JAIDS Journal of Acquired Immune Deficiency Syndromes:
The ADAPT-ITT Model: A Novel Method of Adapting Evidence-Based HIV Interventions
Wingood, Gina M ScD, MPH*†; DiClemente, Ralph J PhD*†‡
From the *Department of Behavioral Sciences and Health Education, Emory University Rollins School of Public Health, Atlanta, GA; †Center for AIDS Research, Emory University, Atlanta, GA; and the ‡Division of Infectious Diseases, Department of Medicine, Emory University School of Medicine, Atlanta, GA.
Supported in part by the Social and Behavioral Sciences Core of the Emory Center for AIDS Research (grant P30 AI050409).
Correspondence to: Gina M. Wingood, ScD, MPH, Rollins School of Public Health, Emory University, 1518 Clifton Road, NE, BSHE/Room 556, Atlanta, GA 30322 (e-mail: email@example.com).
The Institute of Medicine (IOM) recommends the use of HIV prevention interventions with proven efficacy to avert new infections. Given the time and cost associated with the development, implementation and evaluation of efficacious HIV interventions, adapting existing evidence-based interventions (EBIs) to be appropriate for a myriad of at-risk populations may facilitate the efficient development of new EBIs. Unfortunately, few models of theoretic frameworks exist to guide the adaptation of EBIs. Over the past few years, the authors have systematically developed a framework for adapting HIV-related EBIs, known as the “ADAPT-ITT” model. The ADAPT-ITT model consists of 8 sequential phases that inform HIV prevention providers and researchers of a prescriptive method for adapting EBIs. The current article summarizes key components of the ADAPT-ITT model and illustrates the use of the model in several case studies.
The Institute of Medicine (IOM) recommends the use of HIV prevention interventions with proven efficacy to avert new infections.1 In accordance with the IOM recommendations, the Centers for Disease Control and Prevention (CDC) require CDC-funded health departments and community-based agencies to use evidence-based behavioral interventions (EBIs) defined as effective through the CDC's Synthesis of Prevention Research project.2
The CDC's first review of EBIs from 1988 to 1996 was published in the Compendium of HIV Prevention Interventions with Evidence of Effectiveness.3 The compendium was updated to include proven interventions published through 20004 and was recently updated again to include proven interventions published through 2004.5 Although identification of EBIs is an important public health priority, ultimately, to avert further escalation of the HIV epidemic, it is required that EBIs be “scaled up” for wider dissemination and adoption by a diverse array of HIV prevention providers. Adoption of EBIs often requires agencies to modify existing interventions to facilitate implementation, encourage ownership, and enhance acceptability of the intervention for new target populations.6 Thus, the CDC has developed a 3-phase process in which EBIs are identified, packaged in a user-friendly format, and disseminated nationally to HIV prevention providers.5 Although the CDC has identified EBIs for several high-risk populations, there remains an urgent need to develop additional EBIs for other populations at high risk of HIV transmission or acquisition.
Given the time and cost associated with the development, implementation, and evaluation of efficacious HIV interventions, adapting existing EBIs to be appropriate for a myriad of at-risk populations may facilitate the efficient development of new EBIs. The process of modifying an EBI without competing with or contradicting its core elements or internal logic is referred to as “adaptation”.6,7 Without attention to the cultural context and HIV-related risks of a new target population, adapted interventions may remain faithful to the underlying theoretic framework and core elements on which they were originally developed but, unfortunately, may lack relevance, sustainability, and acceptability for the target population.8
Unfortunately, few models or theoretic frameworks exist to guide the adaptation of EBIs. One study reporting on the replication of several EBIs for adolescents described the processes used to adapt the interventions. Although informative, discussion of the replication and adaptation of the EBIs lacked a coherent model.9 Other investigators have discussed the importance of ethnography for guiding the adaptation process10 and the importance of culturally adapting interventions;8 however, implications for model development were limited. In an article on advancing translation, Solomon et al11 discussed principles for guiding researchers' adaptation efforts. In an effort to systematize guidance for adapting EBIs, the CDC has articulated the map of the adaptation process (MAP) of the adaptation process.6 This approach includes an assessment phase, to assess the new target population's HIV risk, the appropriateness of the EBI being considered, and the agency's capacity to implement the EBI. The assessment phase is followed by a preparation phase, during which the agency prepares to adapt the EBI and, finally, an implementation phase, in which the agency implements and pilots the adapted intervention. Throughout these 3 phases, there are specified action steps, feedback loops, and activities associated with monitoring and evaluating the adaptation process. Although it is a valuable addition to the adaptation literature, the MAP is quite an involved model, which may limit its utility for some community-based organizations, the primary providers of HIV prevention education in the United States.12
