We conducted a cross-sectional survey using convenience sampling, in partnership with the Consulate General of Mexico in Portland, OR. Recruitment occurred between April-June 2021. Participants were approached while waiting at the Consulate to process documents (e.g., passports, birth certificates) or to visit the “Ventanilla de Salud” which provides information and healthcare navigation services. Our bicultural and bilingual Mexican-American research team recruited 500 men and women. Inclusion criteria were: self-identified as Latina/o, 18–59 years old, and lived in the state of Oregon. Participants provided oral consent and chose preferred language for the survey (Spanish or English). Participants could fill out the survey on their own or have it read to them by the research team, to reduce language and literacy barriers. Surveys were completed on paper, responses were then entered into a Research Electronic Data Capture (REDCap) [13] database by trained study members. Participants were given a pen to thank them for their participation and a resource sheet with up-to-date information about the public charge. This study was approved by the Oregon Health & Science University Institutional Review Board.
We developed our survey using published questions from the California Health Interview Survey and the Urban Institute’s Well-being and Basic Need Survey [14] and tailored to Oregon public health programs [12]. We received input on the survey from staff of the Mexican Consulate, and two local non-profits focused on legal aid and the Latino community respectively. Our survey contained 20 questions and included socio-demographics, awareness about the public charge; how confident they were in their understanding; where they learned about the public charge; awareness of health care programs in Oregon; whether participants had disenrolled themselves or a family member from specific Oregon health care programs due to the public charge; and whether they were hesitant to seek care due to the public charge. We also included an optional open-ended question: in your words, what do you know about the “public charge” rule? Respondents reported their understanding of the public charge rule as “not confident”, “somewhat confident”, “confident”, or “very confident”; we created a binary variable or lower (not or somewhat confident) vs. higher (confident or vert confident).
We collected several sociodemographic characteristics. We included gender, age (categorized as 18–24, 25–29, 30–39, 40–49, and 50 or older), and years in the US (less than five, five-nine, ten or more, and “all of my life”). We included household size (four people or less versus five people or more), marital status, and education level. We included having a job or regular income and language spoken at home (Spanish, English, both Spanish and English, or other -- n = 1, indigenous language Q’anjob’al). Finally, we included self-rated ability to speak English (not at all, somewhat, or very well). All covariates except age were missing data, so we included categories for missing in Tables and for regression modeling.
We compared respondent sociodemographic characteristics by awareness of the public charge using Pearson’s chi-squared test and determined the proportion of respondents who had heard of health programs or services in Oregon and whether they had enrolled themselves or a family member in those programs. We tabulated the proportion of respondents who had heard of the public charge rule for each information source. We graphically assessed the percent of respondents who had heard of the public charge rule who had disenrolled themselves or a family member from a health program or who were hesitant to seek care due to concerns about the rule, overall and by their confidence in understanding the rule (lower vs. higher confidence). We developed a logistic regression model with awareness of the public charge rule as the outcome to determine what characteristics were associated with awareness of the rule. In order to improve the interpretability of the modeling results [15], we calculated the adjusted probability (average marginal effect) of having awareness of the public charge rule for each covariate in the final model. All analyses were performed in State version 15.1 (Stata Corp, College Station).
We extracted all open-ended responses about awareness of the public charge into an excel sheet; two study team members classified open-ended responses into preliminary themes, which were discussed and refined with team members to arrive at key themes.