Abstract
In Africa, HIV infection is considered a family disease. A retrospective cohort analysis was performed to describe the characteristics and outcome in 35 parent–child pairs taking antiretroviral therapy (ART) in separate adult and pediatric HIV clinics. In 26 pairs, ART was first initiated in children. Baseline median CD4 count was 122/mm3 in adults and 376/mm3 in children. World Health Organization stage 3 or 4 disease affected 49% of adults and 83% of children. In all, 3 parents and 1 child died. Hospitalization, poor adherence, missed appointments, or regimen change affected >50% of pairs on ART. Following tuberculosis diagnosis in their parents, 2 of the 5 children were not investigated. By week 104, 29 (83%) pairs remained on ART, and 69% of patients on ART were virologically suppressed. Parent–child pairs with advanced HIV infection had good outcomes when cared for in separate clinics. Establishing lines of communication between clinics is important. Family-centered services may provide more integrated care.
Introduction
In sub-Saharan Africa, the predominantly heterosexual nature of HIV transmission and perinatal transmission has resulted in parents and children often requiring concurrent antiretroviral therapy (ART). In Africa, HIV is considered a family disease, and a major aim of ART in such a situation is to keep the family unit alive and well, among which the benefits include improved well-being of the child. 1 In the early years of expansion of treatment in sub-Saharan Africa, the organization of therapy usually entailed separate adult and pediatric clinics for logistic reasons, either because of the different expertise and service organization required or because parent and child qualified for the treatment months or years apart in an era when treatment criteria were very strict. Outcomes of ART have generally been reported separately for adults and children. 2 –6 Family-centered services have been developed for HIV-infected children and reported outcomes in children. 7,8 Care is offered for their parents, caregivers, and other family members on-site, which should be convenient for them, but difficulties have been reported in getting the parents into care. 7,8
The sequential introduction of ART into a single family has been detailed in the study by Fielder and Kwatampora. 9 Both parents and 4 children were newly diagnosed with HIV. The parents started ART followed by the children in descending order of age. The family used 31 different medications taken in 49 individual doses per day. Long-term clinical and social outcomes were excellent, though the oldest, adult child failed his first regimen. The ART program was well resourced and had excellent community networks and intensive home-based follow-up and as such was considered to have limited generalizability. There have been no other published articles on the virologic outcomes of ART in parent–child pairs, including regarding prevention of mother-to-child HIV transmission (PMTCT) initiatives that offer postnatal ART. We aimed to describe the characteristics and outcome in parent–child pairs treated with ART, in Soweto, in separate clinics in proximity to each other.
Methods
Setting
The study was conducted at the outpatient Adult and Pediatric HIV Clinics at Chris Hani Baragwanath Hospital, a 2700-bed public-sector university hospital, which is the main hospital serving Soweto. Free ART was introduced into the public sector in South Africa in 2004 and, in the Soweto area, was initially hospital based. The adult HIV clinic and the pediatric HIV clinic have operated separately since their inceptions in 1989 and 1998, respectively, and at the time of the study, these clinics had approximately 5000 and 3000 patients, respectively, on ART. Clinic days for adults and children were not coordinated partly because the adult HIV clinic operated formally on only 3 days/week as the facility was shared with the hematology–oncology service. This would increase the travel costs for attendees who commonly use a comprehensive network of minibus taxis in which children <6 years of age are not charged a fare.
Study Population and Design
A retrospective cohort analysis of parent–child pairs was performed based on adults who had initiated ART from 2004 to 2006, irrespective of the duration of follow-up. Demographic data collected from the adult HIV clinic included the HIV serostatus of children and whether HIV-infected children were receiving ART. The addresses of adults receiving ART and whose children were also noted to be receiving ART were matched with the pediatric HIV clinic database. There was a match with 55 pediatric patients. Adequate records confirming that an adult and a child were receiving ART were available for 35 parent–child pairs.
For adults, the CD4 count <200 cells/mm3 and/or World Health Organization (WHO) stage 4 disease defined ART eligibility, per the national guidelines at that time of the study. The usual starting regimen was stavudine (d4T), lamivudine (3TC), and efavirenz (EFV). The ART was started for children with a WHO stage 3 or 4 disease or a CD4 count <20% for children <18 months of age and <15% for older children. The first-line regimen comprised d4T and 3TC together with lopinavir/ritonavir (LPV/r) for children <3 years of age and EFV for those ≥3 years and >10 kg of weight.
