Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

High Risk of ART Non-Adherence and Delay of ART Initiation among HIV Positive Double Orphans in Kigali, Rwanda

  • Kimiyo Kikuchi,

    Affiliation Department of Community and Global Health, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

  • Krishna C. Poudel,

    Affiliation Department of Community and Global Health, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

  • John Muganda,

    Affiliation Division of Obstetrics and Gynecology, King Faisal Hospital, Kigali, Rwanda

  • Adolphe Majyambere,

    Affiliation HIV/AIDS, STIs and Other Blood Borne Infections Division, Institute of HIV Disease Prevention and Control, Rwanda Biomedical Center, Kigali, Rwanda

  • Keiko Otsuka,

    Affiliation Department of Community and Global Health, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

  • Tomoko Sato,

    Affiliation Department of Psychology, The Meiji Gakuin University, Tokyo, Japan

  • Vincent Mutabazi,

    Affiliation HIV/AIDS, STIs and Other Blood Borne Infections Division, Institute of HIV Disease Prevention and Control, Rwanda Biomedical Center, Kigali, Rwanda

  • Simon Pierre Nyonsenga,

    Affiliation HIV/AIDS, STIs and Other Blood Borne Infections Division, Institute of HIV Disease Prevention and Control, Rwanda Biomedical Center, Kigali, Rwanda

  • Ribakare Muhayimpundu,

    Affiliation HIV/AIDS, STIs and Other Blood Borne Infections Division, Institute of HIV Disease Prevention and Control, Rwanda Biomedical Center, Kigali, Rwanda

  • Masamine Jimba,

    Affiliation Department of Community and Global Health, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

  • Junko Yasuoka

    Affiliation Department of Community and Global Health, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan



To reduce HIV/AIDS related mortality of children, adherence to antiretroviral treatment (ART) is critical in the treatment of HIV positive children. However, little is known about the association between ART adherence and different orphan status. The aims of this study were to assess the ART adherence and identify whether different orphan status was associated with the child’s adherence.


A total of 717 HIV positive children and the same number of caregivers participated in this cross-sectional study. Children’s adherence rate was measured using a pill count method and those who took 85% or more of the prescribed doses were defined as adherent. To collect data about adherence related factors, we also interviewed caregivers using a structured questionnaire.


Of all children (N = 717), participants from each orphan category (double orphan, maternal orphan, paternal orphan, non-orphan) were 346, 89, 169, and 113, respectively. ART non-adherence rate of each orphan category was 59.3%, 44.9%, 46.7%, and 49.7%, respectively. The multivariate analysis indicated that maternal orphans (AOR 0.31, 95% CI 0.12–0.80), paternal orphans (AOR 0.35, 95% CI 0.14–0.89), and non-orphans (AOR 0.45, 95% CI 0.21–0.99) were less likely to be non-adherent compared to double orphans. Double orphans who had a sibling as a caregiver were more likely to be non-adherent. The first mean CD4 count prior to initiating treatment was 520, 601, 599, and 844 (cells/ml), respectively (p<0.001). Their mean age at sero-status detection was 5.9, 5.3, 4.8, and 3.9 (year old), respectively (p<0.001).


Double orphans were at highest risk of ART non-adherence and especially those who had a sibling as a caregiver had high risk. They were also in danger of initiating ART at an older age and at a later stage of HIV/AIDS compared with other orphan categories. Double orphans need more attention to the promote child’s adherence to ART.


Worldwide, the AIDS epidemic puts children at grave risk. The Millennium Development Goals (MDGs) call for stopping the spread of HIV/AIDS by 2015, but progress has not been as fast as expected and approximately 370,000 children under 15 years old were newly infected with HIV in the world in 2009 [1]. In particular, HIV infection has severely hit Sub-Saharan African countries, where the number of HIV positive children represents 90% of all HIV infected children in the world [1]. In this region, HIV/AIDS has been the main cause of the highest under-five mortality rate [2], [3].

