key: cord-0065679-132pa9qc
authors: nan
title: Abstracts from the Joint Health Literacy Annual Research Conference and Health Literacy in Action: October 19–21, 2020
date: 2021-06-22
journal: nan
DOI: 10.3928/24748307-20210428-01
sha: fc58e922e9b6cdfa36a9fe4f489d167154f49b3d
doc_id: 65679
cord_uid: 132pa9qc

nan

Background/Research Question: Improving the delivery of patient-centered health information is especially important for vulnerable patient populations. Targeted, health information technology-based education interventions provide ways to help reach these patients, who often receive care in under-resourced, community-based health clinics. Methods: A low literacy, multimedia information intervention was tested in clinical practice to determine effects on enhancing patient-centered care and improving patient outcomes. Ambulatory cancer patients were enrolled at three medical centers that provide safety net services. Eligibility criteria included: 18 years or older, English-speaking, recent diagnosis of Stage I-III breast or colorectal cancer, starting adjuvant or neo-adjuvant therapy, and having sufficient abilities to interact with a multimedia computer. Prior computer experience was not required. This prospective, parallel-group, randomized controlled trial had two arms: CancerHelp-Talking Touchscreen (TT) intervention (Figure 1) vs. control (standard verbal and written patient education). Outcomes were assessed using a multimedia TT data capture system in the clinic four times: after starting treatment (baseline), during treatment, at end-of-treatment (approximately 6 months), and at the first follow-up visit after treatment. Outcomes included health be-liefs, cancer knowledge, self-efficacy, satisfaction with communication about cancer and its treatments, and cancer-specific health-related quality of life (HRQOL). Health literacy was measured with the Health Literacy Assessment Using Talking Touchscreen Technology (Health LiTT). Efficacy analyses were conducted with mixed effects models to estimate differencein-differences in score change from baseline to after treatment (two timepoints), and with individual growth curve models to compare trajectories over multiple timepoints. Cohen's conventions for magnitude of effect were used: small = 0.20 < d < 0.50; medium = 0.50 < d < 0.80; large = d > 0.80. Results: A total of 129 patients were enrolled (65 intervention and 64 control). The majority were African American (56%), female (83%), had a breast cancer diagnosis (74%), and had low health literacy. The study was stopped early because there were fewer than expected early-stage disease patients at the cancer care centers, the length of treatment was much greater than six months for many participants, and there were insufficient resources to continue following participants. Thus, only 49% of participants completed an assessment at the end of treatment, and 38% at the end of follow-up; completion was similar across study arms. Those randomized to the CancerHelp-TT intervention increased 1.1 more points on the cancer knowledge scale than control participants (effect size = 0.48; p = .05). Effect sizes for differences between randomized groups in self-efficacy, health beliefs, satisfaction with communication, and HRQOL ranged between 0.10 and 0.31 (p > .05). Although individual growth trajectories for satisfaction with care and HRQOL across timepoints were not significant, trends favored the intervention arm, with larger increases in satisfaction and HRQOL scores in comparison to the control group. The majority of intervention participants rated CancerHelp-TT favorably. Conclusions/Implications: In this randomized control trial testing a targeted, health information technology-based intervention for cancer patients in community settings, improvements in cancer knowledge were association with using the intervention, but not health beliefs about cancer, satisfaction with communication about cancer and its treatment, or HRQOL. Interventions like CancerHelp-TT may be able to improve some outcomes, but likely require additional tailoring and easy implementation at home in order to reach the most vulnerable patients.

Waite, Katherine, Salerno, Kathleen, Henault, Lori, Paasche-Orlow, Michael Boston Medical Center Background/Research Question: Discussing advance care directives is a complicated subject, especially if the patient has advanced cancer. Often, patients are unsure what types of care are available and what such procedures would involve. The objective of this cross-sectional study is to examine health literacy and the discordance of patient preferences for having CPR (cardiopulmonary resuscitation) and ventilation. Methods: Participants were from outpatient and inpatient settings of a large urban safety-net hospital with a diagnosis of any type of advanced cancer. After obtaining consent, a baseline pre-video questionnaire was conducted which contained two health literacy measures; the Cancer Health Literacy Test short version (CHLT-6) and the Brief Health Literacy Screener (three item Chew assessment).

Participants were asked what their preferences were for CPR and ventilation. After viewing a brief educational video describing approaches to end-of-life care, a tablet computer was used to video record participants (called a video declaration) as they described their wishes for medical care including specific medical treatment including CPR and intubation. Descriptive statistics were calculated and comments made during the video recording were summarized and grouped by thematic category. Results: There were 32 participants; mean age was 61.4 (SD = 9.9) years. About half were female (47%); the most prevalent race/ethnicity group was Black/African American (44%), followed by White (34%), non-Hispanic other (13%), and Hispanic (9%); 28% had a high school degree. About half (47%) reported their health was fair or poor and just over half (53%) reported having a previous conversation about end-oflife wishes with family/clinicians. According to the Chew items, almost 60% had inadequate health literacy but only 19% had inadequate cancer health literacy. Overall, 41% of participants described discordant preferences about having CPR in the pre-video questionnaire as compared with the video declaration and 44% expressed discordance in preferences for ventilation. There was no significant relationship between health literacy and concordance of preferences. Of those who had inadequate literacy as categorized by the Chew assessment, 28% were concordant when describing preferences and 31% with adequate health literacy were concordant. Conclusions/Implications: Findings show that approximately a third of participants were discordant when describing their preferences for CPR and/or ventilation, regardless of health literacy classification. Further research is needed to understand why this discordance takes place in patients with advanced cancer and to identify strategies to help patients' express their preferences so that they are clear to family, friends, and the medical team.