Procedures
Data of adults and children receiving ART were collected from clinic and laboratory records for 96 weeks or until they were no longer in care at the clinics. A standardized data collection form was used. Baseline information comprised age, sex, history of opportunistic infections, WHO stage, baseline CD4 count and viral load, ART initiation date, and regimen choice. Indicators for progress on ART included sequential CD4 and viral load results at 24, 48, and 96 weeks (or closest available result), hospitalizations, tuberculosis (TB) events on ART, drug toxicity, and any drug changes. Viral load <400 copies/mL defined virological suppression and was included even if the patient was on a salvage regimen. Outcome was assessed by death, loss to follow-up, and retention in care. Adherence was assessed based on specific questioning at each visit concerning patient or parent recall of missed doses over the preceding 3 and 7 days and more generally since the last visit.
Ethics
The study was approved by the Human Research Ethics Committee, University of the Witwatersrand, Johannesburg.
Results
Thirty-five parent–child pairs were studied. Three fathers were among the parents receiving ART (Table 1). In all, 19 (54%) of the parents were married or had a steady partner and 1 had been widowed. In all, 8 parents had a single child, 13 had 2 children, and 14 had ≥3 children. Five parents were employed. Of the 24 children with a documented reason for HIV testing, 22 were tested due to illness. For the 26 parents with a documented reason for HIV testing, 10 were tested during pregnancy (and received preventative therapy), 6 were tested after their child had been diagnosed with HIV infection, and 6 were tested due to ill health. World Health Organization stage 3 and 4 disease was present in 49% of adults and 83% of children at the time of initiating ART. In all, 31 adults started on d4T, 3TC, and EFV, while 2 received zidovudine (ZDV) in place of d4T and 2 had a different regimen. The children were treated by age per the national guidelines.
Baseline Characteristics of Parents and Their Children on ART.
Abbreviations: ART, antiretroviral therapy; WHO, World Health Organization.
While receiving ART, various factors such as hospitalization, episodes of poor adherence, missed appointments, or regimen change affected at least 1 member of >50% of all parent–child pairs (Table 2). In all, 2 parents and 1 child were taking the incorrect drug dose of ART. Poor adherence was noted in 1 child while her mother was hospitalized. In the 2 pairs who had a default before returning to the treatment, the default period was different for parent and child. Children alone usually had a default of 3 to 4 months, while the default period for parents was shorter, the same, or longer. Of the 7 patients with virological failure, 5 had a history of defaulting therapy. Those 7 patients had a regimen switch, while another 25 had substitutions of individual drugs. There were 29 (83%) pairs still on ART at the designated clinics after 2 years of therapy.
Treatment Events of 35 Parent–Child Pairs While on ART.
Abbreviations: ART, antiretroviral therapy; IRIS, immune reconstitution inflammatory syndrome.
Charts were checked to see whether the pediatric clinician knew of, and acted on, information about a parent having TB or of a parent being lost to follow-up while the child was on ART. Parent outpatient hospital records were self-retained and so could be shown at the pediatric clinic if requested. In all, 5 parents developed active TB, and all 5 children’s records reflected this, although 2 children were not evaluated for TB, 1 of whom developed pulmonary TB. Another parent developed TB immune reconstitution inflammatory syndrome (IRIS), which was known, but the child was not evaluated. One further parent was mistakenly noted in the child’s file to have TB, although this was not borne out in the parent’s record; that child was evaluated and given isoniazid prophylaxis. Additionally, 1 child had sputum smear-positive TB IRIS, which was not reflected in the parent’s chart. Three parents were lost to follow-up over the course of the study. This was reflected in only 1 of the pediatric records, and there was no note of the reason the parent gave or of any attempt to put the parent back in contact with the adult clinic.
By week 104, 69% of the patients still on ART were virologically suppressed and 57% of the total cohort were suppressed (Table 3). Of the total cohort, at week 52, 18 (51%) parent–child pairs were virologically suppressed, as were 21 (60%) parents and 25 (71%) children. By week 104, 16 (46%) pairs were virologically suppressed, as were 18 (51%) parents and 22 (63%) children.
Number of Patients Still on ART and Proportion of Patients with Virological Suppression (VL < 400 copies/mL) Reported for Patients Still on ART and for the Total Cohort (n = 70) at Weeks 26, 52, and 104.
Abbreviations: ART, antiretroviral therapy; VT, viral load.
