The advent of antiretroviral treatment (ART) has dramatically improved the prognosis and quality of life for HIV/AIDS patients, reducing the rate of disease progression and death [1–5]. Adherence to ART, however, is critically important for successful treatment of HIV/AIDS. Previous studies have suggested that ≥ 95% adherence results in virologic success of > 80%, while the success rate drops to 60% in patients with 80–94% adherence .
However, since ART is a lift-time treatment, maintaining an almost full adherence level over such a long term poses a significant challenge for both patients and healthcare providers. Moreover, studies in developed countries indicate that HIV drug resistance develops with suboptimal adherence, and leads to diminished drug efficacy .
In China, a nationwide free ART program was initiated in 2002. Since the “Four Free and One Care” policy was promulgated in 2003, 127 projects sites of the China Comprehensive AIDS Response (China CARES) program have been established, providing free ART and related health care services. At each site, ART is distributed to registered patients on a 2-weekly or monthly basis at village clinics. Village doctors are responsible for monitoring the therapeutic effectiveness, drug toxicity and reactions, and also facilitating patients' adherence to ART.
However, increased access to ART has been accompanied by increasingly unsatisfactory adherence levels and the potential risk of drug resistance . As of 30 June 2005, a total of 19 456 patients had received or were on free ART in China, among whom 8% dropped out, with the most commonly cited reason being intolerable side effects and difficulties with adherence (NCAIDS, unpublished data, 2005).
Many studies in developed countries have assessed rates of adherence to ART; however, the results vary significantly, ranging from 30 to 80% , depending on the study design, study population and measurement tools. In developing countries, less than 60% of patients in Brazil  and Botswana  were ≥ 95% adherent.
Factors associated with adherence can be grouped under four main categories: (1) patient-related factors, e.g. demographic characteristics, psychosocial parameters, knowledge, personal skills; (2) treatment regimen-related factors, e.g. years on treatment, pill burden, side effects; (3) provider-related factors and the patient-provider relationship; and (4) environmental and social factors, such as supervision of treatment, HIV-related stigma and social support .
Unfortunately, there have been few studies on ART adherence and its associated factors among patients in rural settings in China where the majority of HIV/AIDS individuals reside. With the rapid scale-up of free HIV treatment and care in China, it is imperative for policy makers and healthcare providers to understand the actual adherence level in specific populations and settings, and how complex factors affect patients' adherence so as to develop feasible and cost-effective interventions to improve the program and to sustain long term treatment benefits.
Setting and participants
A cross-sectional study was conducted in two neighboring project sites supported by the China CARES program, Shenqiu in Henan Province and Fuyang in Anhui Province, where the free ART programme started in late 2003. The majority of HIV-infected individuals in this area were former commercial blood/plasma donors (for a detailed description see ).
Patients aged 18 or above taking free ART between January and April 2005 were selected for the study. In each village, patients were contacted by their village doctors for an appointment and then introduced to a trained interviewer from the National Center for AIDS/STD Control and Prevention (NCAIDS). Those who agreed to participate in the study provided signed informed consent. The questionnaire was interviewer-administered in a separate room in the village clinic. Those not accessing services regularly were contacted through home visits and interviewed at home. The questionnaire we used was developed by NCAIDS based on the modified adherence baseline and follow-up questionnaires created by the Adult AIDS Clinical Trial Group .
There is some evidence that the patients' self reports of missing pills in the recent short-term are almost always reliable . Thus, in this study we only asked patients to recall their medication taking in terms of prescribed doses in the previous three days before the interview. Adherence was calculated as the number of pills actually taken divided by the number of pills prescribed over the past three days. Patients who reported an intake of ≥ 95% of the prescribed medication were considered adherent, those < 95% were assessed as non-adherent.
We studied six groups of factors that might potentially have an impact on adherence to ART in the study population: (1) socio-demographics; (2) clinical characteristics of ART; (3) knowledge and perception of ART; (4) behavioral characteristics; (5) medication management skills; and (6) health services-related characteristics and social support. We also gathered information about reasons for missing medications, and symptoms of side effects.
Since the participating patients received free ART and free treatment of opportunistic infections, the cost factor of medication was not investigated in this study as an exposure variable. Neither could we evaluate the potential risk factors of psychological problems as depression considering the limited abilities of rural patients to understand questions and rating-scaled response options in a predefined depression questionnaire.
All quantitative data was entered into the database twice, by different persons, using EpiData 3.0 (EpiData Association, Odense, Denmark) and analyzed using SPSS 11.5 for Windows (SPSS, Inc., Chicago, Illinois, USA).
