Citation numbered differently when listed alone vs with adjacent refs

I have 3 refs and 2 sentences that look essentially like this:

Sentence 1: Text text text {Ref A;Ref B;Ref C}.

Sentence 2: Text text text {Ref C}.

When I format/update citations, Refs A and B format just fine and are assigned their appropriate citation numbers (“7” and “18”, respectively), but Ref C is assigned citation number “16” when it’s all by itself in Sentence 2 but “116” when it’s adjacent to the others in Sentence 1. When I break apart the group of citations in Sentence 1 (ie, Text text text {Ref A} {Ref B} {Ref C}), then Ref C formats to its correct “16” but as soon as I link the 3 refs back together with semicolons inside the brackets, it goes back to “116”.

To fix this, I have tried (1) to clean up the field codes, (2) to delete Ref C from my library (it had been pulled off PubMed) and recreate it manually, (3) to switch around the order of Refs A, B, and C prior to formatting, (4) to view the field codes in my Word doc then delete everything having to do with the citations at the end of Sentence 1 and reinsert them (individually or as a batch), but nothing has worked and I am absolutely stumped!

Has anyone else encountered this problem and/or know how to fix it?!

what output style are you using?  Does it happen with any numbered style?  

I’m using the style for the journal Neuroepidemiology, downloaded from EndNote’s style library. And no, it doesn’t happen with other numbered styles (eg, Vancouver). I also just realized that in my comment above, when I said that reference “18” was fine, I was incorrect. It actually should be reference “8” but an additional digit got added to this one, too.

I downloaded the output style but I can’t get this to happen in test documents.  Can you attach a few pages of the formatted document (you probably will need to ‘zip’ it so the document can be attached to the forum).  You might first want to try to download a fress copy of the output style, just in case it got corrupted?  I attach the version I used.  

Neuroepidemiology.ens (12 KB)

Downloaded the style again and problem is now fixed. I will touch base again if it recurs. Somehow the style must have gotten corrupted. Thanks!