Using the MoCA in a memory clinic — scoring edge cases I keep running into
I've been administering the moca test for about three years in a geriatric memory clinic and I wanted to start a thread on the scoring edge cases because the manual doesn't cover everything. Our clinic sees roughly 15-20 patients a week and there are a handful of situations that come up repeatedly where I'm genuinely unsure whether to award the point or not.
The biggest one is the clock drawing task. The instructions say you award 1 point each for contour, numbers, and hands — but what counts as acceptable number placement? We get patients who cluster the numbers in the right half of the clock or write them outside the circle but in roughly the right positions. Our team has different interpretations and I've seen us score the same drawing differently depending on who's doing the assessment. Would love to know what others are doing.
The naming task also comes up. If a patient says "rhinoceros" instead of "rhinoceros" or gives a phonologically similar answer, do you prompt? The standard says you don't give feedback, but I've had supervisors disagree on whether a mispronunciation constitutes a wrong answer or just a motor speech issue that shouldn't penalize the cognitive score. We've started noting the response verbatim and flagging it in the chart but it still creates inconsistency in our aggregate data.
Education adjustment is the other one — the +1 point for 12 or fewer years of education seems straightforward but we get a lot of patients who did their schooling in other countries and it's genuinely difficult to verify or compare years-of-education across systems. Anyone have a protocol for handling that?
The education adjustment issue is something our clinic ran into constantly with a large immigrant population. We ended up implementing a brief educational history form that asks about literacy level and years of formal schooling separately, with a field for country of education. It's not perfect but it gives us something to reference when the years don't map cleanly.
Clock drawing scoring is genuinely one of the weakest points in the MoCA's standardization. Our clinic uses the CLOX standardized scoring criteria as a secondary reference when we're unsure, even though it's technically a different instrument. It at least gives us something objective to point to when there's disagreement between raters.
For the naming task and mispronunciations, our neuropsych team made a rule that phonologically related errors get scored as incorrect but we document them separately in the notes. The concern is that being too lenient on naming creates upward score inflation that can delay referrals. It's an imperfect call but at least it's consistent within our site.
Inter-rater reliability on the MoCA is a real problem even within single sites. We do a calibration exercise every six months where two raters independently score the same recorded administration and then reconcile disagreements. It's tedious but our IRR went from 0.81 to 0.94 after we started doing it consistently.