Opsasnick, Lauren A, Curtis, Laura, Wolf, Michael S, Yoshino, Julia, Eifler, Morgan Northwestern University Background/Research Question: Several known or suspected risk factors for cognitive impairment (CI) and related dementias manifest during middle age, including chronic conditions that may be inadequately managed, entrenched lifestyle behaviors (e.g., smoking, obesity), psychosocial stressors (e.g., depression), and inadequate health care use. Additionally, many adults reach middle age lacking proficient health literacy: the capacity to gather accurate self-care knowledge, make informed health decisions, and appropriately use health services. Using an ongoing cohort study, we explored associations between health literacy and risk factors for cognitive decline among adults younger than age 65 years. Methods: Patients ages 55 to 64 years were recruited from an academic internal medicine clinic and six federally qualified health centers in Chicago (N = 563). Health literacy was measured by the Newest Vital Sign. Patients self-reported chronic conditions, body mass index, and health care utilization. Smoking was assessed via the Behavioral Risk Factor Surveillance System, and mental health symptoms were assessed using PROMIS (Patient-Reported Outcomes Measurement Information System) short forms. Chi-square statistics and t tests, as appropriate, were used to evaluate associations. Results: Participants' mean age was 59 (SD = 2.9); 69% were female, 48% were African American, and 51% had limited health literacy. Compared to patients with adequate HL, those with limited HL were more likely to smoke (28.5% vs. 5.1%, p < .001), had higher rates of obesity (46.6% vs. 37.3%, p = .03), and more chronic conditions (M = 2. Background/Research Question: There are 3 million refugees living in the US today whose health and well-being may be diminished by not being able to understand and use health information. Little is known about these barriers to health in multi-ethnic refugee communities. The purpose of this study is to examine (1) the relationship between English proficiency, health literacy, length of time in the US, and health status and (2) differences in poor health status by limited English proficiency and low health literacy individually and in combination to better understand which barriers might be addressed in improving refugee health. Methods: Refugees (n = 136) ages 18-65 years were recruited using health clinics and refugee resettlement agencies. Survey questions included demographics, health status, health literacy, English language proficiency, social determinants of health, and barriers to getting health care. Interpreters were used as necessary. We utilized a crosssectional study with purposeful sampling. Results: There is a high correlation (Pearson's r = 0.77) between health literacy and English proficiency; they were moderately correlated with health status (0.40 and 0.37, respectively) ( Table 1) . Length of time in the US only modestly correlated with health status (0.16). Health literacy and English proficiency taken individually were strong predictors of health status (health literacy odds ratio = 4.0; confidence interval (CI) [1.6, 9.9]; English proficiency odds ratio = 3.6, C.I. [1.5, 9.0] but not significant. Their interaction, however, was significant and accounted for most of the effect (log odds for interaction = 1.67, odds ratio = 5.1, p < .05). English proficiency and health literacy individually and in combination facilitate poor health and present health-related barriers for refugees. Length of time in the US for refugees may not correlate with health status despite studies that suggest a change in health over time for the larger immigrant population (Table 2 ). Conclusions/Implications: This study has two key findings: (1) both English proficiency and health literacy are individual barriers to poor health but some combination of the two may be more indicative of health-related challenges for refugees and (2) length of time in the US for refugees may not correlate with health status despite studies that suggest a change in health over time for the broader immigrant population. Consideration of both language skills and health literacy must be given to resettlement perceptions about what health literacy skills emerge as in context to their self-care behaviors, and as they navigate through the healthcare system. Methods: The study was conducted in Singapore. A purposive sampling method was used to recruit participants from a pool of patients who received support services from two Singapore-based non-profit organizations-Diabetes Singapore and the Singapore Heart Foundation. Twelve focus group discussions with 44 patients and 26 caregivers from the two clinical populations-coronary artery disease (19 patients and 17 caregivers) and diabetes (25 patients and 9 caregivers)-in both English and Mandarin were conducted. Focus group interview data were transcribed and analyzed using grounded theory methodology, using NVivo10. The Institutional Review Board (IRB) of Nanyang Technological University approved this study (IRB-2015-03-042) . Results: The results from our analysis are presented in the context of the "patient's journeys." In all six important themes were identified: (1) Broad Nature of Health Information: Establishing Credibility and Relevancy of Health Information, (2) asking the right questions and being active communicators in doctor-patient encounters, (3) communicating assertively with health care professionals, (4) knowing when to question health information that patients accept to be authoritative, (5) finding and understanding the invisible connections between disparate bits of health information and understanding how this can affect their health, (6) using health information to resolve health issues, (7) navigating through the health care system levels aspects such as insurance paperwork, choosing healthcare options, etc. Conclusion/Implications: Evidence from this study provides evidence to support the notion that healthy literacy is an outcome of the complex interaction between individual level skills and the contextual factors within the health care system. Findings from our study show that it is important that health professionals, practitioners and policy makers pay attention to what patients perceive as important health literacy skills and how they enact these skills in their day-to-day encounters with health information. This can help in the development of health literacy interventions, and communicative practices that dovetail well with real life practices of the patients. combined intercultural competence (ICC) and health literacy educational intervention, along with clinical fieldwork, were effective at changing second-year nursing students' self-efficacy, beliefs, and knowledge in cultural competence and health literacy. Methods: The study was conducted with a secondary data set of 157 responses from second-year nursing students from two different programs affiliated with the same university. The first program was a four-year Bachelor of Science program located in an urban downtown area (n = 60), and the other (n = 97) was an Associate's degree program in a small urban city. Students completed a preand post-survey and received an educational intervention as part of a scheduled class session. The survey consisted of the Transcultural Self-Efficacy Tool (TSET), which has three subscales (cognitive, practical, affective) and the Health Literacy Beliefs and Knowledge survey. The education intervention consisted of a two-part lecture that focused on ICC and health literacy as well as a panel of experts discussing ICC. Following the intervention, students completed up to 10 weeks of work in clinics in their local communities. The post-survey was completed at the end of the 10 weeks. Results: There was a statistically significant difference in all three of the intercultural competency subscales after an educational intervention and spending time in the local community clinics (cognitive z = 7.681, p < .001; practical z = 7.319, p < .001; affective z = 6.533, p < .001). The health literacy belief post-test score (Mdn = 98) showed a statistically significant difference than the pre-test (Mdn = 97). There was also a statistically significant difference in knowledge scores (t = 2.900, df = 152, p < .004). Conclusions/ Implications: Our study results indicate that an educational intervention and clinical fieldwork increased second-year nursing students' confidence and self-efficacy in interacting with patients of different cultural backgrounds and health literacy levels. Combining ICC and health literacy training into second-year nursing student curriculum provides students with comprehensive tools to help patients of differing needs achieve better health outcomes. Background/Research Question: Organizational health literacy has been identified by the National Academy of Medicine (NAM) as a key part of an effective health care system. Many health care organizations, however, fail to integrate health literacy principles into its structure. This study aims to assess the organizational health literacy of an academic medical center (AMC) through the perspective of the Department of Surgery and to identify areas in need of improvement. Methods: To assess organizational health literacy, the 10-item Health Literate Health Care Organization survey (HLHO-10) was administered to all the attending staff in the Department of Surgery at a single AMC from December 2018 to May 2019. The 10 items are based on the 10 attributes of a health literate organization as defined by the NAM. These include effective leadership, integration with health care system, employee training, patient inclusion, patient accommodation, clear interpersonal communication, easy access/navigation, media variety, informed consent, and clearly explained benefits/coverage. Each item is scored on a scale from Statistical Analysis of the 10-Item Health Literate Health Care Organization Survey Results from the Department of Surgery at a Single Academic Medical Center (scores range from 1 to 7, with 1 being absolutely not to 7 being to a very large extent) one to seven, with one being "absolutely not" to seven being "to a very large extent. " Data was analyzed to estimate the organization health literacy level of the AMC through the perspective of the Department of Surgery by each domain. Results: Among the 118 attending physicians of the Department of Surgery who received the survey, 65 (55%) responded. The mean health literacy level of the AMC as rated by Department members was 3.84 out of 7.00. The domains "employee training" and "patient inclusion" scored the lowest with an average of 3.08 and 3.23, respectively. "Employee training" evaluates employee health literacy training with an emphasis on "expert educators" and progress monitoring. "Patient inclusion" assesses the involvement of patients in the development of health information and materials. Assessment of "easy access/navigation, " defined as access to health information and navigation within the healthcare facility, scored the highest with an average score of 4.98. Other scores are summarized in Table 3 . Conclusions/Implications: Opportunities for improvement exist in all domains of organizational health literacy. From the perspective of the Department of Surgery, employee training in health literacy and patient inclusion in health material development at the AMC scored the lowest. Major opportunities exist to address health literacy at an organization-level with several points of needed action. 

