Published in the Bulletin of Experimental Treatments for AIDS January 1999 issue, by the San Francisco AIDS Foundation.

January 1999 Table of Contents

Main Page

beta@sfaf.org

The Challenge of Adherence

Margaret Chesney, PhD

To reach this hall, we walked by exhibits from pharmaceutical companies that are investing millions of francs toward the development of new drugs. At the scientific sessions at this conference, we have learned about promising new therapies. Yet all of this investment and all of these findings will be for naught if people do not adhere to or comply with therapy.

The theme of the 12th World AIDS Conference is "bridging the gap," and in this session there are two gaps we must bridge. The first is a gap between the treatment regimens as they are prescribed, and the people who want to avail themselves of these treatments but also want to have a life. The second is a gap between the developing and the developed worlds. One might say it is a luxury that those in the developed world can worry about timing medications with meals or drinking enough water. In the parts of the world where 90% of new AIDS cases are found, people don’t have medications, and often they must struggle for meals and water.

Adherence is a problem throughout the different areas of medicine and throughout the world. It became a focus in HIV therapy two years ago, when combination anti-HIV therapy including protease inhibitors was introduced. If adhered to, these therapies are remarkably effective for many people. There has been a striking decline in AIDS cases and AIDS deaths in the developed world coinciding with these new treatments.

No sooner had successful results of protease inhibitor therapy been reported, than we became aware that treatment breakthroughs or treatment failures represented a real problem. For example, significant virologic breakthrough -- that is, a return to a higher viral load after suppression -- was observed in a cohort of 136 patients from San Francisco General Hospital. All had been on combination therapy for more than 24 weeks. Fifty-three percent experienced a failure of virologic control. In this study, treatment breakthrough was predicted not only by initial viral load and previous treatment history, but also by non-adherence to therapy.

To understand how non-adherence leads to virologic breakthrough, we need to examine viral dynamics. HIV replicates at a rate of ten billion new virus particles each day. Mutations occur in about 1 in 10,000 of these particles. Each day, every single point mutation is produced. Some of these mutations will confer drug resistance, rendering a virus no longer susceptible to a specific antiretroviral drug. In this manner, HIV replication becomes the source of new viral strains and, ultimately, is the engine that drives the development of drug resistance.

To avoid the development of resistance, we must stop viral replication. If treatment is effective and the person is adherent, the virus will be held in check. If the drugs are not adequately absorbed or the person is not sufficiently adherent, we will see viral replication and the emergence of HIV that is resistant to the drugs being used.

The take-home message is that, to maintain potent antiviral activity, people must adhere to treatment. AIDS treatment guidelines, such as those from the International AIDS Society-USA, state that less than excellent adherence may result in virus breakthrough and the emergence of drug-resistant strains.

Drug resistance is important for two reasons. First, if a person's virus becomes resistant to a drug, then the treatment is no longer beneficial for him or her. But there is another reason to be concerned about resistance, and that is that a person who develops drug-resistant strains of HIV can pass that resistant virus on to others. Evidence of the transmission of multidrug-resistant HIV has been presented at this conference (see BETA, October 1998).

Non-Adherence: A Widespread Problem

All of this points to a key question: how extensive is non-adherence to HIV therapy? Fortunately, we now have several key studies on adherence, and they tell a similar story. I'll briefly describe a few of them.

The first of these studies was conducted by the Recruitment, Adherence, and Retention Subcommittee of the AIDS Clinical Trials Group (ACTG). This committee, along with the ACTG Patient Care Committee, surveyed 75 people taking combination therapy. We administered self-report questionnaires that asked about recent non-adherence, and we also looked at some predictors of non-adherence.

Previous research on adherence has indicated that how people are asked about adherence can make a great deal of difference in the results. We asked about adherence in a very specific manner. To maximize accuracy, we focused on the recent past, and to reduce bias, we asked the question in a way that would make it easier for people to state that they had missed a dose. Specifically, we asked, "How many doses of your antiviral or anti-HIV medication did you miss yesterday? The day before yesterday? Three days ago?" and so on. Among the 75 people we surveyed, slightly more than 10% reported missing at least one dose each day. When we combined the results, approximately 20% had missed a dose in the last two days.

A similar survey, directed by Frederick Hecht, MD, included 134 people with HIV on protease inhibitor regimens at the AIDS clinic at San Francisco General Hospital. Again, we measured self-reported adherence over the past three days. In addition to the survey, we drew blood so that we could measure viral load. The reports of missed doses were almost the same as in the ACTG survey.

