Wong, Ilene Y*; Lawrence, Nicholas V BA; Struthers, Helen MSc, MBA‡; McIntyre, James MBChB, MRCOG‡; Friedland, Gerald H MD†
Widespread availability of antiretroviral medications (ARVs) is a new reality in sub-Saharan Africa, and there is still limited literature regarding adherence among populations there. Most of the available data indicate high levels of adherence,1,2 but much of such data is largely from specialized centers and highly selected populations. Some studies have reported adherence levels less than 80%. For example, at the Perinatal HIV Research Unit (PHRU) at Chris Hani Baragwanath Hospital in Soweto, South Africa, 28% of patients had adherence rates less than 95%.3
In another arena, the literature points to an anecdotal history of poor adherence to tuberculosis treatment regimens in sub-Saharan Africa, particularly in disadvantaged communities.4,5 It is believed that adherence to therapy in the African setting is complicated by difficulties in communication and lack of medical resources (J. McIntyre, MBChB, MRCOG, G. Gray, MBChB, unpublished data, 1999). Preliminary interviews performed by the authors in rural South Africa in 2001 suggest that nonadherence may also be attributable to lack of medication-taking experience. Clearly, room exists for patient education regarding basic pharmacology and medication-taking skills.
In recent years, a small body of literature indicates that several innovative educational programs in Africa have been effective in improving patient adherence to medication for non-AIDS-related illnesses. Patient empowerment, information sharing, social networking support, and educational materials such as “photonovels” have all proven to be effective and may be useful in the promotion of ARV adherence (J. McIntyre, MBChB, MRCOG, G. Gray, MBChB, unpublished data, 1999).6,7
The use of a videotape as a means to improve patient adherence is a relatively new notion, although a substantial body of literature supports the efficacy (and cost-effectiveness) of a videotape as a means to influence patient knowledge and attitudes, particularly in resource-poor areas. One meta-analysis concluded that videotapes can convey complicated information more effectively than written materials, regardless of the race or education level of the audience.8 A growing body of literature attests to the widespread use of videotaped HIV education programs in resource-limited settings. One survey of English language cancer education materials suggested that written patient education materials often outstripped patients' abilities.9 Educational videotapes may circumvent the problem of illiteracy in South Africa, which has been estimated to range from 27% in metropolitan areas to 50% in rural areas.10
The literature on the effect of videotapes on adherence is somewhat smaller, and a 1988 literature review on the efficacy of videotapes in patient education concluded that although videotapes are as good as, and often more effective than, traditional methods of education in increasing knowledge, they offer no advantage in promoting long-term adherence.11 More recent studies, however, have shown that a videotape intervention had a significant effect on drug adherence to gonorrhea treatment and on tuberculosis therapy in Cambodia.12,13 Videotape education has also been shown to improve behavioral adherence with basic health maintenance.14
Another potential benefit of videotape-based education in resource-limited settings is the ease with which digital educational materials may be replicated and disseminated. For example, in rural areas of South Africa, where televisions are scarce, videotapes may be viewed on a battery-powered laptop computer.
Finally, a number of studies articulate the desire of underrepresented minorities to have educational materials specific to their culture. In a well-controlled study, a culturally specific videotape was rated by African Americans as more credible and attractive and of higher quality than an identical videotape with a multicultural cast and message. Educational videotapes made by producers who used focus groups to assess the language, values, and style of thinking of the target groups have been shown to result in greater retention of knowledge.15,16
We report on the development and preliminary testing of a culturally sensitive educational videotape intended to promote better awareness of the importance of maintaining adherence to ARV regimens among HIV-positive patients in South Africa. The study was performed at the PHRU of the University of the Witwatersrand at Chris Hani Baragwanath Hospital, a 2964-bed hospital in Soweto, South Africa. It was developed in 2002 through 2003, when more than 200 patients were enrolled in ARV treatment studies at the PHRU, providing one of the largest cohorts of patients in treatment in an African setting at that time. The videotape can be viewed at http://ymtdl.med.yale.edu/theses/available/etd-08202004-175255/.
To determine the best structure for the educational videotape, the study authors (IYW and NVL) critically reviewed 3 sets of HIV educational videotapes. Two of these, “Introduction to Anti-Retroviral Therapy” from the Beat It!: HIV/AIDS Treatment Literacy Series and the “Inmate Adherence Video Series” produced by Albany Medical Center, directly addressed matters of adherence.17 The videotapes were evaluated for their ability to engage the audience, their pacing, and their use of narrative techniques. Portions of these videotapes that were thought by the screenwriter to be particularly effective were earmarked as models for videotape education.
