Measuring Adherence to Antiretroviral Treatment
Two hundred and ninety eight HIV + adults had been included in the study and 292 were analyzed. The study's enrollment process is shown in Fig. 1.
(Enlarge Image)
Figure 1.
Algorithm of the flow of subjects' enrollment to the study
Most (70.2%) of the 292 subjects were male; their age ranged from 27 to 70 years (median 43; mean 44.3; standard deviation (sd) 9.03); 33.6% self-classified themselves as afro descendant. Most (64.9%) men had sex with men and most women (97.2%) were exposed to HIV through heterosexual contact.
The great majority (80.9%) of the subjects had initiated ART with CD4 + T cells counts <350 cells/mm; 45.6% had had HIV-related symptoms prior to ART and 46.2% presented at least one opportunistic disease during the follow-up.
Most of them were very experienced in ART: the length of ART ranged from less than 1 to 20 years (median 8.0; mean 8.12; sd 3.31); the number of ART regimens used ranged from 1 to 21 (median 4; mean 4.21; sd 2.9); and 42.9% had used non-HAART regimens in the past. At the inclusion in the study, 49% subjects were taking two nucleoside or nucleotide reverse transcriptase inhibitors (NRTI) plus a non-nucleoside reverse transcriptase inhibitor (NNRTI); 42.7% were taking two NRTI plus a protease inhibitor (PI); 7% were taking 2 NRTI plus a NNRTI plus one or two PI; and 1.4% were taking only NRTI-based regimens.
The rates of subjects' adherence to ART based on self report and the proportion of subjects with undetectable HIV viral load according to adherence to ART are presented in Table 1. Reporting of taking "all prescribed ART tablets in the past 3 days" and "all doses during the weekends" were both significantly associated with undetectable viral load (Table 1).
The rates of ART withdrawals from the pharmacy in the past 12 months and the proportions of subjects with undetectable HIV viral load according to ART withdrawals are presented in Table 2. The mean rate of ART withdrawals for the 290 subjects for whom the data was available was 83.5% (IC 95, 81–85.9%). A strong positive association among rates of ART withdrawals in the last 12 months and undetectable HIV viral load was observed (Table 2).
There was a strong positive association between self-report and rates of ART withdrawals from the pharmacy: the mean rate of ART withdrawals in the past 12 months was 86.1% for those who reported having taken all pills and doses in the past 3 days compared to 62% for those who did not report all doses (Mann–Whitney test; p < 0.001). The pharmacy records of ART withdrawals was considered a more reliable measure of adherence and used in the analyses of association of adherence with the variables of interest (Table 3).
Older age, having any HIV-related symptoms prior to ART, CD4 + T cell counts <350/mm prior to treatment and satisfaction with the health service were significantly associated to better adherence to ART in the univariate analyses. Depression, more than four ART tablets per day, reporting difficulties in taking ART, getting drunk, missing medical appointments and having HIV + children were associated to poor adherence (Table 3).
Gender, educational level, occupational situation, family income, housing status, having children, perceived social support, disclosing HIV diagnosis, being in a stable relationship, the partner's HIV status, drugs use in the last year, chronic conditions other than depression, length of ART, agreement between ART described by the subject and medical records, knowledge of the mechanism of action of ART drugs, having doubts regarding ART or HIV, perceived effects of ART and confidence in the physician did not show association to adherence. (Table 3).
The following variables were included in the logistic regression: age, schooling, getting drunk, HIV-related symptoms prior to ART, CD4 prior to ART, number of ART pills, missing medical appointments, depression, doubts about HIV or ART, results of treatment, disclosed HIV diagnosis to someone, and satisfaction with the health service. The ART withdrawal was categorized in <80 or ≥80%.
The logistic regression analysis showed that having HIV-related symptoms prior to ART (p = 0.039), taking fewer ART pills per day (p = 0.003) and not missing medical appointments (p < 0.0001) were independently associated to higher adherence.
