Clinicians have ample reason to be concerned about their patients' ability to adhere to these complex treatment regimens. One of the most consistent findings of research on compliance is that there is a direct association between poor compliance and the complexity of drug regimens, the number of specific medications that are prescribed, and the extent to which an assigned regimen interferes with daily life (1). The multidrug antiretroviral regimens that are now being prescribed, with their exceedingly strict schedules and their detailed directives regarding such factors as timing of meals and intake of fluids, are the most complicated that have ever been prescribed for continuous and open-ended treatment of such a large patient population.
To get an accurate picture of just how demanding the typical AIDS patient's regimen actually is, we must take into account the fact that most of these individuals are also taking medications to prevent or treat AIDS-related opportunistic infections, to stimulate appetite and/or libido, to ablate pain, to supplement suspected nutritional insufficiencies, and to combat depression, fatigue, and insomnia. In addition, many of our patients with late-stage HIV disease are also taking a number of over-the-counter medications, either prescribed or self-prescribed, to counteract such side effects of therapy as nausea, diarrhea, bloating, and neuropathies.
The true extent of the polypharmacy seen in advanced HIV infection cannot be appreciated unless one also takes into account the host of herbs, drugs, and other chemical compounds -- sometimes identified, sometimes unacknowledged -- that AIDS patients may be using. This list includes investigational drugs obtained overseas or through buyers' clubs in this country, alternative or complementary therapies, multivitamins and dietary supplements, illicitly obtained opiates, and marijuana.
Only when such a tally has been made is the daunting complexity of the typical patient's daily dosing regimen revealed. In this light, the high degree of compliance that is attained by so many AIDS patients is nothing short of miraculous. And given the effort that all patients on these regimens must expend, day in and day out, week in and week out, to remain compliant with their assigned therapy, we are under a particular obligation to assist them in every way we can.
Clinicians can promote compliance with antiretroviral regimens -- by devising dosing schedules that can be integrated into patients' daily routines, by encouraging compliance in every encounter with every patient, and by providing patients with tools that will help them achieve and maintain a high level of compliance (see "Starting Your Patient on an Antiretroviral Regimen," the PULL OUT AND SAVE feature in Vol. 3, No. 1, of HIV Newsline).
The fact that patients have problems adhering to antiretroviral therapy was recognized long before these patients began taking protease inhibitors -- which work effectively only when patients are highly compliant with therapy. With this class of potent antiretroviral agents, poor compliance can lead to rapid emergence of drug-resistant viral strains, strains that may prove cross-resistant to one or more drugs in the class. In such situations, the paradoxical result of multidrug therapy is to accelerate, rather than retard, disease progression.
This is the clinical conundrum of current antiretroviral therapy: When taken conscientiously, regimens that include a protease inhibitor suppress viral burden to undetectably low levels and stimulate a rebound in CD4 cells. But when compliance with these multidrug therapies is less than optimal, fluctuating serum drug levels can actually promote the development of drug resistance -- thereby restricting the clinician's therapeutic options and reducing the patient's chances of remaining clinically stable.
Studies of outpatients and of individuals participating in clinical trials indicate that more than 70% of all patients omit some drug doses (2). Preliminary indications from clinics treating patients with multidrug regimens that include protease inhibitors suggest that adherence to these multidrug regimens is actually better than the adherence seen in patients taking less complicated antiretroviral regimens. Pilot studies conducted by the San Francisco-based Center for AIDS Prevention Studies (CAPS) indicate that only 10% to 15% of patients are skipping doses on any given day.
These findings suggest that a significant percentage of treated patients appreciate the importance of strict compliance with their assigned antiretroviral regimen. Patients are trying to achieve perfect compliance, and the reasons they fall short of that mark have as much to do with the sheer complexity of the regimen as they do with fallibility, forgetfulness, and frailty of resolve.
Although these preliminary data on compliance are encouraging, they probably underestimate the extent of the problem. For one thing, these estimates are based on pilot assessments that asked patients to document their own compliance -- a method of data-gathering that is notoriously subject to reporting bias. For another thing, these pilot studies involve a carefully selected patient population, and it will be necessary to exercise caution in extrapolating these data to the general patient population. And, finally, these assessments were made relatively early in the course of care, and we know that adherence diminishes over time.
Indeed, clinicians should expect that adherence to any multidrug antiretroviral therapy will decline with time. One published account, of seven patients on protease inhibitor monotherapy, supports this assertion. Moreover, it reveals that even brief episodes of skipped therapy are associated with increases in viral load (3). Specifically, this study showed that whenever patients took short "drug holidays" (defined as three or more days of little or no medication) their plasma HIV RNA levels rose.
When the results of this study are considered in conjunction with the CAPS data, it becomes apparent that, on any given day, at least 10% of all patients on protease inhibitor therapy are at risk for viral rebound -- and therefore for developing resistance to their assigned protease inhibitor.
