In countries where highly active antiretroviral therapy (HAART) is widely available, this treatment has transformed outcomes in HIV;1-3 however, its success is conditional on patients accepting a timely treatment recommendation.4 Once treatment has been initiated, high rates of adherence are required to ensure efficacy and prevent the emergence of viral resistance.5 The importance of understanding patients' perceptions and likely reactions to a treatment offer is illustrated by the fact that up to a third of people offered HAART in accordance with contemporaneous treatment guidelines refuse.6 High levels of adherence are often hard to achieve, with many studies using objective adherence measures reporting an average of around 70% of prescribed doses taken.7-12 Although several studies have attempted to enhance treatment adherence in HIV, the interventions have achieved limited success.13 A better understanding of the factors influencing the decision to accept or reject an evidence-based treatment recommendation, and of those underpinning adherence, is essential to develop patient-centered methods for facilitating informed choice and optimizing clinical outcomes.
A number of interview-based studies exploring patients' reasons for refusing HAART have identified the potential importance of patients' beliefs about antiretroviral treatment. Patients reported a number of negative perceptions about HAART, including fears about side effects, concerns about the need for strict adherence, inconvenience and practical problems associated with the regimen, distrust of conventional medicines, fear of long-term damage to body organs, and the perception that there is no reason to start in the absence of symptoms.14-22 Similar beliefs have emerged in studies of adherence, where nonadherence was linked to doubts about treatment efficacy, concerns about side effects and long-term toxicities, scheduling demands and personal capacity to adhere, concerns about the impact of HAART on self-identity, and the possibility that taking treatment might lead to disclosure of the individual's HIV status.23-34
It is clear that many patients have a more negative attitude to HAART than might be assumed from consideration of clinical indicators alone. To develop pragmatic interventions to facilitate informed choice and optimize adherence, however, we need a suitable theoretic framework for summarizing key beliefs and identifying patients who would benefit most from interventions. Indeed, one reason why previous interventions to facilitate adherence have met with only limited success is that they have failed to utilize theory-based methods for identifying and addressing the main perceptual barriers (eg, beliefs, attitudes).35
Research in a variety of chronic illnesses suggests that the salient beliefs relating to patients' medication decisions can be grouped under 2 categories: perceptions of necessity or personal need for treatment and concerns about potential adverse effects.36-38 Using a validated questionnaire to quantify patients' necessity beliefs and concerns,39 this “necessity-concerns” framework has been found to predict adherence to medication across a range of chronic illnesses, including asthma, renal disease, heart disease, cancer, and cardiac failure.39,40 A recent cross-sectional study also suggested the utility of this framework in explaining nonadherence to HAART.41 This model offers a convenient framework for clinicians to elicit and address key beliefs underpinning patients' attitudes and decisions about treatment.
To our knowledge this is the first prospective longitudinal study assessing uptake and adherence to HAART. This study was designed to explore the utility of the necessity-concerns framework in predicting patients' decisions to accept or decline HAART after a clinically indicated treatment recommendation and in explaining subsequent adherence among those who accepted the treatment recommendation and initiated treatment.
This was a prospective longitudinal study of uptake and adherence to HAART among HIV-positive individuals attending HIV clinics in Brighton, United Kingdom. Data were collected between January 2000 and May 2004.
Adults attending an HIV outpatient clinic in Brighton were eligible for the study if they were not currently taking antiretroviral medication. Participants were followed up over time, and those who subsequently received a clinically indicated offer of HAART formed the study sample. Exclusion criteria included having insufficient understanding of English or being too ill to complete the questionnaires.
Consecutive patients meeting the inclusion criteria were referred to a research assistant, based at the clinic, by their HIV physician. Standard procedures for consent, as set out by the research ethics committee, were followed. Treatment recommendations were tracked through weekly clinical meetings, where patients eligible to receive HAART, on the basis of contemporaneous treatment guidelines,42,43 were routinely identified. A standardized questionnaire was given to participants, at the clinic, immediately after the appointment at which the treatment offer was made. Medical files and pharmacy records were checked to identify those who initiated treatment and those who deferred or declined. Participants who initially accepted the treatment offer and subsequently started treatment formed the “acceptors” group (n = 98). Those who declined or deferred HAART formed the “decliners” group (n = 38).
