Drug Use: The Effect on HIV Progression, Adherence and the Relationship With Medical Providers
Injection drug use accounts for about a third of HIV cases in the United States, and alcohol and other drug use is common among many people living with HIV. In 2001, the Archives of General Psychiatry published the results of a national survey of 2,864 people with HIV who were accessing medical care. Nearly 40% reported using an illicit drug other than marijuana during the previous year, and over 12% reported drug dependence. Questions arise about the influence these substances may have on the progression of HIV and about the impact substance use may have on access, adherence, and response to treatment.
Progression of HIV
Most long-term studies of HIV-positive people in the pre-HAART (Highly Active AntiRetroviral Therapy) era find no difference in the progression of HIV to AIDS and death between those who use illicit drugs, including the injection of heroin and cocaine, and non-drug users, although there are a few in vitro (laboratory based) studies suggesting that drugs, including heroin and cocaine, increase HIV replication.
A study published by Rompalo and colleagues in the International Journal of STD and AIDS in 2004 is particularly convincing as the analysis distinguishes between former and current drug injectors. About 640 HIV-positive women were followed for up to seven years -- 52% had injection drug use as a risk factor, and 35% injected drugs during the study period. The analysis factored in the effect of initial CD4 count, viral load, and age. Over the seven years, there wasn't a difference in progression between the women who had past, current, or no history of injection drug use.
It is common for drug users to be told that they've made their HIV worse by taking drugs, further stigmatizing drug users; however, there is no good evidence that these illicit drugs actually do this. Past and current injection drug use has not been found to be associated with progression of HIV disease, thus initiation of treatment can be based on standard guidelines including clinical indications and readiness for treatment among drug users.
The recent case of rapidly progressive, multi-drug resistant HIV in New York City has raised speculation that methamphetamine use may lead to more rapid progression of HIV. However, a 2003 study in the Journal of Infectious Diseases comparing HIV viral loads among users and nonusers of methamphetamine found that HIV viral loads weren't different between the two groups unless the subjects were taking HAART. When comparing those taking HAART, the methamphetamine users had significantly higher viral loads. This is consistent with a behavioral or biological impact of methamphetamine on HAART, but not on viral replication. Similarly, studies of the impact of alcohol on HIV sometimes find that alcohol may promote viral replication in the lab but, in the real world, it appears to have more impact on the ability to take or have a good response to HAART than on the virus itself. Marijuana has not been found to have an unfavorable impact on HIV progression.
To conclude, there is no strong evidence that alcohol and other drugs by themselves have a significant biological impact on the progression of HIV disease. But several of these studies suggest that some substances have a negative effect when HAART enters the picture, implying that drug users may have difficulty obtaining, adhering to, and/or benefiting from treatment.
Drug Users' Access to HAART
Numerous studies have found that alcohol and injection drug users (IDUs) are less likely to be prescribed HAART even when their CD4 counts and viral loads indicate that it's time to start HIV treatment. There are many reasons why this may be so, including poor access to medical care, reluctance on the part of providers to prescribe to these populations because of concerns about adherence, and, perhaps, the users' refusal to take the medications. Little is known about drug users' beliefs about HAART, but at least one study suggests that there may be fear about dangerous interactions between HIV medications and illicit drugs and alcohol, causing at least some people to refuse medication or to adhere poorly to the medications if they're prescribed.
Hepatitis C may pose another barrier to the receipt of HAART. Hepatitis C (HCV) is extremely common among HIV-positive injection drug users, and people with HCV are found to be somewhat less likely to receive HAART. However, HIV/HCV co-infected patients can benefit from HAART; while liver inflammation is a risk, it is rarely a cause for discontinuation of HAART. HCV infection isn't a valid reason to delay or defer therapy; in fact, data from a study published in Hepatology in January 2005 suggest that maintaining higher CD4 counts with HAART may delay progression of HCV, underlining the importance of treating injection drug users for HIV in a timely manner. Given that many co-infected patients are unlikely to have a successful response to HCV treatment, HAART may be particularly important in helping to prevent progression of liver disease caused by HCV.
