The Family Health Project: An Investigation of Children Whose Mothers are HIV-Infected

Rex Forehand, PhD, Lisa Armistead, Phd, Edward Mose, PhD,
Patricia Simon, PhD, and Leslie Clark, PhD

The Family Health Project is a collaborative effort among the University of Georgia, Tulane University, and Louisiana State University Medical School and is supported by the Centers for Disease Control and Prevention (CDC).


The prevalence of HIV/AIDS in women is and will continue to be a societal problem of major proportion. Women represent an increasing percent of the total AIDS cases. Moreover, approximately 55 percent of women with AIDS are African-American, and most of these women hail from economically poor urban communities. The vast majority of HIV-infected women are in their childbearing years (CDC, 1996) and over half of them currently have children (Schable et al., 1995). It has been estimated that, by the year 2000, as many as 125,000 children in the United States will have been affected and orphaned by maternal HIV (Michaels & Levine, 1992). As a result of the physical and psychological sequelae of HIV infection and the risks present in their communities of residence, these children may well be at high risk for a variety of emotional and behavioral problems (Armistead & Forehand, 1995). However, research is not available to determine whether the children of infected women have psychosocial adjustment difficulties and, if so, in what areas. Only through research efforts can prevention and intervention programs be developed for these children and public policy formulated for the orphans of the HIV epidemic.

Goals of The Family Health Project

This project currently tracks and regularly assesses one of the few samples of HIV-infected women and their non-infected 6- to 11-year-old children (sample=107) in the United States. Eligibility for participation required that each woman have a CD4 count of less than 600 upon entry into the project indicating a relatively advanced stage of disease. The sample is being assessed on four occasions across a six-year time span, during which approximately half of the HIV-infected women are expected to die as a result of their illness. The families reside in inner-city New Orleans, are African American in ethnicity and of low socioeconomic status. A demographically similar group of mothers who are not HIV infected and their 6- to 11-year-old children (sample=150) also are being followed.

The Family Health Project was designed from the viewpoint that familial factors play a considerable role in child psychosocial adjustment. Specifically, the central tenet of the project is that it is important to identify if these children are likely to need psychosocial prevention or intervention programs and, if so, which child and family process variables could be effectively targeted in such programs. Thus, the first question to be answered is: Do children whose mothers are HIV infected display more difficulties in psychosocial adjustment relative to children whose mothers are not HIV infected? If the answer to this question is "yes", then the next question becomes: What family and child process variables serve as mechanisms in the relationship between maternal illness and child functioning?

Another goal of the project is to follow these children as their mothers become increasingly ill. As HIV progresses, many of the mothers will likely experience increasingly negative physical and psychological states (e.g., physical debilitation, cognitive decline, repeated hospitalizations, and depressive symptoms). A mother?s declining physical and psychological state may impair her ability to effectively care for her children, resulting in even further impairment in child psychosocial adjustment. Specifically, a mother may find herself too ill to care for her child(ren) independently, and repeated hospitalizations may compromise parenting abilities. The living arrangement for the child may require alterations in the form of a co-caregiver moving in with the family, the family moving in with a co-caregiver, or the children being placed with another adult.

Yet another consideration for HIV-infected mothers as they become increasingly ill is placement for their child upon death (Armistead & Forehand, 1995). Custody planning is perhaps the most challenging task that an infected mother faces and often is made more difficult by complicated legal and social service regulations (Levine, 1995). An assessment of how, if at all, these mothers arrange for alternative care for their children, the barriers they face in doing so, and the resources necessary for facilitating this most difficult task are additional foci of the Family Health Project.

The final major goal of The Family Health Project is to follow these children after the death of their mothers to ascertain the impact of this loss. The loss of a parent, particularly a mother, to death can play a significant role in childhood development because of exacerbating problems such as depression and acting out. In addition to the stress generally associated with maternal loss, orphans of the HIV epidemic face a number of unique stressors, including stigma, isolation, and instability, all of which may be traumatic for these children. The Family Health Project has a unique opportunity to investigate the psychosocial adjustment to orphanhood of children of the HIV/AIDS epidemic.

What Do The Preliminary Data Say?