Background/Research Question: Chronic kidney disease (CKD) affects up to 10% of the Australian population, with greater prevalence, higher mortality, and more rapid progression in people with lower health literacy and from culturally and linguistically diverse (CALD) backgrounds. While educational programs for CKD patients are standard practice, no models of care exist that seek to promote global health literacy skills applicable across multiple contexts, or which target CALD groups. Methods: This study assessed the impact of a multi-component app-based intervention (the SUCCESS App) on different domains of health literacy using a pre-post study design. Patients aged 18 years and older with kidney failure requiring hemodialysis were invited to use the SUCCESS App for 12 weeks. Health literacy was measured pre-and post-intervention using the Health Literacy Questionnaire (HLQ). Results: A total of 110 patients were recruited from three Local Health Districts (Sydney, Nepean Blue Mountain, and Northern Sydney) and 98 patients completed baseline surveys. To date, 57 (58.2%) completed pre-and post-intervention surveys and were included in the analysis. There were no significant differences in demographic characteristics between those who completed the study and those who did not complete the study (Table 4) . Reasons for non-completion included 1 (1.9%) started another study, 3 (5.7%) death, 3 (5.7%) started peritoneal dialysis, 2 (3.8%) received a transplant, 3 (5.7%) relocated to another dialysis centre, 7 (13.2%) uninstalled the app, 17 (13.2%) were disinterested to continue and 17 (32.1%) were contacted but COVID-19 interrupted research proceeding. The mean age was 57.2 years (14.9), the majority were males (56.1%), 32 (56%) were born in Australia, 32 (56.1%) spoke English at home and 30 (52.6%) had adequate health literacy. There was no change in the overall scores for the health literacy domains, having sufficient information to manage my health (pre: 12.4 ± 2.3, post: 12.6 ± 2.5) and understanding health information (pre: 20.1 ± 3.8, post: 20.6 ± 4.1). However, there was a statistically significant improvement in the ability to actively engage with health care providers post-intervention (pre: 19.7 ± 4.1, post: 21.1 ± 3.5, p = 0.014). Conclusions/Implications: This feasibility study showed that there was demand for the SUCCESS App and that the app may help build some health literacy skills. A randomized-controlledtrial is warranted to assess the impact of the SUCCESS App on patient reported psychosocial and clinical outcomes. Americans have sickle cell trait (SCT); 2,000 are born annually with sickle cell disease (SCD), an autosomal recessive disorder leading to early mortality and primarily affecting minorities. To recognize the chance of having children with SCD, individuals with SCT must be tested and understand SCT, SCD, and their reproductive partner's SCT status. Despite newborn screening identifying SCT, >80% of people with SCT are unaware. Nationwide Children's Hospital previously offered an individual in-person education program for parents of 800 children with SCT born yearly in central Ohio; 63% attended. While many achieved high SCT knowledge immediately after the program, those with low health literacy (HL) were less likely to do so; whether knowledge gains were sustained is unknown. These are crucial since parents can use SCT knowledge to ascertain their status, inform reproductive decisions, and explain SCT to their children. With COVID19, the in-person program ended, creating a need to apply recent evaluation findings to establish an individually and virtually delivered (video-conferencing) program that is accessible, engaging, health literate, and promotes sustained SCT knowledge. Methods: The program evaluation included recruiting parents of infants with SCT in June-July 2019 to consent to videotaping their in-person sessions and complete verbally administered surveys on: demographics, self-reported SCT status, HL level, SCT knowledge, and education effectiveness. A comprehensive evaluator team reviewed the videos together for content and delivery. Three trained evaluators independently scored session print materials using the PEMAT-A/V and CDC-CCI-Modified, then met to finalize materials' scores. Results: Among recruited parents (Table 5) , three had limited HL; four had low baseline SCT knowledge. Five were unsure of their SCT status. Participants reported high session satisfaction (understandability, questions answered, reassuring). Participants' median pre-/post-SCT knowledge scores were 75% and 100%, respectively. Evaluators observed that parents appeared comfortable, received consistent non-directive genetic counseling, asked few questions, and encountered occasional distractions. They noted undefined technical term use, need-to-know concept omission, closed questions without teach-back, and limited explanation of visual materials that scored low for understandability, actionability, and clarity (Table 5) . Findings were used to create SCT knowledge objectives, a plain language talking points guide, visual materials, and an Educator training curriculum. The guide organizes content to ensure virtual sessions include: all key concepts, HL-informed strategies (e.g., teach-back), and prompts to encourage questions. An experienced design team is collaborating on accompanying electronic visual materials grounded in reader-friendly print material guidelines. An educator is undergoing SCT, genetic counseling, and HL training for the program. Conclusions/Implications: We used the in-person SCT education program evaluation and HL principles to create virtual SCT education. This has broad implications since virtual education has potential to improve access, flexibility, and satisfaction, and will likely increase. Background/Research Question: College students represent a unique population at increased risk for sexually transmitted infections (STIs) and unintended pregnancies given low In-person SCT Education Program Evaluation Summary, June-July 2019 sexual and reproductive health (SRH) literacy and high sexual risk-taking behaviors. The use of technology (including eHealth applications) is widely used among college students to address SRH questions and needs. This study explored how student health services within a college campus setting could capitalize on technology to facilitate SRH literacy, ultimately improving SRH outcomes. Methods: A mixed-methods study was conducted with students (n = 55) at a large, metropolitan university. Quantitative data from a brief, online survey was collected through Qualtrics and included sociodemographic and Technology Acceptance Model (TAM) items. Guided by the Integrated Model of Health Literacy domains (access; understand; appraise; apply), in-depth interviews were conducted to elicit students' ideas on how an eHealth application could be used to address their SRH literacy needs. Interviews were audiorecorded and transcribed; a codebook using a priori and emergent codes guided thematic analysis via MaxQDA. Intercoder agreement was discussed between two coders and the team agreed on final themes. Results: Most participants identified as female (71%), White/Caucasian (56%), heterosexual (73%), and single/uncommitted (44%). Responses to the TAM items suggest that students find an eHealth app to be acceptable and useful; highest TAM item: "Overall, eHealth will be useful in managing my health care" (M = 4.10, SD = 0.60); lowest TAM item: "I will find it easy to get eHealth to do what I want it to do" (M = 3.56, SD = 0.79). Regarding access to information, participants reported an eHealth app could be a reliable, private way to learn more about SRH. Participants also felt an eHealth app could increase access to services by integrating the app within the college clinic system (e.g., appointment scheduling, history forms, health education/counseling). When asked how the eHealth app could be used to understand information, some broadly reported the app could provide infographics, videos, or images to supplement facts and statistics. To appraise information, participants wanted to learn about their peers' experiences exploring different SRH options (e.g., birth control, STI testing) through open forums or discussion boards. Participants also reported the eHealth app could be used to apply information and communicate with providers (e.g., provide direct line of contact, facilitate informed conversations during visit), as well as partners and friends. Conclusions/Implications: Findings suggest that an eHealth app has potential to address college students' SRH literacy needs. Specifically, technology could streamline access to services, centralize important SRH information, and facilitate patient-centered communication during clinical visits. Future research should elicit clinic staff and providers' acceptability of the app. Such formative research is critical to inform the development and evaluation of a theory-and evidence-based eHealth app that can improve patient, provider and clinic system outcomes, ultimately decreasing adverse SRH outcomes among college students. System), less than one-fifth of dialysis patients are on the transplant waiting list. Among the key barriers to pursuit of transplant and LDKT are lack of knowledge of the option, overly complex educational materials, misconceptions about the risks/benefits of transplant/donation, who can be a donor, how long the process takes, and long-term outcomes. Health literate, culturally competent interventions that can be applied within and outside of dialysis and transplant centers are needed to reach and motivate more ESKD patients. With a rich history in health education, first-person storytelling is one solution. Methods: Using innovative digital story capture technology, compatible with any device with a camera, we collected diverse stories from the transplant community, ethically reviewed and removed PHI, and made them available in a searchable library at explorelivingdonation.org. Using open-ended prompts, we solicited educational information from storytellers, including change talk, where storytellers explain how they overcame emotional and logistical challenges to LDKT. This can serve as a roadmap for others interested in or considering LDKT. Results: We gathered 145 stories from participants of diverse race/ethnicities and education levels ( Table 6) . We created compilation videos, told in plain language by real people, covering common questions, concerns, and misconceptions about LDKT that kidney patients and their families experience. Storytellers shared openly about donor health, safety, and quality of life, motivations for transplant/donation, the evaluation and recovery process, life post-donation/transplant, and candid emotional reflections. For example, one recipient shared: "I had many questions about how this would impact [my donor] including the cost and her recovery time but... [I learned] that she would pretty quickly be able to heal and go back to work within a reasonable time. " Conclusions/Implications: First-person video stories can be used to address the educational barriers faced by kidney patients and their potential living donors. When told by diverse storytellers in their own words and style, video stories may have a natural health literacy and cultural relevance advantage over traditional print or provider-given oral education. These video stories can be used by dialysis and transplant centers to supplement traditional education, embedded into patient portals, or shared widely through social media. This model could be used to enhance education for many other chronic and complex disease areas such as cancer, diabetes, or hypertension. 

