Use of highly active antiretroviral therapy (HAART) has resulted in clearly decreased morbidity and mortality for persons living with HIV/AIDS. For these reasons everyone who meets the criteria for treatment with these regimens must have access to them. Once an individual has access to HAART, it is also critically important that she or he adhere closely to the regimen in order to maximally benefit.

Unfortunately, women and others living with HIV/AIDS from historically disadvantaged communities may be denied access to anti-HIV medication because their providers may assume that they will not or can not be adherent to HAART. A March 1997 New York Times article that discussed the complexities of HAART regimens first brought this problem to public attention. The authors interviewed several providers in the New York City area who stated that they were less likely to prescribe HAART regimens for women, particularly if they were poor, belonged to a racial minority, or had a history of drug use.

Although three years have passed since that article was published, women continue to be less likely than men to receive treatment with HAART.

Evidence proving that women and minorities are less likely to receive antiretroviral medications to treat HIV/AIDS when indicated goes back a decade. In a national study of persons being treated for HIV/AIDS in 1991, Michael D. Stein, MD, of Brown University School of Medicine and colleagues found that women, non-Whites, and injection drug users were half as likely as others to be offered antiretroviral therapy (in this case, AZT [Retrovir] monotherapy).

More recent studies document that this problem has continued. The most definitive data come from a study called the HIV Cost and Service Utilization Study (HCSUS), which used a national probability sampling technique and surveyed nearly 3,000 persons who were representative of all those in care for HIV/AIDS in the U.S. They found that 23% of those surveyed were women. Women receiving care for HIV were twice as likely to be Black, to have an annual income of less than $5,000 per year, and to lack private health insurance than were men in care for HIV.

When the one-year HCSUS study began in December 1996, women were significantly less likely to be on HAART than men (49% vs. 61%). Considerable improvements occurred in the percentages of all persons on HAART over the course of the year. Yet although access to HAART had improved for women by study end, the gap between women and men on HAART had remained essentially unchanged (78% vs. 87%).

In a smaller study that I led, which examined access to HAART among persons in five HIV care sites in Boston, MA, and Providence, RI, in 1997, we found that 73% of men were on HAART when indicated, compared with only 50% of women.

No one knows what proportion of these access differentials are due to provider decisions versus the preferences of the individual being treated. Furthermore, if any amount of access differentials are due to provider decisions, it is not known what proportion of these decisions are based on unfounded assumptions or biases about HIV-infected women and what proportion are based on more defensible reasons (for example, concerns about liver toxicity in a woman with HIV and severe hepatitis C [HCV] coinfection). In one recent national survey of HIV providers, 30% reported that they think women with children will have poorer adherence than others with HIV/AIDS.

In addition to providers' assumptions regarding adherence to HAART, many other factors contribute to the disproportionately high number of access problems that affect women with HIV/AIDS (see sidebar). Women with HIV/AIDS are more often impoverished and likely to be racial minorities. As a result, they may experience access problems due to language differences, competing life stresses, racism, or being uninsured or underinsured.

No matter what their causes, these differentials in treatment for women are critically and vitally important. Clear evidence shows that the decreased deaths and rates of opportunistic infections (OIs) due to AIDS are very much associated with the use of HAART. Yet while overall deaths due to AIDS have decreased dramatically in the past four years, the death rate for women has decreased less than that for men. Could the relatively high ongoing death rate for women with HIV be due to problems related to HAART access?

Certainly impaired access to HAART is likely to be a major contributing factor. Providers must move beyond assumptions about women and their adherence to HAART; instead, providers must proactively prescribe HAART and then assist women in their efforts to adhere.

What Is Known about Women and Adherence to HAART?

Since 1997, many HIV/AIDS providers have been forced to relearn what has been known for decades regarding tuberculosis therapy: that they [providers] are ineffective at predicting their own patients' adherence to a medication regimen. In addition, a tremendous amount of research has been conducted since 1997 regarding adherence to HAART, so that a great deal more is known today than in 1997.