The schematic presented in Figure 1 provides an overview of some of the findings in The Family Health Project thus far. It is important to emphasize that most of the findings are based on initial analyses of the first wave of data collection and, thus, should be considered preliminary analyses. Therefore, both the findings and their implications should be viewed as tentative. The findings discussed below will be presented in far more detail in various professional journals in the near future.

As presented in Figure 1, The Family Health Project occurs within the context of an inner-city environment which is characterized by crime and poverty. Within this environment, families in the two samples (HIV infected women and their children; non-infected women and their children) are characterized by single parenthood (85%), low educational achievement (44% < H.S.), and unemployment (70%). Thus, the impact of maternal HIV infection on children occurs within community and family contexts that have been traditionally characterized as high risk for children.

Is maternal HIV infection associated with child psychosocial adjustment difficulties? Our findings suggest that, relative to children from the same environment whose mothers are not infected, those whose mothers are infected demonstrate difficulties in the four areas of psychosocial adjustment typically studied by behavioral scientists: externalizing problems (e.g., aggression, disruptive behavior), internalizing problems (e.g., depression, anxiety), pro-social competence (e.g., peer relationships), and cognitive competence (e.g., academic performance).

Given that many of the children of HIV-infected mothers are evidencing difficulties, the next issue is the identification of mechanisms which may account for these disruptions in adjustment and which can be incorporated into prevention and intervention programs. Our preliminary analyses suggest two levels of variables: first level (i.e., those which directly impact on the child) and second level (those which impact on the child indirectly?through their relation with first level variables).

Three first level variables have been identified thus far: the child's receipt of social support, stage of maternal HIV illness (i.e., asymptomatic versus symptomatic/AIDS), and parenting. Psychosocial adjustment is facilitated by social support received from mother and siblings, a positive parent-child relationship, and parental supervision of a child?s activities. Furthermore, when mothers are grouped by severity of their illness (asymptomatic vs. symptomatic/AIDS), child psychosocial adjustment appears to be less affected at the earlier stages of HIV infection.

We have studied several second level variables which appear to work indirectly to influence child psychosocial adjustment. In particular, a mother?s well-being, defined as an optimistic outlook and/or low levels of depressive symptomalogy, relates to adaptive parenting, which relates to child outcome. Other second level variables appear to influence maternal well-being which, in turn, influences child psychosocial adjustment. These variables can be divided into resources and stressors. Resources include social support received from friends and neighbors, as well as use of prayer and problem-focused coping (i.e., identifying and attempting to solve problems). Stressors, particularly victimization (e.g., having been raped, beaten up, or mugged), has been found to negatively impact the psychological well-being of the mother and appear to accelerate the severity of her physical symptoms.

Also identified in Figure 1 are several primary questions which we will address in the future. One question focuses on child psychosocial adjustment, as well as maternal well-being and parenting, as the mother becomes increasingly ill. A second question examines the role of disclosure of maternal HIV status to the child. Less than 15 percent of our sample had disclosed their HIV status to their 6- to 11-year-old child at the first assessment. The relationship between disclosure and child adjustment is yet to be determined. A third question addresses the role of permanency planning for the child?s future prior to the mother's death. A final question focuses on the relationship between number of transitions in living situations and the child's adjustment after the mother?s death. We hypothesize that without permanency planning, a child will experience more transitions after her or his mother?s death and that these transitions, along with the family environment in the new home, will relate to the child's adjustment after the death of the mother.

What Do The Preliminary Data Mean?

Perhaps the first implication of our research for clinicians is a recognition of the importance of the total environmental context in which children whose mothers are HIV infected often live. Maternal HIV infection is only one of many stressors faced by many of these mothers and their children and, until the latter stages of the illness, may well be less pressing than other problems. Anecdotally, mothers tell us that concern about their illness must often take a "back burner" to other more pressing issues, such as safety, food, and shelter. Prevention and intervention programs which focus on maternal HIV must consider this environmental context of the families they serve.

