After my parents died: The effect of HIV on the mental health of children: a clinical review

Contributors and reviewers

This clinical review is kindly supported by the Diana, Princess of Wales Memorial Fund.

Dr Lucie Cluver, Department of Social Policy and Social Work & Centre for AIDS Intedisciplinary Research at Oxford, University of Oxford, UK & Cape Town Child Welfare Society; Dr Karilyn Collins, Palliative Care Works, UK & Tanzania; Chris Green, Spiritia Foundation, Indonesia; Dr Beena Thomas, Tuberculosis Research Centre, India; Dr Liz Corbett, London School of Tropical Medicine, Malawi; Dr Michelle Meiring, Hospice and Palliative Care Association of South Africa (HPCA), The Bigshoes Foundation; Sue Boucher, International Children’s Palliative Care Network (ICPCN); Dr Annelies Van Rie, University of North Carolina.

Who is an AIDS orphan and a vulnerable child, and are their problems so much worse than other children?

Several years ago, UNICEF’s Global Monitoring and Evaluation Group proposed the following definitions for orphans and vulnerable children:

An orphan is a child below the age of 18 who has lost one or both parents (ignoring any distinction between single orphans (who have lost one parent) and double orphans (who have lost both parents).

A child made vulnerable by HIV/AIDS is below the age of 18 and;
a)  has lost one or both parents, or
b)  has a chronically ill parent (regardless of whether the parent lives in the same household as the child) or
c)   lives in a household where in the past 12 months at least one adult died and was sick for three of the 12 months before he/she died, or
d)   lives in a household where at least one adult was seriously ill for at least 3 months in the past 12 months, or lives outside of family care (i.e. lives in an institution or on the streets)

However, there are millions more orphans due to other causes (143 million total worldwide) and countless children dealing with especially difficult circumstances, challenged by poverty, disabilities or social instability in resource-limited settings. Are AIDS-affected children really that much worse off, and should they be singled out for special treatment?

“A growing body of evidence does not clearly demonstrate that orphans (especially single orphans) suffer greater deprivation,” said Dr Penelope Campbell during a presentation at the International  AIDS Conference last year in Mexico City.²⁴  She based this conclusion on a multivariate analysis of Demographic and Health Survey (DHS) data (from 2000-2005) from eleven southern and eastern African countries. Looking at individual indicators, she did tend to find that double orphans have poorer outcomes, for instance, orphan girls— especially those in urban areas— were more likely to quit school, and orphans were more likely to have an early sexual debut.

However, in the multivariate analysis “the overall conclusion from this DHS analysis is that while orphan hood is clearly one important dimension of child vulnerability, other factors are not only important but in some cases have a much stronger quantitative association with the child’s developmental outcomes,” she said. Namely: factors such as poverty, whether the child was from an urban or rural setting and gender are often more significant variables when considering whether a child is “vulnerable.”

But the omission of South Africa, which has the largest population of children orphaned by AIDS, from this analysis is problematic because there are specific problems children have in this setting because of their difficulty in getting birth registration (and thus access to social services). Furthermore, there are very different HIV epidemics occurring in different African countries which are at different stages of maturity, and cultures and approaches to coping orphans and vulnerable children differ in these countries. A multivariate analysis pooling all data may obscure or even negate differences that are very real and important at the local/national level — and local programme responses should respond to local needs.

That being said, Dr Campbell’s point that a number of factors “should be included in targeting criteria” to make certain that vulnerable children (regardless of cause) are not excluded from psychosocial support services is well made. While recognising the heightened need and stigma associated with HIV, programmes should be inclusive when it comes to considering which children are vulnerable and cast the net as wide as possible. Children orphaned by tuberculosis or other diseases may face similar hardships, as do children who have lost homes or families due to military conflict.

Richter and Rama argued a similar point in Save the Children’s Building Resilience: A rights-based approach to children and HIV/AIDS in Africa:²⁵

“The terms used to speak about children affected by AIDS have evolved from AIDS Orphans to Orphans and Vulnerable Children (OVC) to Children Affected by AIDS (CABA) and most recently to Children Living in Communities Affected by HIV/AIDS. The categories OVC and CABA are both used as if there was wide agreement about the children included – when, in fact, there are often very different views about which children are vulnerable and even which children are orphans. The definition of Children Living in Communities Affected by HIV/AIDS is general and embraces the many different categories of children affected by HIV/AIDS.”

