Worldwide, 16.6 million children <18 years of age have lost 1 or both their parents to HIV; 90% of these orphans live in sub-Saharan Africa.1 Given that orphanhood has been associated with poorer health outcomes, educational attainment, and economic disadvantage, this population is of key concern.2 Moreover, there is a larger group of children and adolescents who have increased vulnerability due to severe illness in the family or overall household poverty, affecting their overall well-being and development.2
Prior estimates in Kenya have found that approximately 3.6 million3 children are orphaned or are vulnerable, and represent almost one-fifth of the total population aged <18 years.4 It has been estimated that 1.1 million, or 44%, of these children have been orphaned due to HIV,5 having lost either 1 or both of their parents to the disease. With an HIV prevalence of 5.6% among adults6 and antiretroviral therapy (ART) being taken by 61% of people who are clinically eligible,7 the number of orphans and vulnerable children (OVC) will likely continue to increase well into the future particularly as AIDS remains the leading cause of death of adults in Kenya.8
Previous studies conducted in Kenya have examined the association between OVC status and immunizations,9,10 nutrition,10–13 and mental health14,15 outcomes. Results obtained have been mixed, with some studies showing an association between OVC status and poor immunization and nutritional status.10,13,14 Vulnerable children with HIV-infected parents were significantly less likely to attend school compared with children of HIV-uninfected parents.10 Few studies have examined the relationship between OVC status and HIV risk behavior, but 1 study conducted among persons aged 10–18 years in Nyanza, the region with the highest HIV prevalence in Kenya, found no significant differences between orphans and non-orphans with regard to the rates of occurrence of sexual activity or sex-related self-efficacy.15
The Government of Kenya views OVC as a priority population in the national response to the HIV epidemic. The Kenya National AIDS Strategic Plan III for 2009–2013 outlines strategies for improving the welfare of OVC through educational, economic, and social support, and 8.4% of the plan's total budget was allocated to OVC programming.16 Social safety net and protection programs targeting OVC households have also been established throughout the country, and cash transfers have been used to promote school attendance and health service use by OVC.17 Nevertheless, despite this recognition of the need to provide age-appropriate services and protection to the OVC population, much is still unknown about its size, its characteristics, the proportion receiving support services, and what its social and developmental outcomes are. Even less understood is the proportion of OVC due to HIV/AIDS and how this group's profile may differ from that of children who are vulnerable due to other reasons.
The second Kenya AIDS Indicator Survey (KAIS 2012) provides a unique opportunity to describe the OVC population in Kenya and to establish national population estimates of OVC using a nationally representative survey. The main objective of this analysis is to present the profile of OVC with regard to their demographic characteristics, use of support services, and their population size.
KAIS 2012 was a nationally representative population-based household survey conducted between October 2012 and February 2013. The survey used a 2-stage cluster sampling design to derive HIV prevalence estimates for the adult and adolescent (aged 15–64 years) and child (aged 18 months–14 years) populations. Because of regional security concerns, the North Eastern region on the Somali border was not surveyed. Full details of the study design and methodology are described elsewhere.18
The study population for this analysis comprised all children from birth to age 17 years who resided in an eligible household. Only children who resided in a household where a competent household head or other designated respondent was found were included in the survey. Eligible household heads were adults aged 18–64 years or minors aged 15–17 years who had been married, had children, or were pregnant. Given the sensitivity of conducting research among orphans and to ensure the protection of this vulnerable population, households that were headed by orphans aged <15 years were not eligible to participate in the survey.
We defined orphans as children who were <18 years of age who had lost 1 or both their parents, as per the national definition,19 and we further classified orphans into double orphans (both parents dead) or single orphans (1 parent dead). Maternal and paternal orphans were used to describe whether the child had lost the mother or the father, respectively. We defined a vulnerable child as one whose parent(s), living in or outside of the child's household, or any adult (aged ≥18 years), living in the child's same household, had been chronically ill. Chronic illness was defined as the condition of being very ill for at least 3 months in the past 1 year before the survey.
