Objectives: To estimate adherence to highly active antiretroviral therapy (HAART) and its determinants in HIV-infected adults followed in field conditions in Abidjan.
Methods: A standardized questionnaire was administered to all consecutive adults on HAART who attended 3 urban HIV outpatient clinics. Patients were asked to self-report their pill intake during the previous 7 days, and, when necessary, to explain the reason(s) why they missed at least 1 intake. The adherence rate was estimated as the number of pills actually taken divided by the number of pills that should have been taken. The association of incomplete adherence (adherence rate <90%) with patients' characteristics was studied using multivariate logistic regression.
Results: Three hundred eight patients (male/female ratio: 1:1, mean time on HAART: 22 months) were interviewed. The median self-reported adherence rate was 78% (interquartile range: 65%-90%), with 76% of patients considered as incompletely adherent (adherence rate <90%). The most frequent self-reported reasons for missing at least 1 intake were an antiretroviral drug being out of stock in the clinic pharmacy (28%), the fear of drug side effects (27%), the impossibility of paying the drug's price (20%), and the intervention of traditional practitioners (18%). The only variables significantly independently associated with incomplete adherence were a school level ≥ secondary (odds ratio [OR] = 1.88; P = 0.03) and the absence of a patient's long-term formal commitment to adhere to HAART (OR = 3.08; P = 0.01).
Conclusions: These data illustrate the difficulty in obtaining high levels of adherence in field conditions in Abidjan. Sustainable access to treatment should be promoted by combating access barriers such as running out of drugs and costs that are too high.
From the *Service des Maladies Infectieuses et Tropicales, Centre Hospitalier Universitaire de Treichville, Abidjan, Côte d'Ivoire; †Département de Statistiques, Unité de Formation et de Recherche, Faculté des Sciences Biologiques et Chimiques, Université de Coccody, Abidjan, Côte d'Ivoire; ‡Unité de Soins Ambulatoires et de Conseil, Centre Hospitalier Universitaire de Treichville, Abidjan, Côte d'Ivoire; §Centre de Traitement Antituberculeux d'Adjamé, Abidjan, Côte d'Ivoire; and ∥INSERM U593, Université Victor Segalen, Bordeaux, France.
Received for publication November 14, 2006; accepted March 16, 2007.
Reprints: Serge Eholié, MD, PhH, Service des Maladies Infectieuses et Tropicales, Centre Hospitalier Universitaire de Treichville, BP V 3, Abidjan 01, Côte d'Ivoire (e-mail: firstname.lastname@example.org).
For the past 6 years, an increasing number of studies have illustrated the feasibility and efficiency of highly active antiretroviral therapy (HAART) in sub-Saharan Africa.1 Most of these studies reporting good adherence rates2-4 were also pilot studies, cohort studies, or therapeutic trials with high-standard conditions of follow-up that cannot be replicated in real life.
In the present cross-sectional study, we estimated the adherence to antiretroviral drugs among adults receiving HAART in 3 public HIV care centers of Abidjan and we looked for factors associated with incomplete adherence. All patients had access to HIV care and treatment under the national standard and were not participating in a scientific longitudinal study.
PATIENTS AND METHODS
The study was conducted in March 2002 within 3 public outpatient clinics in Abidjan: the Department of Infectious Diseases of the Treichville University Hospital, the outpatient HIV clinic of the Treichville University Hospital, and the Antituberculosis Center of Treichville. All 3 clinics were participating in the Ivorian national initiative of access to antiretroviral therapy. When the study started, it was estimated that 2231 patients had been prescribed HAART at least once in these 3 care centers. Patients on HAART attending these care centers during the study period were systematically proposed participation in the study. The inclusion criteria were: (1) 18 years or more of age, (2) HAART initiated for more than 1 month, and (3) informed consent. Patients included in the study were interviewed and examined by a survey doctor who was not part of the care center team. Drugs were supplied by the national drug procurement agency (Pharmacie de Santé Publique de Côte d'Ivoire, Abidjan) at prices negotiated with the major pharmaceutic companies. During the study period, patients with a family income less than US $50 per month and members of nongovernmental organizations (NGOs) fighting AIDS had access to government-subsidized antiretroviral treatment, with a monthly package price of US $7.50.
The standardized questionnaire recorded data on adherence (number and type of antiretroviral drugs taken during the last 7 days) and on the following explanatory variables: (1) sociodemographic characteristics (age, gender, monthly income [low, <US $100; medium, US $100-$600; and high >US $600], and schooling [≤primary school level or ≥secondary level]); (2) clinical data (Karnofsky score, Centers for Disease Control and Prevention [CDC] 1993 clinical stage, and last available CD4 cell count and plasma HIV-1 RNA level); (3) antiretroviral treatment, type (3 nucleoside reverse transcriptase inhibitors [NRTIs], 2 NRTIs plus 1 protease inhibitor [PI], or 2 NRTIs plus 1 nonnucleoside reverse transcriptase inhibitor [NNRTI]), date of first prescription, number of pills taken daily, time of pill intake in relation to meals, and most frequent reasons for omitting a drug intake; (4) patient's past counseling on adherence (once received counseling vs. never received counseling); and (4) patient's commitment to lifelong adherence to treatment (the patient had to answer yes or no to the following question: do you commit to adhere to HAART in the long term?).