When it comes to mental health services, it really doesn’t matter what causes the child’s vulnerability — any child suffering from bereavement, depression, trauma or post-traumatic stress, whatever the cause, deserves care and treatment, and actually, any child suffering from these problems in a high burden setting may also be at risk of HIV. However, most of the following research pertains to children orphaned or in danger of orphanhood due to HIV/AIDS — and suggests that there are significant differences in outcomes. 

Does AIDS-orphanhood or being a child in danger of orphanhood due to HIV/AIDS affect mental health?

Bereavement is a natural part of life, and if managed well (see Addressing grief and bereavement below), should not result in any clinically significant mental disorder. But the impact of other traumas and factors already mentioned on mental health is under-studied.

“The world of a child living in a family affected by HIV/AIDS goes through many changes,” wrote Orne-Gliemann and colleagues in AIDS last year.²⁶  “These children have to witness the physical deterioration and pain of their HIV-infected parents, especially when they are unaware of the nature of the illness affecting their parent(s) or refuse to acknowledge HIV infection. Children are anxious about their source of livelihood and their ability to retain the family home after the parent’s death; separation from siblings is a frequent and important source of trauma. The serial loss of adult figures and carers such as parents, teachers or mentors is also likely to create a sense of insecurity or abandonment.”

However, Dr Cluver, who has conducted some of the most rigorous research and also co-authored what is perhaps the definitive review on the subject believes there is clear evidence of psychological distress in AIDS orphans, but not so much regarding AIDS-affected children: “I think that we now know quite a lot about children who have lost a parent - the impacts on mental health, physical health and education are important. But we know far less about children whose parents are unwell with HIV/AIDS (or other illnesses like TB) — and it's essential that we understand what their needs are and how best we can help them,” Dr Cluver told HATIP.

Cluver & Gardner’s review — which again dealt primarily with orphans — found a paucity of quality research. Including both controlled and uncontrolled studies, in 2007, they could only identify 24 empirical studies (some of which were unpublished and a few ongoing) worldwide (including 5 from the US).²⁷   Most of the studies were rather small as well. While most of these papers identified significant mental health problems in these children, there were wide variations in sampling characteristics, outcome measurements and control groups.

For instance, there were inconsistencies between control groups (such as the lack of any, or the failure to include non-AIDS orphans) that make it difficult to determine the effects of AIDS-orphanhood. Some studies only look at the caregivers’ reports on the child’s mental health (which would tend to under-estimate internalising problems) while others interviewed the child. The tools used to assess mental health problems differed from study to study and they haven’t been adapted for African populations — or for use by non-clinically trained interviewers who tend to be conducting the surveys. 

Finally, they wrote, “we must be cautious in assuming generalisability between studies conducted in diverse parts of Africa, due to variations in social, cultural and economic circumstances.”

That being said, most of the studies found a high degree of internalising disorders such as depression, suicidal ideation, anxiety and especially post-traumatic stress rather than externalising disorders (behavioural problems, anti-social behaviour) in children orphaned by AIDS. Different disorders have been found to be more common in different studies. For instance, depression was the most significant finding in orphaned children in Zimbabwe,²⁸  while a study in Cape Town, found that orphans more often had constant nightmares (p = 0.01), with 73% scored above the cut-off for post-traumatic stress disorder (PTSD).²⁹

Many studies found that other stressors, such as food insecurity, increased the likelihood of psychological distress. For instance, an Ethiopian study compared 479 children who had lost their mother to AIDS with a control group of 574 children orphaned for other reasons. The AIDS orphans had significantly more emotional and social adjustment problems — with the girls faring the worst.³⁰  Notably however, there were other important factors in the poor outcomes, including whether the father was present, whether the child could attend school, household income, clothing conditions, food security and emotional support within the fostering family. While many papers highlight the importance of having one parent still living, as the first case study indicated, sometimes orphans in single-parent families have a harder time economically than children placed with other relatives. Though that case study was from China, the same observation of greater financial hardship among single orphans was also made in one Zimbabwean study — and found to be associated with poorer psychological outcomes.³¹