Background/Research Question: Past studies have shown that there is a health disparity in the deaf community including access to health information, which has implications on their health literacy development. When facing health communication barriers, deaf individuals may resort to seeking information online although this has not been well studied. The main aim of our larger study is to understand the mechanisms deaf people use to obtain health information online. In this specific study, we investigated the deaf community's access to the internet and whether they are able to seek, find, understand, and evaluate health information online (eHealth literacy). Methods: Participants included 892 individuals from Rochester, NY; Flint, MI; and Chicago, IL and were recruited to participate in this study between 2016 and 2019. Our sample includes 447 deaf (mean age = 46.1; SD = 16.9 years) and 445 hearing (mean age = 41; SD = 16.8 years). A translation work group of experts created an American Sign Language (ASL) version of the eHealth Literacy Scale (eHeals) for deaf participants who sign. Participants were interviewed in their preferred language (ASL or spoken English) about their technology access and use and administered the ASL adaptation of eHeals. Univariate and bivariate analyses were conducted using SPSS to evaluate technology and internet access, sources of health information, and eHealth literacy between deaf and hearing participants. Results: Significantly more deaf participants reported that they accessed the internet for information the last time they had a serious health issue (66.9% vs. 58.3%, p < .001), while hearing participants sought information from a doctor or healthcare professional (90.1% vs. 80.9%, p < .01). Significantly more deaf individuals have a desktop/laptop or a mobile device (e.g., iPad) (81.7% vs. 71.1%, p < .001 and 65.8% vs. 46.1%, p < .001, respectively) but there was no difference in ownership of smartphones/ cellphones. Deaf participants were more likely to report having high speed internet at home (95.2% vs. 79.4%, p < .001). Despite this, deaf participants struggled with lower digital health literacy skills as demonstrated by a lower ASL-eHeals mean score compared to their hearing peers (M = 21.3, SD = 7.3 vs. M = 27.2, SD = 4.9, p < .001). Conclusions/Implications: Deaf participants report high ownership of electronic devices and access to highspeed internet but face difficulties navigating and processing health information online. Future studies are needed to find ways to improve eHealth literacy among deaf individuals. , Ashley, Panko, Tiffany L, Hill, Joseph, Cucilick, Jessica, Hauser, Peter C Rochester Institute of Technology Background/Research Question: Health literacy utilizes numerous skills, knowledge, and functions that enables a person's potential to acquire and apply basic health in- Storyteller Characteristics (N = 145) formation to improve health outcomes. While research has shown that young adults with a college education may struggle with health numeracy, most seem to have adequate functional health literacy, with only 6.2% having less than adequate health literacy as assessed by the Newest Vital Sign (NVS). However, past studies have shown that deaf individuals are 6.9 times more likely to have limited health literacy than those who are hearing. It was hypothesized that deaf college students would be less at risk for developing limited health literacy because this subsample is assumed to have greater academic and functional literacy skills coupled with access to resources than those who are not college-bound. Methods: Our sample included 54 deaf and hard of hearing students (Mage = 23.2, SDage = 2.6; 60% female; and 50% POC (person of color) from a larger 2017 data collection. Participants completed a background questionnaire, a health questionnaire developed by the team, and the ACHA (American College Health Association) National College Health Assessment. Participants' health literacy was assessed using a version of the NVS adapted by McKee for use with American Sign Language (ASL). This study focused only on the sample's NVS-ASL performance. Univariate and bivariate analyses were conducted using SPSS to evaluate the status of health literacy in this population. Results: In our sample, 59.2% of the deaf and hard of hearing students had less than adequate health literacy. The proportion in our sample is approximately 10x more than found in the aforementioned study of hearing college students (59.2% vs. 6.2%). Further, significantly more students of color had less than adequate health literacy, 72% vs. 28%, χ 2 (1, N = 54) = 15.032, p < .001. Conclusions/ Implications: Health literacy challenges found in the deaf community extends to young deaf adults in college while this trend is not seen in non-deaf adults. The deaf community's low health literacy is a national public health issue. Future health studies with this population needs to take into consideration the separate effects of being deaf and being a person of color, as well as the interaction effects at the intersection of race and disability. Glick, Alexander F, Trachtman, Howard, Trasande, Leonardo, Dreyer, Benard, Mendelsohn, Alan L, Velazquez, Jessica, Pehrson, Laura Jane, Yin, H. 

