Non-word reading fluency predicts reading difficulty for ELL students and native English speakers equally well

The correlation analyses [in this study] expand upon other studies by demonstrating that the large correlation of NWF [Nonsense Word Fluency] with real-word reading (Swanson et al., 2003) occurs with ELL students as well as with native English-speaking students

Vanderwood, M., Linklater, D., & Healy, K. (2008). Predictive Accuracy of Nonsense Word Fluency for English Language Learners. School Psychology Review, 37 (1) p. 5-17.

So you don't need a different test to identify ELL students as at-risk for reading.

In general, current findings concur with the DIBELS criterion that a score of 50 or better on NWF at the end of first grade indicates low risk for future reading problems (Good & Kaminski, 2002). These data suggest educators can use NWF in first grade to identify ELL students who need additional reading instruction regardless of the student’s English language proficiency.

Vanderwood, M., Linklater, D., & Healy, K. (2008). Predictive Accuracy of Nonsense Word Fluency for English Language Learners. School Psychology Review, 37 (1) p. 5-17.