In contrast to Dr Campbell’s multivariate analysis of selected African countries, some studies found orphanhood to be the only significant predictor of poor mental health outcomes. One example was a study by Atwine et al in rural Uganda, which interviewed 123 orphaned children and 110 matched non-orphaned controls aged 11–15, using the Beck Youth Inventory.³²  Orphans were more likely to suffer from anxiety (OR = 6.4), depression (OR = 6.6) and to display anger (OR = 5.1, and were also significantly more likely to think about suicide. In a multivariate analysis that considered other living circumstances, only orphan status remained significant. Likewise, a national survey in Zimbabwe including 5,321 orphans and non-orphaned children aged 12–17 found that psychosocial disorders, especially depression were significantly more common among orphans, even when controlling for poverty, gender, age of household head, school enrolment and adult support.

In the largest South African study to date, which involved 1,061 children (455 orphaned by AIDS, matched controls of 278 non-orphaned children and 328 orphaned due to other or unknown causes), children orphaned by AIDS reported significantly more depression, peer problems, PTSD, conduct problems and delinquency than the non-orphans or orphans due to other causes.³³ Even when controlling for socio-demographic factors were controlled for, such as age, gender, poverty, migration and household composition, these differences remained.

Even so, some factors had strong ‘strong mediational effects” and should be noted because they may offer potential targets for interventions that may alleviate or even protect against of some of the mental health consequences of orphanhood. The following table is adapted from one of Dr Cluver’s presentations (see http://www.hsrc.ac.za/Document-2033.phtml).

What factors predicted the likelihood of AIDS-orphaned children being more likely to have a clinical-range disorder?

When do these problems start?

Many mental disorders may have their inception while the parent is still alive but their health is failing. Of note, studies by van Rie and colleagues have found that simply being born to a mother with HIV may affect a child’s neurocognitive development, even when the child is uninfected.³⁴  These effects could be the product of HIV’s impact on the mother and the unborn child’s physical health during pregnancy or its impact on the family’s living conditions, and the decreased opportunity for maternal-infant interaction. A review by Dr René Brandt notes that the parental-child relationship is likely to be disrupted as the mother progresses to AIDS, and she also concluded that “children of HIV-positive mothers are at greater risk for emotional and psychological problems than children of uninfected mothers, especially internalising problems such as depression,” though she notes that impoverished children suffer similar problems.³⁵  One factor to consider is the HIV-infected mother’s own mental health status, which is often poor. Possibly counte-balancing this, she notes that caring for children in African societies is often distributed across multiple carers.

Cluver and Gardner write that “More longitudinal studies or studies with a control group of children living with caregivers who are ill with AIDS are also necessary. These can improve understanding of the chronology of distress for children within the process of parental illness and death.” Dr Cluver is currently running a national study in South Africa of children whose parents are sick with HIV/AIDS. It’s called the Young Carers study, which she says is the first of its kind, “a government-academic-kids-NGO collaboration". (See - http://www.youngcarers.netau.net.)

Addressing grief and bereavement

Grief due to the loss of a parent is natural. “Much like the process of physical healing, the grieving process is a series of tasks that one must work through before fully adjusting to the loss,” wrote Li and colleagues, “but characteristics unique to AIDS [such as stigma and discrimination] may create additional cognitive and social barriers for grief resolution among AIDS orphans,” ³⁶  placing these children at heightened risk of prolonged mental and behavioural problems.

A useful overview to helping children cope with death and grief can be found in the recently published guide Children’s Palliative Care in Africa, (which can be downloaded for free, see resources) from which the following words of advice are excerpted or summarised.

“Children who are not allowed to grieve normally are at higher risk of developing problems and complications at the time and later in life,” Amery and colleagues write.³⁷  They note that in Africa, children often have to confront the deaths of “many other family members who have died as a result of the HIV epidemic, malaria or other causes” during their early years. 

Bereavement assistance should start as soon as someone is diagnosed with a terminal illness because “children need to understand and make sense of their experiences in order to help learn that dying and death are part of life... and may later resent their exclusion — their involvement will assist in grieving.”

“Never lie to children or prevent them from talking. It is immoral, unethical and negligent. Children have as much right to ask questions and be given honest, accurate information as adults do,” Amery and colleagues write.