Background/Research Question: Health literacy is associated with adverse child health outcomes through mechanisms that align with the Health Beliefs Model, including perceived risk (e.g., knowledge of severity of health issue) contributing to motivation to take action and behavior change. To date, there has been limited study of these mechanisms in the context of health literacy related to environmental health issues; this could inform intervention strategies to reduce health literacyassociated environmental health disparities. The objective of this study was to examine the association between parent health literacy and environmental health-related perception of risk and motivation to take action. Methods: Cross-sectional analysis of data collected from English-and Spanish-speaking parents of children at 2 sites: urban public hospital clinic and private hospital-affiliated subspecialty site; questions adapted from existing surveys and piloted with low literacy families. Outcome variables include: (1) perceived risk assessed in two ways, (a) being "very worried" about specific exposures (e.g., pesticides), and (b) incorrect knowledge of severity of issue (thinking exposure to specific toxins "never" associated with health problems); and (2) motivation to take action (e.g., spend money, buy carbon monoxide alarm) (Table 7) . Predictor variable include parent health literacy (Newest Vital Sign; 3 categories: low [score 0-1], marginal [2-3], adequate [4] [5] [6] ). Multiple logistic regression with generalized estimating equations to account for repeated measures (3 models [perceived risk-worry, perceived risk-knowledge, motivation] adjusting for child age, chronic disease, insurance type; parent age, race/ethnicity, language, education; site). Results: Participants included 300 parents (82% mothers; 51% Latinx, 12% Black, 26% White; 67% Medicaid; 24% low, 34% marginal, 42% adequate health literacy); 34% children with chronic disease. Perceived risk varied across specific exposures (32%-54% "very worried") (Table 8) . Fifty-five percent made >1 knowledge error. Motivation to take specific actions varied (20%-94%). Those with lower health literacy had an increased odds of being "very worried"; this was not statistically significant in adjusted analyses. Parents with low health literacy had an ~2x increased odds of having incorrect knowledge of potential risk (low vs. adequate: AOR [adjusted odds ratio] = 1.8 [1.02-3.0]). For most specific actions, there was no difference in motivation by health literacy; in adjusted analyses, there were no statistically significant differences by health literacy. Conclusions/Implications: Parents with low health literacy recognize environmental health risks less frequently-yet they share similar levels of concern and motivation to take action as parents with adequate health literacy.

