Download Full Text (659 KB)


Background: Clinical presentation, imaging, and ophthalmologic findings are important factors in distinguishing between noninflicted and abusive head trauma (AHT) in infants. However, little is known about agreement between pediatric subspecialists regarding diagnosis and timing of AHT in infants. Hypothesis/Objects: The primary outcome was differences in rates of AHT diagnosis among different types of pediatric subspecialists using case examples of infant head injury. The secondary outcome was qualitative trends in clinical reasoning related to injury timing. Methods: Four case examples of infant head injury were developed into an online survey. Cases were categorized as: (1) “bilateral mixed density subdural hemorrhages (SDHs)”; (2) “hyperdense right/interhemispheric SDH”; (3) “bilateral SDHs with membranes”; and (4) “hyperdense right SDH”. Each case assessed diagnosis regarding infant head injury, likelihood of AHT diagnosis, and timing of head injury given clinical presentation, laboratory, neuroradiology, and ophthalmologic findings. Participants evaluated at least 1 possible case of AHT during their career and identified as being a part of 1 (or more) of the following 5 pediatric subspecialties: Child Abuse Pediatrics (CAP), Pediatric Emergency Medicine (PEM), Pediatric Critical Care (PCC), Neurosurgery (NS) and Pediatric Hospital Medicine (PHM). The response selected by most CAPs was the reference and compared across subspecialties using Chi-square or Fisher’s exact tests. Bonferroni corrections were used to determine statistical significance. Qualitative responses were coded individually, with more frequent responses being grouped into such categories as clinical presentation (i.e., symptoms, labs, or other information provided in clinical vignette), neuroimaging findings (i.e., SDH characteristics) and ophthalmologic findings (i.e., presence and/or type or retinal hemorrhages). Results/Conclusion: A total of 288 participants completed at least 1 case. Roughly 26.7% of participants were CAPs, and 64% of participants reported evaluating at least 26 cases of possible AHT during their careers. For case 1, 100% of CAPs diagnosed AHT with no significant difference detected across subspecialties (100% PEM, 83.3% PCC, 100% NS, and 97.1% PHM diagnosed AHT). For case 2, 57.9% of CAPs did not diagnose AHT, but significantly fewer PHM (34.9%, p=0.0061), PEM (28.0%, p<0.0001) and NS participants (24.0%, p=0.0033) did not diagnose AHT. For case 3, 72.2% of CAPs diagnosed AHT with no significant difference detected across subspecialties (73.6% PEM, 66.7% PCC, 80.8% NS, and 84.1% PHM diagnosed AHT). For case 4, 39.4% of CAPs reported uncertainty with AHT diagnosis, with no significant difference detected across subspecialties (34.8% PEM, 50.0% PCC, 42.3% NS, and 34.3% PHM reported uncertainty with AHT diagnosis). In case 1, most participants reported neuroimaging findings as most helpful with injury timing. In cases 2 through 4, most participants reported clinical presentation as most helpful with injury timing. Statistically significant diagnostic variability across pediatric subspecialties was detected in only 1 case example of infant head injury; however, variability persisted across the 3 remaining cases. Further research and/or education regarding the determination of AHT diagnosis and timing of infant head injuries is warranted to aid in the medical decision-making process and decrease diagnostic variability.

Publication Date




When and Where Presented

Presented at the Annual Ray E. Helfer Society Meeting; San Diego, California; April 10-13, 2022

Examining Diagnostic Variability Among Pediatric Subspecialists in Case Examples of Infant Head Injury

Included in

Pediatrics Commons