Scientists at the Center for AIDS Prevention Studies and the Center for AIDS Research, both associated with U.C.S.F., have been studying adherence to antiviral medications through a program known as Partnership in AIDS Clinical Trials, or PACT (4, 5). PACT investigated compliance in patients enrolled in AIDS Clinical Trials Group protocols of multidrug antiretroviral therapies. Interviews with participants in these ACTG trials -- which did not include patients taking protease inhibitors -- indicated that approximately half of these study subjects were skipping some of their doses of antiretroviral medications. In order, the four most common reasons for skipping doses were: simple forgetfulness, a change in daily routine, being too busy with other things, and being away from home.
Interestingly enough, most respondents ranked "to avoid the side effects of therapy" well down on their list of reasons for non-compliance. This might be expected in such a highly motivated cohort of patients, all of whom chose to participate in these ACTG studies. What was unexpected in this patient population was the sheer number of skipped doses. If compliance is this inconsistent among self-selected, well-motivated patients, clinicians can expect that it will be even less consistent in the general patient population.
The stress of living with HIV disease can also interfere with adherence to therapy and other health-promoting behaviors. U.C.S.F.'s PACT program has found that HIV-infected persons who report high levels of psychological distress adhere less well to their treatment regimens -- a finding that confirms the results of an earlier study (6). In both of these studies participants who were under significant stress missed doses of their antiretroviral medications more often than their less distressed counterparts.
The PACT investigators also found that patients who used alcohol and recreational drugs were more likely to be non-compliant. This correlation between non-adherence and substance abuse may in fact be a maladaptive manifestation of stress-related non-compliance: interviews with these patients indicate that alcohol and drugs are being used as a means of coping with stress.
Before patients can comply fully with their assigned drug regimen, they must fully understand it. Research conducted as part of the PACT program indicates that significant numbers of treated patients are not clear about their regimens -- even when they are questioned about that regimen immediately following the clinic visit at which the regimen was prescribed. Approximately 10% of the patients interviewed in these circumstances did not know the number of medications their physician had prescribed. An equal number -- although not always the same individuals -- did not know how often they were to take their medications.
According to the PACT investigators, 18% of these interview subjects did not know the names of their drugs, and 15% did not know when they were supposed to return to the clinic for their next visit. Of particular relevance to the issue of compliance with therapies that include a protease inhibitor, the data indicate that 22% of the respondents could not cite the special instructions they had been given about when and how to take their assigned medications.
Given that saquinavir has little or no antiretroviral activity unless it is taken with food -- and that indinavir is poorly absorbed unless it is taken on an empty stomach -- these findings are a matter of real concern. The clear conclusion is that clinicians cannot take too much time explaining how each drug in a multidrug regimen should be taken.
Healthcare professionals often assume that they can identify potentially non-compliant patients by their patient's educational level, income, gender, or personality. One survey indicated that 76% of physicians attributed non-adherence to one or more of these characteristics -- but in fact the accuracy of predictions based on these variables is no better than chance (7). In general, there are no characteristics that consistently identify non-compliant patients -- with one exception: Across all studies and all diseases, individuals who are experiencing depression or psychological distress tend to be less compliant with treatment.
Research conducted at U.C.S.F. has confirmed the importance of stress as a factor in non-compliance. In a study of patients with advanced HIV disease, investigators found a significant association between high levels of stress and/or pessimism about their prognosis and non-adherence to therapy. These researchers also found that younger patients had more difficulty remaining compliant, perhaps because they have had less experience with treatment regimens.
While there are no absolutely reliable guidelines for predicting which patients will be non-compliant with their antiretroviral therapy, there are factors that influence compliance, and these hold across a wide variety of therapeutic regimens. They include the complexity of the dosing schedule, the degree to which that schedule interferes with the patient's daily routine, and how easily and effectively the patient and his or her primary care providers communicate with one another (Table).
Becoming actively involved in their own care is a skill that people with HIV can develop. The literature contains numerous studies which demonstrate that people who have developed positive health habits are better able to adhere to new treatments -- because they have learned the skills that enable them to do so. Where living with HIV is concerned, there are specific tasks that every patient needs to master in order to achieve optimal compliance:
Counseling on compliance can be provided by a physician, a nurse or nurse practitioner, a health educator, or a medical assistant. In the PACT study, specifically trained "facilitators" who had bachelor's degrees in health sciences and some counseling experience were trained in this area, and they worked closely with patients to achieve and maintain optimal adherence to treatment regimens. Their experience has led to the development of specific strategies to assist patients in attaining and maintaining high-level compliance with antiretroviral therapy (see "Strategies to Establish and Maintain Optimal Adherence, the PULL OUT AND SAVE feature).