Gan, Jennifer M, Bryant-Moore, Keneshia, Haynes, Tiffany, Ruvalcaba, Maria, Caballero, Alison University of Arkansas for Medical Sciences Background/Research Question: About 1 in 5 adults (about 44 million) live with a mental illness. The prevalence of mental illness is especially concerning in rural Arkansas; the nonprofit Mental Health Association ranked it as 33rd (high rates of mental illness and low access to care). With grant funding from the Network of the National Library of Medicine, the team aimed to address limited knowledge about emotional wellness and lack of confidence to engage in self-and professional care, with a focus on rural Arkansans. Our evaluation questions sought to determine changes in knowledge and attitude after a community learning session, and whether outcomes differed based on health literacy (HL) status. Methods: A printed How to M.O.V.E. toward Emotional Wellness (M.O.V.E.) guide was developed using HL and plain language best practices: https://healthliteracy.uams.edu/for-healthcareprofessionals/patient-engagment-resources/how-to-m-o-v-etoward-emotional-wellness/. We reviewed each iteration of the guide using validated formulas and instruments to assess readability, understandability, and actionability. Two focus groups were conducted to capture community members' perspectives about the guide's organization, design, formatting, tone, and ease of comprehension. One was held in an urban area and another in the Delta region, where the prevalence limited HL and poor mental health is elevated compared to other areas in the state. A complementary Group Facilitator's Manual was also developed. A "train-the-facilitator" session was held with 10 community partners from the Faith-Academic Initiatives for Transforming Health (FAITH) network, who subsequently hosted their own community learning sessions and collected pre-and post-session surveys (N = 356). Results: Surveys completed before and after attending a community learning session were compared using paired-samples t-tests which revealed that agreement with all of the following statements increased significantly (P < .0001): "Taking care of my emotional wellness is an important part of my overall health and wellness, " "If a friend needs help improving their emotional wellness, I know how to help them, " "I know things I can do to improve my emotional wellness or to keep it on track, " and "I can recognize when my emotional wellness is not at its best. " 

A 2 × 2 repeated-measures analysis of variance was used to explore gains in knowledge for the total, adequate HL, and limited HL samples. While there were differences between the adequate and limited HL groups in their pre-and post-session knowledge, both groups showed the same degree of improvement in their mean knowledge scores (Table 9 ). Conclusions/Implications: These results suggest that the program was successful in improving emotional wellness knowledge and related communication skills among participants across HL levels. Results underscore the importance of development and dissemination strategies that incorporate plain language expertise with community input and engage established community partners with direct access to the populations of interest. M.O.V.E. materials have been adapted into Spanish and disseminated in Arkansas, Louisiana, and Oklahoma. Interactive digital modules are in development and will be publicly available in English and Spanish. The current pandemic has added to stress for many Americans, and early studies have documented worsening mental health. The University of Arkansas for Medical Sciences Center for Health Literacy is currently developing a follow-up module "MOVING Beyond COVID" to help address these new needs.