Data were collected using a household-level questionnaire administered to the head of the household, or if the household head was not present, it was administered to another designated respected household member. The household questionnaire consisted of 4 components: (1) sociodemographic characteristics of each household member, (2) household characteristics and possessions, (3) household availability of and access to food, and (4) support received for OVC residing in that household. The sociodemographic section included questions about the living and health status of the parents for all children from birth to 17 years of age, and about the health status of all household members aged ≥18 years. The household wealth index was calculated using the Demographic and Health Survey wealth index based on the household characteristics and possessions.20 The Household Hunger Scale was used to determine the level of hunger within the household.21 The OVC support section asked households with at least 1 OVC whether the household received any medical, psychological, material, social, or educational support for that child.
All analyses were performed in SAS version 9.3 (SAS Institute Inc, Cary, NC). We used the SAS sample survey procedures to account for stratification, clustering, and weighting due to the survey's 2-stage cluster sampling design. We conducted descriptive analyses to examine the demographic and socioeconomic characteristics of the OVC households across all OVC categories. The HIV status of the parent(s) was based on the report provided by the head of the household. All analyses were weighted to account for sampling probability and nonresponse at the household level. The national population estimates for OVC, orphans, and vulnerable children were determined by applying the non-normalized survey household weights, based on the 2012 population projections in the 2009 Kenyan Population and Household Census, to derive the population-level frequencies.4
We also compared the prevalence of orphanhood due to all causes in Kenya from 1993 to 2012 among children <15 years of age. We used data from the Kenya Demographic and Health Surveys (KDHS) for 1993, 1998, and 2003 and from KAIS for 2007 and 2012. KDHS 2008/2009 was not included as questions on the living status of the child's parents were not included. Because KDHS only asked about parents' status for children aged ≤14 years, an orphan was defined as any child aged ≤14 years who had 1 or both parents not alive for this comparison. The z-test statistic was used to test for differences between estimates by the year of survey, and the difference was considered to be statistically significant if P was <0.05.
Ethical approval was obtained from the Kenya Medical Research Institute's Ethical Review Committee, the University of California, San Francisco's Committee on Human Research, and the US Center for Disease Control and Prevention's Institutional Review Board. Informed consent was obtained from the head of the household for the household interview.
A total of 9189 households were considered to be eligible for participation in KAIS 2012, and of these, 87% (n = 8035) consented to participate in the household interview. Based on information provided in the household interview, there were a total of 16,126 children aged <18 years living in these households, where 14.4% [95% confidence interval (CI): 13.1 to 15.8] (n = 2362) of these children met our definition of OVC. OVC households, or households with at least 1 OVC, comprised 11.7% (95% CI: 10.8% to 12.6%) of all the households interviewed.
Characteristics of OVC Households
Among the 1104 OVC households, more than half fell in the lowest 2 quintiles for household wealth, and approximately one-fifth of OVC households had experienced moderate or severe hunger (Table 1). The average size of an OVC household was 5.4 members, compared with 3.9 members for non-OVC households (P < 0.0001). Thirty-nine households had both orphaned and vulnerable children, and approximately half (52.3%, 95% CI: 49 to 55.6) of all the OVC households had ≥2 OVC (data not shown).
Characteristics of OVC
Among all OVC, 71.1% (95% CI: 67.1 to 75.1) were orphaned (single and double), and 28.9% (95% CI: 24.9 to 32.9) were found to be vulnerable. The estimated median age of OVC was 10.3 years [interquartile range (IQR), 9.9–10.7] compared with 6.7 years (IQR, 6.5–6.9) for non-OVC, whereas 37.1% (95% CI: 35.0 to 39.1) of all OVC were aged between 10 and 14 years (Table 1). Similar to the non-OVC, three-fourths of the OVC lived in rural areas, and approximately half of all the OVC resided in the Nyanza (27.4%, 95% CI: 22.5 to 32.4) and Rift Valley (26.0%, 95% CI: 21.1 to 30.7) regions. Over 90% (93.8%, 95% CI: 91.3 to 96.4) of school-aged OVC (aged 5–17 years) had ever attended school. Among all OVC, 8.3% (95% CI: 5.6 to 11.1) had parents who had (if dead) or currently have (if sick) HIV infection based on the report provided by the head of the household (Fig. 1).
