Based strictly on these numbers, it doesn't immediately follow that there is some characteristic inherent to former nurses that puts them at a disadvantage in medical school admissions. At the very least, you would want to stratify your numbers by, say, MCAT, undergrad GPA, and race/ethnicity -- so you can get an idea of whether the applicant pool of former nurses somehow systematically differs from the general applicant pool.
(If, for example, the applicant pool of former nurses is disproportionately comprised of non-racial/ethnic minorities with lower undergrad GPAs and lower MCAT scores, then those three attributes alone might explain the 15 percentage-point difference in acceptance rates. To conclude that they were unsuccessful 'because they were formerly nurses' would be a mistaken inference -- because you would be mistakenly attributing to their being former nurses an effect which you should be attributing to their racial/ethnic grouping and lower mean GPA/MCAT. But that's just one example. Whether the applicant pool of former nurses somehow systematically deviates from the overall applicant pool is an empirical question that we can't answer in the absence of further data.)
-AT.