Bridging the Digital Divide: A Bilingual Interactive Health Kiosk for Communities Affected by Health Disparities
Although many factors contribute to the gaps in health outcomes and access to health care, the minimal access to health information on the Internet experienced by underserved communities has contributed to health disparities (Kreps, 2009; Moiduddin & Moore, 2008). The U.S. Department of Health and Human Services (HHS) has identified health information technology as a means to improve the health and health care of the underserved (Moiduddin & Moore, 2008). This suggestion was based on the association found between receiving health information through technology and increased health status. For example, racial and ethnic minorities and people with low-income tend not to use technological sources to obtain health information and have worse health outcomes than racial and ethnic majorities (Lorence, Park, & Fox, 2006; Moiduddin & Moore, 2008). The significant difference found in the use of technology to receive health information is often referred to as a digital divide. The digital divide is the gap between people who have and do not have access to the latest technology tools and the educational opportunities tied to this access. The divide is growing along racial lines (Kreps, 2005). Kreps (2005) noted that the digital divide has led to significant health disparities, or differences in health and health care across racial, ethnic, and socioeconomic groups.
The use of computerized kiosks containing health information in accessible spaces for communities affected by health disparities is a new concept. Moiduddin & Moore (2008) reported that there is minimal research that has assessed the prevalence of use and attitudes of underserved communities towards the use of information technologies. Only a few studies on the use of interactive kiosks for underserved communities were found. Two research studies on the use of bilingual, interactive kiosks showed significant differences among Spanish-speaking users; one study had only 16% Spanish-speaking respondents (Leeman-Castillo, Corbett, Aagaard, et. al., 2007), while another had 50% Spanish-speaking respondents (Guanipa, Nolte, & Lizarraga, 2002). These studies do not explain reasons for the significant difference between Spanish-speaking users and what aspects of the different kiosks affected the prevalence of use.
There are several reasons why underserved communities do not use information technologies. Some reasons include concern about confidentiality, disclosure, and privacy. Zeng, Reynolds, and Sharp (2009) reported that there is a trade-off between the accessibility that technology provides and confidentiality of personal information. For example, community members often need support when they are using a technology for the first time. If a community member is typing confidential information into a computer and needs support, the person supporting them could possibly see their confidential information. Technologies that provide health information should explicitly reveal where the information was gathered. A study of seventeen Internet users revealed that none of them checked for disclosure statements while searching for health information on the Internet (Eysenbach & Kohler, 2002). It appears that Internet users blindly accept information gathered via the internet, while research shows that non-users are more cautious. A study of patients with type 2 diabetes showed that patients who did not regularly use the Internet were 20% more likely than Internet users to have privacy concerns related to technology (Watson, Bell, Kvedar, et. al., 2007).
Underserved and minority communities are often served by Community Health Centers (CHCs) with limited funding and training to provide electronic forms of communication. In general, social service professionals have not used technology as often as other professions due to their belief that technologies are not helpful to their work and their lack of self-efficacy with technology use (Zhang & Gutierrez, 2007). CHCs provide medical services to people who are uninsured or insured by Medicaid because of limited income. A study of CHCs throughout the United States revealed that 75% of these people did not use electronic health records, which are the most commonly used form of health information technology in medical offices (Shields, Shin, Leu, et. al., 2007). It is important to assess the effectiveness of interactive health kiosks in providing health information for underserved communities, because social service professionals could benefit from clients' utilization of interactive health kiosks to obtain health information. Social service professionals often have large caseloads and limited time to explain health information to clients. Although social service professionals have been concerned that using information technology could prevent the maintenance of a therapeutic relationship with clients, they should consider the benefit that technology brings to geographically isolated clients (Csiernik, Furze, Dromgole, et. al., 2006). For example, eighty-five percent of social workers work in metropolitan areas, while there are many people in need of services outside metropolitan areas (National Association of Social Workers, 2006). The participants in this study lived in areas with limited access to services, which is associated with health disparities. There are also limited bilingual social workers (National Association of Social Workers, 2006); therefore, a bilingual interactive health kiosk could greatly benefit Spanish-speaking clients' access to health information in their first language.