Characteristics of Orphans
Among the orphans, 15.1% (95% CI: 11.9 to 18.3) had lost both their parents. Double orphans tended to be the oldest across all the OVC subgroups, with an estimated median age of 12.2 years (IQR, 11.4–13.0) (Table 1), and almost half (46.7%, 95% CI: 39.1 to 54.3) of all double orphans were aged between 10 and 14 years (Table 1). Forty-two percent (41.5%, 95% CI: 30.4 to 52.5) of these double orphans lived in the Nyanza region. Ninety-six percent (96.3%, 95% CI: 93.1 to 99.6) of school-aged double orphans had ever attended school. More than one-third of the double orphans (34.5%, 95% CI: 23.1 to 46.0) had at least 1 parent who was infected with HIV (Fig. 1).
Single orphans represented 84.9% (95% CI: 81.7 to 88.1) of all the orphans. Similar to the double orphans, most single orphans were aged between 10 and 14 years, and the estimated median age of the single orphans was 11.0 years (IQR, 10.5–11.5) (Table 1). The majority resided in the Nyanza (28.2%, 95% CI: 21.7 to 34.7) and Rift Valley (27.9%, 95% CI: 21.3 to 34.4) regions. Overall, 93.9% (95% CI: 91.4 to 96.3) of the school-aged single orphans had ever attended school. Maternal orphans comprised 17.3% (95% CI: 13.9 to 20.7) of all single orphans, and paternal orphans were 82.7% (95% CI: 79.3 to 86.1) (data not shown). Among single orphans, 5.9% (95% CI: 2.4 to 9.4) reported that the parent who died had HIV infection (Fig. 1).
Characteristics of Vulnerable Children
Vulnerable children were younger than were double and single orphans but were still older than non-OVC children, and their estimated median age was 7.9 years (IQR, 7.0–8.8) (Table 1). More than two-thirds of all vulnerable children resided in rural areas, and the largest proportion of vulnerable children lived in the Rift Valley region (26.7%, 95% CI: 19.1 to 34.4). For vulnerable children of school-eligible age (5–17 years), 90.6% (95% CI: 83.5 to 97.7) had ever attended school. Among vulnerable children who had a sick parent, 5.5% (95% CI: 2.1 to 8.9) had a sick parent who had HIV infection (Fig. 1). These children with an HIV-infected sick parent comprised 3.7% (95% CI: 1.3 to 6) of all vulnerable children (data not shown).
Support Services for OVC
Additional information on receipt of OVC support services was provided for approximately two-thirds of all the OVC (n = 1520) by their heads of household. Of these children, very few OVC had received any medical (3.7%), psychological (4.1%), or material support (6.2%) in the past 12 months (Table 2). Although still low, school support was the most common type of support received for school-aged OVC aged 5–17 years, at 11.5%.
Population Size Estimates of OVC
Using our definition of OVC, we estimated that there were 2.6 million children aged <18 years who were orphaned or were vulnerable in 2012 (Table 3). Males accounted for 1.33 million and females accounted for 1.27 million OVC. Over 1.8 million (71.1%) of these children were orphans, of whom approximately 1.6 million were single orphans and 280,000 were double orphans. Vulnerable children accounted for 749,479 or 28.9%, of the OVC population. Across the 4 age groups, the 10 to 14 year age group had the greatest number of OVC, at 962,475. The regions with the largest populations of OVC were Nyanza (∼713,000) and the Rift Valley (∼674,000). We also estimated that approximately 217,000 OVC had a parent or parents infected with HIV (data not shown).
Orphanhood in Kenya From 1993 to 2012
Orphanhood among children aged ≤14 years in Kenya increased between 1993 (7.0%, 95% CI: 6.2 to 7.7) and 2003 (11.1%, 95% CI: 10.2 to 12.1) (P < 0.01) but declined between 2003 and 2012, to 8.8% (95% CI: 7.8 to 9.8) (P < 0.01) (Fig. 2). The proportion of orphans among all the children aged ≤14 years was the highest in 2003.
