628 
Inadequate Adherence to Antiretroviral Treatment and Prevention in Hospital and Community Sites in Burkina Faso and Mali
Catherine Boileau*1,2, S Ag Aboubacrine3, P Niamba4, M V Zunzunegui5, V K Nguyen1, and S Rashed5
1McGill Univ, Montreal, Canada; 2Clin Med l' Actuel, Montreal, Canada; 3Hosp Natl du Point G, Bamako, Mali; 4Univ of Ouagadougou, Burkina Faso; and 5Univ of Montréal, Canada
Background: Rapid expansion of ART programs currently
underway in sub-Saharan Africa underlines the
relevance of adherence for ensuring optimal treatment outcomes, reducing the
risk of drug resistance, and boosting prevention efforts with sero-status specific interventions. Pilot studies showing
high levels of adherence may not, however, reflect “real world” adherence
levels. This study tested adherence in a population to test the correlation of
adherence with in ongoing programs with rates shown in the pilot studies.
Methods: To test this hypothesis, we conducted a
cross-sectional study of adherence was conducted using a questionnaire to
document socio-demographic, clinical, immunologic, and virologic
status as well as adherence to treatment, condom use, and partner notification
for patients having received >6 months ART in hospitals and community-based
organizations in Bamako, Mali, and Ouagadougou, Burkina Faso. Health care
providers interviewed pt and reviewed charts to complete the questionnaire.
Results: A total of 270 patients (94 men and 176 women)
were studied. Bivariate analysis showed that 51.5% of
pt reported always taking their ART medications and did not miss a single dose
in the week prior to report. Pt were more adherent in Bamako
(68.8%) than in Ouagadougou
(39.5%, p < 0.0001), and those followed
in hospitals were more adherent (58.2%) than those followed in community-based
organizations (CBO) (42.0%, p < 0.0001).
Individual demographic factors such as age, sex, education, occupation, and socio-economic
status were not associated with adherence. Adherence was also not correlated to
partner notification or condom use (-0.053 p
= 0.438; 0.009 p = 0.898,
respectively). Adherence was, however, found to be associated with better
clinical outcome. Fewer adherent patients lost weight (5.1%) than non-adherent
patients (18.6%, p = 0.001). Fewer
adherent patients had opportunistic infections (1.5%) than non-adherent
patients (7.1%, p = 0.023). No
association was found between adherence and treatment regimen, but pt on their
first regimen were more adherent (62.5%) than those on their second or third
regimen (45.9%, p = 0.080). HIV
status was disclosed to only 56.6% of partners. Condom use at last intercourse
with regular sexual partner was 58% overall. Men were more likely to use
condoms (80%) than women (57%, p = 0.019)
Conclusions: Significant efforts must be urgently
undertaken to improve adherence to treatment and prevention in this group.
Inter site variability suggests that determinants of adherence may vary
according to factors related to health care organization and political
involvement.
Keywords: adherence ; antiretroviral treatment ; Africa
