Testing remains one of the most effective strategies for reducing the spread of COVID-19. COVID-19 rapid antigen tests are well-suited for at-home use; however, several barriers, including language, hinder the widespread adoption of self-testing. The purpose of the present study is to evaluate the impact of a bilingual digital health intervention on self-efficacy, intentions, and willingness to self-administer a COVID-19 antigen rapid test.
Participants (
Participants in the bilingual digital health intervention reported significantly higher intentions to self-test for COVID-19 (
Our findings suggest that the bilingual digital health intervention may have a positive impact on enhancing intentions and willingness to self-administer a COVID-19 antigen rapid test. These findings highlight the potential of culturally and linguistically tailored digital health interventions to reduce barriers and promote access to at-home COVID-19 testing in Hispanic communities. Notably, to help alleviate cost barriers associated with purchasing COVID-19 rapid antigen test kits, the research team collaborated with a highly qualified team of Community Health Workers (CHWs) to disseminate approximately 4,500 free COVID-19 test kits throughout the community.
Coronavirus disease 2019 (COVID-19) is a respiratory infection caused by SARS CoV 2. It is transmitted by airborne particles and droplets released into the air (
Research shows that people from racial and ethnic minority groups are more likely to be infected with COVID-19. Black and Hispanic people have the highest incidence and death rates due to COVID-19 compared to White people and continue to experience higher rates of infections, hospitalization, and death (
One of the most effective methods for reducing the spread of COVID-19 is through testing. To date, there are two main types of tests, namely the Polymerase Chain Reaction (PCR) test and the Antigen test. The PCR test is the ‘gold standard’ for COVID-19 tests. It is a type of nucleic acid amplification test (NAAT), which is more likely to detect the virus than antigen tests (
There are many different brands of COVID-19 antigen rapid test kits that can be used in home tests to identify positive cases for COVID-19 quickly. During COVID-19 surges, the US Department of Health and Human Services sent out up to four free COVID-19 antigen rapid test kits (the BinaxNOW ® or iHealth® brand) to each American household to help reduce cost and access barriers to testing (
To increase the self-efficacy, intention, and willingness of individuals for self-testing using the COVID-19 antigen test, an intervention was required. Therefore, the current study developed and tested a bilingual digital health intervention that can be widely disseminated, impacting not only vulnerable Spanish-speaking populations but English speakers as well. Three specific aims were investigated: (1) to develop a bilingual digital health intervention for COVID-19 antigen rapid tests; (2) to conduct pilot focus groups to evaluate the clarity and acceptability of the intervention; and (3) to recruit community members to test the intervention in a real-world setting. We hypothesized that the bilingual digital health intervention will increase intentions to self-administer a COVID-19 rapid test, willingness to self-administer a COVID-19 rapid test, and self-efficacy to self-administer a COVID-19 rapid test.
The iHealth® COVID-19 rapid test instructions were transformed into a 6–8th grade reading level, which is recommended for the general public. First, the instructions were reviewed and simplified using basic language. Second, readability statistics were calculated, and the language was adjusted until the scores were lower than 8.0, which is the cutoff for general public reading scores. Specifically, the Flesch–Kincaid reading score is a commonly utilized readability formula used to measure the estimated reading grade level of written scripts. The Flesch–Kincaid examines the length of the sentence based on the number of words in the sentence, and the length of the words based on the number of syllables in the words. A score equal to 6.00 to 6.99 is indicative of a 6th-grade reading level. A score equal to 7.00 to 7.99 is indicative of a 7th-grade reading level. A score equal to 8.00 to 8.99 is indicative of an 8th-grade reading level. Two different scores are calculated to determine the readability of the text: first, the Flesch–Kincaid reading ease score, in which a higher score means a text is easier to read, and a Flesch–Kincaid grade level score, which groups text reading level into different grade level groups based on the US education system. It is important to aim for a reading grade level of 8 or less to ensure the majority of Americans can comprehend the content easily (
The benign digital health multi-modal and bilingual intervention developed above was evaluated by three focus groups to assess usability and ease of understanding before conducting the randomized controlled trial.
Focus groups were conducted with 23 participants. There was a total of three focused (
Participants were asked 10 questions related to the written instructions and DIY video. The first five questions pertained to “written instructions” for the COVID-19 rapid test. Sample items:
The three focus groups were scheduled to occur on campus at the University of Texas at El Paso (UTEP) campus and temporary parking permits were made available to participants. Upon arrival, participants were instructed to read and sign the Institutional Review Board (IRB) approved informed consent form. Next, participants were assigned a numerical value (to protect their identity) and were asked to view the developed rapid test printed instructions in English and Spanish, as well as the DIY video in English and Spanish. After viewing each, participants were asked probing questions by the facilitator related to the translated materials. Participants were required to review the proposed educational materials and provide feedback. Lastly, participants were thanked and compensated with a $30 electronic gift card, which was sent to their email addresses after participating in the study. A facilitator and a note-taker were present, and focus groups were audio-recorded using the microphone on a laptop. The study was approved by UTEP’s Institutional Review Board (IRB #: 1914943–5).