At the same time, caregivers need to be aware that it may be taboo to mention death in some cultures.

Caregivers should know that children may experience extreme fluctuations in grief between sudden sadness and then suddenly appearing happy. “Young children may experience bed wetting, loss of appetite, tummy upsets, restlessness, disturbed sleep, nightmares, crying, attention-seeking behaviour, difficulty in concentrating, increased anxiety and clinginess. These only become a cause for concern when they occur over a prolonged period of time.  Older children often display changes in personality and alterations in their normal behaviour including signs of depression, anger and rage, sleep and appetite disturbances etc... Health workers should advise families to be on the lookout for changes in behaviour.”

The time of death is critical, and it is best to have “a rehearsed plan for all the worst case scenarios in advance and to practice exactly what you will do for each.”

After death, it is important not to just treat the child but the whole family.

The chapter also provides some advice about making memory boxes (to collect items that remind them of the person who has died and times they shared); a family record to help a child or young person gain a sense of where they and the person who has died fits into the family. This is particularly important when a child is to be removed from their old home, or separated from siblings or cousins, and hence when there is a danger of losing a sense of his or her ‘roots’; telling the story byhelping the child write or tell (with an adult writing) their story so that they remember clearly what happened — which can provide carers an ideal opportunity to pick up misconceptions and misunderstandings; handprints:a print of their hand and their parents and other loved one’s touching; writing a “children’s will”:children are sure to have some items that they treasure and they might have a clear idea of what they wish to happen to these; permanency planning so that children have a clear idea about what will happen to them (if they are bereaved); a “bereavement tree” (a practical tool that sensitises people to the feelings and behaviours of individuals and expectations of society, to create awareness about bereavement in order to encourage community support to all bereaved people — there is an appendix with this tool in the book).

The book also addresses the importance of spiritual development (as well as play), and describes a number of beliefs and practices surrounding death in Africa. 

Some other useful resources can be found in the April 2009 issue of Together Now , the newsletter of the International HIV/AIDS Alliance in India including using art therapy for children and child-centred counselling.

Screening for mental health problems in AIDS-affected children

As Cluver and Gardner’s review noted, there is great need for standardised psychological tools that have been adapted for African populations. Some key questionnaires used in some of the studies include the Child Depression Inventory,³⁸  the Revised Children’s Manifest Anxiety Scale,³⁹  the Child Behaviour Checklist⁴⁰ and the Children’s PTSD Checklist⁴¹  but these tools need to be translated and adapted to most local settings.

Other tools, such as the Catholic Relief Services ‘OVC Well-Being Tool’, have recently been developed and validated in five African countries to serve as a comprehensive screen of AIDS-affected children’s wellbeing, looking at the domains of: nutrition and food security, shelter and environment, protection, family, health, spirituality, mental health, education, economic and community cohesion.⁴² The tool is administered to children themselves, however, it has only been validated in children above the age of 13.⁴³

Although also not a mental health screening tool per se, another interesting well-being tool, described by Dr Paul Hutchinson of Tulane University at the HIV Implementers’ Meeting this year assesses the cost-effectiveness of interventions for AIDS-affected children, and includes measures of the child’s psychological health (see www.hivimplementers.org for his presentation).

Some mental health interventions for AIDS-affected children

Basic services

There are a large number of basic psychosocial support interventions and programmes targeting the needs of children in communities affected by AIDS, though coverage is nowhere near adequate — and there is a clear need for national governments in resource-limited settings to take responsibility for the security and health of their children, rather than depending upon the charity of others. As Nelson Mandela said, “There can be no keener revelation of a society’s soul than the way in which it treats its children.” One notable barrier to accessing government services, in many African countries at least, is birth registration, which is very low across the region.

Describing the wide range of basic support services being implemented in HIV-affected communities goes well beyond the scope of this document. However, we should note that a recent review of more than 50 studies and programmes by the Joint Learning Initiative on Children and AIDS, concluded that interventions are needed targetting all poor children in communities heavily affected by HIV and AIDS; and that children's overall wellbeing can be improved by policies and programmes that strengthen families, offer social protection for children and income support.⁴⁴ By addressing the co-stressors for poor mental health — such as food insecurity for vulnerable children, programmes could reduce the risk neglect and trauma. But psychosocial interventions do not go far enough for the child with established emotional or mental health problems.

