V. Adherence to HIV Therapies

Adherence to HIV therapies is critical if patients are to achieve and maintain undetectable viral loads and avoid preventable opportunistic infections. Initial clinical trials of highly active antiretroviral therapy (HAART) demonstrated that 80–90% of patients receiving protease inhibitor therapy could achieve viral loads < 400 copies/mL (Gulick, 1997). However, currently in real-world clinical practice only 50% of patients achieve this goal (Deeks, 1997; Lucas, 1999). The primary reason is nonadherence with therapy. Adherence with medications in any chronic disease is a challenge for most patients. As a general rule, only 50% of patients with chronic illnesses maintain “good adherence” (taking > 80% of medication doses) over time. Patients with HIV infection have similar rates of adherence. Unfortunately in this infection, “good adherence” is not good enough, and the consequences of nonadherence are severe.

This chapter addresses the important issues regarding adherence with HIV medications. Beyond this specific issue, there are other aspects of adherence with medical care, including adherence with recommendations for laboratory monitoring for toxicity and clinical response and adherence with appointment keeping, that are also critical factors in successful HIV treatment.

Adherence with medications is critical in HIV infections for several reasons involving both the nature of the virus and the drugs to combat it. First, the virus has a very high replication and mutation rate. If drug doses are intermittently missed, the virus quickly begins to replicate. In the presence of low levels of drug, viral mutations that confer drug resistance will thrive. Second, the most potent drugs, both protease inhibitors and nonnucleoside reverse tran-scriptase inhibitors, have broad class resistance. That is, when resistance to one drug in the class occurs, often resistance has developed to all the drugs in that class. Thus, nonadherence to one regimen can result in virus resistance to many antiretrovirals.

Finally, a patient needs to have near-perfect adherence, as shown in Figure 5-1, to achieve an undetectable viral load necessary to prevent the development of resistant virus (Patterson, 1999). For most chronic diseases, 80% adherence is considered “good adherence.” However, only 50% of patients achieved undetectable viral load with this level of adherence.

Thus, in HIV infection, patients must have exceptionally high levels of adherence to therapy to prevent the emergence of resistant virus. Additionally, unlike other chronic diseases, such as hypertension, if HIV-infected patients are nonadherent, even for a few weeks, they may severely limit their future treatment options because of broad class resistance.

Adherence is a complex behavior. In analyzing predictors of adherence, it is important to consider factors related to the patient, the treatment regimen, the doctor-patient relationship, and the system of medical care. Several patient-related factors are consistently predictive of adherence or nonadherence among HIV-infected individuals. Depression and active substance abuse (including alcohol, cocaine, or heroin) correlate with nonadherence, whereas the patient’s belief that they can both make HIV medication regimens fit into their life and be adherent with the regimen is predictive of adherence (Table 5-1, Cheever, 1999). In addition, one of the best predictors of adherence is adherence with previous therapy.

In most chronic diseases, demographic characteristics, such as sex, race, income, education, and age, are not predictive of adherence (Haynes, 1979).

TABLE 5-1: PATIENT FACTORS ASSOCIATED WITH NONADHERENCE

STRONGLY ASSOCIATED

Depression
Alcohol/substance abuse
Self-efficacy (belief in one’s ability to take medication as instructed)
Belief medications can be fit into their day
Understanding the relationship of viral resistance and adherence
Previous adherence

INCONSISTENTLY ASSOCIATED

Sex (female<male)*
Race (African American <white)
Age (younger <older)
Stage of disease (patients with higher CD4 <patients with lower CD4)

NOT GENERALLY ASSOCIATED

Education, income, employment
HIV risk factor
Belief that medications will improve health or cause symptoms

* In studies where an association is found, females are less adherent than males.

However, in the initial studies of these factors of HIV-infected patients, the data are inconclusive. In the HIV Cost and Services Utilization Study, a national probability sample survey, the investigators found that women, African Americans, Hispanics, younger adults, and patients with higher CD4 counts reported lower rates of adherence (Wenger, 1999). Other studies report conflicting results, and the relationship between these demographic variables and adherence needs further exploration.

In terms of the treatment regimen, adherence decreases as the number of doses per day, pills, or medications increase. Likewise, medications with food restrictions are harder to take as directed. Length of time on therapy also impacts medication adherence. Adherence tends to wane over time, which is a compelling reason to delay therapy until the risk of progression to AIDS is significant. Pharmacy and social work counseling of patients whenever HAART is initiated or the regimen is changed can be helpful in ensuring that patients understand dosing and side effects and can assist in developing strategies to avoid missing doses.