Content analysis was utilized to analyze audio-recorded focus group data and derive themes. Content analysis is the process of detailing things to make descriptive associations so that themes emerge as complete ideas (
A small subset of community members (
As part of the data collection process, common data elements from the National Institute of Health (NIH) Rapid Acceleration of Diagnostics–Underserved Population (RADx-UP) were utilized. The NIH developed these data elements to ensure consistency in data collection and simplify data analysis, and is an initiative to help implement COVID-19 data testing and better understand the impact of the pandemic, such as the outcomes, disparities, and possible solutions. The measures are provided within the
Participants were prescreened to determine eligibility criteria, which included being 18 years of age or older, residing in the border region (e.g., El Paso, TX), and speaking English or Spanish. One hundred and fifty community members were recruited to complete a brief 25-min survey assessing self-efficacy in administering the rapid tests. Additionally, participants were randomly assigned to one of two conditions using block randomization: (1) a treatment condition in which participants completed the self-efficacy assessment to administer the COVID-19 rapid test after viewing the developed bilingual instructions and video, and (2) a control condition in which participants completed the self-efficacy assessment to administer the COVID-19 rapid test after viewing the standard iHealth® COVID-19 rapid test instructions that are provided from the supplier. Participants completed the survey on-site (e.g., at a tabling event) using a project iPad. Informed consent was obtained from all subjects in the study. All methods were carried out in accordance with relevant guidelines and regulations by UTEPs IRB (#: 1914943–5).
Participants were recruited from local health clinics, food banks, and tabling events in the community. Notably, Community Health Workers (CHWs) assisted in recruiting participants at tabling events. CHWs or Promotores de Salud are public health workers at the forefront of serving their communities. They are trusted members of their community with a great understanding of their community (
All statistical analyses were conducted using SPSS version 29.0.2.0. Fundamental descriptive statistics were conducted for all the demographic variables in the experimental and control groups. Means and standard deviations were calculated for continuous variables and percentages for categorical variables. Composite scores were created for variables that included multiple items. A Cronback alpha was calculated for each of these variables and if the score was less than 0.70, the items were not combined. Independent samples
A Flesch–Kincaid reading score of 7.9 grade level was obtained along with a Flesch–Kincaid reading ease score of 70.9. The newly translated English instructions were then translated into Spanish instructions using a certified translator. The contracted local media firm (BarracudaPR) developed two iHealth® COVID-19 rapid test DIY videos, one in English and another in Spanish.
The focus groups revealed two areas of opportunity for improvement: (1) Language barrier—not having instruction in Spanish and at a reading level that could be easily understood, and (2) Combination—visualization of the demonstration with text and voice (spoken word) were very helpful, making it easy for anyone to understand and conduct the rapid test given the clarity of the video.
Language as a barrier was a recurring theme throughout all three focus groups. Participants were asked questions to assess the clarity of the written instructions, for example, “In your opinion, how clear and easy are the COVID-19 rapid antigen test instructions?” One participant stated the following: “The words that they use should be words that people use (common language) that are more commonly used.” The second participant stated, “…the words need to be simpler to better understand the instructions. Simple and short. The easier the better.” Another participant stated, “I would like the instructions to be in Spanish because you could understand them and you will not make a mistake.” The majority of participants expressed similar sentiments.
The second theme that emerged among all of the study participants was the perceived benefit from the combination of visual and spoken words through the DIY video. Participants were asked, “In your opinion, how clear and easy is the video to understand the instructional COVID-19?” One participant responded, “It was very clear. It was to the point. It showed you how to do it, step by step. It did not skip any information. It was very good. She was giving the instructions while she was doing it in the video.” Another question about the video asked, “How easily do you think these instructions will be understood by members of your community, your friends, neighbors?” A participant responded, “Personally, feeling like there is someone with you guiding you, you feel more confident.” Another participant responded, “In another language other than your own, it’s 95%, but in your own language, it’s 100%!”