A model for introducing more targeted interventions

“Basic psychosocial support has its own advantages and uses. It can be delivered by trained volunteers, is inclusive and can be offered to large numbers of affected children. Little training is needed and it provides a basic level of support for children who might otherwise have nothing, but it also has its limits,” said Carl Henn, Palliative Care Advisor for USAID Zambia at this year’s HIV Implementers’ Meeting. “In many cases it’s the only mental health option available. Children generally receive a 'one size fits all' package, regardless of need, and volunteers generally have limited training and skills and little or no training or tools to assess mental health needs. And often they have nowhere to refer children who need help.”

So a number of implementing partners, including Catholic Relief Services, in Zambia came together to develop a model for providing mental health services for children with basic psychosocial support services as the foundation.⁴⁵ More specialised services (paediatric support services (PSS)) are targeted to children who themselves are infected, but community caregivers can also be trained to use a simple evidence-based locally validating assessment to look for the presence of more serious psychosocial needs in AIDS-affected children and either offered trauma-focused cognitive behavioural therapy (for severe needs) (TF-CBT) or interpersonal psychotherapy (IPT-G).

Dr Nancy Scott of Boston University explained the evolution of this model as the result of a four-phase methodological process that has been used in several countries now.⁴⁶  This began with an ethnographic or qualitative survey to determine the local needs. The results were used to develop quantitative assessment tools, which were locally adapted and validated. Then the feasibility of  introducing the evidence-based treatment (necessary for that particular programme’s needs) was piloted. Then the process either concludes with a randomized controlled study of the effectiveness of the intervention or — as in the case of Zambia, simply moving to scale up.

The ethnographic study in Zambia consisting of interviews of caregivers and children (already mentioned above) found that sexual abuse and related post-traumatic stress was far more common than expected, while disorders like depression were somewhat less common^_.47  “We really did go into it thinking that we were going to find depression, and would choose IPT-G for the evidence-based treatment that we would implement. But the qualitative work obviously guided us differently,” she said.

Next they chose assessment tools that were translated and adapted to be usable by Zambians of all education levels, and which could also be used as a tool to measure impact. (Previously, there had been no validated MH assessment tools for the Zambian context.) Working with the University Teaching Hospital, Child Sexual Abuse One-Stop Center in Lusaka; the tool was then validated in about 700 youth and caregivers.

The tool can be used by trained volunteers and was found to reliably differentiate between youth needing different services (HIV Treatment, Medical Services, Drug/Alcohol Services and/or one of the two psychosocial treatments. The validation experiment also detected a great need for the psychosocial services. Those with limited needs received group counselling (IPT-G) (generally depression).

About 20% of the youth assessed had symptoms of depression, according to Henn. Group therapy was chosen as the key intervention for these children on the basis of two controlled studies. One found it to be very effective in reducing depression and dysfunction in the Masaka and Rakai districts of rural Uganda.⁴⁸ The second found that it was effective for depression in the adolescent survivors of war and displacement in northern Uganda.⁴⁹ According to Henn, use of IPT-G in Zambia was piloted in Mazabuka and Chainda (small peri-urban areas), and almost any trained individual can lead the groups, though children have to be 12 years of age or older to participate.

Although he cautioned that the results were still preliminary, Henn said “95% of those diagnosed and participating in IPT-G, improved significantly. It has had a positive effect on families, caregivers and in communities. Are there barriers? Yes, IPT-G involves a certain amount of cost and complexity. But the costs are manageable and the barriers are not insurmountable.”

The Trauma-Focused Cognitive Behavioral Therapy (TF-CBT) is offered for those “who need therapy or treatment to move past this traumatic experience in their life,” Dr Scott said.

TF-CBT was selected on the basis of USAID’s prior global experience with Victims of Torture Fund in the Democratic Republic of Congo, Indonesia and Haiti and over 10 randomised controlled trials.⁵⁰

“It can be used with 3 to 18 year olds over a discrete 3-month time period; and it’s been adapted and shown to be effective across different cultures,” said Dr Scott. “It’s not just for sexually abused children, it’s effective for all types of trauma histories. But it reduces the effects of child abuse, neglect or violence in the short term and symptoms like depression, self-efficacy, behavior problems, nightmares. Over the long term has been proven to reduce drug/alcohol use and/or abuse; violence in relationships: clinical depression and also long-term HIV risk behaviors.  So it really is a prevention tool.”