Over the past few years, Drs. Wingood and DiClemente have systematically developed a framework for adapting HIV-related EBIs, known as the “ADAPT-ITT” model. The ADAPT-ITT model consists of 8 sequential phases that inform HIV prevention providers and researchers of a prescriptive method for adapting EBIs. The ADAPT-ITT model has evolved over repeated applications from adaptations of several of the authors' CDC-defined EBIs.5,13-15 Through this process, each application has informed the evolution and development of successive iterations of the ADAPT-ITT model. Additionally, ADAPT-ITT has been applied with diverse populations of adolescents and adults in domestic and international settings. Moreover, observing the implementation and national dissemination of these EBIs16 also assisted the authors in developing and refining the ADAPT-ITT model. This iterative and experiential process has resulted in the development of a pragmatic framework for adapting EBIs. The current article summarizes key components of the ADAPT-ITT model and illustrates the use of the model in several case studies.
THE ADAPT-ITT MODEL
Phase 1, Assessment, involves conducting focus groups, elicitation interviews, or needs assessments with the new target population. Several researchers have documented the importance of assessment as part of the adaptation process.6,11,17 Formative evaluations are necessary to assess the HIV-associated behavioral and psychosocial risks of the new target population, their preferences for intervention content and delivery, and their perceived need for HIV prevention. Identification of HIV-associated risks that differentiate the new target population from the population on which the original EBI was developed and evaluated is critical. Obviously, it may not be feasible to adapt an HIV intervention to every risk profile. Identifying risks associated with HIV vulnerability is critical for effective adaptation, however.16 Focus groups or elicitation interviews should also be conducted with agency staff involved in implementing the HIV prevention intervention and key stakeholders. These formative evaluations are conducted to assess the capacity of an agency to adapt and implement an adapted EBI and to assess the availability of potential resources (eg, fiscal, space, staff) that could be provided by the agency and other key community stakeholders. At the end of this phase, the results from the formative evaluations should be analyzed. Thus, the assessment phase informs the next phase of the adaptation process.
Phase 2, Decision, involves: (1) reviewing the HIV interventions defined as EBIs in articles and publications written by the CDC,3-5 (2) deciding which EBI to select for the new target population, and (3) deciding if the EBI should be adopted or adapted. Restricting the selection of interventions to EBIs is consistent with the emphasis on using evidence-based decision making heralded by the IOM and adopted by the CDC.1,2 Deciding which EBI to select requires examining the “goodness of fit” between the original EBI and the proposed adapted EBI, with respect to the primary outcome targeted (eg, reduction in drug use, increase in condom use); the demographic characteristics of the target populations (eg, age, gender); the riskiness of the population, content, and delivery of the intervention; the agency's capacity to adapt and implement the EBI; and the resources available from key stakeholders and agency staff to assist in the adaptation process and implementation. Subsequently, the agency decides whether to adopt or adapt the EBI. If the EBI can be utilized without modification, the agency is encouraged to adopt the EBI. If, however, the agency believes that modifications would optimize the efficacy of the intervention, enhance its relevance for the new target population, and facilitate the agency's ability to implement the EBI, the agency is encouraged to adapt the EBI and proceed to the next phase of the adaptation process.
Phase 3, Adaptation, involves using an innovative pretesting methodology known as theater testing to adapt the EBI. Theater testing is a type of pretesting methodology that is commonly used to test products, such as television advertisements, videos, print advertisements, and public service announcements.18 Using this methodology, participants typical of the intended audience (the new target population) are invited to a central location to respond to a demonstration of a product (ie, the HIV intervention). At the end of the demonstration, participants receive a questionnaire and answer questions designed to gauge their reaction to the product. An important strength of this methodology is the opportunity to obtain reactions to messages, concepts, and visual materials in a relatively short period. Furthermore, this methodology closely resembles what is experienced by the target population; thus, an accurate assessment of their reactions to the product can be obtained.
