Longitudinal Adhesion Test for Patches

Hello dear colleagues.

I have to compare the adhesion of two drugs in the form of transdermal patches. According to EMA's (European Medicines Agency) guidelines, the adhesion should be measured several times for each treatment, during the 24 hour period of administration of the drug, using scores from 0 (less than 70% of the patch adhered) to 6 (more than 95% of the patch adhered). The design that I am using is the 2 sequence, 2 period, 2 treatment, 2-way crossover, where subjects will receive the treatments in a randomized order, either Test drug first and Reference drug second or Reference drug first and Test drug second. In my study I'm using only 4 evaluation points per subject and treatment. So I have 4 score outcomes per subject and treatment, one for each evaluation point (e.g. 0, 6, 12 and 24 hours after attachment of the patch).
Then I'd have a ordinal outcome, measured through time (longitudinally), in a crossover manner.
Which statistical technique would be applicable for the statistical comparison of Test and Reference outcomes?

Thanks in advance.