Acceptors and decliners were followed up over a year. Those who initially declined a treatment recommendation but subsequently accepted completed a “baseline” questionnaire before they initiated treatment. These participants were also followed up over their first year of treatment (Fig. 1). Adherence was assessed after 1 (1M), 3 (3M), 6 (6M), and 12 (12M) months of treatment. Participants were telephoned in advance of the follow-up assessment, and a questionnaire was sent out with a stamped addressed envelope (SAE). Telephone reminders were given every 2 days where applicable.
Clinical and demographic information was obtained from medical records at baseline and at the 12M assessment. This information included most recent CD4 lymphocyte cell counts, HIV viral load, symptomatic status (asymptomatic HIV, symptomatic HIV, and AIDS), number of months since first diagnosis of HIV, details of previous antiretroviral treatments, and HIV acquisition risk.
Perceptions of HAART were assessed using the Beliefs about Medicines Questionnaire-HAART-specific version (BMQ-HAART),39 which has previously been used with people living with HIV.40 The BMQ-HAART comprises 2 scales: a HAART-necessity scale and a HAART-concerns scale. The HAART-necessity scale consists of 6 items assessing patients' beliefs about their personal need for HAART for controlling their HIV, maintaining their health, and preventing illness, whereas the HAART-concerns scale consists of 7 items that bring together a range of separate concerns about the potential adverse effects of HAART that have been identified across studies, including fears about short- and long-term side effects, concerns about the timing of pills and the disruptive effects of the HAART regimen on daily life, more abstract concerns about becoming too dependent on HAART, feeling that HAART is a “mystery,” and a general feeling of worry about taking the treatment.22,41 Participants were presented with a series of statements about which they were told, “These are statements that other people have made about combination therapy.” They were then asked to rate their level of agreement with each item on a scale, where responses ranged from strongly agree (scored 5) to strongly disagree (scored 1). Scores for the individual items within each scale were summed to give a total scale score. To facilitate comparison of scores between scales, a mean scale score was computed by dividing each scale by the number of items, giving a range of 1 to 5 for necessity and concerns scales.
The Hospital Anxiety and Depression Scale44 was used to measure depression. This measure was designed to detect the presence and severity of depression among patients attending outpatient clinics without the possible contamination of scores by reporting of physical symptoms. Items were scored from 0 to 3, with possible total scores on each scale ranging from 0 to 21. Higher scores indicate greater depression. This scale has good psychometric properties in studies of medical outpatients.44
Adherence was measured using a visual analog scale (VAS) from the Medication Adherence Self-Report Inventory (MASRI),45 where participants were asked to estimate the percentage of HAART medicines they had taken as prescribed over the previous month. The MASRI is worded in a nonjudgmental manner in an effort to minimize socially desirable responses. The VAS has demonstrated good validity against objective measures (electronic monitoring: r = 0.63, P < 0.001; pill count: r = 0.75, P < 0.001; and viral load: P < 0.01).45
Participants were dichotomized into “low-adherence” and “high-adherence” groups on the basis of whether their average adherence score on the MASRI was 95% or greater (high adherence) or less than 95% (low adherence). This adherence categorization was in accordance with the findings of contemporaneous clinical research, which suggested that at least 95% adherence to HAART is required for clinical efficacy.5 Participants who made a unilateral decision to stop their treatment and those who were lost to follow-up were also included in the low adherence group.
Data were analyzed using SPSS 12.0 (SPSS, Chicago, IL). Continuous variables were examined for normal distribution using sample data (Q-Q plots) and estimates of population normality (KS-Lilliefors test). A log transformation was used to establish a linear scale of measurement for viral load. Independent sample t tests were used to investigate differences in beliefs between those accepting and those declining a treatment offer and between those reporting high and low adherence. To test whether the effect of beliefs about HAART on uptake and adherence were independent of demographic and clinical variables, a hierarchic logistic regression was performed. Clinical and demographic predictors identified in the bivariate analyses were entered first, and BMQ-necessity and -concerns scores were entered second. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported.