Numerous studies have found that women are just as likely as men to adhere to potent HAART regimens. For example, in my study in Boston and Providence, we found that women were statistically no more likely to have skipped a dose in the last day than men (23.1% vs. 20.9%). The same was true for the relative percentages of each group when it came to skipping a dose in the past three days, taking unstructured "drug holidays," and overall adherence.

In this study (as in nearly all other studies of adherence to HAART), however, it became clear that predictors of suboptimal adherence did exist. Regardless of gender, individuals who had 1) active alcohol abuse or dependence, 2) active substance abuse or dependence, 3) depression or other major mental health problems, or 4) low literacy or lower educational attainment were likely to have difficulty adhering to their HAART regimens. Of particular importance, as with gender, none of the other key personal characteristics (such as age and race/ethnicity) were found to be predictive of an individual's adherence to HAART.

How Much Adherence Is Enough?

The bottom line is that adherence is difficult for everyone living with HIV/AIDS, not just certain people. It is well documented that excellent adherence is critically important to ensure that HAART therapy is successful. Suboptimal adherence is known to result in incomplete viral suppression (i.e., very low levels of detectable virus), emergence of drug-resistant virus, ultimate "failure" of the current HAART regimen (resulting in rapidly escalating HIV RNA levels), and the restriction of future therapeutic options. Most people taking HAART and their providers know this, but how much adherence is necessary to ensure that these problems do not occur, and that the person being treated will succeed on the HAART regimen?

Information presented by D. Paterson of the Veterans Administration (VA) Medical Center of Pittsburgh, PA, and colleagues at the 6th Conference on Retroviruses and Opportunistic Infections (CROI) held in Chicago in 1999 suggests that greater than 95% adherence is essential. In their study of an 84-person cohort taking anti-HIV therapy, these researchers found that of those individuals with greater than 95% adherence, 81% had undetectable viral loads (limit of detection 400 copies/mL). When adherence was lower, the percentage that achieved an undetectable viral load was significantly lower. For example, among persons with 80-90% adherence, only 50% achieved an undetectable viral load.

These required levels of adherence are much higher than what many people on therapy are achieving. Of the group of persons on a protease inhibitor (PI)-containing HAART regimen in our Boston/Providence study, 81% had missed a dose of their PI since starting the regimen. Moreover, 21% had missed a dose in the last day, and 34% had missed a dose in the last three days. The most commonly reported reasons for missing a dose were forgetting the dose (47%), feeling too sick to take the dose (17%), being too busy (8%), and not having the medication at dosing time (6%).

What Can Be Done to Maximize Adherence?

Given the need for near-perfect adherence to HAART, devising a strategy for maximizing adherence is necessary. The best strategy entails certain responsibilities for the provider as well as certain responsibilities for the person taking the HAART regimen. Overall, optimizing adherence to HAART has two key components: 1) choosing a regimen that is as simple as possible while incorporating the needed potency, and 2) providing intensive, individualized education and logistical support to optimize the likelihood of maximal adherence.

Choosing an individual regimen should result from a collaboration between the provider and the person who will take the regimen. Understandably, though, the provider often has the most input in the choice because he or she will likely have more information about the range of available medication options.

It is important, however, for the person living with HIV to consider what her preferences are, what side effects she would find most bothersome, and how often she would be willing to take pills. Medical requirements sometimes may be stronger than preferences; in this case, the person who takes the therapy must be particularly sure to make the provider aware of any specific issues she may feel strongly about.

This information will help the individual and the provider to choose a regimen that makes sense for the person taking therapy and that hopefully takes into account the individual's lifestyle needs.

As this choice is being made it is important to make sure the regimen is as simple as possible, with particular attention to areas in which the person on treatment may have real preferences. The issues to consider in constructing a kinder, simpler regimen are (minimizing) the number of different medications, the number of pills, the number of doses per day, and the number of pills per dose, as well as any food-medication timing and fluid intake requirements.

