Background: All infants exposed to human immunodeficiency virus (HIV) prenatally, even those who do not become infected, are at risk for developmental problems because of poverty, prenatal substance abuse, and maternal illness.
Objectives: The purpose was to describe the development of infants of mothers with HIV and to determine, using hierarchical linear models, the longitudinal effects of child characteristics, parental caregiver characteristics, family characteristics, and parenting quality on development.
Methods: Eighty-one infants born to women with HIV and their primary parental caregivers were followed-up until 18 to 24 months of age; 53 infants were always cared for by their biologic mothers, 16 were always cared for by kin or foster parents, and 12 had primary caregiver changes. Predictor variables and developmental outcomes were obtained at enrollment and 6, 12, 18, and 24 months.
Results: Mental development and adaptive behavior scores decreased over age. Infants with changes in their primary caregiver had lower motor and adaptive behavior scores than infants remaining with consistent caregivers. Higher mental, motor, and adaptive behavior scores were associated with more positive attention and more negative control, whereas better language abilities were associated only with more positive attention. Child, maternal, and family characteristics had lesser effects. HIV-infected infants and infants of mothers with more education had lower mental, motor, and adaptive behavior scores. Male gender and more family conflict were associated with lower motor and adaptive behavior scores. Infants from smaller families had lower mental scores.
Conclusions: Because both parenting quality and consistency of the primary caregiver influenced developmental outcomes, interventions with the mothers of these infants need to focus both on improving the quality of parenting and reducing the frequency of primary caregiver changes.
Diane Holditch-Davis, PhD, RN, FAAN, is Professor and Director, Doctoral and Post-doctoral Program.
Margaret Shandor Miles, PhD, RN, FAAN, is Professor.
Margaret Burchinal, PhD, is Director, Design and Statistical Computing Unit, Frank Porter Graham Child Development Center.
Karen O'Donnell, PhD, is Associate Clinical Professor, Psychiatry and Pediatrics, Duke University School of Medicine, Durham, North Carolina.
Ross McKinney, MD, is Associate Professor and Division chief, Pediatric Infectious Disease, Duke University School of Medicine, Durham, North Carolina.
Wilma Lim, MD, is Clinical Associate Professor, Pediatrics, School of Medicine.
Unless otherwise noted, all authors are from the School of Nursing University of North Carolina at Chapel Hill, Chapel Hill, NC.
Accepted for publication September 9, 1999.
The authors acknowledge the support of Grants MH R01 51019, RR 0046GCRC, and NIH 2U01A127535, and P30 HD27360 from the National Institutes of Health and wish to thank Steve Magers, Statistical Programmer, Frank Porter Graham Development Center, University of North Carolina at Chapel Hill; and Joann Haggerty, Research Instructor; Tanya Kewson, Research Assistant; Sally Laliberte, Clinical Assistant Professor; Yvonne Wasilewski, Project Director and Research Assistant Professor; Susan Brunssen, Research Instructor; Charlene Krueger, Doctoral Student; and Deborah Assad Lee, Doctoral Student; all at the School of Nursing, University of North Carolina at Chapel Hill; for technical assistance.
Address reprint requests to: Diane Holditch-Davis, PhD, RN, FAAN, CB#7460 Carrington Hall, School of Nursing, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7460; e-mail: firstname.lastname@example.org