Production of the Soweto-specific videotape began after individual interviews with health care providers at the PHRU. Additionally, the PHRU's 6 adherence counselors (HIV-positive individuals chosen to lead support groups and act as counselors for patients who struggled with adherence) were interviewed individually and in focus groups regarding their experience with patients on ARVs and strategies used to promote medication adherence. In addition to the information collected in interviews and focus groups, the team extracted population-specific data from PHRU adherence surveys conducted in 2001 to address barriers to adherence. Motivational and behavioral skills important to the enhancement of adherence18,19 were identified and embedded into and emphasized in the videotape.
A draft script was produced by the study authors (IYW and NVL) in conjunction with an experienced screenwriter after the decision was made to use a number of unusual narrative techniques. After review of the existing videotape products, it was decided to use 3 different “voices” in the videotape: (1) a role play with a physician and patient discussing the most common concerns that patients have expressed about adherence to ARV therapies and their side effects, (2) voice-over narration contrasting the major health outcomes experienced by adherent compliant and nonadherent HIV-seropositive patients, and (3) a documentary section that showed adherence counselors giving practical advice on medication-taking strategies. As a pacing/transition element, 2 short humorous “commercial” segments were included between the 3 sections, which depicted medication adherence during real-life dramatic situations (sexual encounter and chase through a township area). The documentary portion of the videotape consisted of clips of adherence counselors discussing different strategies they recommended to improve adherence.
Focus Groups With Adherence Counselors
The preliminary script was distributed to the adherence counselors for editing and review and was then brought to a focus group for reading. During this time, the counselors identified segments that seemed confusing, poorly worded, or culturally inappropriate and added additional material based on their own adherence support practices.
Explaining the Concept of Resistance
The counselors noted that they often use a Zulu metaphor, amaso tsha omzimba, meaning soldiers of the body, to explain what HIV does to the immune system. They described the virus as a “naughty” virus that can hide and hurt the soldiers of the body. They often emphasized that the virus is clever and can wear camouflage or dress up as soldiers to deceive the body's army. Additionally, they described how the virus can “learn” to defeat the soldiers of the body if you do not take your medications correctly, such that “the drugs no longer work in your body.” They also frequently depicted HIV as a monster. When asked about which animals were considered to be particularly frightening in their culture, the counselors mentioned snakes as a potentially strong metaphor for Africans.
Reasons for Nonadherence
The adherence counselors identified myriad reasons why their clients in Soweto were not adherent to their medication regimens. These included the following:
1. Attitude of “depression and loss of hope”: successfully adherent clients were more likely to have a positive outlook and an ability to “take control of their life.”
2. Lack of education, including limited understanding of the importance of taking ARVs correctly: many of their nonadherent clients were not fully educated on key elements of drug-taking knowledge, namely, the dosages, medication names, and major and minor side effects. Related to this was a lack of understanding regarding the consequences of nonadherence; one counselor commented that her patients often say they “feel so good they don't need it [medications],” not understanding that they feel good because of their medications.
3. Lack of family support: because of the stigma associated with HIV/AIDS in African society, a large number of their clients had not disclosed their status to their family members and often thought that they had to hide their drugs from their families to avoid being “chased out.”
4. Side effects: lack of tolerance for minor side effects, not knowing that many side effects can be treated, and not knowing that they should return to the clinic if they have side effects.
5. Lack of food and money: contributing to increased side effects and an inability to afford transportation fees to attend routine clinic visits or support groups.
6. Lifestyle: issues such as alcohol, the lack of scheduled daily activities and routine, and high levels of unemployment resulting in lack of mental memory clues to take their medication.
Medication-Taking Strategies/Advice for Nonadherent Clients
1. Education: the counselors uniformly agreed that it is imperative to educate patients fully about all aspects of their disease and treatment. They thought that patients must be aware of the kinds of medication they are taking, including specific dosages and side effects, as well as having an understanding of viral replication.
2. Disclosure: all adherence counselors believed that disclosure to one's family and close friends is essential to promoting adherence. Family members who are aware of the patients' HIV status and ARV regimen can act as in-home reminders and help to organize the family routine around medication administration. Family members and friends can also provide vital moral support for patients.
3. Memory aids: the counselors advocated the use of memory aids to enhance adherence. These included the use of pillboxes near commonly accessed places, such as toothbrush holders and bedsides, and the use of calendars on which daily doses are checked off as taken. Patients were encouraged to synchronize their medication taking with daily routines, such as daily television shows.
4. Income generation and financial aid: the PHRU offered a number of income-generation projects to patients, including jewelry making and food gardening, and counselors were also extremely supportive of patients' efforts to secure disability grant funding. In addition, travel stipends and lunches were provided for patients who attended the clinic and their support group meetings.