Adherence to and Benefit From HAART
There appears to be a hesitancy on the part of at least some medical providers to offer HAART to active drug users. To some extent, this hesitation is based on literature which shows that, as a group, active users are significantly less likely to report good adherence or to achieve undetectable viral loads. On the other hand, most studies also find that many users are highly adherent and successful in achieving undetectable viral loads with the expected clinical benefit of slower HIV disease progression.
This was well illustrated in an article published in the Journal of Acquired Immune Deficiency Syndromes (JAIDS) in 2001 because the study differentiated between current and past drug users. The study followed a group of 764 patients over the course of a year, looking at both self-reported adherence to HAART and virological response to treatment. While a significantly lower number of injection drug users adhered to medication and achieved an undetectable viral load compared to former users and nonusers, fully one third of the users did so.
In 2004, another study in JAIDS compared disease progression while on antiretrovirals among 827 IDUs and 1,314 non-IDUs. The study found that, while the incidence of AIDS-defining illness was higher among the IDU group compared to the non-IDU group, this was not true for the IDUs with undetectable viral loads. These findings indicate that, given equal access to care, IDUs are less likely to have undetectable viral loads, suggesting other barriers to successful treatment. However, over 40% of the IDUs were successful in achieving undetectable viral loads which were durable over more than one measurement, and this group experienced the same positive clinical impact as nonusers.
Active methamphetamine users have also been noted to be significantly less likely to achieve an undetectable viral load. In the study discussed above, only 39% of current methamphetamine users did so compared to about 60% of past and nonusers. As with heroin and cocaine, we see that many methamphetamine users can, indeed, achieve an undetectable viral load, though less do so.
There are few studies of the impact of alcohol on adherence. It appears that heavy drinking is associated with a lower likelihood of adherence, but the level of safe drinking hasn't been determined. A couple of studies have found little impact of light to moderate alcohol use, while a publication in Alcoholism: Clinical & Experimental Research in 2003 reported that any alcohol use, at least in people with a history of problem drinking, could decrease adherence.
There are also few studies that address the impact of marijuana on adherence. Two studies found that marijuana use is associated with poorer adherence, while one found that use of marijuana to control nausea may have a positive impact on adherence.
In summary, it appears that commonly used illicit substances and alcohol don't have a direct impact on the progression of HIV or on the response to HAART. However, for some patients, substance use will interfere with adherence to medication. Those patients will have reduced or no benefit from HAART, leading to a poor clinical outcome.
The Role of Stigma
We have little information about what differentiates those drug users who are able to adhere to their medications from those who aren't. Certainly, the level and patterns of drug use, as well as those factors that affect adherence in drug and non-drug users alike, such as housing and mental health, all play a role. But drug use may also have an impact on individuals' interactions with healthcare providers, as drug use is often stigmatized and associated with many stereotypes. Physicians and probably other healthcare providers are poorly equipped to care for many substance users. A paper published in Academic Medicine in 2001 outlined many of the reasons why. Medical schools provide little education about drug use and addiction, and there are few role models in providing care for these patients. Like the general public, physicians often have negative attitudes about substance users, making it difficult to develop strong, beneficial patient-provider relationships.
It is widely believed that a good relationship between the healthcare provider and patient has a positive impact on adherence. A 2004 study published in the Journal of General Internal Medicine lends support to this belief. A survey of 554 patients in 22 HIV practices examined satisfaction with healthcare providers on a variety of qualities and the subjects' self reported adherence. It was found that various patient-provider relationship qualities, including trust and communication, were significantly associated with adherence.
To summarize, many (but not all) studies find that patients who use heroin, cocaine, methamphetamine, and/or alcohol are less likely to adhere successfully to HAART. But these studies also find that a substantial number of users have excellent adherence.
Opioid maintenance such as the provision of methadone can be an important tool in promoting adherence among heroin users. Among HIV-positive patients with a current or past history of heroin use, methadone maintenance is associated with more consistent use of HIV medications and fewer hospitalizations. However, because methadone is available only in specialized clinics and has long been stigmatized as "just another drug of addiction," many patients are reluctant to access or continue to use methadone despite the significant benefit it offers.
