|Home | About | Journals | Submit | Contact Us | Français|
Advances in antiretroviral medications have resulted in precipitous declines in HIV-associated morbidity and mortality; however, high levels of adherence are crucial to the success of HIV therapies. This article reviews published studies in the United States on HIV-infected youth (ages 13 to 24 years), focusing on adherence to antiretroviral regimens and interventions designed to enhance adherence. A systematic search yielded 21 articles published between 1999 and 2008 that reported data on medication adherence in HIV-infected youth, of which 7 described unique interventions to enhance medication adherence. Five thematic areas were identified to classify factors associated with adherence. Findings suggest psychosocial factors, in particular depression and anxiety, were consistently associated with poorer adherence across studies. Three types of adherence interventions with HIV-infected youth were found. Results suggest that examining adherence within the broader contextual issues present in the lives of youth, including HIV stigma and disclosure, caregiver stress, peer relations, mental health and substance use, and length of time on medications, may be most important to understanding how best to intervene with adherence among this population. Secondary HIV prevention interventions for youth represent a possible mode through which to deliver individually tailored adherence skill building and counseling to improve medication adherence.
According to the Centers for Disease Control and Prevention, an estimated 5259 young people aged 13 years to 24 years received a diagnosis of HIV infection or AIDS in the United States in 2006, a 25% increase from estimated diagnosed cases among youth in this age range in 2003 (n = 4209).1 These youth represented 25% of the estimated 475,871 persons living with HIV or AIDS in 2005 in the 33 states with long-term, confidential, name-based HIV reporting in the United States (n = 19,134).1 Advances in medical treatment, specifically antiretroviral medications, have resulted in precipitous declines in HIV-associated morbidity and mortality,2–5 allowing for HIV-infected adolescents and young adults to manage their HIV infection as a chronic, rather than imminently life-threatening, disease. However, maintaining high levels of adherence (90% to 95%) to antiretroviral therapy is crucial to treatment success,6–10 and promoting adherence remains an essential element of modern HIV care.11,12
In providing HIV care for youth, practitioners may follow the US Department of Health and Human Services guidelines.10 Although substantial advances have been made to simplify regimens and develop combination therapies,12 the behaviors associated with adherence (eg, taking doses at the same time every day, following food restrictions, and not skipping doses as the result of irregularity in routines) remain a challenge, especially for young people living with HIV infection.13
The normal developmental trajectory of adolescence and young adulthood involves behavioral experimentation, risk taking, and confronting a host of difficult choices with regard to romantic relationships, sexual behavior, alcohol and drug use, and identity formation (eg, Arnett, 2004).14 The complexity of these choices is compounded for HIV-infected youth and emerging adults,14 who must negotiate these developmental stages within the framework of having a chronic and stigmatizing disease.15–17 Medication adherence may be particularly challenging at a time of life when adolescents do not want to be different or perceived as different from their peers.10 Moreover, developmental processes, such as concrete thinking,18 may contribute to difficulties in taking medications when adolescents are asymptomatic, particularly if the medications have taxing adverse effects.
Previous reviews of antiretroviral adherence studies in the United States have focused on HIV-infected adults.19–22 This article reviews published adherence studies on HIV-infected youth (ages 13 to 24 years), focusing on rates of adherence to antiretroviral regimens and interventions designed to enhance adherence. Included are possible directions for future research and suggestions for intervention development to improve antiretroviral adherence among HIV-infected youth.
Articles were identified through searches conducted on MEDLINE, PubMed, and PsychInfo using combinations of the keywords HIV/AIDS, youth, adolescents, young adults, adherence (or compliance), nonadherence (or noncompliance), medical treatments, highly active antiretroviral therapy (HAART), anti-retroviral, resistance, and intervention (also keywords associated with specific types of interventions, such as education, telephone, and peer). In addition, bibliographies of relevant articles were reviewed for additional studies.
Included were quantitative and qualitative studies reporting original data on medication adherence among HIV-infected youth (ages 13 to 24 years) and on exercising an intervention technique to enhance antiretroviral adherence among this population. Studies that included children as well as adolescents and young adults were incorporated for review as long as the mean age of participants fell within the 13- to 24-year-old age range; data relevant to adolescents and youth from these studies were reported where available, with the exception of 2 intervention studies23,24 that included data from all participants.
The systematic search yielded 21 articles dating from 1999 to 2008; of the 21 articles reporting data on medication adherence, 7 described unique interventions to enhance adherence among HIV-infected youth. Given the early stage of research in this field, all relevant studies were included in the review, regardless of methodologic rigor. Common methodologic limitations of studies (eg, lack of randomization, lack of control group, or insufficient power) are reported where relevant.