They then performed a feasibility study, training 23 Zambians over six months (working very intensively for five days with a clinical psychologist and, and then completing training as a distance training course. These were people with very different training backgrounds: clinicians, counsellors, students in psychology, and nurse mid-wives.

“It’s been very effective so far, "said Scott, noting that results are being tabulated and will first be presented to the partners in Zambia. “It requires intensive supervision and practice but can be done locally.” She also noted that though they were concerned about finding space to offer therapy confidentially, “during the feasibility study, we found that nobody wanted that. They wanted to meet in churches; they wanted to meet under a tree; they wanted to meet somewhere else. So it wasn’t even necessary to build that space.”

Scott said that they have decided to scale up the intervention rather than perform a larger randomised trial.  

Simple interventions to improve the mental health of AIDS-affected children can be introduced in low-income settings

These interventions don’t work for every child. In the Zambian programme, Henn said some still may require residential and day care options to deal with alcoholism and drug addiction or very severe psychiatric problems.

But the take-home message from the Zambian experience seem clear. There is a clear need for therapeutic interventions, and basic psychosocial support alone is not enough for all children, said Henn.

“You need a variety of methods and services at various levels, to meet a variety of needs. But simple, low-cost methods exist now that are viable in Zambia. And Zambia has shown that the host government can be a strong advocate for the mental health of its children.”

Some selected resources

• Children’s Palliative Care in Africa, Edited by Amery J. Oxford University Press, 2009. Children’s Palliative Care in Africa addresses key palliative care themes, which are crucial to expanding the provision of children’s palliative care, and is focused on experience in Africa (though it should have relevance in other resource-poor settings. Download a free electronic version from ICPCN here.  Purchase hard copies via Hospice Palliative Care Association of South Africa: http://www.hospicepalliativecaresa.co.za/ or from Oxford University Press: http://ukcatalogue.oup.com.
• UNICEF www.unicef.org
• UNICEF’s Demographic and Health Surveys: www.childinfo.org
• UNICEF’s Africa's Orphaned and Vulnerable Generations: Children affected by AIDS, New York, 2006: www.unicef.org/publications/index_35645.html
• The Joint Learning Initiative on children and HIV/AIDS: www.jlica.org  
• The Young Carers' study, a national study in SA of children whose parents have HIV/AIDS: www.youngcarers.netau.net
• ICPCN (International Children’s Palliative Care Network): http://www.icpcn.org.uk/
• Children of Parents with AIDS (COPWA): www.copwa.org/scopeofservices.html
• Building Resilience in Children Affected by HIV/AIDS, by Sister Silke-Andrea Mallmann and Catholic Aids Action, Namibia; downloadable at http://www.caa.org.na/.
• The Big Shoes Foundation: http://www.bigshoes.org.za/
• A Clinical Guide to Supportive and Palliative Care for HIV/AIDS in Sub-Saharan Africa: www.fhssa.org/i4a/pages/Index.cfm?pageID=3361
• Coalition of Children Affected by AIDS. A collective of private and public foundations and re-granting organizations in the North and South concerned with the welfare of children affected by AIDS: www.ccaba.org
• More than a bailout: Social protection for children affected by AIDS (a radio interview with Prof Linda Richter, an expert from Children's Services in Kenya, about the need for child-sensitive, AIDS-inclusive social protection to be included in the agendas of world leaders.): www.hsrc.ac.za/News-1171.phtml

Some resources listed on PEPFAR.gov:

• Plan International http://plan-international.org
• Promoting child rights to end poverty: Circle of Hope: Children's rights in a world with AIDS, published by Plan, July 2006
• Assistance to Orphans and Other Vulnerable Children in Developing Countries Act, signed by the President, November 8, 2005
• State Department Fact Sheet: President Bush’s Emergency Plan for AIDS Relief: Aid to Orphans and Vulnerable Children
• www.pepfar.gov/press/82280.htm
• www.pepfar.gov/press/docs/126827.htm