Development of a Spanish-language drug facts label prototype for a combination mifepristone and misoprostol medication abortion product

To establish the feasibility of moving medication abortion over-the-counter (OTC), a recent U.S. study developed and tested a drug facts label prototype for a combination mifepristone and misoprostol medication abortion product, establishing high levels of comprehension of most key label concepts []. While the U.S. Food and Drug Administration (FDA) does not require drug facts labels to be available or tested in languages other than English, access to medication instructions in languages other than English is increasingly important. Accurate and accessible label instructions can reduce errors in medication use and reduce health disparities. While U.S. pharmacies increasingly use computer-generated translations of medication instructions, they are often inaccurate and incomplete [, , , , ]. To ensure that native Spanish speakers have access to easy-to-understand instructions of how to use medication abortion, the current study translated the drug facts label prototype for medication abortion from English to Spanish and conducted cognitive interviews to assess and improve its readability among native Spanish speakers. The goal of this study is to develop a Spanish-version label, equivalent to the English version which has already been rigorously tested [].

A certified translator first translated materials used in a label comprehension study for a drug facts label prototype for medication abortion [], including the drug facts label and interview guide, from English to Spanish. All materials were reviewed by 4 native Spanish speakers, including three co-authors (SS, MAB, and ESJ). The interview guide was designed to assess participant understanding of 11 primary and 13 secondary communication objectives regarding indications for use, eligibility for use, dosing regimen, contraindications, warning signs, side effects and recognizing the risk of treatment failure, and included open-ended questions soliciting feedback on drug facts label formatting and readability. When participants misunderstood a communication objective, we asked probing questions to better understand the source of misunderstanding and made iterative changes to the label to improve clarity.

From April through August 2021, we recruited participants through Craigslist, Facebook, and Instagram ads to participate in video interviews. We continued recruitment until participant comprehension of key label concepts was consistently high, participant feedback was positive, and the interviews no longer resulted in additional label changes. We also continued interviews until we reached people from a range of ages and geographies. Eligibility criteria included being ages 15−49, assigned female at birth, a native Spanish speaker and able to read Spanish. Participants self-screened for eligibility after reading a study summary, accessed by clicking on social media advertisements. Eligible participants were then contacted with more study details and instructed to schedule an interview by videoconference if interested.

Two native Spanish speaking female interviewers (MAB and SS) conducted all cognitive interviews by videoconference, obtained verbal consent from adult participants, verbal assent from participants ages 15−17 and under, and verbal consent from their parent or guardian who was present at the start of the interview, and assessed health literacy using the Short Assessment of Health Literacy – Spanish (SAHL-S) []. The interviewer asked the participant about label wording, formatting, and key label concepts while displaying the drug facts label onscreen. Whenever the participant indicated confusion or dissatisfaction with label elements, the interviewer asked for alternative suggestions. At the end of the interview, we collected participants’ sociodemographic characteristics and pregnancy experiences.  All interviews were audio-recorded and transcribed in Spanish. The study received ethical approval from the University of California, San Francisco, Institutional Review Board. Throughout the interview process, the research team discussed interviews and proposed and agreed on each label change which we implemented throughout the study. For each new label change, we included a probing question to the interview guide to solicit feedback regarding the changes. We did not assess the proportion of correct responses to each communication objective given that we changed the label throughout the interview process.

Of the 142 people screened, 51 were eligible and scheduled an interview and 38 completed an interview (38/51, 75% response rate). We reached saturation in participant feedback after 10 interviews and continued recruitment until we had representation across all U.S. regions and age groups. Our final sample included 38 people ages 15−48, from 12 U.S. states, Puerto Rico (n = 5), and Ciudad Juarez, Mexico (n = 2, Table 1). While we did not systematically ask people about their native country of origin, the 10 people who shared this information originated from Puerto Rico (n = 5), Mexico (n = 3), Cuba (n = 1), and Colombia (n = 1).

While we did not establish levels of understanding of key label concepts, people generally found the label clear and understood the 11 primary and 13 secondary learning objectives. There were two places (language around blood thinners and menstrual cramps) that participants found confusing that resulted in drug facts label changes (Supplementary Table 2), with most people finding the new phrasing to be clear. A few people found the instructions related to consulting a doctor before use in case of unexplained or abnormal vaginal pain or bleeding confusing and were unclear whether bleeding is normal. While we probed about this phrasing in all interviews, we did not identify clearer wording that would not impact the English version of the drug facts label and therefore made no additional changes. Some people found the instruction to “swallow” one tablet of mifepristone clear but also awkward, stating a preference for “to take” instead of “to swallow”. We chose to maintain the original wording of this instruction so that the route of administration remained clear. For the open-ended questions soliciting feedback on the blue and yellow shapes representing each medication, most (n = 35) found the colored shapes helpful; two people suggested making the colored shapes bolder or brighter and one did not notice them at all.  Everyone found the red and green table formatting helpful (n = 38), one person suggested adding simpler or more concise column headers, and two suggested using brighter or more distinguishable colors. The final drug facts label is presented in Fig. 1.

Participants suggested language improvements to this translated drug facts label prototype for medication abortion. Conducting cognitive interviews with native Spanish speakers was a helpful tool to ensure a high-quality label in a language other than English. We suggest minor label improvements, clarifying that the indication to consult a doctor if you have had abnormal bleeding and adding clearer table column headers. While our sample size of 38 people is small, we found it sufficient to reach saturation in participant feedback and in representation of people from a range of ages, geographic regions, and countries of origin, although an important limitation is the lack of participants with limited literacy scores. We recommend the use of this Spanish-version drug facts label with minor modifications. This Spanish-version drug facts label has the potential to ensure that native Spanish speakers have access to easy-to-understand instructions of how to take medication abortion without clinical supervision.

Conflicts of interest: None

Funding: This research was supported by a grant from the Society of Family Planning Research Fund ( SFPRF12-MA9 ). The funders had no role in the study design; data collection, analysis, and interpretation; or decision to submit the work for publication. The views and opinions expressed are those of the authors and do not necessarily represent the views and opinions of the funders.

The authors thank the members of this study's Advisory Board for their significant input to improve study design and interpretation.