In moving towards the development of kiosks placed in communities affected by health disparities, qualitative data provides more description of community attitudes toward the kiosk technology. One qualitative study of the attitudes towards the use of a kiosk was found. This research analyzed the community's feedback to a cardiovascular disease prevention website and kiosk designed for Latinos. Themes of community feedback include, but are not limited to: information should be easy to read, targeted for the population, and interactive/animated (Padilla, Bull, Raghunath, et. al., 2010). Although these findings cannot be generalized to other communities, they help to guide the development of future kiosks for communities affected by health disparities.
Community Based Participatory Research (CBPR) is an approach that emphasizes equitable partnerships and promotes the sharing of power, knowledge and resources between investigators and research participants in all stages of research and application. This approach highlights a commitment between partners (community and research) and utilizes an ecological perspective that identifies multiple determinants of health. CBPR is built on the principle that when users of research are brought in from the very beginning of the research-planning process, studies lead to stronger findings that are more practical and relevant to community needs. CBPR draws on four key principles in engagement of the community: 1) engage in collaboration and equitable partnerships in all phases of research; 2) build on the resources and goals already present in the community; 3) create and invest in long-term and robust partnerships; and 4) engage in research as a cyclical, iterative process (Schmittdie, et al, 2010).
The Ecological Perspective is appropriate when using CBPR, because it includes an examination of the determinants of health from the individual, community, and policy levels. Rather than looking at isolated events, the Ecological Perspective looks at the context in which events occur (Bronfenbrenner, 1979). Negative health outcomes are not blamed on individuals but they are examined within the context in which they take place. The Ecological Perspective takes into consideration the physical environment, the social, cultural, and historic context of individuals. In this study we proposed that the community be defined as people living in communities affected by health disparities, health system leadership, community stakeholders and clinic staff. This study presents examples of ways in which community feedback and participation may be integrated in research.
The Role of Culture
Cultural relevance and culturally appropriate services are important components to effective healthcare services and research strategies, particularly in CBPR. There are different ways to consider culture in underserved populations. The most common way is to consider the ethnicity of a particular underserved group, such as the Hispanic/Latino community. While ethnic and minority groups are never homogenous, there are always cultural components based on shared norms, values, and beliefs that render one's world view and can conversely also affect one's sense of health and health care. Research focusing on cultural factors in Hispanic/Latino communities has often cited cultural factors such as Familismo, Personalismo, Respeto, and Simpatia as cultural considerations for working with Hispanic/Latino communities.
According to Marin and Marin (1991), Familismo is "a cultural value that involves individuals' strong identification with and attachment to their nuclear and extended families, and strong feelings of loyalty, reciprocity, and solidarity among members of the same family". Personalismo is defined by Paniagua (1994) as an orientation toward people rather than toward impersonal relationships. It is alleged that Mexican Americans relate more to people than to impersonal (institutional) relationships. Huerta and Macario (1999) define "Simpatia" as the need for smooth interpersonal relationships in which criticism and confrontation are discouraged; "Respeto" is defined as the need to maintain one's personal integrity and allow for face-saving strategies. Such cultural characteristics can have dramaic effects on a community's concept of health, health promotion, prevention, and treatment. Other research in counseling Latino alcohol and substance abusers (Gloria, 1996), prevention of AIDS among Hispanics (Marin 1989), treating diabetes in the Hispanic population (2005), cultural constructs of acculturation (Cuellar, 1995), and parental control (Halgunseth, 2006) has shown that cultural characteristics have impacted the medical and health treatment process.
Bilingual interactive kiosks containing information on HIV/AIDS, mental health, and substance use were placed in the waiting areas of two community centers in the Phoenix metropolitan area. A description of the development of the websites and information on the kiosks is described in a previous publication (Davis, McBride, Bean, in-press). The kiosks were strategically placed at these sites based on community partnerships working together with researchers to identify the placement of the technology to best serve communities affected by health disparities. Community centers supported the use of the interactive health kiosk for its additional resource to community members to access health information. The interactive health kiosks included a: 1) website of information on HIV/AIDS, mental health, and substance use, 2) Frequently Asked Questions page where participants would anonymously ask questions and answers would be posted on the website within 3 days, and 3) needs assessment survey requesting demographic and health information.