HIV infection among orphaned children constitutes an important issue in Africa. According to previous studies, orphans aged under 15 who have lost one or both parents constitute an especially vulnerable group of HIV positive children. Compared with non-orphans, orphans suffer from poorer health and nutritional status [4][7]. Furthermore, orphans initiate their antiretroviral treatment (ART) later than non-orphaned children, and they are more likely to be diagnosed as WHO Clinical Stage 4 (most severe HIV infection stage) prior to ART initiation than non-orphans [8].

To reduce children’s HIV/AIDS related mortality and progression, early ART initiation and high levels of ART adherence are particularly critical [9][12]. However, promoting high ART adherence in children is difficult due to various factors which derive from children’s developmental immaturity, their particularity of eating habits, and a lack of knowledge about their disease. Moreover, the adherence of children, especially younger children, is largely dependent on their caregivers, which makes children’s adherence matters more complicated than in the case of HIV positive adults. Well-known barriers identified in quantitative studies in Sub-Saharan Africa include: caregivers' perceived difficulty for administrating ART, child does not know his/her sero-status, and older age of child [13][15]. On the other hand, the relation between adherence and different orphan status (non-orphan, maternal orphan, paternal orphan, and double orphan) needs to be examined more. In most adherence related studies, orphan status was classified as ‘orphan’ or ‘non-orphan’ [16], [17]. This binary categorization may be valid in study setting where single orphans and double orphans were favored with similar social and economic conditions. However, in most resource-limited settings, the death of two parents had a much greater additive effect on the caregiving situation [9]. In addition, caregivers’ support to a child may be different between maternal orphans and paternal orphans due to the difference in each parent’s relationship with the child following their social and cultural setting.

HIV positive orphans are key priorities in Rwanda’s national HIV/AIDS response. In 2009, the number of children aged 0–14 living with HIV was estimated at 22,000 [11,000–34,000], which accounted for 12% of Rwanda’s total HIV positive population [18]. The estimated number of children aged 0–17 orphaned due to HIV/AIDS was 130,000 [98,000–180,000] in 2009 [18], and a large number of these orphans were considered to be infected through a mother-to-child transmission (MTCT). Among all HIV positive children who needed ART in Rwanda, 6,558 had initiated ART by the end of 2009 [19]. This represents 60% of the pediatric coverage, which ranks one of the highest in African countries [1]. However, no official study has been conducted in Rwanda about ART adherence which focuses on HIV positive children.

The objectives of this study were to assess the ART adherence among HIV positive children in Kigali and to identify whether different orphan status was associated with the child’s ART adherence.


Operational Definition

We defined caregiver as a person whom the child turned to when he/she needed to take medication. Caregivers of the child were identified by self-assessment before the interview. Caregivers may include the child’s parents, siblings, grandparents, other relatives, or guardians who directly influenced the child’s medication dosing.

As for orphan status, a double orphan was defined as a child who had no living parents or had no parents whose living status was known. A maternal orphan/paternal orphan was defined as a child whose mother/father had died or they didn’t know their living status. The cutoff point of the child’s age was set at 15 following the definition of UNICEF [20], [21] and that of pediatric patients in Rwanda.

Study Design and Site

We carried out a facility-based, cross-sectional study between March and July 2011 at selected health facilities in Kigali, Rwanda. We counted remaining pills to assess the adherence rate of children currently enrolled in ART programs. To collect data about associated factors to adherence, we also conducted a questionnaire survey for primary caregivers, measured children’s weight/height, and collected clinical records of children. We choose Kigali city as the study site since the city enrolls a high number of pediatric ART patients (n = 1558) and runs a high percentage (23.8%) of Rwanda’s pediatric ART services [19]. To recruit participants from various facility levels, we used two sampling methods. First, all four public hospitals and one clinic in Kigali were selected since they enroll almost half of all pediatric patients in Kigali (TRAC-Plus Clinic, Kigali University Hospital Center, Muhima District Hospital, Kigabagaba District Hospital, and Kanombe Military Hospital). Second, out of 21 health centers which provide a routine pediatric ART service, we randomly selected 10. They were: Kimironko, Kinyinya, Kicukiro, Kabusunzu, Cor-unum, Kacyiru, Gitega, Gikondo, Masaka, and Biryogo.