Results
Two hundred and ninety eight HIV + adults had been included in the study and 292 were analyzed. The study's enrollment process is shown in Fig. 1.
(Enlarge Image)
Figure 1.
Algorithm of the flow of subjects' enrollment to the study
Most (70.2%) of the 292 subjects were male; their age ranged from 27 to 70 years (median 43; mean 44.3; standard deviation (sd) 9.03); 33.6% self-classified themselves as afro descendant. Most (64.9%) men had sex with men and most women (97.2%) were exposed to HIV through heterosexual contact.
The great majority (80.9%) of the subjects had initiated ART with CD4 + T cells counts <350 cells/mm; 45.6% had had HIV-related symptoms prior to ART and 46.2% presented at least one opportunistic disease during the follow-up.
Most of them were very experienced in ART: the length of ART ranged from less than 1 to 20 years (median 8.0; mean 8.12; sd 3.31); the number of ART regimens used ranged from 1 to 21 (median 4; mean 4.21; sd 2.9); and 42.9% had used non-HAART regimens in the past. At the inclusion in the study, 49% subjects were taking two nucleoside or nucleotide reverse transcriptase inhibitors (NRTI) plus a non-nucleoside reverse transcriptase inhibitor (NNRTI); 42.7% were taking two NRTI plus a protease inhibitor (PI); 7% were taking 2 NRTI plus a NNRTI plus one or two PI; and 1.4% were taking only NRTI-based regimens.
The rates of subjects' adherence to ART based on self report and the proportion of subjects with undetectable HIV viral load according to adherence to ART are presented in Table 1. Reporting of taking "all prescribed ART tablets in the past 3 days" and "all doses during the weekends" were both significantly associated with undetectable viral load (Table 1).
The rates of ART withdrawals from the pharmacy in the past 12 months and the proportions of subjects with undetectable HIV viral load according to ART withdrawals are presented in Table 2. The mean rate of ART withdrawals for the 290 subjects for whom the data was available was 83.5% (IC 95, 81–85.9%). A strong positive association among rates of ART withdrawals in the last 12 months and undetectable HIV viral load was observed (Table 2).
There was a strong positive association between self-report and rates of ART withdrawals from the pharmacy: the mean rate of ART withdrawals in the past 12 months was 86.1% for those who reported having taken all pills and doses in the past 3 days compared to 62% for those who did not report all doses (Mann–Whitney test; p < 0.001). The pharmacy records of ART withdrawals was considered a more reliable measure of adherence and used in the analyses of association of adherence with the variables of interest (Table 3).
Older age, having any HIV-related symptoms prior to ART, CD4 + T cell counts <350/mm prior to treatment and satisfaction with the health service were significantly associated to better adherence to ART in the univariate analyses. Depression, more than four ART tablets per day, reporting difficulties in taking ART, getting drunk, missing medical appointments and having HIV + children were associated to poor adherence (Table 3).
Gender, educational level, occupational situation, family income, housing status, having children, perceived social support, disclosing HIV diagnosis, being in a stable relationship, the partner's HIV status, drugs use in the last year, chronic conditions other than depression, length of ART, agreement between ART described by the subject and medical records, knowledge of the mechanism of action of ART drugs, having doubts regarding ART or HIV, perceived effects of ART and confidence in the physician did not show association to adherence. (Table 3).
The following variables were included in the logistic regression: age, schooling, getting drunk, HIV-related symptoms prior to ART, CD4 prior to ART, number of ART pills, missing medical appointments, depression, doubts about HIV or ART, results of treatment, disclosed HIV diagnosis to someone, and satisfaction with the health service. The ART withdrawal was categorized in <80 or ≥80%.
The logistic regression analysis showed that having HIV-related symptoms prior to ART (p = 0.039), taking fewer ART pills per day (p = 0.003) and not missing medical appointments (p < 0.0001) were independently associated to higher adherence.
Source...