Increasingly, students must demonstrate knowledge in mathematics through mathematics writing, yet research lags in understanding how students engage in mathematics-writing tasks. Most available research on mathematics writing focuses on typically achieving students without considering students with mathematics difficulty (MD). In this study, we explored how students with MD who participated in a word-problem intervention randomized-control trial performed on a mathematics-writing task. We sampled 144 third-grade students with MD and evaluated student performance on an explanatory mathematics-writing measure. Overall, students with MD, on a mathematics-writing rubric with five categories, scored between 1 and 2 points out of a possible 5 points for each category. For one of the five rubric categories, Mathematics Content, the students in the word-problem interventions marginally outperformed the students in the business-as-usual condition. On average, students wrote 33.3 words, numbers, and symbols in response to the mathematics-writing prompt with an average of 8.7 mathematics vocabulary words. Of the mathematics vocabulary words used, students most frequently used formal mathematics vocabulary and names for symbolic numbers, with symbolic symbols and general vocabulary used to a lesser extent. The trends in this study will support future research to enhance mathematics-writing instruction.