Balsis Steve, Cully Jeffrey A
Texas A & M University, College Station, TX, USA.
Aging Ment Health. 2008 Nov;12(6):800-6. doi: 10.1080/13607860802428000.
Depression in later life has potential grave implications and contributes to heavy emotional, medical, and economic burdens. Therefore, it is not surprising that identifying depression and its symptoms in later life has remained a sustained concern for professionals who treat older patients. Despite this concern, the current diagnostic gold standard may not identify depression symptoms equally well in older and younger adults. The objective of this analysis is to determine whether older and younger adults with equivalent levels of latent depression are equally likely to endorse particular DSM diagnostic symptoms.
We analyzed DSM depression data using Item Response Theory (IRT)-based differential item functioning analyses. The data came from 1808 older adults (age 65-98 years) and 3734 younger adults (age 18-34 years) who participated in the National Epidemiological Survey on Alcoholism and Related Conditions.
The analyses confirmed our hypothesis. The DSM items identify depression differently in younger and older adults. Specifically, results showed that older adults were more likely to endorse somatic items and less likely to endorse cognitive and suicide items than their younger counterparts with equivalent levels of depression.
These findings provide evidence that the DSM depression items work differently across age groups when controlling for latent depression. It is important to consider, however, that these findings are limited by the sampling methodology and the particular protocol implemented.