The percentages of categorical variables were summarized and proportions were compared using either the chi-square test or Fisher's exact test. Univariate analysis was used for identifying potential risk factors on adherence. Variables that were found to be significant (P < 0.05) were assessed using multivariate logistic regression modeling (forward method). The level of significance considered for the subsequent and final model was α = 0.05, and only those found to be statistically associated with the outcome at a P < 0.10 were retained in the model.
Ethical approval was received from the Institutional Review Board of NCAIDS, China.
Among 273 patients who were on ART during the data collection period, 191 eligible patients (70.0%) were approached and interviewed, resulting in a 100% response rate. Among them, 181 (94.8%) questionnaires were used in the analysis; 10 questionnaires were eliminated for missed responses to key questions.
The descriptive characteristics of 181 patients are summarized in Table 1. Women accounted for 59.7%, and the patients' ages ranged from 27 to 75 with 79.5% in the 30 to 59 age range (mean: 47.8 years, SD: 11.3). In terms of age and sex distributions, the sample was representative of the HIV-infected population treated with free ART resident in the study site. More than half (53.6%) of patients were illiterate and 75.1% were working, either for pay outside the home or on their own farms.
Patients in Shenqiu, Henan Province were receiving an ART combination of zidovudine/didanosine/nevirapine, and those in Fuyang, Anhui Province were receiving stavudine/didanosine/nevirapine. These two regimens were the first line recommended regimens, based on availability, of ART in China and both had dosing schedules of two doses a day and 4–5 pills per dose.
We observed a high percentage of patients (89.5%) who reported having experienced side effects caused by ART during the previous one month prior to the interview. Diarrhea and nausea were the most commonly reported symptoms.
Twenty-nine percent of patients used tobacco and 14.4% of patients drank alcohol. No one reported using illicit drugs, as expected, and this was confirmed by the local doctors.
In the presence of side effects, 59.7% of patients believed that they would not stop taking medication, unless advised to by their doctor; 66.3% of the patients firmly believed that “non-adherence could possibly lead to a therapeutic failure”. However, 24.6% of the patients could not describe the correct methods of taking the medication prescribed.
The majority (89.5%) of respondents claimed that the ART was very effective or somewhat effective for treating their disease. Twenty-one percent of patients claimed that keeping full adherence to ART was difficult and posed a burden to their daily lives. Roughly three-quarters (76.2%) of patients had developed some tools to remind them to take their medicine as prescribed, such as establishing a habit of taking medications twice a day at a specific time, or having either a family member or other patients remind them when a dose was due.
Few patients (11.6%) felt it was inconvenient to visit the village clinics to get their ART because the clinic was located far from their homes. Doctors were reported to regularly review ART regimens with 37.6% patients when dispensing ART; however 32.6% reported that they had never received any information from their doctors about how to deal with side effects. Less than half (45.3%) of patients had been visited by their doctor or other health worker in their home after they had initiated the ART, and 4.4% of patients had little trust in their doctors. Stigma from the community was also common, with 43.8% of patients claiming that their neighbors and friends treated them differently since taking ART.
Based on the definitions of adherence outlined above, 81.8% of the patients in this study reported ≥ 95% adherence in the previous three days; 49.7% of patients claimed to have never missed a single dose over the entire duration of their ART.
Univariate and multivariate analysis
A summary of the univariate analysis of factors affecting ART adherence are given in Table 2 (unadjusted). Adherence was statistically associated with: correct knowledge of side effects; correct knowledge non-adherence leading to treatment failure; perceived effectiveness of treatment; having developed reminder tools to help with the medication; perceived taking medication as no burden to their daily lives; doctor explains regimen each time ART is dispensed; regular home visits by health care staff; and patients' trust in their doctors.
The result of the multivariate analysis of the independence of these above eight factors (Table 2, adjusted) showed that four of them remained statistically significant: patients' knowledge about side effects, knowledge about treatment failure, having developed reminder tools, and patient's trust in his/her doctor were all independently associated with the adherence to ART.
Ninety-eight participants reported having missed treatment doses: 48% simply forgot, 31.6% were too busy with work and 27.6% suffered intolerable side effects. Other barriers to ART adherence included interruption by social or community activities (16.3%), unpleasant taste of the pills (14.3%), stigma (14.3%), food restriction (11.2%) and oversleeping (5.1%).
