A systematic review and meta-analysis of the diagnostic accuracy of the Phototest for cognitive impairment and dementia

INTRODUCTION The recently developed Phototest is a simple, easy and very brief (<3 minutes) test with theoretical advantages over available dementia screening tests. Our objective was to evaluate its diagnostic accuracy under routine clinical conditions. MATERIAL AND METHODS A phase II cross-sectional validation study of diagnostic tests was performed in a sample of 308 patients referred to a general neurology department and in a group of 70 healthy individuals. The diagnostic accuracy (DA) of the Phototest was assessed and compared with that of the Eurotest and a verbal fluency test (VFT) in relation to the clinical diagnosis of dementia (DEM) and cognitive impairment (CI) by calculating the area under the ROC curve (aROC) and determining Sensitivity (Se), Specificity (Sp) and likelihood ratios. RESULTS The total sample comprised 225 subjects without CI (NOR), 58 with CI and without DEM and 95 with DEM. Phototest results showed a normal distribution in NOR subjects (33.4 +/- 3.9 [mean +/- standard desviation]) and were not influenced by educational variables. The DA of the Phototest for DEM and CI (0.95 +/- 0.01 [aRO C+/- Se]) was similar to that of the Eurotest and higher for both tests than that of the VFT. The cutoff points of 25/26 for DEM (Se=0.88 [0.80-0.94], Sp=0.90 [0.86- 0.93]) and 28/29 for CI (Se=0.90 [0.84-0.94], Sp=0.90 [0.83-0.93]) maximised the sum of Se and Sp. CONCLUSIONS The Phototest is a very short test of easy application that is applicable to illiterate subjects, uninfluenced by educational variables and useful to identify CI and DEM in routine clinical practice.