5. Detailed discussion of the patients' habits: the counselors agreed that it was of great benefit to the patient to go through a detailed schedule of his or her day to identify times when he or she is likely to miss a medication dose.
6. Counseling and support groups: all counselors strongly recommended that patients attend support group meetings to identify medication-taking strategies. They noted that because of the intense stigma still attached to HIV disease in South Africa, patients who attended such meetings were generally more optimistic about their illness after speaking with other HIV-positive patients.
The research team cast black South African students who mirrored the language and mannerisms of the videotape's target population, black South Africans living in Soweto. Students who could speak without the assistance of a microphone and who had excellent diction were selected. To play the part of the physician, the research team cast a student who was physically mature and comfortable with medical terminology. On-location sets were secured at Chris Hani Baragwanath Hospital and in nearby townships, including clinic rooms, exterior hospital shots, and township-house interiors. Nonvenomous snakes were procured for the snake metaphor.
The videotape was filmed over the course of 4 days, using a Canon (New York, NY) GL-1 digital camera and tripod, with a boom mike used for the dialogue and documentary style sections. Attention was paid to filming at secure locations to ensure the safety of actors and equipment. More importantly, attention was paid to creative filming, using point-of-view shots, close-ups, and traveling shots to make the videotape a dynamic film rather than a formal documentary.
The score for the film's opening and section sequences was created from the HIV-positive chorus at the PHRU. Additional background music and sounds were provided by the University of the Witwatersrand School of Film's digital music library.
After filming, the digital images were downloaded onto a Fujitsu (Tokyo, Japan) P2110 laptop (Crusoe [Santa-Clara, CA] 900-mHz processor, 256 RAM, 20-GB hard drive, and CD-RW/DVD drive). Editing was performed using Adobe (San Jose, CA) Premiere 6.5. Editing goals included (1) making the videotape seem as cinematic as possible, (2) using layers of background music and sounds to place an emphasis on certain concepts, (3) repeating certain motifs (eg, HIV snake replication) to reinforce concepts, (4) ensuring clarity of sound through boosting and digitally altering unclear dialogue, and (5) keeping the videotape to less than 20 minutes so as to avoid losing the attention of viewers. Selected screen shots from the videotape are depicted in Figure 1.
Videotape Viewing and Evaluation
This was a convenience sample of patients receiving ARVs and patients who were ARV naive. ARV-experienced patients were recruited in a sequential order by the research assistant after being identified through PHRU records as being on an ARV trial, whereas ARV-naive patients attending the clinic were approached randomly. All patients were approached for enrollment while awaiting routine care, and participants gave written consent in accordance with the Human Investigations Protocol of the Committee for Research on Human Subjects at the University of the Witwatersrand. In the case of minors less than 18 years of age, consent was obtained from the guardian or parent. Eligibility criteria included HIV positivity, age between 18 and 50 years of age, and a command of the English language that would allow for understanding of the videotape as determined by a multilingual South African research assistant. Exclusion criteria included unwillingness to watch an educational videotape; evidence of dementia; and inability to comprehend English, Zulu, or Sesotho. After the posttest questionnaire was administered, patients were given food to compensate them for their time and effort.
The patients took the prevideotape questionnaire in a private room under the supervision of a research assistant, who was present to assist with interpretation of the questionnaire. The participants were then shown the adherence videotape. The questionnaire had been pilot tested on 5 patients to identify ambiguously worded or misleading questions. Twelve questions assessed patient knowledge of medication-taking strategies, including basic knowledge of HIV science, management of side effects, and concepts of resistance and adherence. An additional 12-point multiple choice section assessed patient knowledge of the existence of specific side effects and medication strategies that were addressed in the videotape. A final question assessed the amount of HIV education already received by the patient, specifically asking whether he or she had attended general support group meetings, watched a general educational videotape, or met with adherence counselors. The postvideotape questionnaire contained an additional open-ended question asking what the patients had learned from the videotape and how it would change their medication-taking strategies.
The questionnaires were labeled with their study participation number only, and the key was kept in a separate location to ensure confidentiality. Tests were scored by research assistants using a 24-point scale. Statistical analysis, including paired Student t tests, χ2 tests, and regression analysis, was performed using SAS (Cary, NC) software at the Yale University Prevention Research Center.
