Effect of care environment on educational attainment among orphaned and separated children and adolescents in Western Kenya

Background: There are approximately 140 million orphaned and separated children (OSCA) around the world. In Kenya, many of these children live with extended family while others live in institutions. Despite evidence that orphans are less likely to be enrolled in school than non-orphans, there is little evidence regarding the role of care environment. This evidence is vital for designing programs and policies that promote access to education for orphans, which is not only their human right but also an important social determinant of health. The purpose of this study was to compare educational attainment among OSCA living in Charitable Children’s Institutions and family-based settings in Uasin Gishu County, Kenya. Methods: This study analyses follow up data from a cohort of OSCA living in 300 randomly selected households and 17 institutions. We used Poisson regression to estimate the effect of care environment on primary school completion among participants age ≥ 14 as well as full and partial secondary school completion among participants age ≥ 18. Risk ratios and 95% confidence intervals were estimated using a bootstrap method with 1000 replications. Results: The analysis included 1406 participants (495 from institutions, 911 from family-based settings). At baseline, 50% were female, the average age was 9.5 years, 54% were double orphans, and 3% were HIV-positive. At follow-up, 76% of participants age ≥ 14 had completed primary school and 32% of participants age ≥ 18 had completed secondary school. Children living in institutions were significantly more likely to complete primary school (aRR: 1.18, 95% CI: 1.10–1.28) and at least 1 year of secondary school (aRR: 1.28, 95% CI: 1.18–1.39) than children in family-based settings. Children living in institutions were less likely to have completed all 4 years secondary school (aRR: 0.79, 95% CI: 0.43–1.18) than children in family-based settings. Conclusion: Children living in institutional environments were more likely to complete primary school and some secondary school than children living in family-based care. Further support is needed for all orphans to improve primary and secondary school completion. Policies that require orphans to leave institution environments upon their eighteenth birthday may be preventing these youth from completing secondary school. The Orphaned and Separated Children’s Assessments Related to their (OSCAR’s) Health and Well-Being Project is a two-phase longitudinal cohort study investigating the effects of care environment on the physical and psychosocial well-being of OSCA in Uasin Gishu County, Kenya [27]. The study enrolled participants < 18 years of age from May 31, 2010 to April 24, 2013. Phase 1 ran from 2010 to 2015 and Phase 2 ran from 2016 to 2019. The OSCAR cohort comprises participants from 300 randomly selected households caring for OSCA and 19 Charitable Children’s Institutions (CCIs) (of 21 in the county at the time of study start-up). All children, both orphaned and non-orphaned, from each household or institution were eligible to participate in the study. Among the 300 households, recruitment was designed to include 100 households receiving the government cash transfer for the support of orphaned children, 100 households not receiving the cash transfer from within the same sub-Locations as those receiving the cash transfers, and 100 households not receiving the cash transfer from different sub-Locations than those receiving the cash transfers. Sub-Locations are administrative boundaries within Uasin Gishu County, each headed by an Assistant Chief [27]. In-depth details about the OSCAR cohort’s study design, setting, and recruitment have been previously reported [27]. This report includes all participants who were orphaned or separated at the time of enrolment into Phase 1. Separated children were defined as those whose biological mother or father was potentially alive, but functionally not part of the child’s life. Since questions regarding education were introduced in Phase 2, the sample is further restricted to participants who completed at least one Phase 2 visit. Of the 19 CCIs recruiting in Phase 1, two were not eligible to participate in Phase 2 since they provided shorter-term care. Therefore, participants from these two CCIs did not enrol in Phase 2 of the OSCAR study and were not included in this analysis. The Moi University College of Health Sciences and Moi Teaching and Referral Hospital Institutional Research and Ethics Committee, the Indiana University Institutional Review Board, and the University of Toronto Research Ethics Boards approved this study. This study conforms to the principles embodied in the Declaration of Helsinki. Written informed consent for participation was provided by the head of household or Director of the CCI. Individual written informed assent was provided by each child aged 7 years and above. Fingerprints were used for both children and guardians who were unable to sign or write their name. This study utilized community-based, participatory processes to inform the research questions, hypotheses, and methods, as detailed elsewhere [27]. To summarize briefly, the Children’s Officers in the region and representatives from CCIs were initially consulted prior to the funding application. They were requested to provide input as to whether such a study would be important from their perspective, and what their priority questions and concerns were. In addition, traditional community assemblies were held in some of the target communities to identify community concerns and priorities with respect to the care of orphaned and vulnerable children. These assemblies were also held following the initiation of the study to maintain regular contact with the community and disseminate findings. We formed an Advisory Board early on, consisting of representatives from