PURPOSE: Cognitive interviewing techniques were used to adapt existing measures for use with a population of low-literate Spanish-speaking people with diabetes.

METHODS: Five individuals of Caribbean origin with diabetes participated in cognitive interviews for 4 instruments (measuring diabetes knowledge, quality of life, self-management, and depression) adapted for oral administration to low-literate individuals. Audiotaped interviews and handwritten notes were subjected to content analysis to identify problems across the 4 instruments as well as specific to a given instrument.

RESULTS: The following key problems were identified: general instructions were not helpful, items that were not specific enough generated a variety of interpretations, some wording was confusing, abstract concepts were difficult to understand, some terminology was unfamiliar, and interpretation of certain words was incorrect.

CONCLUSIONS: The data illustrate the usefulness of cognitive interviewing as a first step in the process of adapting measurement instruments.