Four hundred ninety-one patients were eligible for the study, of whom 365 (74.3%) were referred by their physician. Three hundred twenty-two (88.2%) of those referred agreed to take part. After recruitment, 153 participants were recommended HAART, of whom 136 (88.9%) returned completed questionnaires. Those who did not complete questionnaires did not differ in terms of demographics (age or gender) or clinical variables (CD4 cell count viral load, symptomatic status, time since diagnosis, HIV transmission risk, or prior antiretroviral use). The sociodemographic and clinical characteristics of the participants are shown in Table 1. These were representative of the overall clinic population not receiving treatment.
Beliefs About HAART
Necessity and concerns scales were normally distributed with a fairly wide distribution of scores, indicating considerable variation in beliefs about HAART among the study participants. After the treatment recommendation, necessity scores ranged from 1.3 to 5.0 (mean = 3.7, SD = 0.7). Concerns scores ranged from 1.9 to 5.0 (mean = 3.2, SD = 0.7). The 3 most prevalent concerns about HAART arose from beliefs about the potential long-term effects (67%), the possibility of side effects (53%), and perceptions that using HAART would be disruptive to daily life (50%).
Predicting Uptake of HAART
Of the 136 participants, 98 (72.1%) accepted the treatment offer and 38 (28%) declined. Participants with a lower CD4 cell count (P = 0.000) and higher viral load (P = 0.001) were more likely to accept HAART, whereas those who had been diagnosed for longer were more likely to decline (P = 0.000). Uptake of HAART was not significantly related to age, symptomatic status, previous antiretroviral experience, transmission risk, or depression (all P > 0.05) (Table 2).
The decision to decline the offer of HAART was predicted by low scores on the BMQ-necessity scale (P = 0.000) and high scores on the BMQ-concerns scale (P = 0.000). The effect of beliefs about HAART on uptake was independent of clinical variables (CD4 cell count, viral load, and time since diagnosis). ORs and significance levels are shown in Table 3.
Predicting Adherence to HAART
In addition to the 98 participants who accepted the original treatment recommendation, a further 22 who initially declined subsequently accepted a treatment offer over the year of follow-up. One hundred twenty patients therefore formed the sample for the adherence study. Of those, 3 died before the 12M follow-up. Their data were removed from the analyses, resulting in a final sample of 117. The sample was composed of predominantly gay men (105 [89.7%]), and most described their ethnic origin as white, United Kingdom (86 [73%]). The age of the participants ranged from 22 to 71 (mean = 37.8, SD = 8.4) years. Thirty-seven participants (31.6%) were asymptomatic, 50 (42.7%) were symptomatic, and 30 (25.7%) had a diagnosis of AIDS. The mean CD4 count was 198.3 (SD = 129.0) cells/mm3.
Validity of Adherence Classification
Of those reporting low adherence, 32 (61.5%) of 52 had an undetectable viral load at 12M compared with 57 (87.7%) of 65 of those reporting high adherence (χ2 = 10.9, df = 1; P = 0.001).
Changes in Adherence Over Time
The number of participants reporting low adherence increased from 25 (21.4%) at 1M to 52 (44.4%) at 12M (Cochran Q = 38.9, df = 3; P < 0.001).
Bivariate Analyses: Baseline Predictors of Adherence at 12 Months
Low adherence at 12M was predicted by baseline beliefs: doubts about personal necessity for HAART (P = 0.006) and concerns about potential adverse effects (P = 0.005). Younger age (P = 0.000) and previous prescription of antiretroviral therapy (P = 0.000) were also associated with low adherence. Adherence was not associated with transmission risk, diagnosis of AIDS, CD4 cell count, viral load, the number of medicines in the antiretroviral regimen, or depression (all P > 0.05) (Table 4).
