On a rank order scale of one to 10, (with one being most likely diagnosis and 10 the least likely), physicians were asked how high suspected child abuse would have to rank before they felt it amounted to reasonable suspicion. Twelve percent of respondents indicated that "abuse" would have to rank one or two to qualify as reasonable suspicion. Forty-one percent set the threshold at three or four, and 47 percent stated that abuse could rank as low as five to ten on the scale, suggesting that for some, any suspicion at all warranted reporting to child protection services.

Using another measurement scale, the same physicians were asked how likely suspected child abuse would have to be before it amounted to reasonable suspicion. Fifteen percent of respondents indicated that abuse would need to be greater than 75 percent likely before reasonable suspicion existed. Twenty-five percent stated that a 60 percent to 70 percent likelihood was needed, 25 percent identified the necessary likelihood as 40 percent to 50 percent, and 35 percent set the threshold as low as 10 percent to 35 percent.

"Not only was there wide variation regardless of physician age, training, or years of experience, but when we compared the individual responses for the two scales, 85 percent were internally inconsistent, meaning that the physicians answered differently depending on the scale used. This is alarming because this suggests there is no consistency in how the majority of physicians surveyed interpret the threshold for when to report suspected abuse," Levi said. "I think we took it for granted that the concept of reasonable suspicion was clear to everyone, but our data show that it in fact has a wide range of meanings not only between physicians, but within the same physician."

Levi and Brown plan to more broadly administer their survey to see if the current result
'"/>


5-Jul-2005