387 - Assessing Reliability of EHR-Documented Pain Scores in the Pediatric Emergency Department

Poster Number: 387
Publication Number: 387.406

Ryan W. Czarnecki, Medical College of Wisconsin, Wauwatosa, WI, United States; Elizabeth Alpern, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States; Blake Nielsen, University of Utah School of Medicine, Sandy, UT, United States; Michael Webb, Biostatistics/research, Salt Lake City, UT, United States; David Brousseau, Medical College of Wisconsin, Milwaukee, WI, United States; Julie C.. Leonard, Nationwide Children's Hospital/The Ohio State University College of Medicine, Columbus, OH, United States; James Chamberlain, Children's National Health System, Washington, DC, United States; Joseph J. Zorc, Children's Hospital of Philadelphia, Philadelphia, PA, United States; Theresa Frey, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Alexandria Wiersma, University of Colorado School of Medicine, Aurora, CO, United States; Samina Ali, University of Alberta, Edmonton, AB, Canada; Bradley J. Barney, Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States; Amy L. Drendel, Medical College of Wisconsin, Milwaukee, WI, United States

Background: Pain scores are documented in the electronic health record (EHR) to characterize pain and the effectiveness of emergency department (ED) pain treatment. Typically, highest ED pain scores are a marker of severity and severe pain is associated with increased opioid use. The last ED pain score prior to discharge represents pain at discharge and is used to evaluate the effectiveness of ED interventions. However, the reliability of EHR pain scores for these pain experiences has never been investigated.

Objective: To determine concordance between parent-reported pain scores via text messaging and EHR-documented pain scores.

Design/Methods: We conducted an analysis of the IMPROVE study, a prospective cohort study of children, 4-17 years discharged home after ED evaluation for a long bone fracture. Parents reported their child’s pain experiences via text for highest ED pain and pain at discharge using a 0-10 scale [none/mild (< 4), moderate ( >=4 and < 7), severe ( >=7)]. EHR recorded pain scores were extracted from the PECARN Registry and used to derive analogous metrics. We tested for differences between sources, and stepwise selection identified characteristics associated with concordance (absolute difference < = 1). Per prior research, a difference in pain scores of 2 or more is clinically significant.

Results:
2381 parents from 7 EDs were included. Children’s mean age was 9.6 (SD 3.5), with majority male (63%), and these race/ethnicities: non-Hispanic White (58%), Hispanic (19%), non-Hispanic Black (12%). Most were upper extremity (87%) and non-reduced (51%) fractures. 95% had an EHR pain score recorded during the visit; only 45% had an EHR pain score recorded during the hour prior to discharge.



Highest ED pain score was significantly higher by parent text [mean 7.4 (SD 2.3)] than by EHR-recorded score [5.3 (SD 3.2); difference (2.1, 95% CI: 2.0, 2.3). There was a higher likelihood of concordance if the EHR indicated moderate pain (aOR, 5.5; 95% CI: 4.1, 7.3) or severe pain (aOR 24.7; 95% CI 18.5, 33.0) compared to mild pain.

ED discharge pain score was also significantly higher by parent text [mean 3.3 (SD 2.4)] than EHR-recorded score [mean 1.3 (SD 2.2)]; difference (2.0, 95% CI: 1.8, 2.2). There was a higher likelihood of concordance if the EHR indicated moderate pain than if it indicated none/mild pain (aOR, 1.7; 95% CI: 1.1, 2.5).


Conclusion(s): Overall, parents’ texted report of their child’s pain was significantly higher than pain scores recorded in the EHR. Concordance appeared worst for low EHR pain scores. The impact of these differences on ED clinical care and pain research should be investigated.