Study Population

The study population was children enrolled in the ART program in the 15 selected health facilities and their primary caregivers. Inclusion criteria for children in the study were: having been under ART for at least 12 weeks, and were aged six months to 14 years. The exclusion criteria were: under only cotrimoxazole prophylaxis. The caregivers were those who were caring for children who matched the above criteria and were over 18 years old.


Child’s adherence to ART.

To measure ART adherence, we counted the remaining pills or syrups which participants presented on the day of the interview [22], [23]. Pill count measurement was used because of its objective and quantifiable advantage [24]. From client files, we collected data about the prescribed regimen, the last prescribed date, the pill quantity prescribed per day, and the number of days prescribed. Adherence was calculated based on the formula: [(pills dispensed/pills prescribed per day)/days between refills]×100. This equation has been used in prior studies to evaluate adherence to ART medication [25]. A child was defined to be adherent if he/she took 85% or more of the monthly prescribed doses. Those who were under 85% were defined non-adherent. This rate was estimated as a poor adherence according to the WHO guideline for adherence [16], [26], and more than a 85% adherence rate was reported as the minimum level to achieve HIV virus suppression and have clinical benefit [27]. For those who were taking multiple medications, a child was defined adherent if his/her adherence rate was at least 85% for each medication.


We collected data from caregivers using a structured pre-tested questionnaire which was adopted from different studies [28][30]. The contents of the questionnaire included: socio-demographic characteristics of children and their primary caregivers, economic status of caregivers (income per month, number of rooms, water source, toilet facility, and electricity equipment), caregiver-child relationship, and caregivers’ knowledge about ART. To enhance the accuracy of translation, the questionnaire was originally developed in English, then translated into Kinyarwanda, and translated back into English. The questionnaire items were reviewed by three Rwandan experts on HIV/AIDS who were based in different health formations in Kigali. Prior to the survey, we provided two-day training for investigators to help them understand each question. The questionnaire was pre-tested for 20 caregivers and revised accordingly.

Table 1. Child Participant Characteristics Overall and Stratified by Orphan Status (N = 717).

Caregiver-child relationship was measured using two of the eight scales from the Parent-Child Relationship Inventory (PCRI) [29]: communication and involvement. The PCRI was originally developed to evaluate the relationship between a parent and a child. However, it has also been used for measuring the caregiver-child relationship in a previous study which certified its validity and reliability [28]. Each score was converted to T-scores, and T-scores less than 40 were classified as low communication and involvement according to the PCRI guideline. Caregiver’s knowledge was assessed using a true/false self-report measure [30]. Caregiver’s knowledge level was estimated by the number of items which were correctly answered.

Table 2. Caregiver Participant Characteristics Overall and Stratified by Orphan Status (N = 717).

Disease and growth status related data.

We collected children’s latest medical records from client files, such as their CD4 counts, WHO Clinical Stage, as well as the date of the first HIV test and ART initiation. We also measured weight (kg) and height/length (cm) in all children. To measure them, we used portable electronic scales and metal measuring rods which were calibrated to 0.1 kg, 0.1 cm, respectively. Z-score of weight-for-age (WAZ), height-for-age (HAZ), and BMI-for-age were calculated using WHO AnthroPlus version 1.0.4 (WHO, Geneva. 2007). Undernutrition and stunting were determined using cut-off points of <−2 Z-score. To calculate WAZ and undernutrition, only children under 10 years old were included due to the data availability of AnthroPlus.

Table 3. Disease and Growth Status Related Factors Stratified by Orphan Status (N = 717).

Data Analysis and Processing

For data analysis, we carried out descriptive analysis for each orphan status to explore all variables of the children and the caregivers. We used the chi-square test for nominal scale variables. ANOVA was used for interval variables. To examine associations between the exposure variables and non-adherence, we conducted a multivariate logistic regression analysis. Variables which showed p-value <.25 in the bivariate analysis were included the model [31]. We also entered additional variables which were associated with ART adherence in previous studies such as gender and age of child and caregiver, relation of caregiver to the child, caregiver’s sero-status, child’s duration on ART, and child’s awareness of sero-status. Statistical significance was set at p<.05. Data entry and analysis were executed using PASW Statistics version 18.0 (SPSS Inc., Chicago, IL).