The assumption that HIV-positive individuals who are substance abusers have difficulty complying with therapy resulted in their exclusion from many early clinical trials (8). Furthermore, anecdotal evidence indicates that physicians who care for patients with HIV disease are not prescribing protease inhibitors to persons who are marginally housed, have a history of substance use, or are, in the physicians' judgment, unlikely to adhere to their medications.
Investigators have examined the assumption that substance users are likely to be non-adherent. Some studies, done before the advent of the protease inhibitors, do indicate that substance users have greater problems with adherence to medication than other groups (9). In a methadone maintenance program, for example, adherence to medications was found to be a problem for half of the patients with HIV disease (10). Direct dispensing of medications can significantly improve adherence in the short term, but the effects do not extend to days when patients are not seen at the clinic (11).
Not all the evidence indicates that substance abusers are unable to achieve compliance, however. A Swiss study found that substance abusers were less willing to begin antiretroviral treatment than non-users, but once on treatment these patients adhered to their assigned therapy just as well as patients in other risk groups (12).
Since none of these studies examined adherence to regimens that included a protease inhibitor, further study is needed to develop and test effective programs for this patient population.
Compliance with therapy has always been an issue in HIV treatment -- and with the adoption of multidrug regimens that include protease inhibitors as the standard of care for patients with advanced HIV disease, compliance has moved front and center. Research is needed to identify specific factors that are associated with non-adherence to these regimens and to develop specific strategies to address these factors. However, we already know that healthcare professionals can work closely with their patients, as "partners in care," to achieve high levels of compliance, prevent the development of resistance, and obtain optimal benefits from these new therapies -- thereby slowing the progression of HIV disease and improving quality of life.
1. Dunbar-Jacob J, Burke LE, and Puczynski S. Clinical assessment and management of adherence to medical regimens. In: Nicassio PM and Smith TW, eds., Managing Chronic Illness (Washington, D.C.: American Psychological Association, 1995), 313-50.
2. Chow R, Chin T, Fong IW and Bendayan R. Medication use patterns in HIV-positive patients. Canad J of Hosp Pharm 1993, 46: 171-5.
3. Vanhove GF, Schapiro JM, Winters MA, Merigan TC, and Blaschke TF. Patient compliance and drug failure in protease inhibitor monotherapy (letter). JAMA 1996; 276: 1955-6.
4. Chesney MA and Folkman S. Psychological impact of HIV disease and implications for intervention. In: Zegans L and Coates T, eds., Psychiatric manifestations of HIV disease. Psychiatric Clinics of North America 17 1994;1: 163-82.
5. Chesney MA and Folkman S. The psychosocial management of HIV disease. In: Holmes K, Sparling P, Mardh P, Lemon S, Stamm W, Piot P, and Wasserheit J, eds., Sexually Transmitted Diseases, 3rd ed., (New York: McGraw Hill, in press).
6. Singh M, Squier C, Sivek C, Wagener M, Hong Nguyen M, and Yu VL. Determinants of compliance with antiretroviral therapy in patients with human immunodeficiency virus: Prospective assessment with implications for enhancing compliance. Intl J AIDS Care 1996; 8: 261-9.
7. Meichenbaum D, Turk C. Facilitating Treatment Adherence: A Practitioner's Guidebook (New York: Plenum, 1987).
8. Hughes JR. Exclusion of "noncompliant" individuals from clinical trials. Controlled Clin Trials 1993; 14: 176-7.
9. Samet JH, Libman H, Steger KA, Dhawan RK, Chen H, Shevitz AH, et al., Compliance with zidovudine therapy in patients infected with human immunodeficiency virus, type I: A cross-sectional study in a municipal hospital clinic. Am J Med 1992; 92: 495-502.
10. Wall TL, Sorensen JL, London J, Delucchi KL, Ferrando SJ, Abbott C, Morris P. and Batki SL. Adherence to zidovudine (AZT) in HIV-infected injection drug users. In: Harris LS, ed., Problems of Drug Dependence 1992: Proceedings of the 54th Annual Scientific Meeting, The College on Problems of Drug Dependence (Washington, D.C.: NIDA Research Monograph Series No. 132, U.S. Government Printing Office, 1993), 292.
11. Wall TL, Sorensen JL, Batki SL, Delucchi KL, London JA, and Chesney MA. Adherence to zidovudine (AZT) among HIV-infected methadone patients: A pilot study of supervised therapy and dispensing compared to usual care. Drug and Alcohol Dependence 1995; 37: 261-9.
12. Broers B, Morabia A, and Hirschel B. A cohort study of drug users' compliance with zidovudine treatment. Arch Int Med 1994; 154: 1121-7.
Margaret A. Chesney, Ph.D., is Professor of Medicine, U.C.S.F. School of Medicine, and Director of Behavioral Medicine Core, Center for AIDS Research, San Francisco, CA.