Thirty-four patients viewed the videotape and participated in its evaluation. The demographic characteristics of the patients are shown in Table 1. Most of the interviewed patients (82%) were women, because they were recruited from the PHRU. The mean age of the patients was 31.2 years, with a range from 13 to 48 years. Patients had completed a mean of 8.8 years of school, and 7 patients had education levels of 6 years or less. Twenty-three patients were ARV experienced, and 11 were ARV naive. The ARV-experienced patients had been taking ARVs for a mean of 19 months, with a range from 1 to 48 months. In the whole group of patients, 17 (50%) had watched an HIV educational videotape, 24 (71%) had been to an HIV patient support group, 4 (41%) had been to an adherence support group, and 19 (56%) had met with an adherence counselor.
Table 2 shows the mean pre- and posttest scores for all patients as well as the mean scores for the ARV-naive and ARV-experienced patients. Figure 2 graphically depicts the improvement shown by each patient cohort. The overall increase in scores for all patients was statistically significant (P = 0.02). In addition, analysis of the 2 patient subgroups using SAS (paired Student t test) shows a significant increase in knowledge after the intervention for ARV-naive (P = 0.003) and ARV-experienced (P = 0.001) patients. Before viewing the videotape, the average score of all participants was 17.8 of 24 points (74.2%, ±4.02 points SD). The average score of the 11 patients who were ARV naive was 15.6 of 24 points (65%, ±4.5 points SD) compared with 19.0 of 24 points (79.1%, ±3.4 points SD) scored by patients who had experience with ARVs.
Overall, 28 (82.3%) of the 34 patients showed improvement after watching the videotape. After watching the videotape, the average score of all participants was 20.1 of 24 points (83.8%, ±4.03 points SD), with an average improvement of 2.2 points of 24 (9.2%, ±1.9 points SD). Subset analysis shows that the 11 ARV-naive patients had an average increase of 3.0 points of 24 (12.5% improvement, ±2.5 points SD), whereas the 26 patients who had prior experience in taking ARVs averaged an increase of 1.8 points of 24 (7.6% improvement, ±1.5 points SD). Although the difference between the ARV-naive and ARV-experienced groups was not significant (P = 0.35), this measurement was likely hampered by the lack of power of the study; a larger sample size (n > 65) is needed to show a difference between intergroup improvement.
The most significant gains in knowledge were seen in the areas of basic adherence concepts, side effects, and medication-taking strategies.
All 34 participants in the study indicated that they thought the videotape was clear in its educational purpose. Primary observation of participants as they viewed the videotape showed that they rapidly engaged with the videotape's content, often reacting with a sharp breath or gasp when the snakes were shown and reacting with outward sympathy when the character Joseph lies ill after being nonadherent to his drugs. After viewing the videotape, most patients described changes they would make to their medication-taking strategies, with disclosure being the intervention most frequently mentioned; others commented on the knowledge they had gained about HIV replication.
This project demonstrates that it is feasible to develop culturally sensitive videotape educational materials in South Africa using focus groups and familiar metaphors, which can have a significant effect on HIV-positive patients' knowledge of ARV-taking concepts. Key to the success of the videotape were the focus groups with ARV adherence counselors. These were critical in creating culturally sensitive materials for use in the videotape. The open-ended question format allowed the counselors to express in detail the specific barriers to adherence experienced by the Sowetan patients in their care. Further, discussions with adherence counselors, patients, and health care professionals identified a core set of strategies that led to the success of the videotape as an intervention, including (1) dramatization of concepts using cinematic motifs and humor; (2) use of a question and answer format; (3) use of culturally specific metaphors, music, and local actors/locations; (4) use of repetition; and (5) provision and illustration of medication-taking skills and strategies. The discussion of the types of metaphors used to demonstrate resistance was particularly important because of the editorial decision to include within the videotape an arresting visual depiction of HIV replication and resistance. It was thought that the concept of resistance often depicted in the literature and in videotapes was a difficult and abstract one that would benefit from a concrete visual representation. The observed response to these metaphors of “soldiers of the body” and snakes on the part of viewers supports this view. We suggest that future educational interventions might benefit from using these strategies to ensure a culturally sensitive product.
At this preliminary stage of the assessment of our videotape, it is promising that analysis shows a statistically significant increase in postviewing knowledge for ARV-naive and ARV-experienced patients. Even more heartening is the increase in knowledge shown in patients who were naive to ARV therapy. We recognize that these results are preliminary and that a larger sample is necessary to demonstrate that ARV-naive patients demonstrate a statistically significant increase in comparison to ARV-experienced patients. Nevertheless, current data from the evaluation do suggest that the videotape intervention might be particularly helpful for patients initiating ARV therapy who are less familiar with medication-taking strategies.
One conclusion that can be drawn from this study is that even ARV-experienced patients who have received extensive PHRU educational interventions may still learn more from watching the videotape.