Multivariate Analyses: Baseline Predictors of Adherence at 12 Months
Associations between adherence and beliefs about HAART remained significant after controlling for age and previous treatment experience. ORs and significance levels are shown in Table 5.
In common with previous studies,6,14 a quarter of participants declined HAART after a clinically recommended treatment offer, and more than a third met our criteria for low adherence after 12 months of treatment. Our findings show that although the refusal of a treatment offer may be at odds with the clinical evidence, it is often a logical and understandable response to the patient's preexisting beliefs about HAART. Although the “clinical necessity” for HAART was confirmed by a treatment offer made according to evidence-based guidelines, patients varied in the degree to which they accepted their personal need for treatment at the time of the treatment offer. Moreover, most people had strong concerns about the potential adverse effects of HAART, ranging from fears of experiencing side effects to more abstract concerns about long-term effects, dependence, and disruption to daily life. Other widely held concerns, such as worry about possible adverse effects of HAART on sexual functioning,46 may also be important in this respect. Some of these concerns may be addressed by newer antiretroviral drugs, which seem to have fewer side effects and long-term toxicities.47
Our findings indicate that patients' perceptions of the necessity for HAART may not be consistent with the clinical imperative for treatment. We therefore need a better understanding of why patients remain unconvinced about their personal need for HAART. Research in other illnesses has identified the importance of patients' common-sense understanding of their illness in this respect and the degree to which the treatment offer is consistent with their underlying beliefs about the illness. Common-sense beliefs about medicines are strongly influenced by subjective experiences of the illness, including symptoms.38 Interviews with the same subjects reported elsewhere22 revealed that receiving “abstract” information about CD4 cell count and viral load laboratory results was often less persuasive than more “concrete” symptom experiences. A common reason given by interviewees for refusing HAART was that they were experiencing few, if any, of the symptoms that they associated with HIV/AIDS. Their experiences (of feeling fine) did not match their common-sense beliefs about when HAART would be necessary.
Perceived necessity for HAART was initially high among those accepting HAART. Variation in necessity at this early stage predicted high adherence 12M later, however, whereas stronger concerns predicted low adherence. These findings highlight the importance of patients' views about HAART, even before they begin treatment, in determining whether adherence is maintained over the long term.
In common with previous research,27 this study indicated the importance of previous treatment experiences for future adherence to treatment, with previously treatment-naive individuals reporting higher adherence. Although we found that having been prescribed HAART in the past was not related to uptake, we did not collect information regarding reasons for stopping previous treatments, such as side effects, problems with the medication regimen, medical advice, or resistance, which may have influenced patients' subsequent treatment decisions.
The study has several limitations. Additional variables that may predict uptake and adherence, such as the quality of the patient-health care provider relationship,18,22 were not explored in this study. The recruitment procedure, where patients were referred through their physician, may have introduced selection bias. The sample consisted predominantly of white gay men. Other studies suggest that women48 and people of African origin49,50 might be more skeptical of HAART for various reasons (eg, fears of disclosure, concerns regarding the safety of antiretroviral therapy in pregnancy, the fact that clinical tests of the drugs were carried out largely on white men). The study would require replication with a more representative sample to be applicable to the wider population. Nonetheless, the clinical characteristics among patients initiating HAART are consistent with those reported in clinical trials and cohort studies.1,51
It is possible that administering the study questionnaires immediately after a treatment recommendation may have acted as an intervention, perhaps influencing the patient in his or her decision about whether or not to accept the treatment offer or, indeed, subsequent adherence. We believe this to be unlikely, however, because the questionnaire contains positive and negative items, and therefore is unlikely to have influenced participants' views in a single direction. Furthermore, necessity and concerns have previously been associated with adherence to HAART in a cross-sectional study, where adherence and beliefs were measured at the same time,41 and across a range of other illnesses36,40 with no evidence of influencing opinions.
