Developing a measure of stigma-by-association with African-American adolescents whose mothers have HIV

Background and Purpose: HIV disease is one of the most stigmatized illnesses in modern history. People with HIV report stigma as a primary stressor and family members experience courtesy stigma or stigma-by-association. As African-American women account for a growing percentage of people with HIV and as people with HIV are living longer, increasing numbers of African-American youth are affected by their mothers' HIV. Unfortunately, several studies report that HIV-affected adolescents who know their mother's status have higher rates of emotional distress, thought problems, substance use, and sexual risk, than those who do not know mother's status. Stigma may be an important mechanism that explains these youths' increased rates of mental health problems and risk behavior. A literature review, however, yielded no measures of HIV stigma-by-association relevant to adolescents or comprehensive in scope.

Methods: To address the need for a comprehensive and valid stigma measure for use with HIV-affected adolescents, the authors followed a multi-stage process. To begin, the authors -- experts in HIV, stigma, adolescents, families, and instrument development -- modified the HIV Stigma Scale for HIV-positive adults (Berger, Ferrans, & Lashley, 2001) for relevance with secondary stigma or stigma-by-association and with adolescents. In the second stage, nine African-American HIV-affected teens participated in cognitive interviews (Willis, 2004) approximately 2 hours in length. During cognitive interviewing, adolescents gave a Likert-type response to a scale item and then were asked to ‘think aloud' about their response, e.g. “when you answered this question, what kinds of things did you think about?” The teens were also asked about specific words in the items, e.g. disgusting or energy, to determine if the words had the same meaning for them. These methods assessed comprehension, content validity, and cultural and developmental relevance. Items were deleted or modified based on this process. The last stage was data analytic with a pilot sample (N=27) of African-American teens living with HIV-positive mothers. Using item analysis and examination of item-total correlations, the measure was further refined and initial psychometric properties established. Qualitative data from four open-ended questions were also examined for domains or items not captured by the instrument.

Results: Through multiple stages of testing, the HIV Stigma-by-Association Scale (Adolescents) went from 45 items to a 23-item measure with good internal reliability (α = .866) with African-American teens between the ages of 13 to 18 who are living with mothers with HIV. An additional item, suggested by qualitative data, will be added to the measure and used in further testing.

Conclusions and Implications: The resulting scale measures HIV stigma-by-association with good reliability. The measure was developed with African-American youth, as future studies by the research team will focus on the largest numbers of HIV-affected at-risk youth – African-American inner-city teens. The measure will benefit from testing with larger samples, including factor analysis, and an assessment of construct validity. Additionally, the authors will use the measure to explore the relationship between stigma-by-association, adolescent risk, and adolescent/parent characteristics, with findings informing potential interventions for affected teens and their positive parents.