Table 4. Factors related to ART Non-Adherence among All Participants (N = 717).

Table 5. Factors related to ART Non-Adherence among Double Orphans (N = 113).

Ethical Consideration

We obtained ethical clearance from the National Ethics Committee of Rwanda and the Ethics Committee of the University of Tokyo. An official agreement of data sharing was issued from the Institute of HIV Disease Prevention and Control/Rwanda Biomedical Center to carry out the data collection in each facility.

We started interviews after obtaining participants’ written consent. As the children are non-emancipated, written consent was carried out at two levels. First, the caregivers gave written consent. Then, when the children assented to participation, the caregivers signed an informed consent in place of the children.


Participant Characteristics

Out of 1,301 children under an ART program in the 15 study sites, 767 children and the same number of their caregivers were available during our interview period (Table 1). Among them, 50 participants were excluded from analysis due to uncountable remaining pills caused by mixing with other pills (n = 22), incomplete data (n = 17), and/or missing client file (n = 11). Finally, 717 were included in our study.

Socio-demographic Characteristics of Children

Out of the total, double orphans, maternal orphans, paternal orphans and non-orphans were, 113 (15.8%), 89 (12.4%), 169 (23.6%), and 346 (48.3%), respectively (Table 1). Approximately half of the children were males (50.7%), and the mean age of children was 9.3 (SD 3.6) years old. Regarding duration under ART, the mean was 3.7 (SD 2.0) years. Those who knew their sero-positive status (p<0.001) were fewer among non-orphans than other orphan categories.

Socio-demographic Characteristics of Caregivers

Of all caregivers, 85.6% were female and about three-fourths (74.1%) were HIV positive (Table 2). The mean age was 39.3 (SD 10.5) years old and the mean duration of child caregiving was 7.9 (SD 10.5) years. The relation of caregiver to the child differs significantly between the four orphan status (p<0.001). The majority of caregivers were the child’s parent among non-orphans and paternal orphans, whereas among maternal orphans’ and double orphans’ caregivers varied. Fewer paternal orphans’ caregivers had other caregivers who helped with child care (p<0.001). Regarding economic status, we found that double orphans were in a higher economic status and paternal orphans were in a lower economic status through the results of number of rooms in a household (p<0.001), water source (p<0.001), toilet facility (p = 0.002), and electricity equipment (p<0.001). Involvement of caregivers in child care was rather lower among double orphans (p = 0.002), while communication between the caregivers and the child was lower among non-orphans (p<0.001).

ART Adherence and Disease/growth Status Related Characteristics of Children

Of all 717 children, almost half (50.5%) were taking 85% and more of their prescribed doses (Table 3). Those who didn’t attain 85% were categorized as the group of ART non-adherent. The non-adherent rate of double orphans, maternal orphans, paternal orphans and non-orphans were 59.3%, 44.9%, 46.7%, and 49.7%, respectively.

CD4 count at first examination (p<0.001) differed significantly between the four orphan status groups. Moreover, double orphans were more likely to be older at the first sero-status detection (p<0.001), as well as at ART initiation (p<0.001). According to the responses related to how children’s sero-status was detected, double orphans were more likely to be detected by the child’s frequent illness (p = 0.006), whereas non-orphans were more likely to be found by the PMTCT (prevention of mother-to-child transmission) program (p<0.001). Children’s growth status did not differ significantly across the orphan status.

Factors Associated with ART Adherence

The multivariate analysis indicated that maternal orphans (AOR 0.31, 95% CI 0.12–0.80), paternal orphans (AOR 0.35, 95% CI 0.14–0.89), and non-orphans (AOR 0.45, 95% CI 0.21–0.99) were less likely to be non-adherent to ART compared to double orphans (Table 4). Children who were prescribed three or more daily pills (AOR 1.56, 95% CI 1.09–2.23) and those whose growth was stunted (AOR 1.46, 95% CI 1.01–2.09) were more likely to be ART non-adherent. In addition, children whose caregiver scored lower on the involvement scale were more likely to be ART non-adherent than those whose caregivers scored normative (AOR 2.14, 95% CI 1.13–4.06). Among double orphans, children whose caregivers were their aunt/uncle (AOR 0.14, 95% CI 0.03–0.84) and children whose caregivers were a grand-parent (AOR 0.09, 95% CI 0.10–0.96) were less likely to be non-adherent compared to children whose caregivers were a sibling (Table 5).


