02662nas a2200169 4500000000100000008004100001653001500042653001100057653002400068653001200092653003200104653001800136100001300154245015100167300000700318520216700325 2017 d10aAdaptation10aBrazil10aCultural validation10aleprosy10aSocial Distance Scale (SDS)10aSocial stigma1 aBraam JF00aAdaptation and cultural validation of the Social Distance Scale (SDS) to evaluate social stigma related to leprosy in Olinda, Pernambuco (Brazil). a633 a

Introduction Stigma can have negative impact on the lives of people affected by neglected tropical diseases (NTDs), of which leprosy is an typical example. Data on stigma can be used to increase understanding of community stigma and can contribute to reduction of community stigma attached to persons affected by NTDs. In order to assess stigma, instruments should be culturally valid in order to be used in a particular setting. The aim of this research is to conduct an instrument validation of the Social Distance Scale (SDS) that assesses community stigma attached to people affected by leprosy in Olinda, Pernambuco (Brazil).

Methods

A process of translation, discussion and training was followed by 20 semi-structured interviews to assess conceptual, item, semantic and operational equivalence. Thereafter, 96 community members were selected through convenience sampling, and 50 repeated measures were obtained to assess measurement equivalence.

Results

The concepts and items seemed equally relevant and important in the target culture. The respondents understood the vignette, questions and answer options. No changes needed to be made in the scale format or setting. Cronbach’s alpha was 0.87. ICCagreement was 0.73 (95%CI 0.56-0.84) and Kappa measure of agreement was 0.84 for intra-interview reliability and 0.47 for inter-interview reliability. SEMagreement was 2.75, SCDindividual 7.6 and SDCgroup 1.07. A floor effect was found (21.9%), but no ceiling effect.

Discussion

The findings of this study indicate that the SDS scores adequate for conceptual, item, semantic and operational equivalence. All psychometric properties were considered to be sufficient in the current context, except for a floor effect that was found. Differences between the two measure moments of the questionnaire interviews seemed to be caused by respondents who changed their answers after learning more about the disease and/or respondents who did not want to appear to be prejudiced in the company of the local healthcare agent.