The doctor-patient relationship is also an important variable in adherence and should not be overlooked. Several studies have shown that patients who trust their doctor are more likely to adhere to therapy, including one new study among HIV-infected patients (Kaplan, 1999). Most medical providers believe that they can predict which patient will be able to adhere to a medical regimen. However, studies have repeatedly shown that physicians, nurse practitioners, and nurses cannot predict with reliability which patients will adhere to therapy (Patterson, 1999). Table 5-2 lists the most common reason patients say they miss their medication (Chesney, 1997; Eldred, 1998). These reasons vary significantly from the reasons that providers cite as reasons their patients miss doses (Gallant, 1998).

The system of care can impact adherence to a medication regimen. Situations that facilitate patients’ access to medication, in terms of refill requests, reminders to refill medication, and mechanisms to obtain medications when gaps in medication insurance occur, can significantly impact a patient’s ability to adhere. Similarly, systems that help patients overcome barriers to seeking medical care, including problems with transportation, childcare, and paying for services, also assist in the patient’s adherence efforts. Finally, patient education is a critical part of improving adherence. Within the system of care, personnel and resources need to be dedicated to this important endeavor.

Incarcerated patients have unique challenges to medication adherence. Medical regulations vary significantly among correctional systems, and these regulations influence patients’ medication adherence. Correctional institutions may provide directly observed therapy of each dose of medication, which may result in excellent medication adherence. On the other hand, lockdowns and other security concerns supercede medical concerns and may result in missed doses. In correctional systems where HIV-infected inmates are identified and called by name to receive medications, many patients forego treatment rather than risk disclosure of their HIV infection. Additionally, the incarceration process itself provides many opportunities for potential disruption in access to medications. Most patients do not immediately access their antiretroviral medication on entry to the system or at times of transfer within the system. When patients are discharged from the penal system, they are provided with a limited supply of medication; often patients run out of medication before they can establish medical care in their communities, resulting in nonadher-ence with their regimen.

It is often assumed that homeless patients face insurmountable barriers to adherence with medications. Certainly, homelessness itself creates many challenges to adherence. Additionally, many homeless patients have mental illness or substance abuse that may further contribute to nonadherence. However, homeless patients can be successful in taking HAART. In a study of homeless and marginally housed patients from San Francisco, 38% of the 32 homeless patients taking protease inhibitors had high levels of adherence (Bangsberg, 1999). However, the group appears to be highly selected since only 34 of 153 patients in this marginally housed cohort were prescribed HAART.

Young children are dependent on the adults around them for their medication adherence. As with many HIV-infected adults, HIV-infected children may be in tenuous social situations due to poverty and unstable housing, in addition to illness among their parents. In addition, many children are unable or unwilling to swallow unpalatable medication or pills, making dosing administration take hours each day. In one study, gastric tubes were placed in children who had extreme problems with taking medication. These children subsequently showed improvement in adherence, viral load, time necessary to administer doses, and interpersonal relationships with the persons administering the medication (Shingadia, 1999). Adolescents who have not disclosed their HIV status may miss doses because of lack of a private place to take their medication. Furthermore, adolescence is a time of testing limits and feelings of invulnerability, both of which may contribute to the likelihood of nonadherence.

There are many methods to measure adherence including electronic devices, pill counts, and drug assays. However, in clinical practice, the most efficient method to measure adherence is simply to ask the patient (Icovicks, 1997). When asked in a nonjudgmental way, most patients (80% in several studies) are truthful about their medication taking (Sackett, 1976). To get the most reliable information, patients should be given permission to have missed doses, asked in a nonjudgmental way, and given a specific time frame. For example,

“Everyone misses doses some of the time. In the last 2 weeks, how many doses have you missed?”

Pharmacy records can also give important information regarding adherence when available to the clinical staff. Serum drug levels are now commercially available. These are most useful when patients professing adherence are not responding well to therapy; however, low drug levels may also be due to poor absorption or drug-drug interactions as well as to nonadherence.

Given the importance of adherence in HIV infection and the relatively low rates of adherence in any chronic disease, it is of utmost importance to support efforts to increase patient adherence. Although there is little information specific to HIV-infected individuals, many interventions have been shown to increase adherence with pill taking in chronic disease. Most interventions only have a modest effect on adherence. Those that work best are multifaceted (working on more than one aspect of the pill taking behavior) and repetitive (having the intervention reapplied over time) (Haynes, 1996; Roter, 1998). With antiretroviral medication, it is critical to start the intervention before nonadherence, and viral resistance, have occurred.

In initiating a successful regimen, the most important thing to consider is whether the patient is ready to start antiretroviral therapy. The patient is usually the best judge of when to start therapy as long as she is well informed regarding the rigors of antiretroviral therapy and the consequences of nonadherence. Thus, the bulk of patient education efforts should be made before initiation of therapy. Given that the patient’s “first shot [of antiretrovirals] is the best shot,” most patients benefit in the long run by delaying therapy if they are not ready to start. The urgency of therapy increases as the CD4 count falls and the viral load increases. Many physicians feel pressured by practice guidelines for starting therapy. The best time to start therapy is when the patient is ready.