This study is one of the first to report pediatric ART adherence in Rwanda. This is also one of the few studies which found an association of ART adherence with four different orphan status. Findings reveal that more than half of the HIV positive children on treatment were adherent to ART. Double orphans were at highest risk of ART non-adherence and especially those who had a sibling as a caregiver had risk of non-adherence. They were also in danger of the latest initiation of ART due to the late detection of sero-status compared to other orphan status children. Furthermore, caregivers’ low involvement in child matters, taking more than three antiretroviral pills daily, and stunted children were positively associated with ART non-adherence.

ART adherence rate of our population was a rather moderate level compared with previous studies which ranged between 29% and 100% [32], [33]. In our study population, mean duration on ART was rather longer compared to previous studies which showed higher adherence rates [9], [16], [17]. The relatively long duration on ART among our study population might have affected low adherence to medication due to treatment fatigue. Our result demonstrated the need for further efforts to enhance ART adherence rate. For its practice, the following suggestions could be taken into consideration.

To the best of our knowledge, this is the first study to reveal that double orphans had higher risk of ART non-adherence compared to not only non-orphans, but also maternal orphans and paternal orphans. Although most of the previous studies compared “orphan” and “non-orphan” only, our result pointed out the importance of distinguishing between having both parents dead and having one parent dead, as well as between having a father dead and having a mother dead.

According to our study results, double orphans’ non-adherence risk was high when the caregiver was his/her sibling. As a previous study revealed, the sibling might not be capable of providing appropriate care for the child due to less matured skills and lower capacity for child care without adult’s supervision [34]. Another point why double orphans tended to be non-adherent could be related to the caregivers’ motivation. The non-biological caregiver of a double orphan may be less motivated to maintain caring for the child’s adherence and less affected by the guilt associated with MTCT compared with a biological caregiver. This was also endorsed by lower caregiver involvement in child care among double orphans’ caregivers.

Furthermore, our result demonstrated that single orphans had lower non-adherence risk compared to double orphans and non-orphans, and especially maternal orphans had the lowest risk of non-adherence among all orphan status in comparison with double orphans. This suggested that having only one surviving parent might affect them maintaining high ART adherence. Also, we could suggest that the living status of the father had an important effect on the child’s high adherence to treatment, which derived from a father’s decisive initiative in the family. Association between father’s high involvement in child care and better adherence was also reported in US study on pediatric chronic disease [35].

Our study identified the late initiation of HIV treatment among double orphans being due to late detection of sero-status. A Ugandan study has reported that compared to non-orphans, an orphan’s age was significantly older and their CD4 count was significantly lower at the initiation of ART [8]. However, different orphan status was not focused on in that study. As our study demonstrated, non-orphans tend to have their sero-status detected during the PMTCT program which monitors a child’s sero-status from their birth. On the contrary, double orphans were not detected till frequent illness emerged, which means they were at an advanced stage of the disease. This point is also endorsed by a significantly lower CD4 count among double orphans at the first examination. This result highlighted the vital need for early HIV detection specifically among double orphans.

Paternal orphans’ sero-status was detected earlier than maternal orphans. Also, their sero-status was detected more by PMTCT than maternal orphans. This suggested that the mother’s living status affected an earliness of the child’s sero-status detection. While a father’s existence influences a child’s better adherence, a mother’s existence might lead to earlier detection of a child’s symptom.

Children whose caregiver was strongly involved in the child’s matters were more likely to be ART adherent. This result is consistent with a study from the US which showed that better medication adherence can be achieved by greater caregiver-child relationships [10], [11]. Caregivers’ involvement may directly affect child care in monitoring ART adherence. Addressing the involvement of the caregiver may be a prominent component of interventions to improve adherence.