Discussion
This study of parent–child pairs with advanced HIV infection, who started ART and were managed in separate busy public-sector HIV clinics on the same hospital premises, was salutary in its illustration that the family unit can be successfully kept together on ART. In 74% of the dyads, the child was the first to initiate ART, in keeping with the fact that at least 22 of the children tested for HIV infection due to ill health compared to 6 of the adults. It would be to the child’s advantage that the caregiver is healthy. There is some evidence that a child who has a caregiver who is HIV infected is protected against mortality 7 and that a child with a caregiver who is on ART has improved adherence, 10 but there is conflicting evidence. 11 Three of the parents were men. There is a dearth of information about men and their families in relation to HIV infection. 12 There has also been limited success in encouraging the attendance of men at family-centered services where the index patient is the child or the mother. 13,14
The study faced a number of obstacles, which affected more than half of the parent–child pairs on ART. One issue was hospitalization, which affected 60% of the pairs. In this study, 1 child had poor adherence while her mother was in hospital. Hospitalization could possibly engender a sense of futility around ART unless adequate support is provided to the family. It is also possible that the parent may not have disclosed his or her HIV status to another adult who could administer the child’s ART during such a vulnerable period. A second obstacle was episodes of poor adherence and missed appointments. There may have been financial reasons contributing as the 2 clinics initially had 1 treatment day a week which overlapped, so that double visits usually needed to be made by the pair. Contrary to that explanation was the fact that routine consultations at the clinics quickly became once every 3 months and that the pharmacy provided drugs for 2 months at a time. It was unexpected that default was less common in pairs than in individuals and that the default time periods varied between child and parent when a pair was involved. Adherence has been related to behavioral issues in the caregiver, such as depression and alcohol use, as well as to life stresses in women, such as child care burden and the number of children in a household younger than 18 years of age. 10,11,15
Communication is important when separate clinics are utilized for adults and children on ART, which is emphasized by the narrative concerning TB and parent withdrawal from ART. Asking questions about the progress of the other member of the pair at every visit and requesting to see self-retained medical records can ameliorate much of this potential problem. Sharing information across clinics by clinicians, with the permission of the patients, would improve the situation. Identifying problems in the family unit can open the way to better supportive care, whether physical, psychological, or emotional. Questions about parents and children with regard to HIV status, testing, and ART should be formalized for all new patients and should be repeated at intervals.
Virological suppression at 24 months was achieved at rates similar to other African and South African studies. 2 –7 This aspect is affected by the small numbers in the study cohort. The pediatric outcomes are subsumed and better represented in a large study from the same HIV clinic, though cumulative virologic suppression was reported. 6 In reviews of ART studies in children in Africa, virologic suppression was achieved in 49% to 81% at 12 months, 50% at 24 months, 4 and 70% at 12 months, which decreased to 53% by intention-to-treat analysis. 5 Among adults in South Africa, viral rebound by 24 months occurred in 17% of the 91% of patients who had virologic suppression at 6 months. 3 There has been limited reporting of virologic outcomes in PMTCT ART programs. In France, pregnancy did not affect the short-term virologic response to ART, with nearly 100% of pregnant women reaching a viral load <400 copies/mL by 6 months. 16 In contrast, a meta-analysis of adherence to ART during and after pregnancy found that 76% of pregnant women had adequate adherence, which dropped to 53% in the postpartum period. By this, it can be inferred that the pregnant women were concerned about their fetus and PMTCT. 17 Longer term outcome studies of PMTCT ART programs involving adults and children are awaited.
There is a movement toward family-centered services, especially in pediatric and PMTCT-plus clinics. The advantage is that all services are under one roof and the family unit is paramount. These clinics have generally showed good outcomes. 7,8,14 There are logistic problems that work against such services in sub-Saharan Africa. These include staffing in general, the relative paucity of pediatric-trained clinicians, the large numbers of people accessing ART, and the fact that many of the family-focused services are well-supported financially by outside resources. When attempting to set up a family clinic at Chris Hani Baragwanth Hospital, we found that pediatric staff trained to manage children and adults were in great demand and had moved on. We were also unable to obtain a site in the hospital suitable for a combined clinic, and the rapid escalation in the number of clients made such a service a dream rather than a manageable reality.
There are a number of limitations to this study. The retrospective nature of the study meant that we were reliant on the extent of documentation done by the attending clinician. The numbers of parent–child dyads studied were small and were cared for in a tertiary hospital setting. We know that many of the caregivers for the children were the parents on ART, but we do not have that data for all pairs. It is unclear whether these findings will be the same for parent–child pairs who access treatment when the CD4 counts are higher and before serious illness has supervened.
In conclusion, parent–child pairs with advanced HIV infection had good outcomes when cared for in separate clinics and most parent–child pairs remained in care. The study highlighted a number of areas of concern around compliance and the effects of hospitalization. Fostering lines of communication between adult and pediatric clinics are of utmost importance. Although one could anticipate that having a parent and child on treatment would result in similar adherence and outcomes, this was not the case in all pairs. The superior rate of virologic suppression among children could conceivably suggest that parents put the health care needs of the child before their own.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: L. Fairlie is an employee of the Pediatric HIV Clinic, Harriet Shezi Children’s Clinic (HSCC), which is funded through USAID; however, no direct financial support for the research, authorship, and/or publication of this article was received.