As part of the adaptation process, we extend the use of the theater pretest methodology to gauge participants' reactions to an HIV prevention intervention. Approximately 15 members of the new target population are invited to participate in the theater test. During theater testing, facilitator(s) implement modules of the original EBI that capture core elements of the intervention. While the members of the target population serve as “participants” for the intervention modules, key stakeholders and agency staff, seated behind participants, observe the implementation of the intervention modules. At the end of each module, participants, key stakeholders, and agency staff complete brief surveys that contain closed-ended and open-ended questions to elicit their reactions regarding the appropriateness of the elements in the module for the new target population, such as role plays, materials, didactic instruction, and other content. The goal is to collect critiques of the material, content, and delivery of the EBI and to identify additional materials and/or activities not in the EBI that should be included to enhance its relevance and efficacy for the new target population.
Facilitators collect the surveys and then, using these surveys as triggers, engage in a group discussion with the participants regarding the relevance of the module content for the new target population. Subsequently, key stakeholders and agency staff are invited to join the discussion and offer ideas for adaptation of the EBI. Analyses of the surveys are used to summarize common themes that emerge from the theater test. This active pretesting methodology closely resembles what participants may experience when they participate in an EBI; thus, this methodology greatly facilitates the adaptation process. Specifically, theater testing highlights what needs to be adapted and solicits guidance on how the content, delivery style, and/or materials should be adapted to enhance the relevance and efficacy of the EBI for the new target population.
Phase 4, Production, involves producing draft 1 of the adapted EBI. Production of draft 1 of the adapted EBI involves balancing the need to maintain fidelity to the core elements, the underlying psychosocial theory, and the internal logic of the original EBI with numerous priorities, including the capacity of the agency to modify and implement the adapted EBI, the resources available from key stakeholders, the results of the theater test, the results of the formative evaluation, and the assistance that could be afforded by consultants (phase 5).
Production of draft 1 of the adapted EBI can be a time- and resource-intensive process. To facilitate documentation of the options considered and decisions made to produce draft 1 of the adapted EBI, it can be useful to create an adaptation plan. This plan outlines (1) the aim of the adaptation, (2) the EBI to be adapted, (3) the CDC publication citing the intervention as an EBI, (4) the new target population or context, (5) the core elements of the original EBI, (6) the aim of the new materials and/or activities for inclusion in the adapted EBI, and (7) the new material and/or activities that may be more appropriate and relevant for the target population. The term core elements, as defined by the CDC, are those components that are critical features of an intervention's intent and design and are thought to be responsible for its effectiveness.6 Core elements are derived from the behavioral theory on which the intervention is based and are essential to the implementation of the intervention and cannot be ignored, added to, or changed. Unlike the core elements, the key characteristics of an EBI can be adapted. As defined by the CDC, key characteristics are important but not essential attributes of the original EBI's recommended activities and delivery method. The key characteristics can be adapted to fit the risk factors, behavioral determinants, and risk behaviors of the new target population and the specific circumstances of the agency and key stakeholders.6 Because adaptation can be a complex process, developing an adaptation plan can assist in documentation of the elaborate process of adapting an EBI. Thus, producing an adaptation plan addresses the questions of why the content and/or materials were adapted and what has been adapted. The adaptation plan provides for a transparent and verifiable adaptation process.
Additionally, during this phase, the agency needs to decide whether the goal of adaptation is to produce a successfully adapted EBI for the new target population or to test whether the adapted EBI produces changes in theoretically important HIV prevention mediators and behavioral outcomes. If the goal of adaptation is to assess the effect of the adapted EBI on mediators and behavioral outcomes, the agency should develop quality assurance procedures and process evaluation measures. These measures are used to monitor the quality and assess the fidelity of recruitment efforts, intervention delivery (ie, fidelity), survey administration, and informed consent activities (if required).
Phase 5, Topical Experts, involves identifying consultants who possess significant expertise in substantive content areas, relevant to draft 1 of the adapted EBI, for which the agency perceives a lack of expertise. Prior research on adapting EBIs has found that consultations and technical assistance greatly facilitate the adaptation process.6,7,17,19 Often, but not always, agencies lack a particular area of expertise that is critical to the adaptation of an existing EBI. For example, if theater testing identified “substance abuse prevention” or “the need to adapt an EBI for a different ethnic group or gender” and the agency does not have this requisite expertise, the agency would attempt to identify experts who could serve as consultants to provide specific content expertise.