The kiosks also addressed ethical issues, such as privacy, confidentiality, and disclosure (Jones, 2009). It employed a privacy screen which blocks the view of surrounding people and is viewed only by the individual standing directly in front of it. Confidentiality was ensured in several ways. No personal information, such as name or address, was collected anywhere on the website on the kiosk. The kiosk also automatically reset to the homepage after sitting idle for a couple of minutes. Because the kiosk included sensitive information on subjects such as HIV/AIDS, it is important that new users do not know what parts of the kiosk previous users have accessed. The website did not save any history of previously viewed pages. Each page on the website of the kiosk included a disclosure statement about where the information was gathered. Interactive parts of the kiosk, such as the Frequently Asked Questions (FAQ) page, also included a disclosure statement and encouraged users to seek professional medical advice.
Between September 2009 and March 2010, the investigators collected qualitative data through kiosk visit observations and interviews. Two investigators conducted bi-monthly visits (N = 26) to the interactive health kiosks. Individual unstructured interviews were conducted with community members in the community center lobbies. Community members waiting in the lobbies during bi-monthly visits were encouraged to use the kiosk and were provided with an explanation of the kiosk's purpose. The community members were asked different questions, such as "Have you ever used the kiosk before?", "Did you take the needs assessment on the website?", and "What would make the kiosk more attractive to use?" We also received anonymous questions (N = 11) through the FAQ page of the kiosk. We posted bilingual responses to the questions on the FAQ page on the kiosk.
The two community health center waiting areas consisted of diverse populations of males and females, Caucasian, African American, Hispanic/Latino, and Native American people at each site visit. Due to many people coming in and out of lobbies for appointments, the exact population of community members that were available to interview is unknown. A sample of community members (N = 100) participated in unstructured interviews and/or were observed in the health center lobbies. Four themes emerged from the unstructured interviews and observations of community members' interface with the interactive health kiosks: 1) market to people with lower health literacy, 2) address community wants and needs, 3) be conscious of computer competency, and 4) be sensitive to participants' response to technological support.Market to People with Lower Health Literacy
Many community members did not understand the function of the kiosk. The kiosk is a black box with a clear screen on the front that shows a computer screen above a keyboard. A large bright yellow sign was placed on the kiosk that simply reported "Receive health information here" in English and Spanish and described how to change the language from English to Spanish. Regardless of the information signs, people continued to confuse the kiosk for other things. One person commented that it looked "like a job application machine at Wal Mart." Another person asked, "Is it to pay bills? What is it?" During the site visits, the site visitors would explain the purpose and function of the interactive health kiosk to community members. The site visitors observed that community members who had previously received explanations of the kiosk would forget the purpose. A young Hispanic woman who had been present for the majority of site visits surprised the site visitors by commenting that the kiosk was for "something like health, right?" Visual signs and verbal explanations by site visitors did not clarify the function of the kiosk to some community members with lower health literacy.
Community members also suggested the kiosk needed to be more visually attractive to attract users. An Accent Health television, which featured short shows with celebrities discussing different health conditions and thought-provoking quizzes about exercising and healthy eating habit, was in one of the kiosk site areas. The Accent Health television engaged the majority of the community members waiting in the lobby, while the black box appearance of the kiosk prevented community members' being interested in it. The homepage of the website on the kiosk had pictures and text describing the website. A young man at the site which housed the Accent Health television observed that "the kiosk looks boring." He proposed it needed "stickers on it or a cool screen saver" to attract people's attention. An interactive health kiosk must include visual and auditory stimulation in order to attract people with lower health literacy.Address Community Wants and Needs
Many community members reported that mental health, HIV/AIDS, and substance use were topics that were important to them. A young African American woman reported that it was good that the website "hit three major topics" [mental health, HIV/AIDS, and substance use] in a short period of time. On the Frequently Asked Questions (FAQ) page, kiosk participants were asked what topic was most helpful on the kiosk. Most participants reported that mental health (N = 6) was the most helpful topic. Community members also asked the site visitors about personal mental health issues (N = 3). One kiosk user asked an anonymous, personal question about her medication for a mental illness on the FAQ page: "It seems to me that I am unable to find a combination, or singular medication, that works for me without a myriad of side effects, mostly nausea... will I ever be able to accomplish this?" Drugs was also reported as a helpful topic (N = 5), but no one inquired about substance use in person or on the FAQ page. While only one participant reported that AIDS was the most helpful topic and no community members inquired about HIV/AIDS in person, three other participants asked individual questions on HIV/AIDS on the FAQ page. One participant asked "What others ways can you get HIV/AIDS… by kissing someone of any races or share something with someone? Or anything so I can keep myself safe."
