294 - Measured by Proxy: Reconsidering indirect assessment in health

Publication Number: 3283.294

Benjamin K. Oelkers, St. Jude Children's Research Hospital, Metairie, LA, United States; Anneliese H. Williams, St. Jude Children's Research Hospital, Essex, MD, United States; Lara Counts, St. Jude Children's Research Hospital, Mempis, TN, United States; Luke Zabotka, Washington University In St Louis: Washington University in St Louis, Glen Carbon, IL, United States; Sara Malone, Washington University in St Louis, Saint Louis, MO, United States; Dylan Graetz, St. Jude Children's Research Hospital, Memphis, TN, United States

Background: Stigma in children with health conditions can negatively affect health-seeking behavior and mental health. In pediatrics, parents are commonly used as a proxy to assess the impacts of health conditions. However, proxy measures have shown variable reliability, and the utility of using proxies is unknown.

Objective: This study aimed to systematically review the use of proxies to assess stigma in various pediatric physical and behavioral health conditions.

Design/Methods: This work is a secondary analysis of a systematic review on stigma measurement in health. This review was registered with PROSPERO and conducted following PRISMA guidelines. Articles were included if they used or developed a quantitative measurement of stigma related to any health condition and measured stigma directly through patients or their caregivers. Data from the articles included were extracted into REDCap and analyzed by narrative synthesis. Extracted data included study overview, proxy characteristics, covariates, and study outcomes.

Results: Of the 2267 articles included in the parent review, 70 articles were marked as pediatric. 17 (24.3%) of the pediatric articles used proxy measures. This represented around 24.3% (17/70) of all stigma literature in children. The median age of patients in proxy articles was 10 years old, with a range of 0-18. In all cases, parents were used as proxies and their median age was 39.5 years old. All but one (94.4%) of these proxy measures were used to assess stigma in physical conditions. The most common conditions were epilepsy (3), HIV (3), and dermatologic conditions (2). Nine unique tools were used, 66% (6/9) of which were designed specifically for pediatric populations. Six articles measured stigma in both patients and their proxies. Of those that measured and compared patient and proxy reports, 66% (4/6) did not find a correlation between patient and proxy stigma scores.

Conclusion(s): Caregiver proxies are commonly used in pediatrics but could have limited or inconsistent validity. Discrepancies between children's report and their proxies suggest that parents might not accurately portray patient stigma, which could be exacerbated if children experience internalized stigma. Researchers should be cautious when using proxy measures in this population, as misrepresentation of pediatric stigma could further widen disparities for children with medical conditions and confuse how we understand stigma that children experience. Researchers should seek additional clarity about the use of proxies and continue to develop methods that allow for children's own voices to be represented in the study of stigma.

PRISMA diagram for the review


Study characteristics stratified by general pediatric and pediatric proxy subgroups
*Percentages may not equal 100 because they could be a part of multiple categories.
~one article in the general pediatric literature did not specify a health condition