Phase 6, Integration, involves integrating content provided by the topical experts into the adapted EBI. The integration of content from the topical experts results in a second draft (draft 2) of the adapted EBI. Integration of intervention content suggested by the topical experts is weighed by prioritizing the capacity of the agency to implement the suggested adaptation and by maintaining fidelity to the core elements and theoretic underpinnings of the original EBI. Additionally, topical experts may provide measures for the study survey that are valid, reliable, and culturally appropriate for the new target population. Finally, readability testing is integrated into the adapted EBI to create a third draft (draft 3) of the adapted EBI.20,21 Utilizing the Flesch-Kincaid Readability Test, a computerized formulaic assessment of the grade level of reading skill that would be required to comprehend the adapted EBI session activities can assist in gauging readability of the adapted EBI materials. Tailoring the adapted EBI's content to a fifth-grade reading level may increase comprehension of the EBI, facilitate its use, and, as a consequence, enhance its efficacy.
Phase 7, Training, involves training personnel, including: (1) facilitators in group management and facilitation skills, (2) recruiters and retention staff in effective recruitment and retention techniques, (3) assessment staff in administering the study assessments, and (4) data management staff in managing the study data. Additionally, topical experts may be actively involved in training facilitators on material they have contributed to the adapted EBI.
Phase 8, Testing, involves 2 discrete steps. The first step involves conducting a pilot test with approximately 20 participants from the new target population of draft 3 of the adapted EBI. The CDC's adaptation process has also identified pilot testing as a key step in the adaptation process.6 The pilot test serves as a “dress rehearsal” during which the agency uses trained staff to implement the adapted EBI. At the end of each session, participants complete exit interviews designed to solicit feedback about whether they think the intervention content, materials, and delivery are relevant, useful, and appropriate. At the end of each session, feedback is also solicited from key stakeholders and agency staff who observed the implementation of the adapted EBI. Thus, the first step of the testing phase assesses “adaptation efficacy” by answering the question of how successful the original EBI adapted for the new target population was. Analyzing the data from this step results in the final adapted EBI and is used in the second step of the testing phase.
The second step of the testing phase involves conducting a phase 2b study and randomizing the new target population to the final adapted EBI (created in step 1) or a control condition to assess the short-term efficacy of the adapted EBI. A baseline survey, process measures, and at least a 3-month postintervention assessment should be administered to gauge study outcomes (eg, changes in knowledge, attitudes, beliefs, and relevant behaviors). Analyzing the pretest/posttest data should answer the question of whether the adapted EBI enhances theoretically important mediators and behavioral outcomes over a 3-month follow-up.
Unique and Salient Features of the ADAPT-ITT Model
Although other adaptation models have been designed and proven valuable, the ADAPT-ITT model has several unique and salient features that should be highlighted. Specifically, throughout the adaptation process, the ADAPT-ITT model (1) directly involves members of the new target population, key stakeholders, and agency staff from the initial phase to the last phase; (2) triangulates diverse measures by using numerous qualitative assessments (ie, formative and process measures) and quantitative assessments (ie, theater test surveys, pilot study pretest/posttest surveys); (3) uses multiple and novel pretesting methodologies, including didactic (ie, focus group, elicitation interviews), action-oriented (ie, theater testing), and computer-based (ie, readability testing) technology, to indicate what needs to be adapted and how adaptation should proceed; (4) attempts to promote a balance between fidelity and adaptation; (5) uses topical experts to assist in creating the adapted intervention; (6) prescriptively indicates when during the adaptation process a draft of the adapted EBI is to be generated and the number of drafts of the EBI that are to be created; (7) facilitates the documentation necessary for efficient adaptation (eg, the adaptation plan); (8) indicates the sources of data (eg, focus group, theater test, exit interviews, pilot study, phase 2 study) that need to be analyzed, which facilitates data analysis; (9) uses a sequential and systematic process to create an adapted intervention; (10) assess adaptation efficacy of the adapted EBI by means of a pilot study; and (11) assesses short-term outcome efficacy of the adapted EBI by means of a phase 2b study. Table 1 summarizes the ADAPT-ITT model.