In this study we measured patient adherence to ART by self-report and found that 81.8% patients were ≥ 95% adherent to their prescribed regimen in the previous three days. However, the study findings should be interpreted cautiously when comparing the adherence rate and risk factors identified in this study with other study results, because the methods used (self-report vs. pill counts or medication electronic monitoring system (MEMS) and settings (free ART vs. non-free, rural vs. urban, plasma donor vs. IDU or other risk groups) where the study was conducted, are quite different across areas in China and the world.
In this study, we did not find an association between adherence to ART and demographic characteristics, duration of treatment, tobacco and alcohol use, patients' perceived treatment effectiveness and medication burden.
Patients' knowledge has previously been reported to be closely associated with adherence to the treatment , as has having a good doctor-patient relationship and quality health care services that assist patients to identify and overcome various barriers to satisfactory treatment adherence . The multivariate analysis in this study confirmed that these are also strong predictors of good adherence in the study population. Our findings also alert us to the importance of regular medical counseling and education related to ART adherence among rural patients on free ART.
Our data showed that the risk of non-adherence among patients with no reminder tools was 4.22 times greater than for those who employed some reminder methods. However, we found very few patients had utilized the memory aids and reminder tools, such as medication organizers or pill diaries [18,19], which suggests is may be necessary to introduce more personal medication management skills into this population.
Side effects have been identified in some studies as significant barriers to good adherence [20,21]. In this study, a high proportion of the patients interviewed (89.5%) reported signs and symptoms of adverse reactions to their treatment, however, the univariate analysis did not identify this factor as being associated with patients' adherence, probably due to the comparatively small number of patients reporting no side effects. Nevertheless, it seems clear that a priority for the free ART program should be to increase the available regimens to include drugs which cause fewer adverse reactions that would presumably limit non-adherence due to side effects.
Stigma and discrimination against HIV-infected people was pervasive in the two study sites. Stigma was anecdotally reported to occur more often in villages where there were fewer HIV-infected individuals. Though we found no significant relationship between such perceived stigma and their treatment adherence, such stigma may lead to patients' unwillingness or fear to take medicine when other people were present.
The three most frequently cited reasons by this study population for missing doses were (1) simply forgetting, (2) being too busy and (3) side effects. Thus it would be beneficial to the patients if ART adherence strategies were integrated into their daily lives, with individualized coping strategies for various barriers such as forgetfulness, work and other life events, which would empower the patients to improve their personal medication management abilities. These strategies have been effective elsewhere [22,23] and are endorsed by the World Health Organisation (WHO) .
There are several limitations to the present study that may affect the validity and generalizability of the results. Firstly, our study investigated patients' adherence to ART over a short period of time – just the three days prior to the interview – so judging whether a patient was ≥ 95% adherent or not was to some extent a matter of chance. For example, missing one dose during the past three days would be categorized as < 95%. In addition, we have only assessed the level of adherence in terms of missing doses and did not ask details on whether the patients take the medication in accordance with time and dietary instructions which could help us to make a more comprehensive assessment of ART adherence. Secondly, social desirability or recall bias in self-reports may have occurred, which may reduce the accuracy of patients' responses and over-estimate the actual adherence to ART, especially among older patients . Thirdly, this study was unable to collect blood specimens for HIV viral load and CD4+ T-cell counts, which would have allowed us to examine patients' adherence levels biologically. Fourthly, the small sample size of this study may limit the generalizability of our results.
In conclusion, poor ART adherence in this area, though low, was associated with poor knowledge of ART, lack of reminder tools and trust in the doctor. Forgetfulness, being busy and side effects are frequent challenges to good adherence. The findings suggest that in addition to education and counseling, comprehensive interventions, such as training for patients on medication self-management skills, tailoring the regimen to the patient's lifestyle, addressing issues related to side effects, and improving adherence monitoring and health care services should be priority approaches to promote adherence to ART among HIV/AIDS patients who received free ART in rural China.
Our study would not be possible without the cooperation of HIV-infected individuals, who took the time to answer the questionnaires.
Sponsorship: This study was partly supported by the China Multidisciplinary AIDS Prevention Training Program with NIH Research Grant # U2R TW06918 funded by the Fogarty International Center, National Institute on Drug Abuse, and the National Institute of Mental Health.
1. Descamps D, Flandre P, Calvez V, Peytavin G, Meiffredy V, Collin G, et al
. Mechanisms of virologic failure in previously untreated HIV-infected patients from a trial of induction-maintenance therapy. JAMA 2000; 283:205–211.
2. Murri R, Ammassari A, Gallicano K, De Luca A, Cingolani A, Jacobson D, et al
. Patient-reported nonadherence to HAART is related to protease inhibitor levels. J Acquir Immune Defic Syndr 2000; 24:123–128.
