Skip to main content
. 2015 Feb 26;18(2Suppl 1):20076. doi: 10.7448/IAS.18.2.20076

Table 1.

Studies conducted between the years 1999 and 2014 on adherence to ART and retention in care among BIY and YKP

Publication (first author, year) Location (cities, country) Study populations Age (mean, range) Sample size (HIV-positive) Measurement of adherence to and/or retention in HIV care Method Intervention Main findings
Belzer et al. 1999 [20] Los Angeles, USA BIY 15–24 31 Self-reported adherence Quantitative (survey) No

  • Medication adherence most significantly correlates with stability of living conditions in BIY.
Martinez et al. 2000 [21] Cook county, Illinois, USA BIY 13–21 25 Self-reported adherence Quantitative (retrospective analysis of medical charts) No

  • 61% of subjects reported >90% compliance with their medications in the previous 90 days.

  • 5 of 10 substance abusing subjects reported adherence to ART.
Murphy et al. 2001 [22] 13 cities in USA BIY 12–19 161 Self-reported adherence, psychological theory to measure anxiety, social support and depression. Viral load (VL). Quantitative (cohort study) No

  • Higher levels of adherence associated with decreased depression, a strong association between adherence and reduced VL.
Rogers et al. 2001 [23] USA BIY N/Aa 288 Viral load and CD4 count Quantitative (evaluation) Yes

  • Only 18 of 288 participants received full TREAT programme, which led to adherence with ART.
Murphy et al. 2003 [24] 13 cities in USA BIY 12–19 114 Self-reported adherence and viral load (HIV-1 RNA level in plasma) Quantitative (survey) No

  • Viral load was significantly associated with self-report of adherence to ART.

  • Only 28.3% of adolescents reported taking all of their prescribed antiretroviral medications in the previous month.
Flynn et al. 2004 [25] 28 sites in the US and Puerto Rico in USA BIY 8–22 120 Self-reported adherence and viral load Quantitative (cohort study) No

  • Adherence to ART was the only predictor of achieving undetectable virus loads.
Murphy et al. 2005 [26] 13 cities in USA BIY 18.4, 12–18 231 Self-reported adherence, behavioural factors associated with adherence and viral load Quantitative (survey) No

  • Adolescents in the later HIV disease stage were less likely to be adherent compared with those in the earlier disease stage.

  • Less alcohol use and being in school were associated with adherence by adolescents on weekends and over the preceding month.
Puccio et al. 2006 [27] Los Angeles, USA BIY 16–24 81 Self-reported adherence Quantitative (pilot intervention study) Yes

  • Most participants found the calls to be helpful and the level of intrusion into their daily lives acceptable.

  • Using cell phone reminders to assist patients does not require an extensive amount of daily staff time.
Naar-King et al. 2006 [28] USA BIY 16–24 24 Self-reported adherence, self-efficacy, social support, psychological distress Quantitative (survey) No

  • Self-efficacy and psychological distress were significantly correlated with adherence but social support was not.

  • Social support specific to taking medications was correlated with self-efficacy.
Rao et al. 2007 [29] Chicago, USA BIY 17–25 25 Self-reported adherence Qualitative No

  • Half of respondents indicated that they skipped doses because they feared family or friends would discover their status, suggesting that HIV stigma impacts treatment for youth on several levels, from the accuracy of communication with medical providers to medication adherence, subsequent health outcomes and the emergence of treatment-resistant strains.
Rudy et al. 2009 [30] USA BIY and blood products. Separate sexual abuse category 12–24 396 Survey instrument to measure adherence and outcome expectancy of adherence Quantitative (observational study) No

  • Non-adherence influenced by not having healthcare insurance, dropped out of school, homelessness and/or spent time in detention facility.
Garvie et al. 2010 [31] Mid-southern USA BIY, blood transfusion and unknown 16–24 60 Routine pharmacy pill count and self-reported. CD4 and VL. Quantitative (survey) No

  • The first study to measure adherence measurement based on both CD4 and VL.

  • Non-adherence was related to off-schedule dosing.
Magnus et al. 2010 [32] Bronx, Chapel Hill, Chicago, Detroit, Houston, Los Angeles, Oakland, Rochester, USA AA, Latino YMSM 16–24 224 Retention defined as programme visits every three months Quantitative (cohort study) No

  • Retention associated with <21 years old, history of depression, receipt of programme services, feeling respected at clinic.
Comulada et al. 2003 [33] Los Angeles, USA BIY 14–29 253 Self-reported adherence, health status, sexual behaviour, substance use and psychological measures Quantitative (survey) No

  • Almost all youth had been offered ART (84%); 77% had ever used it, 54% were currently using and 63% of users adhered to 90% of their medications.