A third rather interesting study was recently reported by Joel Gallant, MD, of Johns Hopkins University. Gallant and colleagues surveyed 665 patients and 100 physicians by telephone. They asked about the occurrence of recent non-adherence and reasons for non-adherence. They found that 26% of patients reported non-adherence to their regimen the previous day. This study is slightly different in that the researchers looked not only at missed doses, but also included in their definition of non-adherence those who took their drugs off schedule, or did not take their drugs exactly on time, or did not follow meal instructions.

Let me focus for a moment on the issue of instructions. In the ACTG study, we also asked the 75 people with HIV if they were aware of any special instructions that they should follow when taking their medications, such as "take on an empty stomach"; we only looked at the responses from those people who were on medications for which there should have been instructions.

We found that 25% of the people did not even know about the special instructions. This does not necessarily mean that their physicians did not tell them. It’s just that the patients were not aware of them. We also found a significant gender difference in knowledge about instructions. Whereas 85% of the men in our sample were aware of their instructions, only 50% of the women were aware. However, when we controlled for income and education, the gender differences was diminished, so it may really be due to socioeconomic factors. What this suggests is that those in lower socioeconomic groups -- those who have fewer resources -- for some reason are not being informed, not hearing, not understanding, or not retaining the information about special instructions, and therefore may not be benefiting as much as they could from their therapy.

So now we can answer the question "How extensive is non-adherence?" The answer is, it is extensive, both in terms of failing to take doses of medication and problems with following the complex regimens with all their instructions.

Adherence and Viral Load

Another critical question is whether non-adherence is linked to viral load. This is particularly an issue when we use self-report measures to assess adherence. Are these measures valid? Are people who report higher levels of adherence more likely to have an undetectable viral load?

To answer this, I’ll return to the San Francisco General Hospital data and look at the association between people who said that they were taking their medication and undetectable viral load. Of the group of people with HIV who reported taking 100% of their medications, approximately two-thirds had an undetectable viral load. In contrast, only one-third of the group who reported taking less than 80% of their medication had an undetectable viral load. So, the answer is that non-adherence is linked to viral load.

Factors Associated with Non-Adherence

In looking at all of this data, we may wonder what are the reasons for these missed doses. It is important to emphasize that when asked what is the number one reason for non-adherence, physicians and patients disagree. And I ask you, who is more likely to really know why they are having difficulty adhering to a regimen -- the physician or the patient?

Perhaps the best way to understand the challenge of adherence is to think for a moment about these regimens. What is it like to take different pills every few hours -- some with meals, some on an empty stomach; some with fat, some with non-fat foods; some with water? When the end of the day comes and the last dose is taken, the daily total may be 28-30 pills. Think of the impact this would have on one's life.

As part of the Johns Hopkins telephone survey I described earlier, both physicians and patients were asked what is the number one reason people that are having trouble adhering to their antiviral regimens. Fifty-nine percent of the physicians believed that the primary reason is the number of doses and the number of pills. Only 16% of the patients cited this as the number one reason. Twenty-eight percent of the physicians said side effects were the primary reason, compared to only 13% of the patients.

People with HIV are saying something quite different. Twenty-two percent identified meal restrictions as a problem, compared to only 5% of the physicians. The leading category among the patients was actually "other," and it included a whole host of issues. One of these was privacy, and the need to be confidential about taking medication. People who have jobs are often very sensitive about taking medication in public because they do not want their HIV status to become known. Cost is also an issue, as is the complexity of the regimens.

The many reasons people miss their medications have been documented in a number of studies. The leading reason is always the same: "I just forgot." Others include falling asleep, being busy with other activities, being away from where medications are stored, feeling stressed or down, and running out of medication.

In another survey, people were asked why they missed recent doses. They did not even mention the number of medications or the number of doses per se, and only 11% mentioned side effects. People with HIV are saying something different. Thirty-two percent said that they missed doses because they were away from home, or they changed their routine, or they were too busy. I know of people coming to this conference who missed doses because their luggage was lost at the airport. So we see a gap between complex regimens and people’s lives.

Factors Associated with Non-Adherence

• Forgetting
• Medication interferes with lifestyle (e.g., meals, other activities)
• Falling asleep
• Change in routine (e.g., weekend, travel)
• Alcohol or drug use
• Stress
• Pessimism about HIV disease
• Need to conceal medication from others
• Younger age
• Complexity of regimen
• Higher number of medications
• Lower levels of coping efficacy
• Depressed mood

Other Causes of Treatment Failure

Two years ago, we were heralding the advance of protease inhibitors. Today, our enthusiasm is tempered by the evidence of virological breakthrough. One cause of this breakthrough -- and it is just one cause among many -- is non-adherence. We should also keep other causes in mind. People may experience early development of drug-resistant HIV due to the prior use of antiviral therapies. They may not be able to absorb their medications. We must not blame people with HIV for breakthroughs in viral load and immediately assume that the problem is non-adherence. But non-adherence is certainly something that we need to attend to, because -- unlike some of the other causes of drug failure -- it is something that we can change.