Buprenorphine is a new option for treating heroin addiction. It is similar to methadone in that it prevents withdrawal symptoms, blocks the effect of heroin, and reduces opioid craving. Because it's more difficult to misuse and it's very hard to have a fatal overdose on this medication, it may be prescribed by physicians outside of the methadone clinic system who apply for a waiver after completing an eight-hour training. As yet, there is little experience with buprenorphine in the United States, but a 2000 study by French researchers published in AIDS showed buprenorphine maintenance to have a positive impact on adherence and on viral suppression.
The U.S. Health Resources and Services Administration has shown support for the use of buprenorphine in HIV settings by funding a Special Project of National Significance to lead the way to integrating the use of this intervention into primary care. The option for physicians to prescribe buprenorphine can both reduce the need to rely on a completed referral to specialty care and, perhaps, promote the development of a trusting relationship as drug treatment becomes normalized. Further information can be found at buprenorphine.samhsa.gov.
We have seen that many drug users can adhere to their HIV therapy without intensive support, but, for those who can't, there are interventions. There have been numerous successful studies using modified Directly Observed Therapy (DOT) programs, which observe patients take their antiretroviral doses daily or less frequently in methadone maintenance treatment clinics. Of note, good results have been achieved with methadone patients who are also cocaine users.
Not all drug users requiring additional support are in maintenance drug treatment. For example, a program in Boston reported on in Clinical Infectious Diseases in 2004 trains peers from the community to work with patients, including drug users, who have very poor adherence to treatment. In varying levels of intensity, the peers accompany patients to medical appointments to facilitate the patient-provider relationship, provide education, and offer adherence assistance that may include DOT, risk reduction, and crisis intervention. According to the study, among the first 15 patients enrolled, 11 achieved a persistently undetectable viral load. While this approach is labor intensive, a recently presented cost benefit analysis found that the cost of the intervention is less than the projected costs of the treatment of complications of AIDS expected with disease progression.
Access to sterile syringes is key in preventing transmission of HIV and other blood-borne diseases. Syringe exchange is highly effective in reducing the transmission of HIV. Gibson and colleagues published an excellent review of studies of syringe exchange programs in the journal AIDS in 2001. Syringe exchange is not the only modality for providing access to sterile syringes. In many states, syringes can be purchased at pharmacies or prescribed by physicians. A helpful guide to syringe access in the United States can be found at www.temple.edu/lawschool/aidspolicy. In addition to preventing HIV, provision of syringes can assist in developing a good patient-provider relationship as the patient realizes that the provider is willing to work within the patient's priorities.
Opioid maintenance is also essential in HIV prevention. Heroin users maintained on methadone have been found to be four to six times less likely to become infected with HIV, either because they stop injecting heroin or are able to have greater control over their heroin use because they aren't constantly avoiding withdrawal symptoms. It's easier to say no to the used syringe or to avoid unsafe transactions of sex for drugs if one isn't fearful of withdrawal. It is likely that buprenorphine will have a similar effect as a tool in prevention of transmission of HIV.
Substance use does not appear to have a biological impact on the progression of HIV, but it is associated with barriers to access and adherence to HAART. In some cases, it may be appropriate to delay the initiation of HAART while patients stabilize a variety of complications in their lives such as substance use, housing and mental health issues, but it is clear that active substance use is not a contraindication to HAART -- large numbers of substance users successfully achieve full viral suppression. In fact, timely initiation of therapy may be vital in HCV co-infected patients.
Primary care providers may be able to improve patient-provider relationships and promote adherence by becoming knowledgeable about substance use as well as by examining their own stereotypes and attitudes about substance use.
Sharon Stancliff, M.D. is Medical Director of the Harm Reduction Coalition in New York City and has worked with HIV-positive drug users in primary care, drug treatment and harm reduction.
This article was provided by AIDS Community Research Initiative of America. It is a part of the publication ACRIA Update. Visit ACRIA's website to find out more about their activities, publications and services.