Evaluation of a shortened version of the Abbreviated Mental Test in a series of elderly patients

Abstract

Objective: To determine whether a shortened version of the Abbreviated Mental Test is as effective as the Abbreviated Mental Test (AMT) itself in assessing cognition in elderly patients.

Design: A shortened four-item version of the Abbreviated Mental Test (AMT4) was constructed using the following items: (1) Age, (2) Date of birth, (3) Place, and (4) Year, with impaired cognition indicated by an AMT4 score of less than four. Patients were assessed with the AMT. The AMT4 scores were then determined and matched against AMT scores. The performance of all 210 possible four-item combinations derivable from the AMT was assessed and ranked according to predictive efficiency (the percentage of patients whose cognition as judged by the AMT was correctly categorized by each four-item combination).

Setting: Inner-city teaching hospital.

Subject: Two hundred consecutive elderly patients seen on domiciliary visits, in the clinic or as orthogeriatric referrals.

Results: The AMT4 score showed a statistically significant correlation with AMT score (Somers' d statistic 0.90: p < 0.001). The AMT4 had a predictive efficiency of 91% and ranked 7 = /210 possible four-item combinations.

Conclusions: The AMT4 may be useful in the initial assessment of cognition in elderly patients, with little loss of accuracy in detecting marked cognitive impairment when compared to the AMT.