Skip Navigation LinksHome > December 1, 2006 - Volume 43 - Issue > Adherence to Antiretroviral Medication Among HIV-Positive Pa...
JAIDS Journal of Acquired Immune Deficiency Syndromes:
doi: 10.1097/01.qai.0000248346.79888.78
Reports From Abroad

Adherence to Antiretroviral Medication Among HIV-Positive Patients in Thailand

Maneesriwongul, Wantana Limkulpong DNSc*; Tulathong, Somchit MSN†; Fennie, Kristopher P PhD‡; Williams, Ann B EdD‡

Free Access
Article Outline
Collapse Box

Author Information

From the *Community Health Nursing Division, Nursing Department, Ramathibodi Faculty of Medicine, Mahidol University, Bangkok, Thailand; †Sena Hospital, Ayutthaya, Thailand; and ‡Yale University School of Nursing, New Haven, CT.

Reprints: Ann B. Williams, EdD, Yale University School of Nursing, PO Box 9740, New Haven, CT 06513 (e-mail:

Collapse Box


Background: The use of antiretroviral (ARV) medications is expanding rapidly in Thailand. The determinants of optimal adherence for HIV-positive patients in Thailand are unknown.

Methods: A sample of 149 Thai patients receiving ARV therapy at Bhumrasnaradura Infectious Disease Institute located near Bangkok completed a structured questionnaire and reported medication adherence on a 30 day visual analog scale. HIV RNA test results were abstracted from the medical record.

Results: Adherence ranged from 25% to 100%. The median was 100% and the mean was 96%. The majority of subjects (114, 77%) had an HIV RNA ≤50 copies/mL. An undetectable viral load was associated with adherence ≥95% (odds ratio [OR] = 3.0; 95% confidence interval [CI] 1.3 to 7.1; P = 0.02) and with a lower mean number of months on ARV therapy (22 versus 32 months; P = 0.03). Gender, educational level, method of payment, use of GPO-VIR, and whether or not the patient was on his or her initial ARV regimen were not associated with an undetectable viral load. In the multivariate analysis, only length of time in months was associated with an undetectable viral load. For each additional month, the odds of being undetectable were 0.975. (OR = 0.975; 95% CI 0.954 to .996; P = 0.02).

Conclusions: Adherence was high in this cohort and was associated with HIV-RNA levels. However, these data confirm that adherence is only 1 factor that determines the effectiveness of ARV treatment. Duration of treatment was associated with virologic failure, controlling for adherence.

Back to Top | Article Outline


In United States and other parts of the developed world, HIV clinical care has advanced rapidly through improved understanding of HIV dynamics and pathogenesis,1,2 the development and widespread use of quantitative tests for plasma HIV RNA levels,3 and the availability of powerful antiretroviral agents (ARVs).4 As a result of these scientific advances, the quality and quantity of life for people living with HIV in the wealthiest countries of the world have improved significantly. These stunning advances have been largely unavailable to most people with HIV, however, many of whom have little income and live in countries with limited health resources. For example, in 2002, the cost of a basic regimen of highly active antiretroviral therapy (HAART) in Thailand was $450 per month and only 5% of those who needed HAART could afford it.5

Fortunately, in Thailand, much progress has been made in providing access to care for the many people living with HIV/AIDS. In 1997, the Thai Communicable Disease Control Department established a clinical research network, and in 2000, health services, including a monitoring system, were developed to enhance the efficacy of triple therapy. In 2002, the Government Pharmaceutical Organization (GPO), a state enterprise run by the Thai Ministry of Public Health, began generic production of a number of ARVs, including zidovudine, didanosine, stavudine, lamivudine, nevirapine, saquinavir, and nelfinavir. In addition, a combination of zidovudine and lamivudine (Combid) and another of stavudine, lamivudine, and nevirapine (GPO-VIR) are available.5 At $35 per month, the cost of the GPO-VIR triple regimen is now less than a tenth of the price in 2002.

As a result of the reduction in cost, the Thai Ministry of Public Health was able to offer an Access to Care (ATC) program that covered 6000 to 10,000 patients with HIV infection in 2002.6 and was able provide more extensive access in 2004 in response to advocacy by the National Access to Antiretroviral Therapy for People with HIV/AIDS Program. Approximately 50,000 Thai patients received triple therapy in the fiscal year 2004,7 and the number is expanding rapidly with additional support from the Thailand Social Security Health Care Scheme in late 2004 and the Universal Health Care Scheme in 2005.

Back to Top | Article Outline


Although combination ARV regimens have proven to be effective in reducing HIV replication and forestalling immune system damage, they require a high level of adherence. Failure to take the medications as prescribed results in subtherapeutic drug levels and frequently leads to partial viral suppression and viral resistance.8-10 The determinants of optimal adherence in Thailand, however, are largely unknown.