A further limitation of this research was our reliance on self-report questionnaires. To minimize the possibility that individuals would give socially desirable responses, emphasis was placed on the fact that the individuals' responses were confidential and questionnaires were carefully worded in a nonjudgmental manner. Self-report measures have advantages over other methods of measuring adherence, including high face validity and high specificity for nonadherence; however, they are subject to self-presentation and recall bias. A recent review found moderate to high concordance between questionnaire measures of adherence and non-self-report methods (eg, tablet count, electronic monitors, plasma drug concentrations).52 We accept that our rating of adherence may underestimate the true extent of nonadherence (some people may be giving socially desirable responses, reporting higher adherence rates than they actually attain in practice). The inherent bias within the self-report measure does not diminish our confidence in the finding that beliefs were related to adherence, however. This is because the direction of bias inherent in self-report measures of adherence mitigates against us finding a relation between our categoric adherence measure (high vs. low) and our medication belief scales. Given the direction of bias, we are confident that the observed relations between beliefs and adherence are real rather than an artifact arising from the limitations of self-report and that these relations would be magnified with the use of more objective measures of adherence.
Although our adherence categorization was significantly associated with virologic outcome, with 87.7% of those reporting high adherence recording a viral load of <50 copies/mL, 61.5% of those categorized into the low adherence group also had an undetectable viral load. In setting our cutoff point for high adherence, we adopted the convention of 95%, current at the time the study was conducted.5 More recent studies suggest that viral suppression may be achieved at lower rates of adherence to boosted protease inhibitor53 and nonnucleoside reverse transcriptase inhibitor (NNRTI)-based regimens.54
Despite these limitations, the data demonstrate the utility of the necessity-concerns framework. Barriers to acceptance of antiretroviral treatment that have been elicited in previous studies,14-22 and those associated with nonadherence23-34 can be conceptualized within this framework. Furthermore, this study dealt with methodologic issues that had not been addressed by previous studies into HAART uptake, showing that the relation between beliefs, uptake, and adherence was independent of clinical variables and depression. Moreover, beliefs were assessed immediately after a treatment recommendation rather than retrospectively, indicating that necessity and concerns predicted subsequent decisions rather than being the result of post hoc rationalization.
The necessity-concerns framework may be used to inform interventions facilitating informed choice and supporting optimal adherence to HAART. Patients' perceptions of necessity and concerns about HAART should be elicited and addressed after a clinically indicated treatment recommendation. Interventions to support informed decision making and adherence should ensure that all patients have an accurate description of the medical model of HIV, including the ways in which the CD4 cell count is used as a marker of disease progression and the action of anti-HIV medication. Clinicians and researchers should also be sensitive to the fact that many patients have experienced earlier treatments for HIV that were later found to be ineffective or have seen others experience problems with antiretroviral treatment and, as a result, may be suspicious of current medical advice.22 In the current study, having previously stopped therapy was associated with low adherence. Many patients stop therapy because of side effects,55 indicating that adherence in this group may be enhanced by more proactive management of side effects.
This study has identified a useful theoretic framework explaining patients' decisions about HAART uptake and adherence and paves the way for the design of evidence-based interventions to support patients who are faced with difficult treatment decisions. Ensuring appropriate uptake of and adherence to HAART should facilitate optimal therapeutic benefits of treatment.
The authors thank David Claxton, Cara Davis, Natalie Lambert, and Jenny Robbins for their help with data collection; the staff at the Lawson Unit and Elton John Centre for all their help with recruitment; and the staff at the Sussex Eye Hospital Pharmacy for access to pharmacy data. They are especially grateful to the participants who took part in the study.