communities, CCIs, and Children’s Officers, and this board met regularly throughout the life of the study. Our study disseminated findings through the monthly Uasin Gishu Children’s Services Forum, through additional traditional community assemblies, and through the study website [28]. Data collection was conducted in situ at CCIs and at the OSCAR Project clinic for participants from households. Annually, participants completed a standardized clinical encounter and those ≥10 years of age also completed a psychosocial encounter. The clinical encounter was an enhanced well-child ‘check-up’ administered by the project medical officer (i.e., physician) that included a complete physical history and review of health symptoms. A psychosocial encounter measured education and employment, material well-being, behaviours and risks, peer and family relationships, and mental health. The psychosocial assessment was self-administered for those who could read and write or psychologist-administered for those that could not adequately read or write. A clinical psychologist was always available during the assessments to assist in case of questions, lack of understanding, or distress. Follow-up of cases requiring individual counselling or case management took place on a case by case basis as needed, by study staff. Household level data, including age and education level of the household head, and other characteristics of the care environments (such as shelter type and source of water) were obtained through annual site assessments administered by the Project Manager for CCIs, and Community Health Workers in the participating households. The primary exposure of interest was care environment (institutional or family-based) upon study enrolment [13]. Sociodemographic characteristics were ascertained at the baseline clinical encounter, including age, sex, orphan/separated status (maternal, paternal, or both), HIV status (positive, negative, unknown), hospitalizations in the past year, area of residence (rural or urban) and time living with caregiver (< 6 months, 6 months-2 years, 2–5 years, > 5 years, all the child’s life). The guardian’s level of education at enrolment (none, primary, secondary, vocational, university) was assessed through a site assessment. Follow-up time was defined as the time between the first and last encounters each individual participated in. The Kenyan education system includes 8 years of primary school from ages 6 to 13 (Class 1 to Class 8) and 4 years of secondary school from age 14 to 17 (Form 1 to Form 4). Participants were asked to identify the highest class they had completed in school, if they had ever attended school, if they were currently attending school, and how many days of school they had missed in the past 4 weeks (none, 1–2 days, 3–5 days, > 5 days). The primary outcomes were completion of primary school (Class 8 or higher among participants age 14 or older), completion of one or more years of secondary school (Form 1 or higher among participants age 18 or older), and completion of all 4 years of secondary school (Form 1 or higher among participants age 18 or older) at the time of the participant’s last follow up visit. Descriptive statistics at baseline were calculated for both the initial study population and the population with at least one Phase 2 visit, overall and by care environment. Mean values and standard deviations are reported for normally distributed continuous characteristics, median values and interquartile ranges are reported for non-normally distributed continuous characteristics, and frequencies and percentages are reported for categorical characteristics. Differences in baseline characteristics by care environment were assessed using Pearson’s Chi-Squared tests for categorical characteristics and two-sample t-tests for continuous characteristics. To assess loss to follow up, Pearson’s Chi-Squared tests were used to compare categorical characteristics of participants who completed a Phase 2 visit to participants who did not complete a Phase 2 visit. Continuous characteristics were compared using a two-sample t-test. Educational outcomes at the last follow-up visit were described by frequency and percentage for each care environment. The effect of care environment on each educational outcome (primary school completion, partial secondary school completion, and secondary school completion) was estimated separately using bootstrapped Poisson regression. Poisson regression was chosen to present a risk ratio, the ratio of the cumulative incidence of school completion in the exposed (children from CCIs) and unexposed (children from FBS) groups. Results are reported unadjusted and adjusted for sex, orphan status at enrolment, HIV status at enrolment, and hospitalization in the past year. A sensitivity analysis was conducted to adjust for area (urban or rural). The risk ratios and 95% confidence intervals were estimated using bootstrap resampling with 1000 replications. Sampling of participants with replacement was conducted within each original sampling stratum (CCI, non-cash transfer household, same sub-Location household, and different sub-Location household) to account for clustering. The regression models were fit using inverse probability-of-censoring weights to reduce selection bias from the differential loss to follow by simulating a pseudo-population where the loss to follow up was random [29]. These weights estimate the probability of each participant completing a Phase 2 visit based on their characteristics. The weights were calculated using generalized additive models (GAM) stratified by CCI, non-cash transfer household, same sub-Location household, and different sub-Location household. The GAMs predicted the probability of a participant completing a Phase 2 visit using a smoothed function on age at enrolment and adjusted for sex, area (urban or rural), orphan status at baseline, time with guardian at baseline, recent hospitalization at baseline, and HIV status at baseline.