A coding manual was developed to extract descriptive information on setting, study design, population and sample characteristics, definition of adherence used, adherence measurement method, key study variables, and reported findings. In accordance with the approach utilized in prior literature reviews (eg, Fogarty et al, 2002),19 names and definitions of variables were extracted verbatim from study authors, generating a list of 46 variables. A combination of content analysis25 and an iterative process of variable sorting and concept formation common in qualitative research26 was employed to identify 9 categories in which all the variables could be classified. These categories were further refined into 5 broad thematic areas associated with adherence: (1) demographic factors (eg, age, sex); (2) psychosocial factors (eg, family/ caregiver, psychologic/developmental); (3) disease factors (eg, clinical status, disease stage); (4) treatment regimen factors (eg, regimen complexity, adverse effects); and (5) practitioner factors. Intervention components and relevant outcomes are also described.
Variables were often worded in both the positive and negative directions (for example, predictors of adherence and predictors of nonadherence). Findings were classified in 1 of 3 ways: (1) variables statistically significantly associated with adherence, (2) variables statistically significantly associated with nonadherence, or (3) variables inconsistently associated or failing to demonstrate an association with adherence.
Accurate measurement of HIV medication adherence presents challenges to researchers, and few studies are consistent in their classification of adherence.27 Three categories were used in this review to classify how studies measured adherence: (1) subjective measures of adherence based on self-report or others’ report of adherence; (2) pharmacologic measures of adherence (eg, pill count, pharmacy refill records, use of mechanical monitors of pill or drug use); (3) physiological methods or indicators (eg, plasma HIV RNA level below detection limits, CD4+ count, plasma assay results, other laboratory reports).
Five broad thematic areas of factors associated with medication adherence among HIV-infected youth were identified (Table 2). Each is described in detail below.
Age, sex, and race were inconsistently associated with adherence across studies.29–33 For example, 1 study found that younger age was associated with poorer adherence;29 others found no association to either adherence or nonadherence.31,32 With respect to education level and socioeconomic indicators, being in school was associated with better adherence,29 whereas having repeated a grade in school and having unstable housing were each associated with poorer adherence to antiretroviral medications among HIV-infected youth.34
Much of the research on adherence among HIV-infected youths (48%) focused on social and psychologic factors.
Several disease factors related to HIV were associated with adherence, namely undectable plasma HIV RNA28,32 and CD4+ count greater than or equal to 500 cells/μL.32 In contrast, detectable plasma HIV RNA33,40 and later disease stage29 were associated with nonadherence.
Reduced regimen complexity (ie, fewer drugs prescribed)32 was associated with improved adherence. Both physical and psychologic medication-related adverse effects were associated with poorer adherence.28 Notably, length of antiretroviral medication treatment (eg, longer term in years)34 was associated with poorer adherence. Self-assessment of adherence by the patient was also strongly associated with decreased reports of adherence, compared with reports of adherence by a caregiver or medical practitioner.33
Few studies explore practitioner factors in investigating adherence among HIV-infected youth. The only study that examined this relationship41 found that maintaining regular follow-up care and treatment with a medical practitioner was associated with increased adherence.
Seven intervention studies targeting improved adherence to antiretroviral medications among HIV-infected youth were reviewed (Table 3). Of these, 3 utilized directly observed therapy (DOT), in which participants met with a medical practitioner who administered their HIV medication, and involved a retrospective analysis of chart information.23,24,42 Two studies applied regimen-related interventions prospectively that ranged in duration from 12 weeks to 96 weeks,13,43 and 2 studies utilized education and counseling sessions to promote adherence to antiretroviral medications with study periods between 8 weeks and 12 weeks.44,45
Hospital-based DOT interventions were associated with substantial changes in the plasma HIV RNA level and CD4+ count of participants. All studies that used DOT involved a retrospective analysis of information from a variety of clinical sources, including admitted children’s or adolescents’ medical charts, electronic medical records, and a clinic laboratory database. Parsons and colleagues,24 who observed 19 admissions between 2000 and 2003, found the mean plasma HIV RNA level at admission to be copies/mL, at discharge to be 5.7 log10 copies/mL, and at 6 months 4.7 log10 copies/mL. after discharge to be 5 log10 A decrease in plasma HIV RNA level for patients on DOT indicates prior nonadherence when the patient was not under direct supervision. Glikman and colleagues23 reported a statistically significant decrease in mean plasma HIV RNA level (0.8 ± 0.55 log10 copies/mL) for the DOT period; they also reported a rise in mean plasma HIV RNA level when the participants or their caregivers again became responsible for maintaining treatment in the absence of DOT. Purdy and colleagues42 observed 5 admissions, of which 4 had a decrease in plasma HIV RNA level while receiving DOT (range, 0.5–2.46 log10 HIV RNA copies/mL; mean, 1.15 log10 HIV RNA copies/mL). All DOT studies showed that plasma HIV RNA level increased as time after discharge increased, suggesting that hospital-based DOT has a limited effect on adherence among HIV-infected youth, as long-term benefits were not observed.