Diversity cautions against transferring public health strategies and interventions from one local to another or from one population to another without attention to the social environment. Using methodologically appropriate tools that are based on sound scientific practice, are culturally congruent, and include the target population can be beneficial, however. This article offers an approach to adapting EBIs, and it implementation could provide an efficient mechanism of designing culturally sensitive and efficacious HIV interventions.
1. Institute of Medicine. Report Brief No Time to Lose: Getting the Most from HIV Prevention
. Washington, DC: National Academies Press; 2001. Available at: http://www.iom.edu/file.aspid=4131
. Accessed May 10, 2007.
3. HIV/AID Prevention Research Synthesis Project. Compendium of HIV Prevention Interventions with Evidence of Effectiveness
. Revised ed. Atlanta, GA: Centers for Disease Control and Prevention; 1999.
4. Kay L, Crepaz N, Lyles C, et al. Update of the Compendium of HIV Prevention Interventions with evidence of effectiveness. Presented at: National HIV Prevention Conference; 2003; Atlanta.
5. Lyles CM, Kay LS, Crepaz N, et al. Best-evidence interventions: findings from a systematic review of HIV behavioral interventions for US populations at high risk, 2000-2004. Am J Public Health
6. McKleroy VS, Galbraith JS, Cummings B, et al. Adapting evidence-based behavioral interventions for new settings and target populations. AIDS Educ Prev
7. Center for Substance Abuse Prevention. Finding the Balance: Program Fidelity and Adaptation in Substance Abuse Prevention
. Rockville, MD, 2001.
8. Castro FG, Barrera M, Martinez CR. The cultural adaptation of prevention interventions: resolving tensions between fidelity and fit. Prev Sci
9. Bell SG, Newcomer SF, Bachrach C, et al. Challenges in replicating interventions. J Adolesc Health
10. Wainberg ML, Gonzalez MA, McKinnon K, et al. Targeted ethnography as a critical step to inform cultural adaptations of HIV prevention interventions for adults with mental illness. Soc Sci Med
11. Solomon J, Card JJ, Malow RM. Adapting efficacious interventions: advancing translational research in HIV prevention. Eval Health Prof
12. Collins C, Harshbarger C, Sawyer R, et al. The Diffusion of Effective Behavioral Interventions Project: development, implementation, and lessons learned. AIDS Educ Prev
13. DiClemente RJ, Wingood GM. A randomized controlled social skills trial: an HIV sexual risk-reduction intervention among young adult African-American women. JAMA
14. Wingood GM, DiClemente RJ, Mikhail I, et al. A randomized controlled trial to reduce HIV transmission risk behaviors and sexually transmitted diseases among women living with HIV: the WiLLOW Program. J Acquir Immune Defic Syndr
15. DiClemente RJ, Wingood GM, Harrington KF, et al. Efficacy of an HIV prevention intervention for African American adolescent females: a randomized controlled trial. JAMA
16. Wingood GM, DiClemente RJ. Enhancing diffusion of HIV interventions: development of a suite of effective HIV prevention programs for women. AIDS Educ Prev
17. Kelly JA, Heckman TG, Stevenson LY, et al. Transfer of research-based HIV prevention interventions to community service providers: fidelity and adaptation. AIDS Educ Prev
. 2000;12(Suppl A):87-98.
18. National Cancer Institute. Making Health Communication Programs Work
. Publication No. 04-5145. US Department of Health and Human Services: Bethesda, MD, 2004.
19. O'Donnell L, Scattergood P, Adler M, et al. Strategic planning for addressing technical assistance needs during replication of effective HIV interventions. AIDS Educ Prev
20. Kincaid JP, Fishburne RPJ, Rogers RL, et al. Derivation of New Readability Formulas (Automated Readability Index, Fog Count and Flesch Reading Ease Formula) for Navy Enlisted Personnel, Research Branch Report 8-75
. Millington, TN: Naval Technical Training, US Naval Air Station, Memphis TN; 1975.