Although the kiosk only contained information on mental health, HIV/AIDS, and substance use, community members inquired in person and on the FAQ page about other issues. Kiosk participants inquired about employment, cancer, flu shots, and pregnancy. Community members also asked the site visitors if the kiosk provided information on jobs and flu shots in the community. Inquiries regarding cancer and pregnancy, which were more personal, were asked on the FAQ page. One kiosk user asked "How do you know that you're pregnant? What symptoms are you supposed to go through if you are pregnant?" The person who posted this inquiry may have been afraid to ask the question in person, so the FAQ page provided the community member with a safe, confidential place to inquire about a personal issue.
Community members also reported their needs in order to use the kiosk. Some people reported that there was no incentive to use the kiosk or take the needs assessment survey on the website. One woman passionately expressed that she had filled out survey after survey and each one offered to bring services to the community, but the services never came. After taking the needs assessment survey, a young man reported: "I answered 24 questions and got nothing." This implies the importance of building rapport and credibility with the community before providing an intervention for the community. The kiosk users also reported that they needed a stool to sit on and large font in order to use the kiosk. The kiosk screen that we used was approximately at eye-level while a person was standing, but community members felt that could not stand long enough to receive all of the information on the website or to take the needs assessment survey. Community members also reported that the font on the kiosk was too small. Although kiosk users had the option to change the size of the font, the primary font was increased to accommodate the needs of kiosk users.Be Conscious of Computer Competency
The major reason why community members did not want to use the kiosk was their reported intimidation of computers. Hispanic/Latino and African American community members were more likely to report their dislike of or intimidation by computers. When invited to use the kiosk, community members reported: "No way, I don't touch those things" and "I don't know how. My kids do, but not me." Although site visitors were there to assist community members to use the kiosk, community members who reported feeling intimidated by computers immediately responded with strong resistance to the idea of using the kiosk Some kiosk users did not know how to use the mouse to scroll up and down to view the entire website. A suggestion to address intimidation in using the kiosk included using a touch screen kiosk.Be Sensitive to Participants' Response to Technological Supports
Overall the community members responded well to the site visitors, were naturally curious about why the site visitors were there, and the purpose of the kiosk. Site visitors received comments from community members regarding the kiosks, such as: "It's great that you're trying to do good for people" and "That's pretty cool!" A Hispanic/Latina woman who had been present in the lobby during many site visits inquired, "How's that computer thing going?" and reported that she took the needs assessment the other day. The community members' friendliness and willingness to interact with the site visitors reflected acceptance of the site visitors and the technological support that the site visitors provided.
Although community members were friendly and inquired about the kiosk while the site visitors were present, they were less likely to use the kiosk and take the needs assessment on the kiosk while site visitors were present. For example, a Hispanic woman who had been present for many site visits reported taking the needs assessment on a day when the site visitors were not present. An electronic tracking system of the needs assessment surveys showed that many surveys were taken after kiosk site visits were conducted. This indicates that community members were more comfortable using the kiosk when site visitors were not present, which may reflect privacy concerns.
This interactive health kiosks study informs future engagement in the development of interactive health kiosks and other informational technologies using CBPR with underserved communities. The implications were discussed based on the four components of CBPR: 1) engage in collaboration and equitable partnerships, 2) build on resources and goals already in the community, 3) create and invest in long-term partnerships, and 4) engage in research as a cyclical process.
