Development of alternative methods of data collection in South Asians with Type 2 diabetes

Previous research in the UK has established the difficulty of recruiting and collecting information from individuals whose main language is spoken and does not have an agreed written form. The aims of this study were (i) to develop culturally competent translations of two questionnaires measuring diabetes self-care in languages with no written form and to establish their face validity and (ii) to develop acceptable methods of delivery with the potential for providing valid and reliable data for use in research studies.

Adults with Type 2 diabetes from two minority ethnic groups whose main language is spoken (Sylheti and Mirpuri) were recruited via the Birmingham Heartlands hospital diabetes centre. Participants were invited to attend five focus groups to consider the content and method of delivery of two questionnaires measuring knowledge of diabetes and confidence in diabetes self-care. Delivery methods were (i) pen and paper self-completion in Bengali/Urdu, (ii) pen and paper assisted completion in spoken language, (iii) partially-assisted completion in spoken language, (iv) independent audio delivery in spoken language.

Culturally competent content was achieved for both questionnaires in both languages. The Mirpuri men and women's groups preferred assisted or partially assisted completion in spoken language. The Sylheti groups' preference was for independent audio delivery in spoken language.

The face validity of two questionnaires measuring diabetes self-care is established for Sylheti and Mirpuri, in four alternative delivery formats. The questionnaires are now ready for psychometric testing in minority ethnic populations and the methods available for use by researchers to establish within-study feasibility.