1. Mocroft A, Ledergerber B, Katlama C, et al. Decline in the AIDS and death rates in the EuroSIDA study: an observational study. Lancet
2. Palella FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med
3. Sterne JAC, Hernan MA, Ledergerber B, et al. Long-term effectiveness of potent antiretroviral therapy in preventing AIDS and death: a prospective cohort study. Lancet
4. Phillips A, CASCADE Collaboration. Short-term risk of AIDS according to current CD4 cell count and viral load in antiretroviral drug-naive individuals and those treated in the monotherapy era. AIDS
5. Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med
6. Maisels L, Steinberg J, Tobias C. An investigation of why eligible patients do not receive HAART. AIDS Patient Care STDS
7. Golin CE, Liu H, Hays RD, et al. A prospective study of predictors of adherence to combination antiretroviral medication. J Gen Intern Med
8. 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
9. McNabb J, Ross JW, Abriola K, et al. Adherence to highly active antiretroviral therapy predicts virologic outcome at an inner-city human immunodeficiency virus clinic. Clin Infect Dis
10. Arnsten JH, Demas PA, Farzadegan H, et al. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: comparison of self-report and electronic monitoring. Clin Infect Dis
11. Liu H, Golin CE, Miller LG, et al. A comparison study of multiple measures of adherence to HIV protease inhibitors. Ann Intern Med
12. Gross R, Bilker WB, Friedman HM, et al. Effect of adherence to newly initiated antiretroviral therapy on plasma viral load. AIDS
13. Gazzard B, BHIVA Writing Committee. British HIV Association (BHIVA) guidelines for the treatment of HIV-infected adults with antiretroviral therapy. HIV Med
. 2005;6 (Suppl 2):1-61.
14. Catt S, Stygall J, Catalan J. Acceptance of zidovudine (AZT) in early HIV disease: the role of health beliefs. AIDS Care
15. Ezzy EM, Bartos MR, de Visser RO, et al. Antiretroviral uptake in Australia: medical, attitudinal and cultural correlates. Int J STD AIDS
16. Gold RS, Hinchy J, Bartouney CG. The reasoning behind decisions not to take up antiretroviral therapy in Australians infected with HIV. Int J STD AIDS
17. Gold RS, Ridge DT.“I will start treatment when I think the time is right:” HIV-positive gay men talk about their decision not to access antiretroviral therapy. AIDS Care
18. Misener TR, Sowell RL. HIV-infected women's decisions to take antiretrovirals. West J Nurs Res
19. Siegel K, Gorey E. HIV-infected women: barriers to AZT use. Soc Sci Med
20. Siegel K, Raveis VH, Krauss BJ. Factors associated with urban gay men's treatment initiation decisions for HIV infection. AIDS Educ Prev
21. Stall R, Hoff C, Coates TJ, et al. Decisions to get tested and to accept antiretroviral therapies among gay/bisexual men: implications for secondary prevention efforts. J Acquir Immune Defic Syndr Hum Retrovirol
22. Cooper V, Buick D, Horne R, et al. Perceptions of HAART among gay men who declined a treatment offer: preliminary results from an interview-based study. AIDS Care
23. Altice FL, Mostashari F, Friedland GH. Trust and the acceptance of and adherence to antiretroviral therapy. J Acquir Immune Defic Syndr
24. Aversa SL, Kimberlin C. Psychosocial aspects of antiretroviral medication use among HIV patients. Patient Educ Couns
25. Catz SL, Kelly JA, Bogart LM, et al. Patterns, correlates and barriers to medication adherence among persons prescribed new treatments for HIV disease. Health Psychol
26. Gebo KA, Keruly J, Moore RD. Association of social stress, illicit drug use, and health beliefs with nonadherence to antiretroviral therapy. J Gen Intern Med
27. Duran S, Spire B, Raffi F, et al. Self-reported symptoms after initiation of a protease inhibitor in HIV-infected patients and their impact on adherence to HAART. HIV Clin Trials
28. Spire B, Duran S, Souville M, et al. Adherence to highly active antiretroviral therapies (HAART) in HIV-infected patients: from a predictive to a dynamic approach. Soc Sci Med
29. Chesney M, Ickovics JR, Chambers DB, et al. Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: the AACTG Adherence Instruments. AIDS Care
30. Eldred LJ, Wu AW, Chaisson RE, et al. Adherence to antiretroviral and Pneumocystis
prophylaxis in HIV disease. J Acquir Immune Defic Syndr
31. Gifford AL, Bormann JE, Shively MJ, et al. Predictors of self-reported adherence and plasma HIV concentrations in patients on multidrug antiretroviral regimens. J Acquir Immune Defic Syndr
32. Molassiotis A, Nahas-Lopez V, Chung WY, et al. Factors associated with adherence to antiretroviral medication in HIV-infected patients. Int J STD AIDS
33. Roberts KJ, Mann T. Barriers to antiretroviral medication adherence in HIV-infected women. AIDS Care
34. Sankar A, Luborsky M, Schuman P, et al. Adherence discourse among African-American women taking HAART. AIDS Care
35. Haynes R, Yao X, Degani A, et al. Interventions to enhance medication adherence. Cochrane Database Syst Rev
36. Horne R, Weinman J. Patients' beliefs about prescribed medicines and their role in adherence to treatment in chronic physical illness. J Psychosom Res
37. Horne R. Representations of medication and treatment: advances in theory and measurement. In: Petrie K, Weinman JA, eds. Perceptions of Health and Illness
. Amsterdam, The Netherlands: Harwood Academic Publishers; 1997:155-188.