One of the 2 regimen-related interventions focused on medication scheduling (ie, reduction to once-daily dosing) and evaluated viral load and CD4+ count as outcomes. Sampling 37 therapy-naive individuals, McKinney and colleagues13 evaluated the efficacy of a regimen that included emtricitabine, didanosine, and efavirenz. The median CD4+ count at baseline was 310 cells/mL with an increase to 673 cells/mL by week 96 of the intervention, resulting in a gain of approximately 18% and demonstrating successful viral load decreases and suppression over time.
A second study incorporated the use of cell phone reminder calls to assist HIV-infected adolescents to adhere to their antiretroviral therapy.43 Of the 8 participants recruited, 5 completed the entire 12-week study period of cell phone reminders. Although the intervention technique was reported as “annoying” by participants, the 5 participants who completed the study experienced clinically important decreases in their viral loads (for example, 1 participant had a plasma HIV RNA level of 342,536 copies/mL at baseline and 242 copies/mL at 24-week follow-up).
One of these studies evaluated the efficacy of an 8-week program that involved antiretroviral therapy education via 2 videotapes, information booklets, and a set of 5 audiotapes using the transtheoretical model to increase adherence across stages of change.44 The video- and audiotapes followed a newly HIV-infected youth coming to terms with her condition as she joins a support group in which other HIV-infected youth receiving antiretroviral therapy answer questions, discuss difficult issues, and model stage-specific processes. Of the 18 of 112 participants who completed the program, two-thirds initiated antiretroviral therapy, and half self-reported maintaining adherence “most” to “all of the time.” An important limitation of this study was difficulty retaining participants.
Lyon and colleagues had more success in retention of study participants.45 Initially recruiting 30 pairs of HIV-infected youths between the ages of 15 years and 22 years and a family member or “treatment buddy,” they retained 23 pairs for the final assessment. The program consisted of 12 weeks of education sessions, 6 of which were exclusive to just the HIV-infected youth, and the other 6 of which incorporated all participants. The curriculum focused on the dynamics of HIV; the purpose of antiretroviral therapy; medication choices and managing adverse effects; nutrition, exercise, and alternative treatments; communication with doctors and health care practitioners; and the media.
On alternate weeks, the youths met to discuss issues with medication adherence in a group psychotherapy format. To further help participants adhere to medication, a new device (such as a pillbox, beeper, calendar, multiple-alarm wristwatch, or gym bag) was introduced to the youths at each youth-only session. Upon entry into the program, 43.5% of the 23 HIV-infected youth had a CD4+ count between 200 cells/μL and 499 cells/ μL, and the other 56.5% had a count of less than 200 cells/μL. By the end of the 12-week study period, 17.4% had more than 500 CD4+ cells/μL, 30.4% had between 200 cells/μL and 499 cells/μL, 26% had less than 200 cells/μL, and 13% were deceased. In addition to the positive changes in the CD4+ counts, 91% of the study participants self-reported increased adherence to antiretroviral medication as a result of the group education sessions.
Consistent with the literature on HIV adherence among adults19 and general adherence literature,27 our review of research on HIV-infected youth suggests that individual demographic factors and readily observable patient characteristics failed to distinguish adherent from nonadherent individuals. No consistent, predictive sociodemographic relationships with adherence to antiretroviral medications emerged. In contrast, psychosocial factors such as depression and anxiety were most consistently associated with nonadherence across studies. Continuing to examine adherence within the broader contextual issues present in the lives of youth is essential to understanding how to improve medication adherence and long-term survival for young people living with HIV.
The most promising strategies for improving treatment adherence among HIV-infected youth involve patient and caregiver education, self-monitoring, peer support, and telephone follow-up. Consistent with adult adherence interventions,21,22,27,46 multicomponent strategies tended to be most effective in improving poor adherence. A commonly cited reason for nonadherence to medication among youths is “simply forgetting.”29,38 Interventions that include simple treatment regimens with once-daily dosing13 seek to address this barrier to adherence.
However, once-daily dosing provides other challenges in a population with adherence difficulty. For example, missing a once-daily dose means 24 hours without medications, whereas missing 1 dose of twice-daily regimens means only 12 hours uncovered. Hosek and colleagues note that non-adherence relates more to difficulty incorporating the medication regimen into patient lifestyle than to regimen complexity itself.38 Thus, interventions might consider skill building around taking medications during a specific time that is integrated into a routine behavior, such as after brushing one’s teeth or eating breakfast.
