Cohort Profile: The Siyakhula Cohort, rural South Africa

Each year millions of children in low- and middle-income countries (LMIC) fail to reach their developmental potential due to factors including poverty, malnutrition, poor stimulation and HIV. Although vertically-acquired HIV can now be prevented, little is known about the impact of HIV exposure in fetal and early life on the development of the many HIV-negative children. The Siyakhula Cohort was established within the Canadian Grand Challenges ‘Saving Brains’ initiative, to support re-enrolment of strategic cohorts in LMIC. This unique cohort in rural South Africa includes 1536 HIV-negative children born to HIV-infected (HIV-exposed) and HIV-uninfected (unexposed) mothers, enrolled from the Africa Health Research Institute (AHRI)–formerly the Africa Centre for Population Health (Africa Centre). The cohort includes data on HIV exposure in fetal and early life, and other early life factors (including breastfeeding) known to impact on later health outcomes. At birth, all children benefited from the early Prevention of Mother-to-Child-Transmission of HIV services in the district, and a subgroup were part of an additional early life breastfeeding intervention, the Vertical Transmission Study (VTS). This cohort pre-dated antiretroviral treatment (ART) roll-out, allowing for examination of outcomes associated with HIV exposure without ART exposure in utero and during breastfeeding. Current assessments at ages 7–11 years collected data on growth, health, cognition (including executive function), education and emotional-behavioural outcomes at primary school age. The Africa Centre, one of 11 sites from low-middle income countries (LMIC) to receive funding from Grand Challenges Canada, established the Siyakhula cohort to examine associations between early life factors and later child development, specifically exposure to a breastfeeding intervention1,2 and later child development, allowing for HIV exposure in fetal and early life. Evidence linking breastfeeding with improved cognition is conflicting, with exclusive breastfeeding (EBF) linked with improved cognition in a meta-analysis3 but inconsistent results in a systematic review, depending on study design and methodology.4 There is also a dearth of evidence on the effect of HIV on HIV-exposed but HIV-uninfected children. A recent systematic review5 examining HIV exposure and child development found data from only 11 studies worldwide (1591 children aged 0–18 years: 650 HIV-exposed; 736 HIV-unexposed; 205 HIV-infected). The review concludes that HIV-exposed children are disadvantaged in terms of child development, in particular emotional-behavioural development, compared with their HIV-negative unexposed peers. However, findings were inconsistent, with most evidence based on small samples with wide heterogeneity in outcome measures. There are few longitudinal studies, almost none with HIV-negative controls or a population norm, and no studies on primary school-aged children. The Siyakhula cohort was established in 2012 from the Africa Centre [www.africacentre.ac.za] research platform, in a rural, high HIV prevalence setting.6 With the scale-up of HIV treatment programmes, parents are surviving to care for their children, and mother-to-child transmission (MTCT) of HIV has been virtually eliminated.7–10 Children in the Siyakhula cohort were born in the pre-ART era, between 2001 and 2006, in the Hlabisa sub-district, and are all HIV-negative. Some children had previously participated in the VTS, which supported mothers with exclusive breastfeeding (EBF) for the first 6 months of life, and demonstrated that EBF reduced the risk of MTCT of HIV compared with mixed breastfeeding.2 Similar-aged children were also enrolled from the Africa Centre Demographic Surveillance System (DSS). These children had been exposed to the same standard of care, including similar messages regarding HIV and early infant feeding, according to national guidelines at the time,11,12 without the additional VTS breastfeeding support. The Siyakhula cohort is well placed to address the question of whether, in the context of HIV, EBF contributes to improvements in the development and health of children. All children in the cohort are HIV-negative. We excluded HIV-positive children, as they have HIV-specific and unique developmental risks.5,13 Children were eligible for enrolment in the Siyakhula cohort if HIV-negative, 7–11 years of age, born and still residing in the study area (the Hlabisa sub-district), if their mother’s HIV status during pregnancy was known, if mothers received antenatal care for the index child in the study area, and if both mother and child were still alive. The children included in the Siyakhula cohort came from two different sources. First we enrolled HIV-exposed and unexposed children, who met the above eligibility criteria, from the VTS. The children in the VTS had their final study visit when they were 2 years of age. At the end of the VTS in 2006, 1289 children were still alive, were known to have mothers who were alive, and were themselves HIV-negative (see Figure 1). The VTS enrolled children from the Hlabisa sub-district between 2001 and 2005 (see Figure 2). Consort diagram for the Siyakhula Cohort. Demographic Surveillance Area within the Hlabisa sub-district, and catchment areas of the Vertical Transmission Study. The second source of children for the Siyakhula cohort came from the Africa Centre Demographic Surveillance Area (DSA), situated in part of the Hlabisa sub-district (see Figure 2).6 Since 2000, the Africa Centre has collected data biannually (trianually since 2012) from almost 90 000 people in 11 000 households per round. In 2003, an annual HIV surveillance was added, with HIV status collected from consenting adults.14 The children from the DSA had been born between 2001 and 2006, as had those in the VTS, but had not taken part in the VTS (so had not received the EBF intervention). Within the DSA in 2012, 1226 