A substantial number of Kenyan children aged <18 years were OVC in 2012. Approximately 70% of the OVC were orphans, and the remaining 30% were vulnerable due to a chronically ill parent or adult in the household. Overall, 8% of the OVC had 1 or both parents who were reported to have been infected with HIV.
To our knowledge, this is the first population-based survey to estimate the size of the OVC population aged <18 years and to attribute the number of children who were orphaned or vulnerable due to HIV/AIDS. Our population estimate for orphans is within the range of the current national estimate of 3.6 million based on modeling.3 Survey-based estimates have been found to be lower than model-based estimates in other contexts, possibly due to the underreporting of orphan status in surveys, overestimates of mortality in the model, or both.22
More than one-tenth of all the households in Kenya had at least 1 child who had been orphaned or was vulnerable due to the presence of a very sick adult or household member. Most households that had at least 1 OVC were in the lowest wealth quintiles and were larger in size than non-OVC households. Approximately one-half of the OVC households were caring for multiple OVC. Half of all the OVC resided in the Nyanza and Rift Valley regions, an expected finding given that the Nyanza region has the highest HIV prevalence6 and the Rift Valley region is the most populous among all the regions.5 OVC were typically young adolescents, and orphans were older than vulnerable children.
Over the past 2 decades, orphanhood among children aged <15 years in Kenya increased from 1993 to 2003 but declined between 2003 and 2012. In 2003, one-tenth of all children aged <15 years were orphans, and this can be seen as a result of high HIV prevalence of 7%–10%5 throughout the preceding decade. ART was widely introduced in the public sector in Kenya in 2003. It is unclear whether the introduction of ART and its scaleup since then is associated with the decreasing proportion of orphans, but the literature suggests that this relationship is plausible.23 One modeling study of 10 sub-Saharan Africa countries concluded that universal ART access by adults would avert 717,382 orphans due to AIDS over a 10-year period in Kenya.24
The majority of OVC and their households were not receiving OVC-targeted support services. These services are intended to address the basic or core needs of the OVC population (food and nutrition; shelter and care; legal protection; health care; psychosocial support; and education), as outlined by the global framework for the protection, care, and support of OVC.2 Our data suggest that the coverage of these services is limited and that there is a need to evaluate the design and delivery of these services to determine how best to reach these children.
Our study had several limitations. First, our OVC population size estimates did not include estimates from the North Eastern region; did not capture those orphans <15 years of age who were themselves heads of households; nor did they account for those OVC who did not live in households, such as those who were homeless and living on the street or living in institutions. Information on the number of household heads who were orphans and aged <15 years was not recorded by the survey, and to our knowledge, there are no estimates available for the number and proportion of OVC living in institutions or who are street children. Nevertheless, we surmise that the OVC population is greater than the estimates we derived. Because of the cross-sectional design of the survey, the sample excluded children who may have died shortly after their parent(s) died, potentially underestimating the true number of orphans in the population. Further, our definition of vulnerable children was limited to those children who had a very sick parent or lived with a very sick adult household member; it did not capture those who were vulnerable due to household poverty, and thereby underestimated the population size and the need for support services for vulnerable children. We were also not able to calculate the number of children who were vulnerable due to a household member having HIV, as HIV status was asked only for parents. The HIV status of parents was based on the report provided by the household head and may have been underreported given the stigma of HIV/AIDS. Due to interviewer misclassification of OVC and head of the household nonresponse, the findings on school attendance among OVC and OVC support received by the household represent approximately two-thirds of the total OVC study population, limiting the generalizability of our findings on the coverage of these OVC services. Because of the design of the questionnaire, we were not able to directly examine the receipt of cash transfers by OVC households, a key OVC program in Kenya, in addition to school retention and secondary school enrollment among OVC. Last, given that the primary intent of this analysis was to describe the OVC population, only descriptive analyses were conducted. Results therefore do not account for factors that may confound the magnitude of the estimates.
