Altered mental status is one of the strongest predictors of clinical deterioration in the hospital, yet it often goes unrecognized. Current tests to assess mental status in the hospital lack the sensitivity to capture small changes in mental status and are cumbersome to administer, making serial administration challenging. Accurate and timely assessment of cognition in the hospital could help identify clinical deterioration earlier and prompt further assessment and therapy.
We developed the Functional Assessment of Mentation (FAM), a self‐administered electronic mobile device application to assess memory and attention. We conducted a prospective observational pilot study at an urban academic medical center. Subjects included a convenience sample of healthy adult volunteers encountered by trained medical student researchers between July and August of 2013. Subjects were assessed via an iPod Touch application using orientation questions and a test of memory and attention.
During the memory and attention portion, subjects were shown a random sequence of flashing numbers followed by a screen prompting them to reproduce the sequence by pressing on the unlit numbers in the correct order. Through computer‐aided testing, correct reproductions of the sequence led to an increase by two in the next sequence (i.e., the difficulty of the sequence increased). Incorrect reproductions of the sequence led to a decrease by one in the following sequence (i.e., the difficulty of the sequence decreased). After two failures to reproduce a sequence, a final sequence was shown to aid in score determination. The final score reflected the highest level completed by the subject. FAM completion time was recorded by the mobile device application.
Of the 652 individuals consented for the study, 624 (95.7%) completed FAM. Subjects were predominantly women (57.0%) with at least some college education (80.0%). The mean age was 52 ± 18 years. Scores on the FAM were normally distributed with a median score of 5 (IQR 4‐7). The median time for completion of the FAM was 41s (IQR 32‐49s). FAM score decreased with increasing age quartiles and increased with increasing education. Specifically, the age quartile representing 18‐38 years of age had a mean score of 6.3 while individuals over 65 had mean scores of 3.4. Subjects with less than high school education had a mean score of 2.2, while individuals with some college education or more had a mean score of 5.3. In multivariate analysis, lower age, higher education, and white race were all independently associated with higher FAM scores.
We found that the Functional Assessment of Mentation (FAM) is quick and feasible to administer, yielding a wide distribution of scores. Similar to other tools assessing mental status, younger age, higher education level, Caucasian race, as well as orientation correlated with higher FAM scores while gender was not significantly associated with FAM score.