Recent studies have reported high proportions of patients with >95% self-reported adherence, including 83% of patients in southern Thailand,12 80% in Nontaburi (a suburb of Bangkok),13 and 53% in eastern Thailand.14 These relatively high rates in comparison to other Asian countries such as India15 and China16 may be partially the result of extensive training workshops on the importance of adherence in ARV management that were conducted for Thai health care workers before widespread initiation of access to therapy.

Despite the high rates of adherence to HAART in Thailand, however, resistance to all 3 classes of ARVs is common.17,18 In a cohort of 88 patients from 4 regions of Thailand, prevalence of resistance to nucleoside reverse transcriptase inhibitors (NRTIs), nonnucleoside reverse transcriptase inhibitors (NNRTIs), and protease inhibitors was 48%, 21%, and 12%, respectively.17 All subjects had prior ARV exposure and were receiving at least 2 classes of ARVs when the plasma samples were collected in 2002. These reports suggest that Thailand needs effective monitoring for drug resistance and support of medication adherence in the free HAART program so as to preserve the effectiveness of the relatively inexpensive medications currently being dispensed.

We set out to describe the level of and predictors of adherence in a population of people receiving a variety of ARV regimens. Such knowledge is crucial if they are to benefit fully from the scientific advances that have transformed HIV into a manageable chronic disease in the world's wealthiest regions.

Back to Top | Article Outline


This was a cross-sectional descriptive study of medication adherence among Thai patients receiving HAART.

Back to Top | Article Outline

Data were collected between July 2003 and February 2004 at the Bamrasnaradura Infectious Disease Institute, a 400-bed infectious disease hospital operated by the Disease Control Department of the Thai Ministry of Public Health. The institute, located in a suburb of Bangkok, currently provides outpatient care to approximately 1200 patients with HIV infection. The clinical staff includes HIV specialists from medicine, nursing, social work, and pharmacy. The standard of clinical practice is consistent with the Thai national guidelines published by the Ministry of Public Health.19,20 In Thailand, HAART is recommended for all HIV-positive patients with CD4 cell counts <200 cells/mm3.

Approval for research involving human subjects was obtained from the Ramathibodi Hospital Faculty of Medicine, Mahidol University, and Bumrasnaradura Infectious Disease Institute in Thailand before data collection.

Back to Top | Article Outline

Patients were invited to participate in this study if they were currently prescribed a combination of at least 3 ARVs, were responsible for self-administration of their medications, were not cognitively impaired, and were Thai speakers. All eligible patients were approached in the clinic waiting room.

Back to Top | Article Outline

An interviewer-administered structured questionnaire was used to collect information about subjects' demographic and background characteristics, including duration of HIV infection, date of starting first HAART regimen, side effects, and reasons for missing doses.

A 30-day visual analogue scale (VAS) was used to measure medication adherence. This scale requested subjects to place a mark at the point on a horizontal scale corresponding to the subject's best estimate of the proportion of medication doses taken over the past 30 days. The scale ranged from 0% to 100% of doses. VAS adherence has been associated with lower plasma HIV RNA levels among patients at a publicly funded London clinic,21 among marginally housed San Franciscans,22 and in a small sample of Ugandan patients.23

HIV RNA test results were abstracted from the medical record. Test results were included in the study if they were obtained at least 3 months after the subject began HAART and within the 6 months before the interview date.

Back to Top | Article Outline
Data Analysis

Adherence was categorized as <95% and ≥95% for the primary analysis. HIV-RNA results were categorized as undetectable (≤50 copies/mL) and detectable (>50 copies/mL). Analyses were conducted using SPSS (version 13.0, SPSS, Inc., Chicago, IL). Frequencies for categoric variables and the mean, median, and standard deviation of continuous variables were used to describe the sample. Bivariate and stratified analyses were performed using t tests, χ2 tests, Fisher exact tests, and the Mantel-Haenszel summary χ2 test. Logistic regression was used to model predictors of an undetectable HIV RNA level (≤50 copies/mL). Backward selection evaluating −2-log likelihood scores was used to select the best model.

Back to Top | Article Outline


Characteristics of the sample are summarized in Table 1. Approximately 8 subjects declined to participate because of lack of time or concern about stigma. Of the 149 individuals participating in this study, 78 (52%) were male. The age of the subjects ranged from 22 to 61 years, with a median of 36 years. Most subjects were employed at the time of data collection (n =118 [79%]). Less than half were on their initial ARV regimen (56 [38%]), and the number of months on ARVs ranged from 5.5 to 132, with a median of 21 months.

Table 1
Table 1
Image Tools

GPO-VIR (stavudine, lamivudine, and nevirapine) was the most commonly prescribed medication; 94 (63%) of subjects were currently taking this medication. Forty-seven subjects (32%) were taking an NNRTI-based regimen other than GPO-VIR. Only 8 subjects (5%) were on a regimen that included a protease inhibitor. For those subjects on a second or subsequent regimen, reliable information about previously prescribed ARVs was not available.