Burkhart, Katherine R, Chesser, Amy K, Woods, Nikki K, Hansford, Hayley, Petersen, Emily Wichita State University Background/Research Question: Low health literacy, when not identified, is associated with poor health outcomes such as unsatisfactory medication compliance, poor disease manage-ment, and increased health care costs. Timely and accurate assessment tools are key to identifying low health literacy rates. The purpose of this study was to compare the validity of the three-question screener to the widely used and validated Short Test of Functional Health Literacy in Adults (STOFHLA). Methods: This was a powered study using a convenience sample of underserved participants in a midwestern state. An electronic survey containing both the STOFHLA and three-question screener was distributed to 225 participants and issued electronically. Participants were recruited through community organizations by email, phone, or in person. Survey order (3 question screener or S-TOFHLA) was randomized and each participant completed both STOFHLA and three-question screener. Participants had to be age 18 years or older, English speaking, and receiving health care in Kansas. This study was approved by the university Institutional Review Board for Human Subjects protection. Results: To determine the validity of the three-question screener as compared to the STOFHLA at identifying those with inadequate health literacy a Receiver Operating Characteristic Curve (ROC) was utilized. The three-question screener as compared to the STOFHLA in detecting inadequate health literacy had an area under the ROC of 0.58. The three-question screener had a sensitivity of 100% and specificity of 16.74% for determining health literacy. A McNemar Test was performed. It was found that there were more people identified as having adequate health literacy by STOFHLA where the three-question screener identified them as having inadequate health literacy (184/225) p < .001, there were only 4 participants that both the STOFHLA and three-question screener identified as both having inadequate health literacy (4/225) p < .001, and zero participants were marked as having adequate health literacy by the three-question screener where the STOFHLA identified them as having inadequate health literacy (0/225) p < .001. Conclusions/ Implications: Although the three-question screener is not specific at identifying low health literacy it did demonstrate high self-reported rates of health literacy skills. Therefore, it did not miss any patients identified as having low health literacy by the STOFHLA but included those who perceived themselves as having low health literacy. It would be more problematic to not identify those with low health literacy then to incorrectly identify them as having low health literacy. It should be noted that there is a distinct difference in the assessment tools. The STOFHLA is assessing skills and knowledge, where the three-question screening tool is assessing self-efficacy. The patient's perception of their health literacy may be more useful and practical in determining health literacy than their actual skills and knowledge in health literacy. In this case, use of the three-question screener in the clinic setting may be more time efficient and identify those with both perceived and actual low health literacy. 