21. Farr JN, Jenkins JJ, Paterson DG. Simplification of Flesch Reading Ease Formula. J Appl Psychol
Case Study 1: Traditionally, African-American faith-based institutions have been instrumental in addressing diverse social issues confronting African Americans. The involvement of faith-based institutions in HIV prevention has the potential to influence an individual's norms, attitudes, and perceptions supportive of risk-reduction practices. A partnership has been formed between faculty at the Emory University Rollins School of Public Health and a nondenominational megachurch and its satellite churches in Atlanta. The goal of this unique partnership is to develop a “faith-based HIV prevention intervention” for young adult African-American women. The 5 counties in which the churches are located have the highest proportion of HIV and AIDS cases relative to the rest of Georgia, thus indicating the population's vulnerability to HIV. Over a 3-month period, the adaptation team proceeded to adapt the EBI known as SiSTA3 using the ADAPT-ITT model (Table 2). Following the ADAPT-ITT model, the adaptation team first conducted focus groups and a needs assessment with young adult African-American women attending the megachurch and elicitation interviews with key stakeholders (ie, pastors, directors of the Health Ministry, the Women's Ministry, and the College Ministry). Subsequently, as part of a theater test, over a 2-day period, the original EBI was administered to 15 African-American women, aged 18 to 29 years, who attend the megachurch. Participants provided feedback on activities that needed to be adapted and direction on how to adapt the activities. While the original EBI was being implemented to participants, key stakeholders observed the implementation process and also provided feedback on how to adapt the intervention. After these activities, biweekly conference calls were initiated to create an adaptation plan (Table 3), to collate the suggestions from the theater test, and to integrate the ideas of the topical experts knowledgeable about HIV prevention for African-American women and the culture of the church. A 3-session adapted curriculum has been developed, and we are in the process of pilot testing the adapted intervention.
Case Study 2: More than 70% of women and girls infected with HIV live in sub-Saharan Africa. Studies of HIV acquisition throughout sub-Saharan Africa show sharply increasing prevalence in women during the teenage years. Drs. Wingood and DiClemente are collaborating with scientists from the Nelson Mandela School of Medicine to adapt the EBI known as SiHLE for Zulu-speaking female adolescents residing in KwaZulu-Natal, South Africa. With discretionary funds from the HIV Prevention Trials Network (HPTN), a meeting was held focusing on adapting the SiHLE intervention to enhance its relevance for female adolescents in KwaZulu-Natal, South Africa. Over a 4-month period, the adaptation team proceeded to adapt the EBI known as SiHLE5 using the ADAPT-ITT model (Table 4). In accordance with the ADAPT-ITT model, the adaptation team first conducted focus groups with adolescents and key community stakeholders and elicitation interviews with HIV prevention scientists in KwaZulu-Natal. Subsequently, as part of a theater test, over a period of 4 days, the original SiHLE intervention was implemented to 15 adolescents from diverse urban and rural HIV/AIDS organizations in KwaZulu-Natal. Adolescents provided feedback on activities that needed to be adapted and provided direction on how to adapt the activities. While the SiHLE intervention was being implemented to the adolescents, representatives from a range of local nongovernmental organizations (NGOs; key informants) observed the implementation process and also provided feedback on how to adapt SiHLE. After these activities, biweekly conference calls were initiated to integrate and collate suggestions from the meeting and an adaptation plan was developed to adapt SiHLE for Zulu female adolescents (Table 5). In addition to the activities outlined in Table 5, as part of the adaptation process, the name of the intervention was changed from the SiHLE, which means “beautiful,” to SIBAHLE, which translates as “We are beautiful” in Zulu, the predominant language in KwaZulu Natal province. We have modified the language and added local cultural concepts and customs to the intervention that more accurately capture African society as being more collective and communal. For example, we have chosen to reframe the intervention by using words such as “we,” “us,” and “our” as opposed to “I,” “me,” or “my.” We have also allowed more time for roundtable discussions and exploration without losing the core elements of intervention. Throughout the intervention, we have allocated time for South African culture to be effectively integrated into the intervention. We have used names, terms, and regional slang that are relevant to South African young adults. A 4-session adapted curriculum has been developed, and we are in the process of planning for evaluation.
adapting; HIV interventions
© 2008 Lippincott Williams & Wilkins, Inc.
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Highlight selected keywords in the article text.
Data is temporarily unavailable. Please try again soon.