Adherence ranged from 25% to 100% of doses taken over those prescribed during the past 30 days, with a median of 100% and a mean of 96%. More than 80% of subjects reported adherence to 95% or more of their prescribed doses. The most frequently reported reasons for missing doses included forgot (36%), had activities outside the home (33%), was worried that others would notice (24%), slept through the dose-time (20%), was too busy (16%), and experienced a change in daily schedule (9%).

Most subjects (n = 114 [77%]) had undetectable viral loads (HIV RNA level ≤50 copies/mL). In bivariate analyses, an undetectable viral load was associated with adherence ≥95% (odds ratio [OR] = 3.0, 95% confidence interval [CI]: 1.3 to 7.1; P = 0.02; Table 2) and with a lower mean number of months on ARV therapy (22 vs. 32 months; P = 0.03). Gender, educational level, method of payment, use of GPO-VIR, and whether or not the patient was on his or her initial ARV regimen were not associated with an undetectable viral load. When stratified by initial or subsequent ARV regimen, however, the statistically significant relation between ≥95% adherence and undetectable viral load was seen only among subjects on a second or subsequent regimen. Although the cell sizes for this analysis were small, the Mantel-Haenszel weighted OR was statistically significant (Table 3).

Table 2
Table 2
Image Tools
Table 3
Table 3
Image Tools

In the multivariate analysis, only length of time in months was associated with an undetectable viral load. For each additional month, the odds of being undetectable were 0.975. (OR = 0.975, 95% CI: 0.954 to 0.996; P = 0.02).

Back to Top | Article Outline


These results suggest that patients receiving HAART through the public hospital clinic in Bangkok are maintaining high levels of adherence and are responding with viral suppression. The cohort included many treatment-experienced subjects; most were not on their first regimen. Unfortunately, data were not available regarding previous regimens or reasons for changing regimens. One subject had been on ARVs for 11 years; the mean and median lengths of time on treatment were 24 and 21 months, respectively. It is encouraging that adherence levels are high after this length of time.

For subjects on their initial regimen, there was no association between adherence ≥95% and an undetectable viral load. The number of subjects for this analysis was quite small, however; only 8 subjects had adherence <95%, and 6 of those subjects had undetectable viral loads. The failure to detect an association could be attributable to the small number of subjects. It is also possible that high levels of adherence are not as crucial to viral suppression when a patient is taking his or her initial regimen, especially when that regimen is a potent combination of NRTIs and NNRTIs.

Although adherence is widely accepted as a necessary component, these data confirm that it is only 1 factor that determines the effectiveness of ARV treatment. We found that duration of treatment was associated with virologic failure, controlling for adherence. Parienti et al24 found that treatment interruptions were associated with virologic failure, controlling for adherence. Although we did not have data on treatment interruptions, treatment interruptions are more likely to occur with extended therapy and may explain the relation between treatment duration and virologic failure in this group.

The situation in Thailand with regard to ARV treatment may not be typical of the rest of the developing world, where large numbers of patients are just beginning therapy. Although the HIV/AIDS epidemic in Thailand is well established, so is the response. Despite economic and political obstacles over the past 2 decades, many Thai people have received high-quality care, including ARVs. Given the number of treatment-experienced individuals in Thailand, studies of virologic response and antiretroviral resistance among those entering the new treatment programs are essential.

There are several limitations to this study. First, the viral load results were obtained from chart review and were not tightly linked to the timing of the adherence interview. Economic constraints meant that HIV RNA levels were not monitored frequently; therefore, the time between self-reported adherence and measurement of viral load was as long as 6 months in some cases. A closer chronologic match between these 2 variables might have revealed a stronger association between adherence and undetectable viral load.

Second, the lack of reliable drug histories for each subject limits the conclusions that can be drawn from the data. For example, in a treatment-experienced group such as the patients in this study, sequential monotherapy, dual therapy, and treatment interruptions could play an important role in determining virologic response. That information was not available, nor was reliable information regarding the specific components of previous ARV regimens. Although most patients in this cohort were receiving NRTIs and NNRTIs, protease inhibitors have also been available in Thailand for some time in the private sector.

Finally, this was a small cross-sectional study conducted in an active clinical setting with a convenience sample of subjects reporting for clinical care. Carefully designed prospective studies to examine the links between adherence, virologic response, antiretroviral resistance, and specific aspects of the medication regimens are needed as Thailand continues to move forward with its program to provide comprehensive care to all its citizens living with HIV/AIDS.

Back to Top | Article Outline


We would like to thank Sigma Theta Tau (Delta Mu) and the Yale University School of Nursing for providing a grant to support this study, as well as the nursing staff and patients at Bamrasnaradura Infectious Disease Institute for their participation.