Findings suggest that providing DOT, while considered impractical for all youth because of its cost, might be important for selected adolescents infected with HIV,10,32 such as those with active substance use disorders. To date, no studies have examined the use of multidisciplinary treatment teams (eg, teams with case managers, physicians, nurses, psychologists) versus physicians alone in working with HIV-infected youth.47 However, multidisciplinary treatment teams are more successful than physicians alone in providing and implementing successful adherence interventions among adults.27,48 Future research with HIV-infected youth may benefit from investigating intervention delivery mode (eg, team treatment vs individual treatment).
An estimated 2% to 6% of US youth have a depressive disorder, and approximately 15% have elevated depressive symptoms.49,50 Among HIV-infected youth, elevated levels of psychologic distress have been documented, with rates of depressive symptoms ranging from 18% to 45%.15,50 Evidence from studies in adults demonstrates the effectiveness of treating depression as a means of improving adherence.51,52 Findings that depressive symptoms are strongly associated with nonadherence among HIV-infected youth29,32,33,39 suggest that treatments for adolescent depression may assist in improving medication adherence. A recent study demonstrated that treating HIV-infected adults with cognitive behavioral therapy for depression and adherence skill building effectively reduced depression over time and improved medication adherence.53
Limitations bear mention when interpreting review findings. Given the early stage of research in this field, all relevant studies were included in our review, regardless of methodologic rigor. The few studies conducted to date, small sample sizes, and paucity of research on specific subgroups (eg, gay and lesbian youth, racial or ethnic minority youth) limit generalizability. In addition, studies often presented adherence and nonadherence as opposing or opposite constructs. However, findings suggest that distinguishing between and understanding the differences between these concepts may yield valuable insight as to the roles that diverse factors play in adherence and may be especially productive in helping develop new interventions. Thus, differences between adherence and nonadherence should not be reduced but should instead be expanded and each concept thoroughly investigated.
Given the range of complex factors associated with nonadherence, different sets of targeted interventions may be warranted that focus on specific populations of youth (eg, homeless, sexual minorities, substance abusing). Randomized controlled trials are needed that incorporate solid theoretic frames, satisfactory sample sizes, psychometrically sound outcome measures, consistent operationalization of adherence, and better adherence assessment measurements.22 Cost-effectiveness data to assess the practical value of different adherence interventions in the long-term would be beneficial.22,54 Finally, because the psychosocial needs of HIV-infected persons are changing to more closely resemble the needs of the chronically rather than the terminally ill individual,15 investigating how psychosocial issues such as distorted body image, substance use, anxiety, history of childhood sexual abuse, and influence of peer norms relate to antiretroviral adherence may be particularly crucial to promoting long-term survival and quality of life among HIV-infected youth.15–17,55 Secondary HIV prevention interventions54,56 may provide useful means of not only reducing HIV transmission through sexual risk taking but also improving health outcomes through the incorporation of strategies to increase antiretroviral adherence.
This review indicates that more research on adherence among HIV-infected youth, as well as more rigorously evaluated interventions, are needed. Maximizing adherence may not only be fundamental to the well being of HIV-infected youth but may also have a far-reaching and broader impact on public health.12 Nonadherence may lead to drug resistance and cross-resistance that may render HIV treatments ineffective and may be implicated in the emergence of drug-resistant strains of HIV.9,57,58 Consequently, gaining a more thorough and contextualized understanding of factors associated with adherence and nonadherence, including individual demographic, social and psychologic, disease-related, treatment-regimen, and practitioner factors, represents an important step in helping people live longer and in intervening to address infectious disease rates. Further culturally tailored, intervention development research for HIV-infected youth is warranted.
The authors thank Jayson Caracciolo, MPH, for editorial help with this project while interning at the Fenway Institute, Fenway Community Health.
Funding/Support: Funding for investigator time for this article came in part from grant R03DA023393 to Dr Mimiaga from the National Institute on Drug Abuse, grant 1F31AA017338-01 to Ms Skeer from the National Institute on Alcohol Abuse and Alcoholism, and grant R01MH084757 to Dr Safren from the National Institute of Mental Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Mr. Sari L. Reisner, Senior Research Associate for Epidemiology and Behavioral Science Studies at the Fenway Institute, Fenway Community Health.
Dr. Matthew J. Mimiaga, Research Scientist at the Fenway Institute, Fenway Community Health, and Instructor in Psychiatry at Harvard Medical School/Massachusetts General Hospital.
Ms Margie Skeer, Research Analyst at the Fenway Institute, Fenway Community Health, and doctoral candidate at Harvard School of Public Health.
Mr. Brandon Perkovich, Undergraduate student at Harvard College.
Mr. Carey V. Johnson, Project Manager for International Research at the Fenway Institute, Fenway Community Health.
Dr. Steven A. Safren, Senior Research Scientist at the Fenway Institute, Fenway Community Health, Director of Behavioral Medicine at Massachusetts General Hospital, and Associate Professor of Psychology in the Department of Psychiatry at Harvard Medical School.