  • Compared to non-users, users were more likely to be female, Latino or AA.
Agwu et al. 2011 [34] 17 US Clinic sites BIY 18–24 3127 Self-reported adherence and clinic visits Quantitative (retrospective study) No

  • Youth PLHIV less likely to report injecting drug use behaviour.

  • They were less likely to initiate ART.
Tapp et al. 2011b [19] Vancouver, Canada YPWID <24 PWID <24 (n=24), N=545 Adherence measured by compliance to prescription refill Quantitative (cohort study) No

  • Younger age (<24), being female, daily heroin injection and daily cocaine injection were negatively associated with 95% adherence while methadone treatment was positively associated with adherence.
Hadland et al. 2012b [35] Vancouver, Canada YPWID Median=37.2, age was dichotomized at 29 545 Self-reported adherence, VL Quantitative (cohort study) No

  • Young adults (age <29) were less adherent and were less likely to achieve VL suppression.
Wohl et al. 2011 [36] Los Angeles, USA AA and Latino YMSM 18–24 61 Retention associated with number of intervention visits, prescription of ART Quantitative (pilot intervention study) Yes

  • Highlights the critical needs of HIV-positive AA and Latino YMSM and demonstrate that a clinic-based YCM can be effective in stabilizing hard-to-reach clients and retaining them in consistent HIV care.
Hightow-Weidman et al. 2011 [37] North Carolina, USA AA and Latino MSM Mean age 21 81 Retention defined as 1 medical visit every four months Quantitative (cohort study) Yes

  • Interventions on adherence need to actively reach out to youth populations.
Bouris et al. 2013 [38] Chicago, USA AA YMSM and TG 16–29 94 Self-reported adherence, VL Quantitative (RCT) Yes

  • Supportive relationships promote retention in care.
Barnes et al. 2013 [39] Baltimore, New York City, Washington, USA BIY, PIY 13–21 166 Assessed HIV knowledge Quantitative (survey) Yes

  • BIY outperformed PIY on questions related to disease awareness.
Gillman et al. 2013 [40] Houston, USA AA YMSM Mean 19.9 47 Retention in care defined as completion of physician visits 90 days after linkage to care Quantitative (survey) No

  • Greater conspiracy beliefs were associated with negative medication attitudes while trust in physicians was correlated with positive medication attitudes; conspiracy beliefs were not associated with poor linkage to care and retention.
Harper et al. 2013 [41] 14 cities in USA YMSM (66% AA, 19% Latino) Mean 21.5, range 16–24 200 Self-reported adherence to medical appointment in the past three months Quantitative (survey) No

  • Ethnic identity affirmation and HIV-positive identity were associated with significantly higher risk for missed appointments in the past three months.
Belzer et al. 2013 [42] Los Angeles, Washington, New Orleans, Fort Lauderdale, San Francisco, USA BIY, YMSM 15–24 37 Self-reported adherence (dichotomized at 90%), viral load data abstracted from medical record Quantitative Yes

  • Intervention of daily cell phone conversation with health care providers.

  • Self-reported adherence was significantly higher in intervention group than in the control group.
Saberi et al. 2014 [43] USA BIY, PIY 12–24 1317 Self-reported adherence in the past seven days (dichotomized at 100%); plasma HIV RNA Quantitative No

  • Pillbox was the most endorsed adherence device.

  • Using adherence devices was inversely associated with having undetectable viral load.

  • BIY more likely to be gay, adherent to ART and never been to jail.
Hussen et al. 2014 [44] Atlanta, USA YMSM 13–24 20 Self-reported adherence Qualitative No

  • Successful transition to adulthood and optimal ART adherence were inextricably linked.

  • Detrimental impact of HIV on development was moderated by the degree of physical illness at diagnosis.
a

Only specify participants as from REACH project

b

these two studies were conducted on the same cohort.

AA=African American; ART=antiretroviral therapy; BIY=behaviourally infected youth and adolescents including sexual behaviour and injecting drug use; HAART=highly active antiretroviral therapy; HIV=human immunodeficiency virus; PIY=perinatally infected youth and adolescents; RCT=randomized control trial; REACH=Reaching for Excellence in Adolescents Care and Health; TG=transgender; TREAT=Therapeutic Regimens Enhancing Adherence in Teens; YCM=youth-focused case management; YMSM=young men having sex with men; YPWID=young people who inject drugs.