Although few studies have been done to examine the hypothesis that simpler regimens lead to better adherence, it does seem intuitive--and some studies have indeed shown that the simpler the regimen is, the better adherence will be. It is important to keep in mind, however, that if someone misses one dose of a twice-a-day regimen, a larger percentage of medication is missed than when someone misses one dose of a three-times-a-day regimen. Therefore, the fewer doses per day, the more important each individual dose is--a point that must be emphasized and made clear to the person on treatment.

Many HIV/AIDS care sites have developed an array of ways to assist persons taking HAART regimens to optimize their adherence. More than likely a given provider will utilize one or more of these strategies for everyone in care at a particular site. It is important, however, for those on therapy to ask about what else is available and to think through what strategies would be most helpful to them personally.

Some of the most frequently used and most helpful physician-led approaches include 1) clearly writing out the medication regimen on a card with photo-stickers of the medications to be posted as a detailed reminder, 2) provision of devices that serve to organize the medications or serve as a reminder of when to take them, such as a seven-day pill box or a beeper, 3) more frequent visits with the provider or with a provider who has special expertise in adherence (often a pharmacist, nurse, or psychologist), and 4) asking about adherence at each visit and brainstorming solutions to adherence problems that are identified.

Early follow-up is important so that any certain stumbling blocks to adherence can be identified, such as a troublesome side effect or a time of day the dose is always forgotten. Once identified, obstacles can be addressed and solutions found early, before the problems damage the effectiveness of the regimen. As mentioned earlier, when individuals taking or initiating HAART have active mental health or substance abuse problems, it will be more difficult for them to adhere to their drug regimen than for others without such issues.

For this reason, try to institute a management plan for the person with a mental health or substance abuse problem before starting HAART. If that is not possible or is not the individual's preference, several of the strategies outlined above should be utilized to provide intensive support for optimizing adherence. Alternatively, these individuals may be particularly well suited for modified directly observed therapy (MDOT) programs, which have been piloted in selected sites throughout the country.

In MDOT programs, persons taking a HAART regimen go to an accessible location on a daily basis to obtain their medication and are observed taking at least one dose per day. While active substance abuse and depression are not reasons for withholding HAART if the individual is committed to taking it, intensive support to assist with adherence in these situations is essential.

Are There Special Strategies for Women to Optimize HAART Adherence?

All of the strategies detailed above apply to women as well as men. A number of studies have shown, however, that women are more likely to succeed in taking and adhering to HAART when they have a longstanding patient-provider relationship and when they truly trust their provider. Therefore, it is very important for women with HIV to find and continue with a provider that they trust. Providers should be aware of the importance of side effects in determining whether women continue on HAART regimens.

In a study performed by myself and Lisa Hirschhorn, MD, MPH, of the Dimock Community Health Center (Roxbury, MA) and Harvard Medical School, women were found to discontinue their PIs most frequently due to problematic side effects. It is therefore essential to find a regimen with no or minimal side effects, since they often cause women to skip doses or stop their regimens completely.

Motivational tools--anything that motivates people to take their HAART medications on a day-to-day basis--are particularly important for women who are living with HIV/AIDS. Women respond more favorably when "cold data" on viral load levels and CD4 cell counts are coupled with information on clinical health status and general health issues. That is, women are more likely to do well on treatment when they are focused on their own health and quality of life or on relationships with important individuals in their lives such as children, partners, or other family members or friends.

Therefore, it is helpful for HIV positive women to identify what their motivators are, and for providers to actively participate in this process. Once their motivating factors are identified, women in treatment should try to keep focused on them as they work to adhere consistently to their particular regimen.

Bangsberg, D. and others. Protease inhibitors in the homeless. Journal of the American Medical Association 278: 63-65. 1997.

Bozzette, S. and others. The care of HIV infected adults in the United States. New England Journal of Medicine 339(26): 1897-1905. 1998.

Broers, B. and others. A cohort study of drug users' compliance with zidovudine treatment. Archives of Internal Medicine 154: 1121-1127. 1994.

