Selection of veterinary students: does the interview do what we think and what we want?

The veterinary profession is rapidly changing and our veterinary curricula evolve and change to keep pace with the demands on and of the profession. At the same time there is a need to ensure that the students we select demonstrate the desired attributes and have the potential, to not only succeed within the veterinary curricula, but to flourish and thrive within the veterinary profession.
As the final step on which we base our offer decisions, the interview performance is critical, but is perhaps the part of the selection process most vulnerable to subjectivity, bias and assumption. One of the main drivers for the increasing use of the multiple mini interview (MMI) in university admissions has been the perceived reliability and robustness that this should offer1, 2.
At the R(D)SVS we have adopted the use of MMIs which is employed for all of our prospective veterinary students (national and international). Over three admissions cycles we have interrogated our data for any evidence of bias in the MMI process and evaluated the utility of the MMIs as predictors of academic performance and practical aptitude in the early years of the veterinary degree programme.
In other words, is the MMI doing what we need and want, is it offering any additional benefits and are there any unwanted and unintended consequences or biases?