Patients who self-reported that they missed more than 10% of drug intake during the previous 7 days were defined as incompletely adherent.5,6 The association between imperfect adherence and explanatory variables was studied using univariate and then multivariate logistic regression.
Three hundred eight patients were recruited in the study, including 162 men. Their main characteristics are detailed in Table 1. The mean age was 38 years (SD = 9.1). Most patients had a modest socioeconomic level, although more than 70% of them had completed secondary education. Most patients had once been symptomatic (80% at stage B or C), and their last available median CD4 count on HAART was 209 cells/mm3 (interquartile range [IQR]: 97-348 cells/mm3). The mean time on HAART was 19.2 months (SD = 5.9), with 60% of patients being on HAART for at least 12 months. Eighty-five percent of patients were still receiving their first-line HAART regimen, and 13% and 2% were on a second-line and a third-line regimen, respectively. The most frequent ongoing regimen at the time of the survey was 2 NRTIs plus 1 PI (58%).
The mean self-reported adherence rate to antiretroviral treatment was estimated at 77% (SD = 15%, median = 78%, IQR: 65%-90%). Two hundred thirty-five patients (76%) had a rate <90%, and were thus classified as incompletely adherent.
The most frequent self-reported reasons for missing 1 pill intake were the following: antiretroviral drugs out of stock (28%), fear of side effects (27%), lack of money (20%), influence of traditional practitioners (18%), complexity of antiretroviral treatments (18%), forgetting to take the drug with no other reason (16%), trips or weariness (14%), and unavailability of the referent doctor (15%).
In univariate analysis, only 3 of the variables listed in Table 1 were associated with incomplete adherence: secondary education or greater, a time on HAART equal to or greater than 12 months, and the lack of a patient's commitment to long-term adherence to his antiretroviral treatment. In multivariate analysis, the absence of commitment and a high schooling level remained independently associated with incomplete adherence (Table 2).
To our knowledge, this study is one of the few in sub-Saharan Africa to evaluate patients' adherence to antiretroviral treatment in field conditions, outside of therapeutic trials, cohort studies, or any high-standard programs managed by NGOs.4,7 The low adherence rate that we observed in our study suggests that adherence to HAART may be a more crucial issue in Côte d'Ivoire than previously hypothesized. Our mean adherence rate was less than the lower bracket of results previously reported from other African settings, where mean adherence rates were close to 90%,2-4,7,8 and from Brazil, where the mean adherence rate was reported at 79%.9 There is still limited information from developing nations on adherence rates and barriers, however.10,11
In our study, most reasons put forward by patients for not taking antiretroviral drugs were similar to those reported by other African investigators: lack of money, repeated stock shortages of antiretroviral drugs at the pharmacy, fear of side effects, and the negative influence of traditional practitioners.2,12 With regard to the latter, our experience is that these patients were asked by traditional practitioners to stop HAART because of an incompatibility between antiretroviral drugs and traditional medications. This suggests that efforts are required to improve the integration of traditional practitioners within the biomedical HIV care and treatment system, with the aim of maximizing adherence to HAART. This is most crucial in rural areas, where traditional healers are more numerous and more easily reachable than physicians. The integration of traditional practitioners within the national system should also help people living with HIV to pay their medical expenses by shortening their therapeutic route.13
Whereas a significant association between a more advanced clinical stage and nonadherence has been reported in Brazil,9 we did not find such an association among our patients. In our study, factors statistically associated with incomplete adherence also differed from those reported in industrialized countries, where the factors most frequently associated with poor adherence are the poor living conditions of patients and the complexity of treatments.5,14,15 Among our patients, incomplete adherence was associated with a high level of schooling, in contrast to reports from developed countries.9,15 In our context, people with high education are likely to be those fearing stigmatization the most. Thus, with their health status improving on HAART, they would flee from HIV care centers. We also found a trend toward an association between incomplete adherence and the time spent on HAART, a result already reported in numerous studies, where adherence decreased over time in relation to the weariness of patients.5,8,9,14,15 Finally, in our study, the factor most strongly associated with incomplete adherence was the patient's refusal to commit to long-term adherence to HAART. Such a simple question is probably inappropriate to be used as a tool for assessing adherence in the future. Nevertheless, this result shows that the patient is likely to be the best witness of self-adherence and that a patient can communicate his or her ill feelings about adherence to a physician who is not his or her personal doctor.