A total of 152 participants (
Demographic and background information (
| Variable | Total responses | Percentage of responses |
|---|---|---|
| Sex assigned at birth | ||
| Male | 27 | 17.8% |
| Female | 124 | 81.6% |
| Prefer not to answer | 1 | 0.7% |
| Ethnicity | ||
| Non-Hispanic | 4 | 2.6% |
| Hispanic | 145 | 95.4% |
| Prefer not to answer | 3 | 2% |
| Missing | 0 | 0% |
| Race | ||
| American Indian or Alaskan Native | 0 | 0% |
| Black or African American | 0 | 0% |
| Asian | 0 | 0% |
| Native Hawaiian or Other Pacific Islander | 0 | 0% |
| White | 125 | 82.2% |
| Some other race | 14 | 9.2% |
| Prefer not to answer | 12 | 7.9% |
| Primary/preferred language | ||
| English | 32 | 21.1% |
| Spanish | 118 | 77.6% |
| Missing | 2 | 1.3% |
| Household income | ||
| Less than $15,000 | 47 | 30.9% |
| $15,000–$19,999 | 18 | 11.8% |
| $20,000–$24,999 | 21 | 13.8% |
| $25,000–$34,999 | 16 | 10.5% |
| $35,000–$49,999 | 12 | 7.9% |
| $50,000–$74,999 | 13 | 8.6% |
| $75,000–$99,999 | 4 | 2.6% |
| $100,000 and above | 2 | 1.3% |
| Prefer not to answer | 18 | 11.8% |
| Missing | 1 | 0.7% |
An independent samples
An exploratory hierarchical regression model was used to assess impact of the intervention on self-efficacy in administering a COVID-19 test, after controlling for age, sex, education, and language. Preliminary analyses ensured no serious violations of the assumption of normality, linearity, multicollinearity, and homoscedasticity. The model was not statistically significant (
To assess perceived ease of understanding written instructions, an independent
An independent samples
An exploratory hierarchical regression model was used to assess impact of the intervention on intention to test self for COVID-19, after controlling for age, sex, education, and language. Preliminary analyses ensured no serious violations of the assumption of normality, linearity, multicollinearity, and homoscedasticity. Age, sex, education, and language were entered at Step 1, explaining 6.9% of the variance in intentions to test self. After entry of the experimental conditions at Step 2, the total variance explained by the model was 9.6%,
An independent samples
An exploratory hierarchical regression model was used to assess impact of the intervention on intention to test family for COVID-19, after controlling for age, sex, education, and language. Preliminary analyses ensured no serious violations of the assumption of normality, linearity, multicollinearity, and homoscedasticity. Age, sex, education, and language were entered at Step 1, explaining 8.3% of the variance in intentions to test family. After entry of the experimental conditions at Step 2, the total variance explained by the model was 11%,
An independent samples
An exploratory hierarchical regression model was used to assess impact of the intervention on willingness to test self for COVID-19, after controlling for age, sex, education, and language. Preliminary analyses ensured no serious violations of the assumption of normality, linearity, multicollinearity, and homoscedasticity. Age, sex, education, and language were entered at Step 1, explaining 8.3% of the variance in willingness to test self. After entry of the experimental conditions at Step 2, the total variance explained by the model was 9.9%,
An independent samples
An exploratory hierarchical regression model was used to assess impact of the intervention on willingness to test family for COVID-19, after controlling for age, sex, education, and language. Preliminary analyses ensured no serious violations of the assumption of normality, linearity, multicollinearity, and homoscedasticity. Age, sex, education, and language were entered at Step 1, explaining 7.1% of the variance in willingness to test family. After entry of the experimental conditions at Step 2, the total variance explained by the model was 8.3%,
Correlational analyses revealed that ease of understanding the video was positively correlated with the perceived benefit of the video in self-administering a COVID-19 test (
Correlates of ease of understanding and perceived benefit of written and video instructions.
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
|---|---|---|---|---|---|---|---|---|
|
Ease of understanding written instructions |
- | |||||||
|
Written instructions in easy to understand language. |
0.93** | |||||||
|
Written instructions are in confusing language |
0.06 | 0.05 | ||||||
|
Video instructions are easy to understand |
0.77** | 0.76** | 0.05 | |||||
|
Perceived benefit of video to self-administer test |
0.77** | 0.77** | 0.77** | 0.81** | ||||
|
Perceived benefit of video to others to test |
0.65** | 0.65** | 0.06 | 0.82** | 0.85** | |||
|
Willingness to recommend video to others |
0.71** | 0.71** | 0.01 | 0.81** | 0.86** | 0.95** | ||
|
Willingness to purchase COVID-19 test if I had access to video |
0.61** | 0.62** | 0.08 | 0.68** | 0.78** | 0.84** | 0.88** | |
|
Mean (SD) |
3.69 (1.37) | 3.73 (1.34) | 2.50 (1.28) | 4.00 (1.27) | 4.05 (1.22) | 4.10 (1.13) | 4.19 (1.04) | 3.99 (1.09) |
|
|
84 | 84 | 81 | 84 | 84 | 84 | 84 | 84 |
Correlations are reported using Pearson Correlations.
