Reasonable suspicion: a pilot study of pediatric residents

To identify pediatric residents' understanding and interpretation of reasonable suspicion, in the context of mandated reporting of suspected child abuse.

A survey was administered to pediatrics and combined medicine/pediatrics residents. An open-ended question plus three operational frameworks for interpreting likelihood examined how residents conceived of reasonable suspicion. Responses were examined for evidence of a group standard, and also compared for internal consistency.

Forty-two of 49 residents completed the survey (86% response rate). There were no significant differences in responses based on age, gender, year of residency, or anticipated practice type. Respondents exhibited wide variation in the thresholds they set for reasonable suspicion. On a Differential Diagnosis scale, 10% indicated that "abuse" would have to rank 1st or 2nd; 45% set the threshold at 3rd or 4th; while 45% stated that abuse could be as low as 5th to 10th and still qualify as reasonable suspicion. Using a Estimated Probability scale, 9.5% indicated that "abuse" would need to be >75% likely before reasonable suspicion existed; 28.5% stated that a 60-70% likelihood was needed; 38% identified the necessary likelihood as 40-50%; and 24% set the threshold as low as 10-35%. In comparing individual resident responses for the two scales, 83.3% were internally inconsistent.

There was no consensus among pediatric residents with regard to (1) a standard meaning for reasonable suspicion, (2) a standard application of reasonable suspicion, or (3) how likely "abuse" must be before reasonable suspicion can be said to exist. Additionally, many residents' conceptions of reasonable suspicion were internally inconsistent.