This interactive health kiosk study engaged in collaboration and equitable partnerships and provided outreach efforts for community participation by visiting the two community kiosk sites in the Phoenix metropolitan area. Since many of the community members who used the kiosks were not computer competent, site visitors invited community members to use the kiosk and helped community members to use them. If needed, community stakeholders, such as social workers, may need to receive training on the kiosk as well. The site visitor was sometimes able to help participants to feel at ease and show them how to use the kiosks. Community stakeholders, who already have built rapport, may be a better resource to provide community members with support on an interactive kiosk. The feedback received from community participants will continue to be used to update the kiosk site and make it more sensitive to the needs of the community. Semi-annual focus groups will engage community members and stakeholders for feedback on experiences with the website and suggestions for change and on-going development of the website.
The goals of the communities with which our study engaged indicated a desire and need for increased health literacy in the areas of mental health, substance use, and HIV/AIDS. This information was originally provided on the kiosks. As community goals change, the intervention should adapt. For example, during the development of the kiosk website, the economy began to limit resources available to the community. Also, a flu epidemic threatened the community. Information was added to the kiosk on resources that were available to the communities that used the kiosks, such as flu vaccine locations, in addition to the original information on mental health, substance use, and HIV/AIDS. Although it can sometimes be difficult and time consuming to adapt interventions based on communities needs, it is critical to CBPR and to fulfilling the needs of the community.
The outreach efforts for the interactive health kiosks strengthened the long-term relationship between the research partners and the community. Community organizations and stakeholders were engaged and asked to express their needs related to supporting the community with health. Community organizations were grateful to receive the kiosks at their community center locations because they were a part of the process of developing the information on the kiosks. The participating community organizations were consulted regularly and their responses to their requests for changes to the kiosks were shown through changes to the kiosk. For example, one community stakeholder and manager of a participating community organization requested an increased font size option on the site. This change was made to the website, and the stakeholder was informed. Continuous contact will be made with community stakeholders will be made to maintain a relationship with the community and provide a needed and effective intervention to the community.
The kiosk may not be effective in impacting health disparities alone. A cyclical relationship with the community will provide information to research other strategies to intervene in communities affected by health disparities. Therefore, although the findings of this study can be used as a guide or starting point for future developments of health kiosks in other communities, not all of the findings will be generalizable and applicable to other communities. For example, some findings were similar to a previous study on kiosks, yet some needs of the community were different (Padilla, Bull, Raghunath, et. al., 2010). Community members should be engaged to learn about the unique needs that the kiosk can address in other communities.
In addition to lessons learned regarding engaging underserved communities during the CBPR process, this study of bilingual interactive health kiosks indicated that information technologies can benefit underserved communities if they are culturally appropriate. The kiosks provided an opportunity to provide health information to people who had limited access to health information. The participants took advantage of the site visitors' support to teach them how to use the kiosk, while participants also benefitted from opportunities to confidentially use the kiosk. The kiosks created a space for participants to ask questions that they would not usually trust to ask in person. This was an unexpected advantage to information technology. While this study demonstrated that diverse communities may use health kiosks, a limitation of this study is that we were unable to track how many users were Spanish- and English-speaking. IP addresses can be used to track this information, yet wireless IP addresses prevented us from tracking how many people accessed the Spanish and English versions of the website on the kiosk.
Barriers to use must be removed before some underserved communities will utilize information technologies. Although it appears that some community participants trusted the information technology and believed that the questions they asked on the FAQ page were confidential, other community participants expressed a lack of trust in the kiosk and technology in general. Social service professionals need to be conscious of certain populations' lack of experience and distrust with technologies. It can be helpful to have a social worker or direct support staff who has built rapport with the community to introduce the technology to establish the community member's confidence in technology. Participants also suggested that touch screen computers may be used for kiosks, which could increase the usability of the kiosk for people who struggle with computer literacy. In addition, content displayed on the computer screen of the kiosk may look like communication tools other than computers, such as books or pamphlets, which could increase familiarity among most people who are not computer literate.
Finally, it can be helpful to employ a community advisory board (CAB) that offers insight into the culture of a community and provides greater access through partnership and communication. A CAB serves as the voice of community partners who are intrinsically connected to the pulse of their communities, and could potentially minimize distrust of technology. Moreover, it is suggested that social service professionals engage underserved communities through community-based participatory approaches that promote collaborative ways for developing health technology literacy, for fostering culture and communication strategies, and for understanding the integration of cultural strengths and partnerships in the intervention of community health technologies.