children were documented at their last surveillance visit to be alive, HIV-uninfected, with mothers who were alive (see Figure 1). It is important to note that all eligible children within the DSA were approached for inclusion in the Siyakhula cohort–some of whom had participated in the VTS and others who had not. However in addition, the Siyakhula cohort included some VTS children who lived in the Hlabisa sub-district but in areas outside the DSA (see Figure 2). Therefore, four groups of HIV-negative children were recruited: HIV-exposed and unexposed from the VTS, and HIV-exposed and unexposed from the DSA. The consort diagram (Figure 1) shows the pool of 2515 potential participants, those who enrolled (n = 1592), and those who completed assessments (n = 1536). Table 1 shows the characteristics of those who were enrolled compared with those who were not enrolled. Of the 1536 children who completed all assessments, 1059 were HIV-unexposed and 477 HIV-exposed at birth. Comparisons of the Siyakhula cohort by enrolled/not enrolled and completed/not completed ‘Enrolled’ refers to children from the original population of children from the VTS and DSS who met the criteria for the Siyakhula cohort and whose parents/guardians provided written informed consent. ‘Not enrolled’ refers to children from the original population of children from the VTS and DSS who were not enrolled for a number of reasons listed in Figure 1, including those who could not be traced, those whose parents/guardians did not provide written informed consent and those who were not eligible for inclusion. ‘Completed’ refers to children who completed all the Siyakhula developmental assessments and other data collected. ‘Not completed’ refers to children who did not complete all the Siyakhula developmental assessments and other data collected. Comparisons of the Siyakhula cohort by enrolled/not enrolled and completed/not completed ‘Enrolled’ refers to children from the original population of children from the VTS and DSS who met the criteria for the Siyakhula cohort and whose parents/guardians provided written informed consent. ‘Not enrolled’ refers to children from the original population of children from the VTS and DSS who were not enrolled for a number of reasons listed in Figure 1, including those who could not be traced, those whose parents/guardians did not provide written informed consent and those who were not eligible for inclusion. ‘Completed’ refers to children who completed all the Siyakhula developmental assessments and other data collected. ‘Not completed’ refers to children who did not complete all the Siyakhula developmental assessments and other data collected. Data for Siyakhula have been collected over three visits between September 2012 and June 2014, when the child was between 7 and 11 years of age. Study consent was obtained in Visit 1, socio-demographic, economic and health data, mothers' mental health and cognitive ability in Visit 2, and children’s cognition and executive function in Visit 3. When the mother was not the primary caregiver, mental health assessments were completed by the child’s primary caregiver during Visit 2. Differences between those lost to follow-up and those who completed assessments are shown in Table 1. Children who are part of the DSS also have longitudinal data available, collected biannually. Table 2 describes the measures used and data collected. Description of data collected in Siyakhula ‘Current’ refers to data collected in the Siyakhula cohort data collection. ‘Measured in Siyakhula’ refers to the data collection for this cohort, when children were aged 7–11 years. aMental health data collected from either biological mother (if she was the child’s primary caregiver) or from primary caregiver of index child if biological mother was not the child primary caregiver. bMeasured either at home or, if mother preferred (for example if there was limited space at home), child was brought to a fixed building, for example a clinic or mobile unit. Description of data collected in Siyakhula ‘Current’ refers to data collected in the Siyakhula cohort data collection. ‘Measured in Siyakhula’ refers to the data collection for this cohort, when children were aged 7–11 years. aMental health data collected from either biological mother (if she was the child’s primary caregiver) or from primary caregiver of index child if biological mother was not the child primary caregiver. bMeasured either at home or, if mother preferred (for example if there was limited space at home), child was brought to a fixed building, for example a clinic or mobile unit. Child cognition was measured using the Kaufman Assessment Battery for Children 2nd Edition (KABC-II), a validated measure of cognitive development in children aged 3–18 years (Table 3).15 The test battery was implemented using the Luria model theoretical approach, well-suited to children in low-income, cross-cultural settings where quality and exposure to school may vary. Eleven subtests were administered, including both verbal and nonverbal tests for all domains. Subtests were scored into four index scales, covering all aspects of cognition and used to calculate a mental processing index (MPI) reflective of general intelligence. Description of test battery (including subtests and scales) used in the measurement of child cognition and executive function All these assessments were conducted when children were aged 7–11 years in the Siyakhula cohort. Assessments took place at the child’s home, or if mother preferred (for example if there was limited space at home), child was brought to a fixed building, for example a clinic or mobile unit. Description of test battery (including subtests and scales) used in the measurement of child cognition and executive function All these assessments were conducted when children were aged 7–11 years in the Siyakhula cohort. Assessments took place at the child’s home, or if mother preferred (for example if there was limited space at home), child was