The consequences of incomplete adherence to HAART in terms of increased morbidity and mortality and increased risk of resistance to drugs have long been highlighted.5,12 The latter is particularly crucial in resource-limited countries, where viral load measurement is not part of the routine follow-up of patients on HAART, where resistance tests are not available, and where the access to second- and third-line regimens is still problematic.16,17 In addition, the risk of drug-induced resistance is higher with drugs with a low genetic barrier such as NNRTIs and lamivudine (3TC), which are currently recommended by WHO experts to be part of most first-line HAART regimens.18
Our 308 patients who consecutively attended 3 care centers during the study period cannot be seen as representative of the overall population of patients receiving HAART in Côte d'Ivoire. Nevertheless, our results highlight the importance of adherence support mechanisms in resource-limited countries. The encouraging experiences reported from Senegal, South Africa, and Haiti show that the following points are crucial: good coordination among health professionals and between health professionals and the community, the establishment of follow-up tools to support patients who find it difficult to adhere to the treatment, and a community approach to the willingness to pay medical expenses.2-4,8,19 This necessary multidisciplinary strategy cannot succeed without the decision makers' commitment to avoid stock shortages and to subsidize patients' treatments in the public system, however, as is already successfully done in the private sector.20
In conclusion, our data show that HIV-infected adults have real difficulties in reaching complete adherence to HAART in field conditions in Abidjan. Adherence is a multifactorial phenomenon that needs to mobilize various actors having a good knowledge of the field realities, including socioeconomic and cultural realities. Better cooperation between doctors and traditional practitioners may improve adherence to HAART. Meanwhile, sustainable access to treatment should be promoted by combating access barriers such as running out of drugs and high costs.
The authors thank Joanna Orne-Gliemann (INSERM U593) for her grammatical and editing contributions.
1. Kartenstein D, Laga M, Moatti JP. The evaluation of the HIV/AIDS Drug Access Initiatives in Côte d'Ivoire, Senegal and Uganda: how access to antiretroviral treatment can become feasible in Africa. AIDS
2. Laurent C, Diakhaté N, Ngom Gueye NF, et al. The Senegalese government's HAART initiative: an 18-month follow up study of feasibility, effectiveness, adherence, toxicity and viral resistance. AIDS
3. Orrel C, Bangsberg DR, Badri M, et al. Adherence is not a barrier to successful antiretroviral therapy in South Africa. AIDS
4. Tassie JM, Szumilin E, Calmy A, et al. Médecins Sans Frontières. Highly active antiretroviral therapy in resource-poor settings: the experience of Médecins Sans Frontières. AIDS
5. Bangsberg DR, Hecht FM, Charlebois ED, et al. Adherence to protease inhibitors, HIV-1 viral load and development of drug resistance in an indigent population. AIDS
6. Farmer CK. Methods for measuring and monitoring medication regimen adherence in clinical trials and clinical practice. Clin Ther
7. Landman R, Schieman R, Thiam S, et al. Once-a-day highly active antiretroviral therapy in treatment naïve HIV-1 infected adults in Senegal. AIDS
8. Lanièce I, Ciss M, Desclaux A, et al. Adherence to HAART and it principal determinants in a cohort of Senegalese adults. AIDS
9. Nemes MIB, Carvalho HB, Souza MFM. Antiretroviral adherence in Brazil. AIDS
10. Gill CJ, Hamer DH, Simon JL, et al. No room for complacency about adherence to antiretroviral therapy in sub-Saharan Africa. AIDS
11. Mills EJ, Nachega JB, Bangsberg DR, et al. Adherence to HAART: a systematic review of developed and developing nation patient-reported barriers and facilitators. PLoS Med
12. Weidle PJ, Malamba S, Mwebaze R, et al. Assessment of a pilot antiretroviral drug therapy programme in Uganda: patient's response, survival and drug resistance. Lancet
13. Hosseinipour MC, Kazembe PN, Sanne IM, et al. Challenges in delivering antiretroviral treatment in resource poor countries. AIDS
14. Carrieri MP, Raffi F, Lewden C, et al. APROCO study group. Impact of early versus late adherence to highly active antiretroviral therapy on immuno-virological response: a 3-year follow-up study. Antivir Ther
15. Gordillo V, Delamo J, Soriano V, et al. Socio-demographic and psychological variables influencing adherence to antiretroviral therapy. AIDS
16. Diomandé FVK, Bissagnéné E, Nkengasong JN, et al. The most efficient use of resources to identify those in need of antiretroviral treatment in Africa: empirical data from Côte d'Ivoire. AIDS
17. Kent DM, McGrath D, Ionnadis JPA, et al. Suitable monitoring approaches to antiretroviral therapy in resource-poor settings: setting the research agenda. Clin Infect Dis
18. World Health Organization. Antiretroviral drugs for the treatment of HIV-infection in adults and adolescents in resource-limited settings; recommendations for a public health approach (2005-2006 revisions). Guidelines Development Group, brief meeting report; Montreux, Switzerland, June 22-23, 2005. Available at: http://www.who.int/3by5/ARVmeetingreport_june2005.pdf
. Accessed September 1, 2005.
19. Koenig SP, Léandre F, Farmer PE. Scaling-up HIV treatment programmes in resource-limited settings: the rural Haiti experience. AIDS
20. Eholié SP, Nolan M, Gaumon AP, et al. Antiretroviral treatment can be cost-saving for industry and life saving for workers: a case study from Côte d'Ivoire's private sector. In: Economics of AIDS and Access to HIV/AIDS Care in Developing Countries. Issues and Challenges
. Paris, France: Collections Sciences Sociales et SIDA, ANRS; 2003:329-346.