Despite these limitations being present, KAIS 2012 provides important population-based information on the profile of OVC, the population sizes of OVC groups, and trends in orphanhood in Kenya. Understanding where these children are most concentrated in the country and information pertaining to their ages, the proportion affected by HIV/AIDS, their household characteristics, and the extent of households with multiple OVC are essential for informing interventions that can effectively target the OVC population. Although the data are not reflective of the full scope of the OVC intervention programming in Kenya, the low coverage of basic OVC support received among those reporting suggests that reaching these children with services has been challenging. Additional interventions that are evidence based, targeted, and age appropriate will be required to advance the development and ensure the well-being of OVC. This will be especially critical as the population ages and transitions into adulthood. Closing the gap of the unmet need for support services must be a key priority for the government and other stakeholders to ensure the protection, care, and support of at least 2.6 million children.
The authors thank the KAIS field teams for their contribution during KAIS data collection and all the children and families that participated in this national survey. The authors would also like to thank Anthony Gichangi, Timothy Kellogg, Samuel Mwalili, Anthony Waruru, John Williamson, and Eddas Bennett for their statistical input; Kevin DeCock, George Rutherford, Amanda Viitanen, and Angele Marandet for reviewing and providing input on the manuscript; Ray Shiraishi, Eddas Bennett, and Paul Stupp for their input in weighting of the data set; and the KAIS Study Group for their contribution to the design of the survey and collection of the data set: Willis Akhwale, Sehin Birhanu, John Bore, Angela Broad, Robert Buluma, Thomas Gachuki, Jennifer Galbraith, Anthony Gichangi, Beth Gikonyo, Margaret Gitau, Joshua Gitonga, Mike Grasso, Malayah Harper, Andrew Imbwaga, Muthoni Junghae, Mutua Kakinyi, Samuel Mwangi Kamiru, Nicholas Owenje Kandege, Lucy Kanyara, Yasuyo Kawamura, Timothy Kellogg, George Kichamu, Andrea Kim, Lucy Kimondo, Davies Kimanga, Elija Kinyanjui, Stephen Kipkerich, Danson Kimutai Koske, Boniface O. K'Oyugi, Veronica Lee, Serenita Lewis, William Maina, Ernest Makokha, Agneta Mbithi, Joy Mirjahangir, Ibrahim Mohamed, Rex Mpazanje, Silas Mulwa, Nicolas Muraguri, Patrick Murithi, Lilly Muthoni, James Muttunga, Jane Mwangi, Mary Mwangi, Sophie Mwanyumba, Francis Ndichu, Anne Ng'ang'a, James Ng'ang'a, John Gitahi Ng'ang'a, Lucy Ng'ang'a, Carol Ngare, Bernadette Ng'eno, Inviolata Njeri, David Njogu, Bernard Obasi, Macdonald Obudho, Edwin Ochieng, Linus Odawo, Jacob Odhiambo, Caleb Ogada, Samuel Ogola, David Ojakaa, James Kwach Ojwang, George Okumu, Patricia Oluoch, Tom Oluoch, Kenneth Ochieng Omondi, Osborn Otieno, Yakubu Owolabi, Bharat Parekh, George Rutherford, Sandra Schwarcz, Shanaaz Sharrif, Victor Ssempijja, Lydia Tabuke, Yuko Takanaka, Mamo Umuro, Brian Eugene Wakhutu, Celia Wandera, John Wanyungu, Wanjiru Waruiru, Anthony Waruru, Paul Waweru, Larry Westerman, and Kelly Winter.
1. Joint United Nations Programme on HIV/AIDS (UNAIDS). Global Report: UNAIDS Report on the Global AIDS Epidemic 2010. Geneva, Switzerland: Joint United Nations Programme on HIV/AIDS (UNAIDS); 2010. Available at: http://www.unaids.org/globalreport/global_report.htm
. Accessed May 24, 2013.
2. United Nations Children's Fund (UNICEF). The Framework for the Protection, Care, and Support of Orphans and Vulnerable Children Living in a World With HIV and AIDS. New York, NY: United Nations Children's Fund (UNICEF); 2004. Available at: http://www.unicef.org/aids/files/Framework_English.pdf
. Accessed May 24, 2013.
4. National AIDS Control Council (NACC), National AIDS and STI Control Programme (NASCOP). The Kenya AIDS Epidemic Update 2012. Nairobi, Kenya: NACC, NASCOP; 2011. Available at: http://www.nacc.or.ke/
. Accessed May 24, 2013.