Pill burden (number of daily pills) was associated with adherence. The same result has been reported in previous studies [36][38]. Taking a larger number of medication and the complexity of administration impair appropriate adherence to the treatment. To promote adherence, we need to identify the difficulties perceived by children prescribed a large quantity of medication, and take necessary steps to address these difficulties. However, there are also other studies which found no association [17], [39], thus there still remains a need for more inquiry about the relation between pill burden and adherence.

The study revealed that stunting (HAZ<−2) was positively associated with ART non-adherence. This result suggested that non-adherence may restrain the child from gaining HAZ, since some studies showed the significance for improving HAZ after the initiation of ART [40], [41]. On the other hand, the association between stunting and non-adherence could also have resulted from a lack of food. Lack of food has been cited as one of the ART adherence barriers among pediatric populations in resource-limited settings [42]. Since Rwanda is one of the countries whose population suffers from insufficient nutritional status [43], lack of food could be an important barrier in non-adherence.

This study has several limitations. First, as the protocols for ART provision differs in each health facility, conclusions about individual adherence rates may be limited. In principle, a patient’s remaining pills were required to be collected by health staff on the day of a patients’ appointment and patients then get their monthly resupply and leave the clinic with the exact number of pills after each visit. However, this wasn’t strictly practiced. Second, as patients were sometimes provided with additional quantities of medicine to avoid unexpectedly missing doses, this may also have affected the resulting adherence rate. To minimize the influence of these two limitations, we asked participants their reasons for excess or deficiency of medicine as much as possible and examined the data.


Our study revealed that double orphan status was at the highest risk of ART non-adherence compared to other orphan status, and especially those who had a sibling as a caregiver had the highest risk of non-adherence. Double orphans were also in danger of initiating their ART at an older age and at a later stage of HIV/AIDS due to delays in their HIV detection. We, thus, highlight the need for particular interventions for HIV positive double orphans by supervising their adherence intensively and detecting their HIV infection at an earlier stage. Moreover, we demonstrated that adherence was associated with various factors such as a poor caregiver-child relationship, pill burden and the child’s poor nutritional status. To improve ART adherence, it is crucial to implement interventions that can promote better involvement of caregivers in child care, support children prescribed a large quantity of pills, and improve child nutrition.


We acknowledge all the heads, health workers of the health institutions and all research staff for their contribution to this study; and particularly acknowledge all participants for their valuable time and commitment.

Author Contributions

Conceived and designed the experiments: KK KP JM AM KO TS VM SN RM MJ JY. Performed the experiments: KK JM AM TS. Analyzed the data: KK KP KO MJ JY. Contributed reagents/materials/analysis tools: KK KP JM AM VM SN RM MJ JY. Wrote the paper: KK KP KO MJ JY.


