Objectives: The research reported here seeks to quantitatively assess the comparability of two widely used sources of information on child health: maternal reports and medical records. The analysis provides a comparison of how well maternal reports and medical provider data agree on 15 types of chronic health conditions, ranging from specific illnesses such as asthma or sickle cell anemia to broader categories such as chronic heart or orthopedic conditions, to impairments such as vision, hearing, or speech problems.
Methods: This study uses data on a nationally representative sample of 6201 preschool aged children whose mothers participated in the 1991 Longitudinal Followup (LF) to the 1988 National Maternal and Infant Health Survey (NMIHS) and whose identified health care providers supplied medical visit data for the children. The LF and NMIHS were conducted by the National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention (CDC), in collaboration with a number of other agencies of the Federal Government. In the 1991 LF survey, the mothers' questionnaire included questions on chronic and acute health conditions, accidents requiring medical attention, and hospital admissions for their children that had been included in the 1988 NMIHS. The mothers were also asked to grant NCHS permission to request medical records from the children's medical providers.
Results: Kappa statistics reveal poor agreement for 12 of 15 conditions studied. Weighted prevalence estimates vary widely across sources. For some conditions, despite apparently similar prevalence estimates from the two sources, mothers' reports and medical records identified very different groups of children as "cases." Concurrence rates were lower for children from poor, less educated, and Hispanic families. Reasons for and implications of these findings are also discussed.