What Can Be Done?

We have evidence that non-adherence to HIV therapy is extensive, and that it is linked to viral load. We used to say that the number of AIDS cases or the number of HIV infections that we knew of were just the tip of the iceberg. I suggest that we have another iceberg. The non-adherence problems that we are discussing today are just the tip of another ominous iceberg, with viral resistance and treatment breakthroughs below the surface.

So, perhaps the real question of the moment is, what can be done to improve adherence? Some physicians have tried to fix the problem by reducing the number of medications and doses, and by managing side effects. Pharmaceutical companies are attempting to develop new drugs that are easier to take. These are very important measures, and we should applaud these efforts, but they alone will not solve the problem, because people with HIV are telling us something else.

Strategies for Improving Adherence

Clarify the regimen -- include pictures and names of pills, how often they must be taken, meal requirements, etc. Physicians should ask patients what medications they are going to take tomorrow, and answer any questions this exercise generates.

• Written instructions -- written instructions can help people remember what was said during an office visit. Treatments are very complex -- often overwhelming -- and people may not be able to remember specific instructions when they are feeling overwhelmed.
  
  
• Drug names -- make sure that various drug names (generic, brand, etc.) are clear. Physicians can help by using the names their patients use.
  
  
• Call with questions -- physicians and pharmacies can establish a toll-free, 24-hour phone line that patients can call if they have questions about their medications.

Motivation -- think about the positive reasons for taking the medications: not only decreasing viral load, but also what taking control of HIV can mean for an HIV positive person.

• Education -- learn about how HIV works, how resistance develops, and why it is so important to maintain good adherence. Physicians and pharmacies can provide accessible educational materials that people with HIV can understand.
  
  
• Monitor viral load and CD4 count -- regular viral load tests and CD4 cell counts can be helpful, but don't over-emphasize these measurements; a person can adhere perfectly and still experience an increase in viral load.

Make taking medications part of a daily routine -- find consistent daily activities that can serve as cues for taking medication (e.g., TV shows, bedtime, brushing teeth, soup kitchen line, exercise routine, walking the dog); plan to take pills before rather than after the activity.

• Daily plan -- write out a daily plan linking medications to specific daily activities; use a personal log to record medication doses.
  
  
• Plan ahead for disruptions -- plan in advance for weekends, holidays, vacations, and other disruptions to the regular daily routines; order special meals in advance.
  
  
• Pill planning -- establish a time to count out and arrange pills for the following week; use compartmentalized pill boxes.
  
  
• Extra pill bottles -- keep a supply at home, at work, and other places where one spends a lot of time; keep pills at the location where they need to be taken (e.g., by the alarm clock or the coffee maker).
  
  
• Reminders -- use devices such as timers, watches with alarms, checklists, and compartmentalized pill boxes to help remember when to take medications.
  
  
• Privacy -- find ways to take drugs in a way that preserves confidentiality.

Manage side effects -- report all side effects to healthcare providers. Medications and other therapies (e.g., acupuncture, medical marijuana) can be used to help manage side effects such as diarrhea, nausea, and peripheral neuropathy. Physicians may adjust doses or use substitute drugs if possible to minimize side effects.

Address stress, depressed mood, and alcohol and drug use -- stress, depression, and alcohol and drug use can have a negative effect on adherence.

• Stress -- many people with HIV/AIDS become stressed or burnt out. Seek social support or counseling to help manage these moods. Living with HIV/AIDS is difficult from time to time; take advantage of available programs.
  
  
• Alcohol and drug use -- use of alcohol and drugs -- particularly cocaine and amphetamines -- is associated with non-adherence. Get counseling to address this issue in order to gain the full benefits of anti-HIV treatments.

Good communications -- communication between people with HIV and their healthcare providers is an important key to successful treatment.

• Patient/physician communication -- maintain good communications with physicians and other healthcare providers. Physicians can encourage questions, provide index cards for patients to write down questions they want to ask, and train staff to help patients maintain adherence
  
  
• Trust and honesty -- report any adherence problems to healthcare providers. Physicians can create a non-judgmental environment in which people are comfortable honestly reporting any adherence problems they may have.

Social support -- enlist family, friends, and other social supports in helping to maintain adherence.

Page last updated 15 January 1999

About AIDS | Treatment | Prevention | Policy | About SFAF | Take Action | Donate
Español | Search | Site Map | Feedback | Email Updates | Home (sfaf.org)