There are, however, a few “antiretroviral emergencies,” pregnancy being one of them. Clearly, for the pregnant woman, the cost benefit considerations are different, as outlined in Chapter VII on HIV and Reproduction. In the case of pregnant women, the time permitted for the patient to prepare to start therapy is considerably compressed. Thus, even more intensive educational efforts are needed in this population.

Patient education is clearly important, with high returns on the investment made before starting therapy. Group education generally works better than one-on-one education, probably because of the support and practical peer advice that is shared (Roter, 1998). Many patients resist groups, but many find them beneficial if they can be convinced to go. Specific interventions to improve adherence are outlined in Table 5-3.

Two controlled intervention trials to improve adherence in the HIV-infected population have been presented. In a pharmacy-based trial from Spain, patients starting highly active therapy received education detailing the importance of treatment, adherence, and side effects management (Knobel, 1998). They were engaged in tailoring the medication into their daily schedule and given a telephone number to call with problems. In follow-up phone calls, the importance of adherence was stressed, and patients were engaged in problem-solving incidents of missed doses. At 48 wk, in the intervention arm, 77% of patients were adherent with medication, compared with 52% in the control arm.

TABLE 5-3: INTERVENTIONS TO IMPROVE ADHERENCE
INTERVENTION	COMMENTS
THE PATIENT
Start when the patient is ready	For pregnant women, the cost-benefit analysis of treatment is different.
Treat substance abuse and depression before initiating antiviral therapy	If there is no antiretroviral emergency, patients with active substance abuse and depression should have these comorbidities addressed before initiation of antiretrovirals.
Engage the patient in medication tailoring	Discuss with the patient in detail how the medications will fit into their daily routine, i.e., when (and if ) meals are eaten, what the patient does on a daily basis that can be linked to dosing times.
Educate (group/individual) regarding: The regimen Side effects management Consequences of nonadherence	Patient education is essential—both group and one-on- one education. Involve caretakers and patient support network in educational efforts. Patients need to know exactly how to take their medication. A daily calendar with pills on it will help a patient to visualize the regimen. Before initiating therapy, patients should know which side effects to expect, what they can to do to manage them, and when to call the medical practice. Patients need to understand the serious consequences of nonadherence and what to do in the event of a late or missed dose.
Increase support	Patients should enlist the aid of family and friends to promote their adherence. The HIV health care team can provide support through office visits, home visits, and telephone calls, especially in the first days and weeks of antiretroviral therapy.
Use skill-building exercises	Have the patient use a trial of jellybeans in a pill box to accustom themselves to their pill-taking schedule before initiating therapy.
Address barriers to adherence	Have the patient consider when medications are likely to be missed and make plans to decrease these events. Some patients store a few doses in places where they spend a lot of time, such as at the houses of friends and relatives.
Use reminders	Alarm clocks, in the form of watch alarms or pill boxes, can decrease missed doses due to simply forgetting. Patients can place medications in locations they will notice them at dosing times, such on the breakfast table.
THE REGIMEN
Simplify as much as possible	Once or twice daily regimens are easiest for patients. Use as few pills and medications as possible. Try to use regimens that can be taken without regard to food intake.
Tailor the regimen to the patient’s lifestyle (and not the patient’s lifestyle to the regimen).	Ask the patient about her daily routine, comfort in taking medications in front of others and at work. Construct a regimen that works for the patient.
Use pill boxes	Use of pill boxes allows the patient to carry her daily medication. Pill boxes allow for patients to easily recognize when they have missed a dose.
Make refills accessible	Develop policies to allow patients ready access to refills.
THE DOCTOR-PATIENT RELATIONSHIP
Develop a trusting relationship	Rarely is initiation of antiretrovirals required at the first visit. Invest in the doctor-patient relationship before initiating therapy.
Ask about adherence	Medical providers cannot predict adherence; you must ask the patient. Ask in a nonjudgmental way, with a specific time frame, to get good information. Give permission for missed doses before asking. Ask repetitively over time.
Use positive reinforcement	Share viral load and CD4 results and reinforce the relationship to adherence.
Listen to the patient	Individualize therapy based on patient preferences regarding fear of specific side effects or specific medication. Negotiate the regimen with the patient.
THE SYSTEM OF CARE
Maintain close follow-up at initiation of regimen.	Have telephone, office, or home contact with patient within first few days of therapy to assess for side effects and accurate understanding of regimen.
Develop patient education program	Consider use of nurses, case managers, pharmacists, and peers in patient education. Have written materials accessible.
Incorporate the adherence message throughout the medical practice	All staff members need to understand and promote the importance of adherence. Have pill boxes, alarms and other adherence aids available to patients.
Source: Adapted from Cheever, 1999.