  1. 1. UNAIDS (2010) Report on the global AIDS epidemic 2010. Geneva: World Health Organization.
  2. 2. Walker N, Schwartländer B, Bryce J (2002) Meeting international goals in child survival and HIV/AIDS. Lancet 360: 284–289.
  3. 3. Newell ML, Coovadia H, Cortina-Borja M, Rollins N, Gaillard P, et al. (2004) Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet 364: 1236–1243.
  4. 4. Watts H, Gregson S, Saito S, Lopman B, Beasley M, et al. (2007) Poorer health and nutritional outcomes in orphans and vulnerable young children not explained by greater exposure to extreme poverty in Zimbabwe. Trop Med Int Health 12: 584–593.
  5. 5. Lindblade KA, Odhiambo F, Rosen DH, DeCock KM (2003) Health and nutritional status of orphans <6 years old cared for by relatives in western Kenya. Trop Med Int Health 8: 67–72.
  6. 6. Ayaya SO, Esamai FO, Rotich J, Olwambula AR (2004) Socio-economic factors predisposing under five-year-old children to severe protein energy malnutrition at the Moi Teaching and Referral Hospital, Eldoret, Kenya. East Afr Med J 81: 415–421.
  7. 7. Panpanich R, Brabin B, Gonani A, Graham S (1999) Are orphans at increased risk of malnutrition in Malawi? Ann Trop Paediatr 19: 279–285.
  8. 8. Ntanda H, Olupot-Olupot P, Mugyenyi P, Kityo C, Lowes R, et al. (2009) Orphanhood predicts delayed access to care in Ugandan children. Pediatr Infect Dis J 28: 153–155.
  9. 9. Vreeman RC, Wiehe SE, Ayaya SO, Musick BS, Nyandiko WM (2008) Association of antiretroviral and clinic adherence with orphan status among HIV-infected children in Western Kenya. J Acquir Immune Defic Syndr 49: 163–170.
  10. 10. Bikaako-Kajura W, Luyirika E, Purcell DW, Downing J, Kaharuza F, et al. (2006) Disclosure of HIV status and adherence to daily drug regimens among HIV-infected children in Uganda. AIDS Behav 10: S85–93.
  11. 11. Naar-King S, Montepiedra G, Nichols S, Farley J, Garvie PA, et al. (2009) Allocation of family responsibility for illness management in pediatric HIV. J Pediatr Psychol 34: 187–194.
  12. 12. Wrubel J, Moskowitz JT, Richards TA, Prakke H, Acree M, et al. (2005) Pediatric adherence: perspectives of mothers of children with HIV. Soc Sci Med 61: 2423–2433.
  13. 13. Polisset J, Ametonou F, Arrive E, Aho A, Perez F (2009) Correlates of adherence to antiretroviral therapy in HIV-infected children in Lomé, Togo, West Africa. AIDS Behav 13: 23–32.
  14. 14. Simoni JM, Montgomery A, Martin E, New M, Demas PA, et al. (2007) Adherence to antiretroviral therapy for pediatric HIV infection: a qualitative systematic review with recommendations for research and clinical management. Pediatrics 119: e1371–1383.
  15. 15. Elise A, France AM, Louise WM, Bata D, François R, et al. (2005) Assessment of adherence to highly active antiretroviral therapy in a cohort of African HIV-infected children in Abidjan, Côte d'Ivoire. J Acquir Immune Defic Syndr 40: 498–500.
  16. 16. Cupsa A, Gheonea C, Bulucea D, Dinescu S (2000) Factors with a negative influence on compliance to antiretroviral therapies. Ann N Y Acad Sci 918: 351–354.
  17. 17. Nyandiko WM, Ayaya S, Nabakwe E, Tenge C, Sidle JE, et al. (2006) Outcomes of HIV-infected orphaned and non-orphaned children on antiretroviral therapy in western Kenya. J Acquir Immune Defic Syndr 43: 418–425.
  18. 18. UNAIDS (2010) Epidemological Fact Sheet on HIV and AIDS. Geneva: World Health Organization.
  19. 19. Center for Treatment and Research on AIDS M, Tuberculosis and Other Epidemics (2009) TRAC ART report for November 2009. Kigali: Center for Treatment and Research on AIDS, Malaria, Tuberculosis and Other Epidemics.
  20. 20. Shetty AK, Powell G (2003) Children orphaned by AIDS: a global perspective. Semin Pediatr Infect Dis 14: 25–31.
  21. 21. UNICEF (2007) Children and AIDS. A stocktaking report. New York: United Nations Children's Fund.
  22. 22. Safreed-Harmon K, Siripong A, Kerr SJ, Gruskin S, Pancharoen C, et al. (2007) Antiretroviral therapy adherence did not differ between Thai children with biological and those with nonbiological parents. Clin Infect Dis 45: 669–670.
