Validation of short screening tests for depression and cognitive impairment in older medically ill inpatients

To investigate the criterion validity of the four-item Geriatric Depression Scale (GDS4) and the six-item Orientation-Memory-Concentration-test (OMC) against longer widely used screening instruments.

Participants were 153 patients (aged 65 or over) admitted to four acute medical wards of a northern UK town. The validity of the GDS4 was determined using the 30-item geriatric depression scale (GDS30) as the comparator; the validity of the OMC was determined using the standardised mini-mental state examination (MMSE) as the comparator. For both screens, the area under receiver operating characteristic (ROC) curve was calculated in addition to the number of true and false positives and the sensitivity and specificity for various cut-off points.

The area under ROC curve was 0.80 for the GDS4 and 0.90 for the OMC. Using a cut-off of 0/1, the GDS4 correctly classified 78.2% of participants, using the GDS30 as the standard. This cut-off gave a sensitivity of 90.1% and specificity of 55.3%. With a cut-off of 1/2 the GDS4 correctly classified 76.8% of participants and had sensitivity and specificity of 78% and 74.5% respectively. The GDS4 and GDS30 were highly correlated (rho=0.63, p < 0.0005). A cut-off of 10/11 on the OMC gave optimum performance. With this cut-off, it correctly classified 85.9% of participants, and had 85.6% sensitivity and 86.8% specificity. There was a significant correlation between the OMC and the SMMSE (rho = -0.827, p < 0.0005).

The GDS4 and OMC appear to be useful instruments for screening for depression and cognitive impairment among older medical inpatients.