Background/Research Question: This study aimed to develop a skills-based measure of health literacy diabetes mellitus (HL-DM) and to explore the relationship between health literacy and glycemic control. Methods: A systematic sampling method was used to recruit 7,019 permanent residents of Beijing for evaluation of HL-DM. According to the definition of health literacy, self-designed measurement tools were used to test reading, calculating, information searching, and communication skills. The stimuli materials included reading blood glucose meter and test sheet, reading drug instructions and communication materials, calculating body mass index and drug dosage, doctor-patient communication, information retrieval and so on. The total score of the scale was 30 points. Classical Test Theory (CTT) and Item Response Theory (IRT) were used to test the degree of difficulty, discrimination, reliability, and validity. Then the relationship between HL-DM and glycemic control was explored by structural equation modeling among 1,194 patients with type 2 diabetes in Beijing. Results: The evaluation results based on CTT showed that the difficulty of each item ranged from 0.32 to 0.88, and the discrimination was in the range of 0.27-0.91. Six factors were extracted by exploratory factor analysis, explaining 65.5% of variance. The factor loading of each item was in the range of 0.399-0.911. The goodness of fit indexes of confirmatory factor analysis was as follows: TLI (Tucker-Lewis index) = 0.861, CFI (comparative fit index) = 0.875, SRMR (standardized root mean residual) = 0.065, RMSEA (root mean square error of approximation) = 0.069. The total Cronbach's α coefficient of the scale was 0.934, and the Cronbach's alpha coefficient of each factor was in the range of 0.740-0.956. The results of IRT showed that the discrimination of each item was in the range of 1.201-3.375, the difficulty was in the range of -3.081-0.576, and the peak value of information function was in the range of 0.387-2.837. The area under the ROC curve (AUC) was 0.803. The maximum point coordinates of sensitivity and specificity were (0.260, 0.723), while the sensitivity and specificity were 72.3% and 74.0%, respectively. The cutoff was at 20 points and more than 20 points was defined having diabetes health literacy. The average HL-DM score of the patients was 16.0 ± 7.6 and the proportion of "having HL-DM" was 33.9%. The average HbA1C was 7.1 ± 1.4, and the glycemic control (HbA1C <6.5%) rate was 48.3%. Analysis of the structural equation model found that higher body mass (beta = 0.109), longer history of DM (beta = 0.179), and lower health literacy (beta =-0.104) are linked to a higher level of HbA1C (Figure 2) . Conclusions/Implications: The skillsbased measure of HL-DM has acceptable discrimination, reliability, and validity. Health literacy of patients directly affects their HbA1C level. Background/Research Question: Limited health literacy (HL) predicts worse health and partially mediates disparities. HL interventions can improve outcomes and reduce disparities. Measurement constraints, including lack of validity assessment across racial/ethnic groups and administration challenges, have undermined the field and impeded scaling of interventions. Methods: We employed computational linguistics to develop a novel HL measure, analyzing language from >300,000 messages sent by 9,527 diabetes patients via a patient portal. This AI approach harnessed "big linguistic data" to estimate HL, applying machine learning to a gold standard of expert ratings of a purposive sample of messages to create literacy profiles (LP). We carried out stratified analyses to determine if LPs have comparable criterion and predictive validities with respect to physician communication (CAHPS), medication adherence, severe hypoglycemia, poor glycemic control (A1c >9%) and emergency department utilization among White/non-Hispanics (N = 2,797), Black/non-Hispanics (N = 1,409), Hispanics (N = 1,374), Asian/Pacific Islanders (N = 2,894), and mixed/other race (N = 1,053). Results: Criterion validity of LP was consistently high across White, Black, Hispanic, and Asian/ Pacific Islanders (c-statistic 0.88, 0.82, 0.85, 0.89). Across racial/ethnic groups, patients with low HL (measured by LP), compared to high HL, generally demonstrated statistically significantly worse: physician communication, adherence, metabolic control, and ED utilization. The only conditions in which consistent HL-related patterns were not observed were (a) communication for those in the other race category, and (b) ED visits for Hispanics and Asian/Pacific Islanders. Conclusions/Implications: This is the first study to validate a HL measure across major racial/ethnic groups. While concerns have arisen regarding bias in AI, automated LPs appear sufficiently valid across race/ethnicity, enabling HL measurement at a scale that could improve clinical care and population health among diverse populations. Background/Research Question: Frailty is a syndrome characterized by multisystemic physiological decline associated with high risk for adverse health outcomes. Frailty and inadequate health literacy are common in the older population. Assessing frailty with questionnaires may be challenging for older adults with inadequate health literacy. Patients with inadequate health literacy might benefit from frailty assessments that incorporate images rather than words. The Pictorial Fit Frail Scale (PFFS) was designed to assess frailty using simple visual images. The aim of this study was to evaluate the concurrent validity of PFFS in measuring frailty in a racially and ethnically diverse population of older Veterans with varying levels of health literacy and cognition. Methods: A cross-sectional study was conducted in an outpatient geriatric clinic at a Veterans Health Administration medical center. Community-dwelling Veterans age ≥65 years old were asked to complete socio-demographic questionnaires, cognitive and depression screenings, the PFFS and FRAIL scales. The PFFS is a visual image-based questionnaire consisting of 14 domains (mood, medications, mobility, function, balance, social connections, tiredness, memory and thinking, vision, hearing, pain, unintentional weight loss, aggression, and bladder control), each one represented by 3-6 images. Patients selected the image that best resembled their usual level of function with scores ranging from 0 (no frailty) to 43 (severely frail). The FRAIL scale is a validated instrument using 5 domains (fatigue, resistance, ambulation, illnesses, and loss of weight). Patients were categorized into non-frail (1-2) and frail (3-5). A 31-item VA Frailty Index (VA-FI) was automatically generated from electronic health record data matched to the study date. The VA-FI was used as the reference for frailty in this study and categorizes patients into non-frail (robust <0.1, pre-frail ≥0.1-0.21) and frail (≥0.21). We evaluated health literacy with the Newest Vital Sign (NVS) consisting of a nutrition facts label with six questions that required the individual to identify and interpret nutritional information. Results were classified as inadequate (<4) and adequate (≥4-6) health literacy. We calculated Pearson correlation coefficients to assess the association between the PFFS, FRAIL scale and VA-FI. To compare the accuracy of the PFFS and FRAIL scale against the VA-FI, we used the area under the curve (AUC) for the receiver operating characteristic (ROC). Results: Participants were 83 older Veterans, mean age of 76.20 (SD = 6.02) years, 65.1% Caucasian, 74.7% non-Hispanic; 48.2% were found to have inadequate health literacy, 20.4% screened positive for cognitive impairment, 9.6% with depression, and 57.83% were frail. The PFFS mean score was 13.60 in the Frail group vs 7.77 in the Non-frail group, p < .001. Significant positive correlations were found between the PFFS and VA-FI (r = 0.550; 95% confidence interval [CI] [0.378, 0.689] and between FRAIL scale and VA-FI (r = 0.499; 95% CI [0.324, 0.655]; p < .001). When compared to the VA-FI, ROC AUC values for the PFFS (0.737; 95% CI:0.629-0.844) and the FRAIL scale (0.724; 95% CI [0.615, 0.824]; p < .001) showed similar satisfactory accuracy (Table 10) . Conclusions/Implications: The PFFS is a feasible instrument in the assessment of frailty in racially and ethnically diverse older Veterans with varying levels of health literacy and cognition. The PFFS exhibits comparable diagnostic accuracy to the FRAIL scale and VA-FI. Smoking history, n (%) 9 (10.8%) 0 (0%) 9 (100%) .009

<9 years of education 1 (1.2%) 1 (100%) 0 (0%) .422

At least 2 ADLS, n (%) 6 (7.2%) 0 (0%) 6 (100%) .037

Significant differences are in bold (p < .05). SD = standard deviation

ADLS = activities of daily living; CI = cognitive impairment