Celentano, D.D. and others. Self-reported antiretroviral therapy in injection drug users. Journal of the American Medical Association 280: 544-546. 1998.

Eldred, L.J. and others. Adherence to antiretroviral and pneumocystis prophylaxis in HIV disease. Journal of Acquired Immune Deficiency Syndromes: 117-125. 1998.

Hayward, R.A. and others. Regular source of ambulatory care and access to health services. American Journal of Public Health 81: 434-438. 1990.

Hecht, F.M. and others. Adherence and effectiveness of protease inhibitors in clinical practice. 5th Conference on Retroviruses and Opportunistic Infections. Chicago. February 1-5, 1998. Abstract 151.

Hirschhorn, L.R. and others. Discontinuation of protease inhibitor therapy: reasons and risk factors. 12th World AIDS Conference. Geneva, Switzerland. June 28-July 3, 1998. Abstract 12443.

Kalichman, S.C. and others. Adherence to combination antiretroviral therapies in HIV patients of low health literacy. Journal of General Internal Medicine 14: 267-273. 1999.

Lerner, B.H. and others. Rethinking nonadherence: historical perspectives on triple-drug therapy for HIV disease. Annals of Internal Medicine 129: 573-578. 1998.

Markson, L. and others. Who receives HAART? 6th CROI. Chicago. January 31-February 4, 1999. Abstract poster F222.

Moore, R. and others. Racial differences in the use of drug therapy for HIV disease in an urban community. New England Journal of Medicine 330(11): 763-768. 1994.

Motashari, F. and others. Acceptance and adherence with antiretroviral therapy among HIV-infected women in a correctional facility. Journal of Acquired Immune Deficiency Syndromes 18: 341-348. 1998.

Palella, F.J. and others. Declining morbidity and mortality among patients with advanced HIV infection. New England Journal of Medicine 338: 853-860. 1998.

Paterson, D. and others. How much adherence is enough? A prospective study of adherence to protease inhibitor therapy using MEMS caps. 6th CROI. Abstract 92.

Samet, J.H. and others. Factors associated with initiation of primary medical care for HIV-infected persons. American Journal of Medicine 97: 347-353. 1994.

Samet, J.H. and others. Compliance with zidovudine therapy in patients infected with human immunodeficiency virus, type 1: a cross-sectional in a municipal hospital clinic. American Journal of Medicine 92: 495-502. 1992.

Shapiro, M.F. and others. Variations in the care of HIV-infected adults in the United States. Journal of the American Medical Association 281: 2305-2315. 1999.

Sontag, D. and Richardson, L. Doctors withhold HIV pill regimen from some; failure to follow rigid schedule could hurt others, they fear. New York Times: 1, 31. March 2, 1997.

Stein, M.D. and others. Differences in access to zidovudine (AZT) among symptomatic HIV infected persons. Journal of General Internal Medicine 6: 35-40. 1991.

Stone, V.E. and others. Strategies for assisting HIV/AIDS patients with adherence to combination antiretroviral therapy. Journal of Care Management 5(1): 58-67. 1999.

Stone, V.E. and others. HIV/AIDS patients: perspectives on adhering to regimens containing protease inhibitors. Journal of General Internal Medicine 13(9): 586-593. 1998.

Stone, V.E. and others. Access to treatment with protease inhibitor (PI) containing regimens: is it equal for all? 12th World AIDS Conference. Abstract 42305.

Stone, V.E. and others. Adherence to protease inhibitor (PI) therapy in clinical practice: usefulness of demographics, attitudes and knowledge as predictors. 12th World AIDS Conference. Abstract 32337.

Wainberg, M.A. and Friedland, G. Public health implications of antiretroviral therapy and HIV drug resistance. Journal of the American Medical Association 279: 1977-1983. 1998.

Wall, T.L. and others. 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: 261-269. 1995.

Weissman, J.S. and others. Changes in insurance status and access to care by persons with AIDS in the Boston Health Study. American Journal of Public Health 84: 1997-2000. 1994.