Measuring health variables among Hispanic and non-Hispanic children with chronic conditions.

This paper addresses two concerns related to differences in the health status of Hispanic and non-Hispanic children: methodological issues in the measurement of health status across population subgroups and the substantive differences in the health of these subgroups. Interview data from a study of chronically ill children in a northeastern inner city were collected using carefully translated measures of health and health-related behaviors. The psychometric properties of the scales were assessed across the subgroups to determine if common interpretation of the scales was possible. After determining that this was the case, group means in health and health-related variables were compared. Despite sociodemographic group differences in variables, there were remarkably few differences among the groups on traditional morbidity measures. However, significant differences were found on four of five scaled health-related measures (the impact of the child's illness on the family, the child's functional status, and the mental health of both mother and child). These findings did not all favor the same group, suggesting that certain areas of function may present more problems for some subgroups. These differences virtually all disappear when multivariate techniques are used to control for variation in important socioeconomic characteristics among the three subgroups. Statements that the health status of one subgroup is better than that of another are too simplistic if they do not indicate the particular aspect of health status being discussed and control for differences among the groups in maternal education, family structure, maternal welfare status, and similar background characteristics.