Several important findings emerged in the current study, emphasizing the importance of language in self-efficacy, intentions, and willingness to administer a COVID-19 rapid test. We hypothesized that the bilingual digital health intervention would increase self-efficacy, willingness, and intentions to administer a COVID-19 rapid test. Willingness and intentions to self-administer a COVID-19 rapid test were higher in the group who received the bilingual digital health intervention in comparison to the control group. However, the current study did not detect statistically significant differences in self-efficacy to administer a COVID-19 rapid test. Studies have shown that visual aids such as pictograms and videos significantly improved health literacy outcomes by increasing adherence to medications and understanding of health information (
Focus groups are effective for obtaining valuable information related to the development of questionnaires and exploring diverse opinions (
Findings from the current study demonstrate the effectiveness of a bilingual digital health intervention at improving intentions to self-administer a COVID-19 rapid test. A goal of this study was to reduce the barriers that individuals face when trying to test themselves for COVID-19 through the development of our bilingual digital health intervention. Developing an effective bilingual digital health intervention was important for several reasons. First, when patients understand how to properly administer their tests, it increases the probability of getting an accurate diagnosis. When instructions are not properly followed, the likelihood of obtaining a false positive or false negative result is increased, which can lead to misdiagnosis. Second, it prevents potentially costly and dangerous repercussions. Communicating health instructions in a language the patient understands can prevent the risk of unknowingly spreading COVID-19 to others due to assuming they have a false negative result. Third, educating patients on how to accurately test themselves for COVID-19 in a language they can easily understand can lead to timely diagnosis, isolation, and treatment. This can reduce the spread of COVID-19 infection and also potentially serious side effects of COVID-19. Fourth, communicating with patients in a language they can understand empowers the patient to feel confident, engaged, and responsible for their health care. It builds trust between me and the patient, thereby allowing them to be receptive to public health recommendations concerning COVID-19. Notably, to help alleviate cost barriers associated with purchasing COVID-19 rapid antigen test kits, the research team collaborated with a highly qualified team of Community Health Workers (CHWs) to disseminate approximately 4,500 free COVID-19 test kits throughout the community at venues including food pantries, local clinics, high schools, libraries, and community centers.
There were several limitations pertaining to this study. The first limitation is that self-reported measures were used which can lead to response bias. Another limitation of the current study concerns the generalizability of the findings given that the study participants were mostly female (81.6%), of Hispanic origin (95.4%), Spanish speaking, and resided in El Paso, Texas. A different quantitative study using a larger and more diverse population could help improve the generalizability of the study. Additionally, the current study did not include objective or behavioral measures which could help identify whether participant’s intentions actually contribute to the behavior of self-administering a COVID-19 rapid antigen test.
Although the current study focused on translating written instructions for the COVID-19 test kit, future studies should investigate the impact of having written and or video instructions available in Spanish for other at-home test kits available on the market, such as Flu, glucose, pregnancy test, and drug tests. Future studies should also include survey questions that assess the participant’s trust in the test kit, as this is a key factor in understanding perceptions of the COVID-19 test kit and any fears or trust issues that need to be addressed.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving humans were approved by University of Texas at El Paso Institutional Review Board (IRB). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
UG: Project administration, Writing – original draft, Methodology, Formal analysis, Resources, Data curation, Visualization, Writing – review & editing, Investigation, Conceptualization, Validation, Software. MG: Writing – review & editing, Writing – original draft, Investigation, Project administration, Methodology. BM: Methodology, Conceptualization, Writing – review & editing, Formal analysis, Writing – original draft, Funding acquisition. TA: Methodology, Writing – review & editing, Writing – original draft, Formal analysis. GF: Investigation, Software, Supervision, Conceptualization, Writing – original draft, Resources, Funding acquisition, Project administration, Data curation, Visualization, Writing – review & editing, Methodology, Formal analysis, Validation.
We thank the Health Communication and Risk Perception (HCRP) laboratory at the University of Texas at El Paso, School of Pharmacy, for their support and assistance with the project. Portions of this research contributed to the first author’s dissertation which was submitted to the University of Texas at El Paso, College of Health Sciences, Interdisciplinary Health Sciences, under the direction of GAF. The citation for the dissertation is provided within the references under Gbugu, U. O. (2025).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author(s) declare that no Gen AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
The Supplementary material for this article can be found online at:
Edited by:
Reviewed by: