Social-support needs among adolescents living with HIV in transition from pediatric to adult care in Cambodia: findings from a cross-sectional study

This cross-sectional study was conducted in August 2016 among 328 adolescents living with HIV, randomly selected from 11 ART clinics throughout Cambodia. The details of the study have been reported elsewhere [34]. Briefly, we applied a two-stage sampling design to identify 598 eligible participants from the 18 major ART clinics countrywide. Twenty-one clinics with less than 10 adolescents living with HIV each were excluded to minimize cost. The sample size was set at 320, calculated to estimate a high level of preparedness of 60% with a confidence interval (CI) of 95% and design effect of 1.4 using a probability proportional to size method [35]. Eligible subjects were selected using a random number table.

Selected adolescents were screened via telephone and informed of research objectives as well as place, date and time of the interview. Adolescents would be asked to participate if they were between 15 and 17 years of age, receiving treatment and care services, able to communicate in Khmer, allowed by a parent or guardian to participate, their parent or caregiver was willing to provide a written consent, able to present themselves on the day of the interview and physically and mentally stable to provide verbal assent to participate. Upon confirming eligibility, a written consent from their parent or guardian was obtained. Recruitment continued until the required sample size for each clinic was achieved.

Questionnaire development was guided by existing literature [36, 37]. Information on socio-demographic characteristics, perceived health status, school attendance, disclosure, stigma experience, psychosocial support and medication supply and ART adherence were collected through self-reported interviews. Clinical and immunological data, including HIV status, ART history and laboratory values for CD4 and viral loads were obtained from medical records.

Data were collected by two teams of interviewers and moderators who were trained on study protocol, questionnaire administration, confidentiality and issues related to research with human participants [38]. Pre-testing was conducted among 20 adolescents living with HIV and 10 parents and guardians, and questionnaire adjustments were made.

Data were classified and scaled continuously or categorically as appropriate and recorded into a computerized database using EpiData version 3 (Odense, Denmark). Double data entry was implemented to minimize errors. Descriptive analyses were performed to determine frequency distribution for categorical variables and mean with standard deviation (SD) for continuous variables. Chi square, Fisher’s exact or t-test tests for significance were used to compare characteristics of male and female adolescents. p-values of less than 0.05 were considered statistically significant.

The study was approved by the National Ethics Committee for Health Research, Ministry of Health in Cambodia (Ref: 297NECHR). Assent from participants and a written informed consent from a parent or guardian was obtained. Data were collected in private locations, and confidentiality protected by removing all personal identifiers from questionnaires. Participants were provided with a token of $2.5–$5 for transportation.

Of the 328 adolescents, 51.3% had completed secondary (7–9 years) school, less than half (42.7%) were living with one or both parents, and 55.2% were male. Parents (40.8%) or relatives (33.6%) were the most frequently reported caregivers. More than half of respondents had at least one deceased parent (mother 50.9%; father 60.5%). Maternal and paternal educational levels were both 20% for high school (10–12 years) or university education (Table 1).

On average, ART refills were dispensed every two months. Forty-five participants (13.7%) reported having to return earlier than their scheduled appointment due to inadequate supply of medicines (22.2%) or the need for an additional check-up (46.7%). A small number (2.7%) had been asked to purchase their own drugs. Notably, sharing of ART was not reported (Table 2).

Thirty-two percent reported negative attitudes or being mistreated (10.7%) because of their HIV status; 8.2% had been denied housing or food (Table 3).

Within the past month, 22.9% experienced disruptions with attendance at school, and 16.8% reported that they were no longer enrolled. Being too ill (34.1%), lack of fees (14%) or poor academic performance (18.6%) were the most frequently reported reasons for being absent. Only a fraction (0.6%) of the 69 adolescents currently working for pay sought employment outside of their hometown. Almost half (46.6%) reported that their family had received social support for their health care, including food support (76.5%), school allowance (62.1%), transport allowance to ART clinic (53.6%), emotional counseling (35.3%), vocational training (22.9%) and home visit (11.1%). Thirty percent belonged to a peer support group (Table 4).