38. Horne R. Treatment perceptions and self regulation. In: Cameron LD, Leventhal H, eds. The Self-Regulation of Health and Illness Behaviour
. London: Routledge; 2003:138-154.
39. Horne R, Weinman J, Hankins M. The Beliefs about Medicines Questionnaire: the development and evaluation of a new method for assessing the cognitive representation of medication. Psychol Health
40. Horne R, Weinman J. Self-regulation and self-management in asthma: exploring the role of illness perceptions and treatment beliefs in explaining non-adherence to preventer medication. Psychol Health
41. Horne R, Buick D, Fisher M, et al. Doubts about necessity and concerns about adverse effects: identifying the types of beliefs that are associated with non-adherence to HAART. Int J STD AIDS
42. BHIVA Writing Committee, BHIVA Executive Committee. British HIV Association (BHIVA) guidelines for the treatment of HIV-infected adults with antiretroviral therapy. HIV Med
43. BHIVA Writing Committee, BHIVA Executive Committee. British HIV Association (BHIVA) guidelines for the treatment of HIV-infected adults with antiretroviral therapy. HIV Med
44. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand
45. Walsh JC, Mandalia S, Gazzard BG. Responses to a one month self report on adherence to antiretroviral therapy are consistent with electronic data and virological treatment outcome. AIDS
46. Gellaitry G, Cooper V, Peters K, et al. The perceived impact of HAART on gay men's sexual relationships: an interview based study. Presented at: Conference on AIDS Impact; 2005; Cape Town.
47. Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naïve patients: a 3-year randomized trial. JAMA
48. McDonald K, Bartos M, Rosenthal D. Australian women living with HIV/AIDS are more sceptical than men about antiretroviral treatment. AIDS Care
49. Erwin J, Peters B. Treatment issues for HIV+ Africans in London. Soc Sci Med
50. Siegel K, Karus D, Scrimshaw EW. Racial differences in attitudes toward protease inhibitors among older HIV-infected men. AIDS Care
51. Sabin CA, Smith CJ, Gumley H, et al. Late presenters in the era of highly active antiretroviral therapy: uptake of and responses to antiretroviral therapy. AIDS
52. Garber MC, Nau DP, Erickson SR, et al. The concordance of self-report with other measures of medication adherence: a summary of the literature. Med Care
53. Gross R, Yip B, Wood E, et al. Boosted PI are more forgiving of suboptimal adherence than non-boosted PI or NNRTI. Presented at: 13th Conference on Retroviruses and Opportunistic Infections; 2006; Denver.
54. Bangsberg DR. Less than 95% adherence to nonnucleoside reverse-transcriptase inhibitor therapy can lead to viral suppression. Clin Infect Dis
55. Mocroft A, Phillips AN, Soriano V, et al, for the EuroSIDA Study Group. Reasons for stopping antiretrovirals used in an initial highly active antiretroviral regimen: increased incidence of stopping due to toxicity or patient/physician choice in patients with hepatitis C coinfection. AIDS Res Hum Retroviruses