A number of lessons were learned concerning the logistics of videotape production and questionnaire administration. First, we recommend that documentary sections involving interviews be structured as a question and answer session in which the interviewed individual repeats the question stem in his answer. In this way, their answers can be taken in context without needing to involve the second role of an interviewer. This is also a more focused method of conversation than ad-lib talking, which can fail to cover certain key points. Second, authenticity was ensured by the use of drama students and adherence counselors only as characters in the videotape. Finally, confidentiality concerns dictated the use of actors and actresses and nondisclosure of HIV status by counselors.
With regard to the videotape assessment and questionnaire, we noted that the knowledge questionnaire was most effective if given orally so as to minimize the impact of relative illiteracy or unfamiliarity on the part of the test takers. More than once a patient admitted to reading the question incorrectly when the test was reviewed after viewing the videotape.
The project has several limitations. First, the study may not be fully generalizable to other populations. The PHRU at Chris Hani Baragwanath Hospital is a center of excellence in South Africa, offering extensive baseline education interventions that are lacking at most other South African health centers. The human resources offered by the PHRU, particularly the presence of a full-time staff of HIV counselors who run daily support group meetings, were invaluable in the development of the educational videotape. The presence of these resources, however, may have affected the outcomes data that we procured regarding patients' knowledge of medication-taking strategies. It is possible that patients without the support offered by the PHRU would score lower because of a lack of basic health literacy. Conversely, it is also possible that patients who see the videotape in health care environments that offer fewer educational interventions may demonstrate an even greater increase in score than the PHRU patient population. The generalizability of the study is further limited because most participants did speak basic English and had education levels higher than elementary school, which are skills that would be less expected in rural areas. Large populations of urban patients with HIV are present in South Africa and eligible for ARVs, however.
With regard to the evaluation, we recognize additional limitations of the study, including the small sample size, short-term nature of the study, and lack of correlation between patient adherence and therapeutic outcomes.
The adherence literature suggests that improvement in knowledge does not, in and of itself, lead to improved adherence or clinical outcome,20-25 which are the most desirable end points of any adherence intervention. The videotape was also designed to speak to larger concerns, to identify barriers, and to emphasize issues of motivation and behavioral skills, however. In addition to HIV/AIDS knowledge, the videotape addressed motivational issues such as maintaining a positive attitude and family support; counseling and support groups; and learned behavioral skills such as diagnosis disclosure, lifestyle strategies, and use of memory aids. The videotape was developed with the intention of providing a culturally relevant and engaging script that would be readily accessible to patients. This and similar videotape-based interventions could have widespread utility as an efficient and inexpensive technique to enhance ARV adherence as life-prolonging medications become increasingly available to patients in resource-poor countries.
Despite the limitations noted, we are pleased to report that after being viewed by local health authorities, the adherence videotape has since been dubbed into the Zulu language and has been extensively used in the urban and rural province-wide rollout of ARVs in the South African province of Kwa-Zulu Natal. In addition, the videotape has been widely used throughout the rest of South Africa.
The logical next step in the evaluation of the educational videotape is a randomized controlled trial comparing adherence among patients who are shown or not shown the videotape before initiating ARV therapy. Because prior studies have shown a decay effect of educational interventions on behavior, it would be helpful to assess the durability of the impact of the videotape by measuring adherence and clinical outcomes over time.
This study demonstrates that the development, production, and testing of a culturally sensitive and specific videotape to identify and address barriers to adherence to ARVs in South Africa is feasible and has a measurable and beneficial impact on patient knowledge of HIV, ARVs, and basic medication-taking concepts and strategies. More widespread use of the videotape strategy and a controlled randomized study to determine the effect of the intervention on patient adherence and clinical outcome are warranted.
The authors thank Nancy Angoff, MD, Associate Dean, Yale School of Medicine, David Kessler, MD, former Dean, Yale School of Medicine, and Kaveh Khoshnood, PhD, for providing initial direction and funds. Thanks go to Dr. Valentine Njike for his invaluable assistance on statistical matters and to Susan Barringer for editorial help. At the PHRU at Chris Hani Baragwanath Hospital, Jane Munyoro, Andreas Pakendorf, Karen James, and Glenda Gray provided significant logistic support and allowed the study to be conducted on their premises. Most importantly, this project could not have been completed without the accessibility and enthusiasm of the PHRU's 6 adherence counselors: Ghadi Makhoba, Lawrence Ndou, Annie Mokonyane, Nonhlanhla Themane, Priscilla Yatha, and Joan Modise.
© 2006 Lippincott Williams & Wilkins, Inc.