brought to a fixed building, for example a clinic or mobile unit. The KABC-II test battery is licensed to Pearson Ltd USA,15 and test kits and forms were purchased. All subtests in the Luria Model battery were retained without adaptation, and the administration manual was translated under license from Pearson Ltd, with fees waived.16 An expert review team, including the authors of the KABC-II, selected the subtests considered most culturally appropriate and one subtest substitution was made (Atlantis and Atlantis delayed tests replaced Rebus/Rebus delayed tests). One additional subtest Riddles from the knowledge scale was included as a supplementary test of vocabulary and general knowledge. Three additional subtests to the KABC-II were added to test executive function capacities: working memory, inhibition and switching (Table 3). These subtests were taken from the Neuropsychological Assessment Battery 2nd Edition (NEPSY-II),17 also licensed to Pearson Ltd USA.18 We used individual subtests in the NEPSY battery (Attention and Executive Function Domain) considered appropriate for focused evaluation of neuropsychological functioning. Test kits and forms were purchased; tests were used in their original format, and auditory stimuli were translated under translation license from Pearson, again with fees waived.18 Children’s emotional and behavioural problems were measured using the Parent Report versions of the Child Behaviour Checklist (CBCL) for children aged 6–12 years, which has been validated in over 30 countries including South Africa.19, 20 The CBCL, licensed to the Achenbach System of Empirically Based Assessment (ASEBA), offers a comprehensive approach to assessing adaptive and maladaptive functioning and was used with permission and translation licence. The CBCL comprises two parts: Part 1 includes a competencies questionnaire, including questions on children’s engagement in academics, sports and hobbies, and the quality of their friendships and sibling relationships. CBCL Part 1 is time-intensive and seldom used in a research context, but mostly provides data for clinical interpretation and treatment. We used an abbreviated version, collecting qualitative data on children’s social and peer competencies which were coded and categorized to be used in quantitative analysis. The CBCL Part 2 behavioural problems rating scale was implemented in full, including a 120-item rating scale which makes up a composite Total problems score; a high score indicating more problems. The parent rated the child’s behaviour on a three-point scale on a series of symptoms which represent eight psychological syndromes. There are 113 numbered items, but item 56 has 7 subitems on somatic symptoms, making 120 items in total. The items are scored as: 0 = not true (as far as you know); 1 = somewhat or sometimes true; 2 = very true or often true. Some items, if endorsed, include qualitative descriptive answers on the child’s problem behaviour. These descriptions are not used in the scoring system and are of clinical interpretative value only. In Siyakhula, CBCL scores were normed using multicultural Rating-to-Score norming software (purchased from to normed scores for the Total the two and the and problems including and and problems including and was high = the = for a Research with years of research the assessments 2 and were conducted by two was a standard for a of of assessments, with for all the data collection the VTS children we that of EBF months was associated with and was associated with improved cognitive development in In addition, HIV-exposed children as well as HIV-unexposed children in the and was associated with children’s later cognitive development, an as is seldom included as a in breastfeeding studies, in This to the cohort of HIV-exposed and unexposed children in Africa who have completed a battery of cognitive and executive function The is similar to most samples in We considered to a battery of tests the of or developmental data in The approach provides an for children to on at two subtests in of including both verbal and the risk that is a of or due to or school We used to test the of the child cognitive measures. This is an important when using child development in where they have not been also to the measured by a number of different The battery has a of and subtests which are based on well established of cognitive We used to subtest one of four Each a of which a subtest was on which of the four of cognition the subtest is to measure (see Figure 3). is a to test the data in the cohort with the theoretical model of cognition within the that subtests measure one of four cognitive In addition, the from subtest (for ability to or a and This is the represent the different which measure the same in different We also the three NEPSY subtests using and found to represent individual The NEPSY are to measure executive function in to the on we included the NEPSY with the the NEPSY executive function tests with the scale of the (Figure 3). The scale a measure of executive the of the test battery a battery of both cognition and executive The of this final model was also of of was in with and was by of over and of were used as they provide a more comprehensive evaluation of model The factors were all with the of these cognitive and the for to In examining children’s Figure shows the scores for the subtests by age, scores from the Siyakhula Cohort. Children’s scores are from the KABC-II which the score for at a subtest and scores for the Siyakhula by Kaufman Assessment Battery for Children 2nd and by child age. the developmental scores of the cohort were with the of the cohort in a similar to the children’s scores in high income countries However, the of children in Siyakhula their Some of this could be for by in school exposure and The are and with age, that these children particular within Since children most one not this poor at a scale to be for by subtest or stimuli One is children’s improved on the number in the which tests a child’s