Maternal HIV infection has disproportionately struck ethnic minority women who are economically challenged, minimally educated, and often single parents. Prevention and intervention programs focusing on the children of these women will need to be derived, considered, and implemented with an appreciation and understanding of the various cultural, ethnic, and socioeconomic conditions under which these children live. Prevention and intervention efforts must be sensitive to and tailored to unique aspects of the particular environment in which the child resides in order to be effective. For example, providing child care and transportation will facilitate participation of many women in prevention and intervention programs.

In light of our findings regarding the impact of first order variables on child adjustment, prevention and intervention efforts can be undertaken to enhance the degree of social support these children receive. This can occur through promoting positive sibling relationships, involving children in organized activities outside of the home and, when deemed necessary, teaching social skills. Furthermore, mothers of these children can be taught procedures to enhance the parent-child relationship and to increase monitoring and supervision of their child's activities. Although we are just beginning to explore the role of other family members, grandmothers and aunts of children whose mothers are HIV-infected appear to be available as co-caregivers and, thus, are potential sources of support.

An important implication which can be derived from The Family Health Project data is that the family can impact a child on many different levels. As such, interventions will need to move beyond first level variables that directly impact the child in order to maximize treatment effects. For example because a mother?s well-being relates to how she parents, intervention efforts will need to involve not only teaching parenting skills but also, at least with some women, alleviating maternal psychosocial distress, such as depressive symptoms.

Other second level variables which deserve attention include increasing the available social support and problem-focused coping skills of mothers, as well as decreasing their probability of victimization of HIV-infected women. In addition, behavioral scientists often have overlooked the potential role that religion or spirituality can play as a source of support and strength for individuals in distress. Preliminary analyses suggest that spirituality, particularly prayer, is a frequently utilized coping procedure for HIV-infected women. Acknowledgment by mental health professions of the role of prayer in the lives of these women may well facilitate therapeutic alliances with the women and enhance their well-being.

The role of disclosure of maternal HIV status to a child has not been examined in this population. However, our work with other HIV infected populations, hemophilic men infected through blood transfusions, suggests that the family context within which disclosure may or may not occur is equally or more important than whether or not disclosure occurs. That is, a home environment characterized by a positive parent-child relationship is a better predictor of a child's psychosocial adjustment than is disclosure (Armistead, Klein, Forehand, & Wierson, 1997). The same findings may also be true for inner-city African-American women. Until research data are available addressing the role of variables like disclosure and permanency planning, it is important that mental health workers be aware of these critical issues and work on an individual case-by-case basis with mothers who are HIV infected. Without research data to guide recommendations, mental health professionals will need to provide a supportive, nurturing, and non-judgmental environment in which mothers can explore these issues.


The Family Health Project has the opportunity to answer important questions which have not yet been addressed. Furthermore, the findings which emerge can provide the basis for prevention and intervention programs for children whose mothers are HIV-infected or who are orphans of the AIDS epidemic. In addition, treatment recommendations for the mothers of these children should be evident from the results. Finally, implications for policy development regarding these children and women will be considered and proposed.


Armistead, L., & Forehand, R. (1995). For whom the bell tolls: Parenting decisions and challenges faced by mothers who are HIV seropositive. Clinical Psychology: Science and Practice, 2, 239-250.

Armistead, L., Klein, K., Forehand, R. & Wierson, M. (1997). Disclosure of parental HIV infection to children in families of men with hemophilia: Patterns, outcomes, and the role of the family environment. Journal of Family Psychology, 11, 49-61.

Centers for Disease Control and Prevention. (1996). HIV/AIDS Surveillance Report, 8 (no.2).

Levine, C. (1995). Orphans of the HIV epidemic: Unmet needs in six US cities. AIDS Care, 7, 557-562.

Michaels, D. & Levine, C. (1992). Estimates of the number of motherless youth orphaned by AIDS in the United States. Journal of the American Medical Association, 268, 3456-3461.

Schable, B., Diaz, T., Chu, S.Y., Caldwell, M.B., Conti, L., Alston, O.M., Sorvillo, F., Checko, P.J., Hermann, P., Davidson, A.J., Boyd, D., Fann, S.A., Herr, M., & Frederick, M. (1995). Who are the primary caretakers of children born to HIV-infected mothers? Results from a multistate surveillance project. Pediatrics, 95, 511-515.