5. Kenya National Bureau of Statistics (KNBS). Kenya Population and Housing Census. Nairobi, Kenya: KNBS; 2010. Available at: http://www.knbs.or.ke
. Accessed May 24, 2013.
6. Kimanga DO, Ogola S, Umuro M, et al.. Prevalence of HIV infection, trends, and risk factors among persons aged 15–64 years in Kenya: results from a nationally representative study. J Acquir Immune Defic Syndr. 2014;66(suppl 1):S13–S26.
7. Odhiambo J, Kellogg TA, Kim AA, et al.. Antiretroviral treatment scale-up among persons living with HIV in Kenya: results from a nationally representative survey. J Acquir Immune Defic Syndr. 2014;66(suppl 1):S116–S122.
8. World Health Organization (WHO). World Health Statistics 2013. Geneva, Switzerland: World Health Organization (WHO); 2013.
9. Radcliff E, Racine EF, Brunner Huber LR, et al.. Association between family composition and the well-being of vulnerable children in Nairobi, Kenya. Matern Child Health J. 2012;16:1232–1240.
10. Mishra V, Arnold F, Otieno F, et al.. Education and nutritional status of orphans and children of HIV-infected parents in Kenya. AIDS Educ Prev. 2007;19:383–395.
11. Akwara PA, Noubary B, Lim Ah Ken P, et al.. Who is the vulnerable child? Using survey data to identify children at risk in the era of HIV and AIDS. AIDS Care. 2010;22:1066–1085.
12. Magadi MA. Household and community HIV/AIDS status and child malnutrition in sub-Saharan Africa: evidence from the demographic and health surveys. Soc Sci Med. 2011;73:436–446.
13. Lindblade KA, Odhiambo F, Rosen DH, et al.. Health and nutritional status of orphans <6 years old cared for by relatives in western Kenya. Trop Med Int Health. 2003;8:67–72.
14. Okawa S, Yasuoka J, Ishikawa N, et al.. Perceived social support and the psychological well-being of AIDS orphans in urban Kenya. AIDS Care. 2011;23:1177–1185.
15. Puffer ES, Drabkin AS, Stashko AL, et al.. Orphan status, HIV risk behavior, and mental health among adolescents in rural Kenya. J Pediatr Psychol. 2012;37:868–878.
16. National AIDS Control Council (NACC). Kenya National AIDS Strategic Plan 2009/10-2012/13: Delivering on Universal Access to Services. Nairobi, Kenya: NACC; 2009. Available at: http://www.nacc.or.ke/
. Accessed February 10, 2013.
18. Waruiru W, Kim AA, Kimanga DO, et al.. The Kenya AIDS indicator survey 2012: rationale, methods, description of participants and response rates. J Acquir Immune Defic Syndr. 2014;66(suppl 1):S3–S12.
19. Department of Children's Services, Kenya. National Plan of Action for Orphans and Vulnerable Children, Kenya, 2007–2010. Nairobi, Kenya: Department of Children Services, Ministry of Gender, Child Social Development; 2008. Available at: http://www.ovcsupport.net/s/library.php?lo=391
. Accessed February 10, 2013.
20. Rutstein S, Johnson K. The DHS Wealth Index. Calverton, MD: ORC Macro; 2004.
22. Grassly NC, Lewis JJ, Mahy M, et al.. Comparison of household-survey estimates with projections of mortality and orphan numbers in sub-Saharan Africa in the era of HIV/AIDS. Popul Stud (Camb). 2004;58:207–217.
23. Nyberg BJ, Yates DD, Lovich R, et al.. Saving lives for a lifetime: supporting orphans and vulnerable children impacted by HIV/AIDS. J Acquir Immune Defic Syndr. 2012;60(suppl 3):S127–S135.
24. Anema A, Au-Yeung CG, Joffres M, et al.. Estimating the impact of expanded access to antiretroviral therapy on maternal, paternal and double orphans in sub-Saharan Africa, 2009–2020. AIDS Res Ther. 2011;8:13.