512 - Designing an Inclusive Demographic Survey for Pediatric Clinical Care and Research
Friday, April 28, 2023
5:15 PM – 7:15 PM ET
Poster Number: 512 Publication Number: 512.116
Elizabeth Nguyen, University of Washington, Seattle, WA, United States; Jacqueline Burgara, Seattle Children's, Seattle, WA, United States; Dureti Hajikedir, Seattle Children's, Seattle, WA, United States; Tommy Nguyen, Seattle Children's, Seattle, WA, United States; Elizabeth Wingfield, Seattle Children's, Seattle, WA, United States; Arti D. Desai, University of Washington School of Medicine, Seattle, WA, United States
Clinical Research Scientist Seattle Children's Hospital Seattle, Washington, United States
Background: While we need to accurately capture family demographic information to examine pediatric health disparities; we lack a standardized and inclusive approach to asking relevant demographic questions. Objective: To design a set of inclusive survey items for common demographic categories that can be used for clinical care, quality improvement, and research. Design/Methods: We used an iterative approach to design the survey in three phases (Fig. 1). First, we conducted a literature review to draft survey items related to the categories: parental education, gender identity, racial and ethnic identity, language of care, and household socioeconomic status. Next, we had researchers and leaders with expertise in equity, diversity, and inclusion at our institution review survey items for all or some of the categories. We modified survey items based on expert reviews, revisions, additions, and item option rankings within each category. We then used purposive sampling and conducted cognitive interviews with patients admitted to our hospital and their caregivers (parent and/or legal guardian) to assess understandability of survey items, gather feedback, and rank item options. We prioritized interviews with Black, Indigenous, and People of Color (BIPOC) families. We conducted a content analysis of interview rankings and comments and selected one item for each category when there was clear consensus among families. For categories without clear consensus, we combined options into a single item by keeping positive attributes and removing negative attributes of each option. Results: We identified 3-6 sources for each category from our literature review. We received feedback from 7/15 invited experts (47%) comprised of directors, clinicians, program managers, graduate students, and research coordinators. We conducted interviews with 33 families; 73% identified as BIPOC and one-third preferred a language other than English for care (Table 1). In general, highly ranked options included those with more inclusive language (e.g., using “caregivers” instead of “parents”), options that created a space to self-identify, and items with granular response options (e.g., including ethnicity categories that mapped to larger racial groups). We ultimately narrowed survey items from 3-4 options to one option for each category.
Conclusion(s): In collaboration with content experts and families from diverse backgrounds, we designed an inclusive demographic survey (Fig. 2). These items may be used in clinical, quality improvement, and research settings to provide a more family-centered and standardized approach to demographic data collection. Figure1_Flowchart_PAS.jpeg