Back to Top | Article Outline


1. Ho DH, Neumann AU, Perelson AS, et al. Rapid turnover of plasma virion and CD4 lymphocytes in HIV-1 infection. Nature. 1995;373:123-126.

2. Wei X. Viral dynamics in human immunodeficiency virus type I infection. Nature. 1995;373:117-122.

3. Saag MS. Use of HIV viral load in clinical practice: back to the future. Ann Intern Med. 1997;126:983-985.

4. Deeks SK, Smith M, Holodniy M, et al. HIV-1 protease inhibitors. JAMA. 1997;277:145-153.

5. Kraisintu K, Eksaengsri A. Generic production of HIV/AIDS-related drugs in Thailand [abstract MoOrG1038]. Presented at: XIV International AIDS Conference; 2002; Barcelona.

6. Bunjumnong O. Thailand: access to antiretroviral treatment under Universal Health Care Scheme [abstract TuPeG5658]. Presented at: XIV International AIDS Conference; 2002; Barcelona.

7. Bureau of AIDS, TB and STI. Comprehensive and Continuum Care Guideline for Persons Living with HIV/AIDS. Bangkok, Thailand: Disease Control Department, Ministry of Public Health; 2004.

8. Vella S. Clinical implications of resistance to antiviral drugs. AIDS Clin Care. 1997;9:45-52.

9. Bangsberg DR, Hecht FM, Charlebosis ED, et al. Adherence to protease inhibitors, HIV-1 viral load, development of drug resistance in indigent population. AIDS. 2000;14:357-366.

10. Arnsten JH, Demas PA, Fazadegan H, et al. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: comparison of self-report and electronic monitoring. Clin Infect Dis. 2001;33:1417-1423.

11. Deleted in Proof.
12. Sittidach M, Chayakul P, Silpapojakul K, et al. Impact of pharmacist counseling on adherence to antiretroviral therapy. Presented at: XV International AIDS Conference; 2004; Bangkok.

13. Tulathong S, Maneesriwongul W, Sirapongam Y. Affective factors to medication adherence among patients receiving highly active antiretroviral therapy. Presented at: Third International Conference on Community Health Nursing Research; 2005; Tokyo.

14. Hanvong S. Knowledge, health belief, and adherence to antiretroviral therapy among people living with HIV/AIDS [thesis]. Thailand: Burapa University, Chonburi; 2005.

15. Safren SA, Kumarasamy N, James R, et al. ART adherence, demographic variables and CD4 outcomes among HIV-positive patients on antiretroviral therapy in Chennai, India. AIDS Care. 2005;17:853-862.

16. Li H, Wang Z, Cui WG, et al. Study on adherence and interrelated factors of acquired immunodeficiency syndrome patients receiving antiretroviral treatment [in Chinese]. Zhonghua Liu Xing Bing Xue Za Zhi. 2005;26:507-510.

17. Jenwitheesuk E, Watitpun C, Vibhagool A, et al. Prevalence of genotypic HIV-1 drug resistance in Thailand, 2002. Ann Clin Microbiol Antimicrob. 2003;2:4.

18. Sutthent R, Arworn D, Kaoriangudom S, et al. HIV-1 drug resistance in Thailand: before and after national access to antiretroviral program. J Clin Virol. 2005;34:272-276.

19. Ministry of Public Health. National Guidelines for the Clinical Management of HIV Infection in Children and Adults. 6th ed. Bangkok, Thailand: Ministry of Public Health; 2000.

20. Ministry of Public Health. Practice Guidelines on Development of Health Service and Monitoring System for Persons Living with HIV/AIDS Receiving Antiretroviral Therapy. Bangkok, Thailand: Ministry of Public Health; 2003.

21. Walsh JC, Mandalia S, Gazzard BG. Responses to a 1 month self-report on adherence to antiretroviral therapy are consistent with electronic data and virological treatment outcome. AIDS. 2002;16:269-277.

22. Giordano T, Guzman D, Clark R, et al. Measuring adherence to antiretroviral therapy in a diverse population using a visual analogue scale. HIV Clin Trials. 2004;5:74-79.

23. Oyugi JH, Byakika-Tusiime JB, Charlebois ED, et al. Multiple validated measures of adherence indicate high level of adherence to generic HIV antiretroviral therapy in a resource-limited setting. J Acquir Immune Defic Syndr. 2004;36:1100-1102.

24. Parienti JJ, Massari V, Descamps D, et al. Predictors of virologic failure and resistance in HIV-infected patients treated with nevirapine- or efavirenz-based antiretroviral therapy. Clin Infect Dis. 2004;38:1311-1316.


adherence; Thailand; VAS

© 2006 Lippincott Williams & Wilkins, Inc.


Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.