HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS REGISTER		Author Keyword(s) Vol Page [Advanced]

This Article Full Text Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Add article to my folders Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Search for Related Content Topic Collections EBM Notebook

Jottings ...

The first 150 words of the full text of this article appear below.

Less is more. At least when it comes to diagnosis, where simplifying a tool can make it more readily usable at the bedside. In this issue are several short or shortened tools for diagnosis and prognosis. Shorter tools are not just timesaving: they can be as accurate or more accurate for 2 reasons. First, adding weak diagnostic information to good information can degrade the overall performance: as we add items to a clinical prediction rule accuracy initially improves but then begins to get worse as we add weaker information. The poor information pollutes the good. Statisticians call this "overfitting" of a model. Second, a more complex model may be less robust when transferred to new settings.

If during a 10-minute general practice consultation the possibility of dementia occurs to me, doing a Mini-Mental State (MMS) exam is usually not feasible. So I use an alternative 2-minute test— the MiniCog (the . . . [Full text of this article]