  23. 23. Nabukeera-Barungi N, Kalyesubula I, Kekitiinwa A, Byakika-Tusiime J, Musoke P (2007) Adherence to antiretroviral therapy in children attending Mulago Hospital, Kampala. Ann Trop Paediatr 27: 123–131.
  24. 24. Osterberg L, Blaschke T (2005) Adherence to medication. N Engl J Med 353: 487–497.
  25. 25. Ma A, Chen DM, Chau FM, Saberi P (2010) Improving adherence and clinical outcomes through an HIV pharmacist's interventions. AIDS Care 22: 1189–1194.
  26. 26. Organization WH (2010) Antiretroviral therapy of HIV infection in infants and children: towards universal access. Recommendations for a public health approach - 2010 revision. Geneva: World Health Organization.
  27. 27. Johnson MO, Charlebois E, Morin SF, Remien RH, Chesney MA, et al. (2007) Effects of a behavioral intervention on antiretroviral medication adherence among people living with HIV: the healthy living project randomized controlled study. J Acquir Immune Defic Syndr 46: 574–580.
  28. 28. Mellins CA, Brackis-Cott E, Dolezal C, Abrams EJ (2004) The role of psychosocial and family factors in adherence to antiretroviral treatment in human immunodeficiency virus-infected children. Pediatr Infect Dis J 23: 1035–1041.
  29. 29. AB G (1994) Parent-Child Relationship Inventory (PCRI). Los Angeles: Western Psychological Services.
  30. 30. Nicholson O, Mellins C, Dolezal C, Brackis-Cott E, Abrams EJ (2006) HIV treatment-related knowledge and self-efficacy among caregivers of HIV-infected children. Patient Educ Couns 61: 405–410.
  31. 31. MH K (2006) Multivariable analysis: A practical guide for clinicians. Cambridge: Cambridge University Press.
  32. 32. Vreeman RC, Wiehe SE, Pearce EC, Nyandiko WM (2008) A systematic review of pediatric adherence to antiretroviral therapy in low- and middle-income countries. Pediatr Infect Dis J 27: 686–691.
  33. 33. Sutcliffe CG, van Dijk JH, Bolton C, Persaud D, Moss WJ (2008) Effectiveness of antiretroviral therapy among HIV-infected children in sub-Saharan Africa. Lancet Infect Dis 8: 477–489.
  34. 34. Yanagisawa S, Poudel KC, Jimba M (2010) Sibling caregiving among children orphaned by AIDS: synthesis of recent studies for policy implications. Health Policy 98: 121–130.
  35. 35. Wysocki T, Gavin L (2006) Paternal involvement in the management of pediatric chronic diseases: associations with adherence, quality of life, and health status. J Pediatr Psychol 31: 501–511.
  36. 36. Biadgilign S, Deribew A, Amberbir A, Deribe K (2009) Barriers and facilitators to antiretroviral medication adherence among HIV-infected paediatric patients in Ethiopia: A qualitative study. SAHARA J 6: 148–154.
  37. 37. da Silveira VL, Drachler MeL, Leite JC, Pinheiro CA (2003) Characteristics of HIV antiretroviral regimen and treatment adherence. Braz J Infect Dis 7: 194–201.
  38. 38. Kastrissios H, Suárez JR, Katzenstein D, Girard P, Sheiner LB, et al. (1998) Characterizing patterns of drug-taking behavior with a multiple drug regimen in an AIDS clinical trial. AIDS 12: 2295–2303.
  39. 39. Mellins CA, Kang E, Leu CS, Havens JF, Chesney MA (2003) Longitudinal study of mental health and psychosocial predictors of medical treatment adherence in mothers living with HIV disease. AIDS Patient Care STDS 17: 407–416.
  40. 40. Verweel G, van Rossum AM, Hartwig NG, Wolfs TF, Scherpbier HJ, et al. (2002) Treatment with highly active antiretroviral therapy in human immunodeficiency virus type 1-infected children is associated with a sustained effect on growth. Pediatrics 109: E25.
  41. 41. Guillén S, Ramos JT, Resino R, Bellón JM, Muñoz MA (2007) Impact on weight and height with the use of HAART in HIV-infected children. Pediatr Infect Dis J 26: 334–338.
  42. 42. Fetzer BC, Mupenda B, Lusiama J, Kitetele F, Golin C, et al. (2011) Barriers to and facilitators of adherence to pediatric antiretroviral therapy in a sub-Saharan setting: insights from a qualitative study. AIDS Patient Care STDS 25: 611–621.
  43. 43. WHO, Nutrition Landscape Information System. Available: Accessed.