3. Gifford AL, Bormann JE, Shively MJ, Wright BC, Richman DD, Bozzette SA. Predictors of self-reported adherence and plasma HIV concentrations in patients on multidrug antiretroviral regimens. J Acquir Immune Defic Syndr 2000; 15:386–395.
4. Haubrich RH, Little SJ, Currier JS, Forthal DN, Kemper CA, Beall GN, et al
. The value of patient-reported adherence to antiretroviral therapy in predicting virologic and immunologic response. AIDS 1999; 13:1099–1107.
5. Bangsberg DR, Hecht FM, Charlebois ED, Zolopa AR, Holodniy M, Sheiner L, et al
. Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS 2000; 14:357–366.
6. Paterson DL, Swindells S, Mohr J, Brester M, Vergis EN, Squier C, et al
. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med 2000; 133:21–30.
7. Richman DD, Morton SC, Wrin T, Hellmann N, Berry S, Shapiro MF, et al
. The prevalence of antiretroviral drug resistance in the United States. AIDS 2004; 18:1393–1401.
8. Zhang FJ, Au M, Haberer J, Zhao Y. Overview of HIV drug resistance and its implications for China. Chinese Medical Journal 2006; 119:1999–2004.
9. Sabaté E. Adherence to long-term therapies: evidence for action. Geneva: World Health Organization; 2003.
10. Pinheiro CA, de-Carvalho-Leite JC, Drachler ML, Silveira VL. Factors associated with adherence to antiretroviral therapy in HIV/AIDS patients: a cross-sectional study in Southern Brazil. Brazil J Med Biol Res 2002; 35:1173–1181.
11. Weiser S, Wolfe W, Bangsberg D, Thior I, Gilbert P, Makhema J, et al
. Barriers to antiretroviral adherence for patients living with HIV infection and AIDS in Botswana. J Acquire Immune Defic Synd 2003; 34:281–288.
12. Simoni JM, Frick PA, Pantalone DW, Turner BJ. Antiretroviral adherence interventions: a review of current literature and ongoing studies. Topics in HIV Medicine 2003; 11:185–198.
14. Chesney MA, Ickovics JR, Chambers DB, Gifford AL, Neidig J, Zwickl B, et al. ACTG Adherence baseline and follow up questionnaire. AIDS Clinical Trials Group (ACTG) Recruitment
. Adherence and Retention Subcommittee. Available at: http://www.caps.ucsf.edu/tools/surveys/#8
. Accessed: October 2004.
15. Chesney MA. Factors affecting adherence to antiretroviral therapy. Clin Infect Dis 2000; 30(Suppl 2):171–176.
16. Ammassari A, Murri R, Pezzotti P, Trotta MP, Ravasio L, De Longis P, et al
. Self-reported symptoms and medication side effects influence adherence to highly active antiretroviral therapy in persons with HIV infection. J Acquir Immune Defic Syndr 2001; 28:445–449.
17. Murphy DA, Wilson CM, Durako SJ, Muenz LR, Belzer M, Adolescent Medicine HIV/AIDS Research Network. Antiretroviral medication adherence among the REACH HIV-infected adolescent cohort in the USA. AIDS Care 2001; 13:27–40.
19. Turner BJ. Adherence to antiretroviral therapy by human immunodeficiency virus-infected patients. J Infect Dis 2002; 185(Suppl 2):S143–S151.
20. Stone VE. Strategies for optimizing adherence to highly active antiretroviral therapy: lessons from research and clinical practices. Clin Infect Dis 2001; 33:865–872.
21. Monforte AD, Lepri AC, Rezza G, Pezzotti P, Antinori A, Phillips AN, et al
. Insights into the reasons for discontinuation of the first highly active antiretroviral therapy (HAART) regimen in a cohort of antiretroviral naive patients. AIDS 2000; 14:499–507.
22. Chesney MA, Ickovics J, Hecht FM, Skipa C, Rabkin J. Adherence: a necessity for successful HIV combination therapy. AIDS 1999; 13:S1–S8.
23. Gifford AL, Bormann JE, Shively MJ, Wright BC, Richman DD, Bozzette SA. Predictors of self-reported adherence and plasma HIV concentrations in patients on multidrug antiretroviral regimens. J Acquir Immune Defic Syndr 2000; 23:386–389.
24. Wu XW, Wu GJ. Adherence measurement among senior inpatients (in Chinese). Chin Med J Commun 2003; 17:512–513.
© 2007 Lippincott Williams & Wilkins, Inc.