entering orders remotely.  Informatics professionals need to be diligent in anticipating, identifying and addressing these negative unintended consequences, but we also experienced positive unintended consequences.

One unexpected positive outcome is technology-spotlighted problems that previously existed.  This allows the institution to quantify and address them.  A specific example is the practice of waiting until the end of the shift for nurses to complete their documentation.  The results of patient assessments can be critical to allow the care team to make informed clinical decisions, and not having the results available can delay the decisions or result in inappropriate decisions made with incomplete information.  Electronic date/time stamps allow the institution to determine how long the average delay is, identify orders that may have been made before that data was available and to track if action plans have been effective in changing the practice. 

Another unexpected outcome we experienced was associated with the implementation of barcoded electronic positive patient identification for blood administration.  The intent was to use technology to assist in assuring that the correct product was being given to the correct patient and to alert the administering nurse if any issues were present.  With this technology, we were able discontinue the use of specific blood bank ID bracelets because the blood product tag contains the electronic mapping of patient MRN and name with the BBID for comparison.

We were also able to remove the need for a second independent check of the patient, tag and product during administration because the barcode scanner is the independent check.  This significantly reduced the amount of time to administer blood because nurses needed a co-worker to check blood with them.  The electronic process also eliminated transcription errors associated with transcribing the product ID into the EHR because the scanning process automatically uploaded that information in the nursing documentation.

A third unintended consequence was the elimination of patient events associated with our pediatric patients eating ID bands. 

A fourth unexpected outcome we experienced was in addressing communication lapses. As noted, one of the unexpected negative consequences is the changes in communication patterns associated with physicians placing orders remotely and nurses being unaware of the new orders. We have long-standing issues with communicating not just to the nursing staff, but also to the performing department. We have been able to implement system rules that now will enhance the communication process by paging and emailing both to the nurse and the performing department when a new order is entered. So now, when a physician enters an order, instead of waiting for the nurse or unit clerk to see the order and call the performing department, once the order is signed, the system will page or mail the performing department to let them know if an order has been entered, changed or canceled. 

It is going to continue to be important that the Informatics Team clearly communicate the expected benefits, anticipate and mitigate negative unintended consequences and celebrate the unintended positive consequences of these projects.               

Cowan, L. (2013) Literature review and risk mitigation strategy for unintended consequences of computerized physician order entry. Nursing Economc$, January-February 2013, 31(1) 23-31,11.

About the Contributor

Tina Dieckhaus is the Director of Patient Care Services Informatics at St. Jude Children’s Research Hospital and has worked in an informatics role at St. Jude since March 2000.  She has a BSN from Mississippi University for Women and her MSN in Nursing Informatics from Tennessee State University.  She has been practicing as an informatics nurse since 1997 and holds ANCC certifications both in Informatics and Nurse Executive.  She has been a member of HIMSS since 2004 and obtained her CPHIMS certification in 2008.