to and and again within a few The scale has three subtests that and and verbal Children well on number but on the other two subtests (Figure This a exposure to number in primary not these are by on the children’s was again poor all this was on the that rural children were more with the stimuli used in this The cognitive in Siyakhula for the developmental potential of children in these Children’s on the which well children and to shows that children are disadvantaged from school age, and that the most scales, with age, children more and this the of or the also to the potential of high quality in the early years, to that are We are funding to the cohort into their years, to examine development and and the effect of EBF on later outcomes. An important area of research on the development of executive function into Executive function is a area of study the life with poor executive function in early and and Children’s executive function outcomes including and social The are to with executive function is to at years of In Siyakhula, and as the areas of mental health risk to child We that children’s mental health problems in the areas of were associated with executive These data provide a which to examine the of executive and mental health as to risk in early There are almost no cohort studies in Africa which have used a battery approach to measure children’s cognition in and very limited data on children’s development in high A recent systematic review of the on the of HIV exposure on child development found only 11 studies from three from one from and from with quality design and measures of behavioural and developmental these studies, cognitive behaviour and developmental were measured with different from 650 HIV-exposed children (including 205 HIV-positive and 736 children. cohort of HIV-exposed and unexposed children includes a number of children all exposed and unexposed children in the 11 studies to particular of the research are by of a comprehensive cognitive additional subtests in areas of cognition as executive examination of emotional and behavioural outcomes and of almost all known including In we have collected data on children’s growth, and all of which later studies examining developmental outcomes were to for factors known to child development, including early infant feeding, HIV exposure and A of studies examining developmental outcomes and early was the to the of EBF on of which have been shown to be We have been to this for the children who received the VTS intervention, which the most of breastfeeding South Africa have one longitudinal the cohort of children born in in the with data collected from pregnancy to has to of development in South Africa and a national but the cohort in the context within which took Children born in were not to from many of the implemented since as health care and the child support from birth. In addition, the cohort is an cohort early in the HIV The children in the Siyakhula cohort are rural, and were born in an and within an surveillance the potential to examine the of national as child support to education and and on their cohort provides a allowing for the of a different and on outcomes. there is in the outcomes of HIV-exposed but HIV-uninfected children, with the of more Prevention of to Child Transmission including during pregnancy and The children in Siyakhula were born HIV treatment was in South was as part of the This cohort includes HIV-exposed children, and provides an important of children who were not exposed to ART in for studies on the impact of fetal and early life exposure to Data are not on mental health during pregnancy or their mental health in the early and be from their measures of and We have no data on or are to child outcomes. The between and follow-up may ability to examine or other factors the between early life and these later outcomes. this was a with a well from the was a and excluded HIV-positive children. and more children with HIV-negative compared with HIV-positive mothers, were enrolled. Differences between who did or did not complete assessments were limited to children born to mothers more to complete all assessments children of can be obtained from the Africa Health Research Institute for who the criteria for to data of the Research Data Africa Health Research The Siyakhula cohort, established in 2012, is an cohort associations between early life factors (including exclusive breastfeeding and HIV and later child The cohort includes 1536 HIV-negative, rural children aged 7–11 years, including 477 HIV-exposed to HIV-positive and 1059 HIV-unexposed to HIV-negative children. The cohort includes a wide of health and developmental outcomes including cognitive development, executive emotional-behavioural development, and for a of and early life factors including infant feeding, HIV school and mental One of data collection has taken place in the Siyakhula cohort including three data collection visits per early life data are on all children, and for children who within a surveillance area, limited additional longitudinal data are biannually since their birth. can be obtained from the Africa Health Research Institute for who the criteria for to data of the Research Data Africa Health Research The Africa Centre for Population now the Africa Health Research Institute where the research took place is by the Africa Centre Current The DSS is by the South of and the South Population Research The original Vertical Transmission Study was The re-enrolment and of the cohort were by Grand Challenges Canada, Challenges is supported by the The support of the Centre of in data is also We the and data the of Health and at the and for allowing to the at the Africa Centre for their and for data and the of the Africa Health Research Institute for their the We for to child development and We the on this We are to Pearson Ltd for the translation for the and NEPSY and to and Kaufman for expert on subtest and and of

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.