In a second controlled interventional trial, a nurse-based intervention to improve adherence with Pneumocystis carinii pneumonia (PCP) prophylaxis was also reported to be successful (Cheever, 1998). Patients in the intervention group received a 10-min nurse-based exit interview at the end of each regular clinic visit. During the exit interview, the nurse discussed the importance of PCP prophylaxis, stressed the importance of adherence, and engaged the patient in medication tailoring. Median adherence in the intervention group was 78%, 16% better than among the control group. More randomized trials are needed to define which interventions work best to improve adherence, and many are under way.

Adherence with medications is critical in HIV infection. Whenever medications are prescribed, special emphasis must be placed on patient readiness and adherence. Many types of interventions can improve adherence, and it is vital that these are employed to give patients the best chance at an optimal response to HIV therapies.

Bangsberg DR, Hecht FM, Charlebois EC, et al. Spontaneous adherence (ADH) audits (SAA) predict viral suppression in the REACH cohort. Program and Abstracts of the 6th Conference on Retroviruses and Opportunistic Infections, Chicago. Abstract 93, 1999.

Cheever LW, Barnes GL, Chaisson RE. A randomized educational intervention to improve adherence with medication. Abstracts of the 38th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego, Abstract I-178, 1998.

Cheever LW, Wu AW. Medication adherence among HIV infected patients: understanding the complex behavior of taking this complex therapy. Curr Infect Dis Rep 1: 401–407, 1999.

Chesney MA. New antiretroviral therapies: adherence challenges and strategies. Abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Canada. Evolving HIV Treatments: Advances and the Challenge of Adherence Symposium, 1997.

Deeks S, Loftus R, Cohen P, et al: Incidence and predictors of virologic failure to indi-navir (IDV) and/or ritonavir (RTV) in an urban health clinic. Abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Canada, Abstract LB-2, 1997.

Eldred LJ, Wu AW, Chaisson RE, et al. Adherence to antiretroviral and pneumocystis prophylaxis in HIV disease. J Acquir Immune Defic Syndr Hum Retrovirol 18: 117–125, 1998.

Gallant JE, Block DS. Adherence to antiretroviral regimens in HIV-infected patients: results of a survey among physicians and patients. J Int Assoc Physicians AIDS Care May 1998.

Gulick RM, Mellors JW, Havlir D, et al. Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior anti-retroviral therapy. N Engl J Med 337: 734–739, 1997.

Haynes RB. Determinants of compliance: the disease and the mechanics of treatment. In: Compliance in Health Care. Edited by Haynes RB, Sackett DL. Baltimore: Johns Hopkins University Press, 49–62, 1979.

Haynes RB, McKibbon KA, Kanani R. Systematic review of randomized trials of interventions to assist patients to follow prescriptions for medications. Lancet 348: 383–86, 1996.

Icovicks J. Behavioral measures. Adherence to New HIV Therapies: a Research Conference. Washington, DC, NIH Office of AIDS Research, 15–18, 1997.

Kaplan A, Golin C, Beck K, et al. Adherence to protease inhibitor therapy and viral load. Program and Abstracts of the 6th Conference on Retroviruses and Opportunistic Infections, Chicago. Abstract 96, 1999.

Knobel H, Carmona A, Grau S, et al. Strategies to optimise adherence to highly active antiretroviral treatment. Program and Abstracts of the 12th World AIDS Conference, Geneva, Switzerland. Abstract 338/32322, 1998.

Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: risk factors for virologic failure and adverse drug reactions Ann Intern Med 131: 81– 87, 1999.

Patterson D, Swindells S, Mohr J, et al. How much adherence is enough: a prospective study of adherence to protease inhibitor therapy using MEMS Caps. Program and Abstracts of the 6th Conference on Retroviruses and Opportunistic Infections, Chicago. Abstract 092, 1999.

Roter DL, Hall JA, Merisca R, et al. Effectiveness of interventions to improve patient compliance: a meta-analysis. Med Care 36: 1138–1161, 1998.

Sackett DL, Haynes RB, Gibson ES, et al. Randomized clinical trial of strategies for improving medication compliance in primary hypertension. Lancet 1: 1265–68, 1976.

Shingadia D, Viani R, Dankner W, et al. Gastrostomy tube for improvement of adherence with HAART in pediatric HIV patients. Program and Abstracts of the 6th Conference on Retroviruses and Opportunistic Infections, Chicago. Abstract 436, 1999.

Wenger N, Gifford A, Liu H, et al. Patient characteristics and attitudes associated with antiretroviral (AR) adherence. Program and Abstracts of the 6th Conference on Retroviruses and Opportunistic Infections, Chicago. Abstract 98, 1999.