The majority (72.0%) of respondents were able to recognize when they were getting sick, and 80.5% knew when they needed to call a health care provider. However, less than half (46.0%) reported being the one responsible for scheduling follow-up visits, and 13.4% indicated that they felt comfortable asking health questions during appointments. An approximately equal number of males and females reported some degree of difficulty with ART adherence. The average visual adherence scale score between the two groups was 95.4% (SD = 9.6%) (Table 5).

Significantly, females were less likely to retain copies of their health records and doctor contacts (32.7% vs. 57.4%, p < 0.001), less likely to report having a method for organizing HIV-related appointments (47.9% vs. 72.8%, p < 0.001) and more likely to report that they would discontinue ART when feeling ill (7.1% vs. 15.1%, p = 0.02) (Table 5).

About one-third (29.6%) had received counseling on the transition to adult services. However, only 2.7% of them had completed a transfer form, and 19.7% had visited an adult ART clinic to prepare for the transition. About one-forth (24.7%) reported that a case manager had been identified to support them during the transition. The majority of them felt supported during the preparation process for transition with 14.5% feeling very supported, and 59.5% felt somewhat supported. Sixty-six percent preferred to receive pediatric HIV-related care. Issues related to health, sexuality or daily life were most frequently discussed with family (47.9%), health providers (20.4%) and friends (13.4%). However, health care providers (79%) were reported as the most trusted source of information for HIV care and treatment (Table 6).

Males were significantly more likely to discuss health problems with medical providers (26.6% vs. 13.8%, p = 0.01), and they were also more likely to trust friends or family members the most for their treatment (14.8% vs. 6.9%, p = 0.02). Preparation strategies among males were more likely to include a visit to an adult clinic (34.9% vs. 21.4%, p = 0.007) despite having a lower likelihood of being assigned a case manager (33.7% vs. 49.7%, p = 0.003) or of including counselors or peer-educators in their transition plans (1.8% vs. 6.9%, p = 0.02). Overall, male participants reported feeling more prepared for the transition to adult services (96.5% vs. 80.5%, p < 0.001), despite being more likely to report feeling inadequately supported (77.6% vs. 92.4%, p = 0.006) (Table 6).

The proportion (22%) of adolescents working for pay is concerning and may explain the lower school attendance rate in our study population. Education exerts a positive influence on adolescents living with HIV [24] and is a considerable protective factor against poor mental outcomes, risky sexual behavior and substance abuse [40, 41]. Our findings offer additional rationale to improve school retention by confronting discrimination in schools and normalizing HIV. HIV-related stigma is a primary determinant in adolescent health and is also a decisive motivator in the transition process [42]. In addition, expansion of peer support to the two-thirds of adolescents who did not belong to such groups is needed, as participation in peer support is associated with improved medical adherence [24, 43]. Given that access to available social protection support was suboptimal, our results argue for adolescent-focused interventions at multiple layers of the sociological context [24] to ensure that every needy adolescent has access to available wrap around services to support successful adolescent transition. In addition, future analyses are needed to explore the socioeconomic factors driving these students out of school and into the workforce.

An inadequate drug supply was common which may affect ART adherence and partially contribute to the high rate of school absences. Improved medication adherence is associated with a daily routine [39], which is reliant upon a consistent supply of medication. Moreover, the transition from pediatric to adult care has the potential for being a significant psychological burden [44] on adolescents living with HIV, which requires adequate supply and adolescent-friendly services, devoid of stigma. In the current context, strengthening supply of ARV will be essential. At the same time, there is a need to strengthen the provision of counselling on transition and enhance adolescent familiarization with adult clinics as part of the transition. Furthermore, despite policies requiring allocation of a transition case manager, only a quarter of adolescents in this study had been allocated one. This will also need to be enhanced in order to support successful transition of adolescents to adult HIV care. Transition managers provide and link adolescents to available social support and protection.

Our study offers a picture of the HIV epidemic among Cambodian adolescents and the sociocultural phenomena they experience while transitioning from pediatric to adult HIV care. However, and despite our sampling methodology, the generalizability of our results to the entire population of adolescents living with HIV in Cambodia may be limited. Potential for bias due to social desirability or recall biases were minimized by validating responses against medical records whenever possible, but may not have been eliminated.