Abstract
Background: Significant differences have been reported in the prevalence of depression between Mexicans and Mexican Americans. Whether the differences in prevalence are real or are the results of cultural bias in instrumentation is not known.
Objectives: To examine the association between acculturation and the responses to the individual depressive symptom item of the Center for Epidemiologic Studies Depression Scale (CES-D).
Methods: The CES-D was administered to 395 Hispanic women during their 22-24 weeks' pregnancy clinic visit. Acculturation was defined by a subject's language preference of English or Spanish when completing the CES-D. Those who preferred English belong to the acculturated group and those who preferred Spanish are considered nonacculturated. Logistic regression was used to conduct differential item functioning (DIF) analysis for each of the 20 CES-D items to determine whether they operated differently between the acculturated Hispanic group and the nonacculturated Hispanic group.
Results: Reliability of the CES-D total score was.86 and.88 for the nonacculturated and acculturated groups, respectively. None of the items of the Depressed Affect Subscale showed DIF. However, DIF was found in four out of the seven items in the Somatic Activity Subscale and in three out of the four items in the Positive Affect Subscale.
Discussion: The results suggest that even within a homogenous Hispanic group of mostly Mexican Americans, responses to the CES-D differ by acculturation. The nonacculturated group are less likely to endorse somatic symptoms but more likely to endorse positive items than the acculturated group. Depression screening in this population needs to account for acculturation differences within the Hispanic group.