Field research for this report was carried out in July and August 2007 and August 2008 in Western province (Kakamega district), Nyanza province (Siaya and Bondo districts and Kisumu), Rift Valley province (Uasin Gishu district and Eldoret), Coast province (Mombasa), and Nairobi, including Kibera and Mathare slums.
We conducted 68 interviews with children, parents, and guardians. Twenty-five interviews were with children, of whom 18 were confirmed to be HIV-positive; 6 others were orphans whose status was not known, and 1 was an orphan who had tested HIV-negative. Of the children interviewed, 16 were girls and 9 were boys. Our interviews also included conversations with 29 HIV-positive mothers, 2 HIV-positive fathers, and 12 guardians of children living with HIV. Adults were in a better position to explain to us the obstacles to testing and treatment both for very small children and in some cases (for the reasons described below), for older children. In situations where children were too ill to be interviewed, we refrained from the interview. For those children who were not aware that they are HIV-positive, we limited our questions to issues that did not touch on the child's HIV status.
In addition, we carried out over 80 interviews with social workers, community health workers, counselors, nurses, clinical officers, doctors, administrative medical staff, and nongovernmental organizations (NGOs) working on health or child rights. In the Kenyan government, we interviewed representatives of the Ministry of Health, the Ministry for Gender and Children Affairs, and the Ministry of State for Special Programs. We also interviewed academic experts and staff of international donors, UNICEF, and humanitarian agencies.
Carrying out interviews on a sensitive issue such as HIV poses a number of methodological and ethical challenges, particularly with children. Children who have experienced a parent's suffering and death may be particularly susceptible to re-traumatization and heightened stress when questioned by an unknown person. For each child interviewed, we explained our work in age-appropriate terms and sought their consent. We took great care to interview adults and children in a friendly and sensitive manner, and ensured that the interview took place in a private setting. Information about the HIV status of the adults and children interviewed was kept strictly confidential. In order to protect the identities of the children interviewed, we have replaced all their names in this report with pseudonyms. The names of the community health workers whom we interviewed have generally been withheld in the report.
In order to avoid false expectations of support or financial assistance, we made clear at the start of each interview that we were not able to provide direct individual support to those who spoke with us. Instead, when we encountered situations where people were in acute need of medical treatment or food, we tried to refer them to local NGOs or other actors who could assist them.
Interviews were carried out in English, Swahili, Kikuyu, and Dholuo. Where interpreting was necessary, it was usually provided by a person well known to the interviewee, such as a community health worker or social worker.
In this report, "child" refers to anyone under the age of 18, in accordance with the Convention on the Rights of the Child. However, when referring to HIV/AIDS statistics, we usually refer to children up to age fifteen, as all authoritative HIV/AIDS statistics by the UN classify children over the age of fifteen as adults. This makes it more difficult to get complete data on children up to age eighteen, and is also problematic in that it might lead policymakers to treat adolescents between the ages of fifteen and eighteen as adults when designing policies.
 We use the term community health worker broadly to mean all persons working in the community, with or without payment, as part of health facilities, agencies, or NGOs with health programs. This includes social workers, outreach workers, counselors, adherence counselors, and others whose main role is to provide information about health issues and accompany patients in the process of seeking health care.
 National Children's Bureau (London), "Guidelines For Research," April 2003, http://www.ncb.org.uk/dotpdf/open%20access%20-%20phase%201%20only/research_guidelines_200604.pdf (accessed September 24, 2008).
 The National Child Traumatic Stress Network (Los Angeles, CA and Durham, NC), "Understanding Child Traumatic Stress," 2005, http://www.nctsn.org/nctsn_assets/pdfs/edu_materials/understanding_child_traumatic_stress_brochure_9-29-05.pdf (accessed September 24, 2008).
 Erick O. Nyambedha, "Ethical dilemmas of social science research on AIDS and orphanhood in Western Kenya," Social Science & Medicine, vol. 10, 2008, pp. 1-9.
 Convention on the Rights of the Child (CRC), adopted November 20, 1989, G.A. Res. 44/25, annex, 44 U.N. GAOR Supp. (No 49) at 167, U.N. Doc. A/44/49 (1989), entered into force September 2, 1990, art. 1.