Children's Health, the Nation's Wealth: Assessing and Improving Child Health (2004)

This appendix examines selected national surveys to assess the current status of children’s health measurement and monitoring at the national level. This review is intended to inform an assessment of the current national approaches to measuring children’s health using survey methods. We examined 12 national surveys that collect information on children’s health and both its social and medical influences. Specifically, we gathered data on the design and reach of the surveys (Table B-1) and on the following aspects of child health:

• Health conditions (Table B-2)

• Children’s health status, functioning, and health potential (Table B-3)

• Influences on children’s health (Table B-4)

The surveys that we reviewed encompass the vast majority of large, nationally representative, and publicly available efforts to collect information about children’s health and the individual, social, economic, and medical care influences of health as defined broadly. The review incorporates both public and private initiatives; one-time and ongoing surveys (only the most current version); longitudinal and cross-sectional designs; and surveys focusing on different child age groups. While many other nationally representative studies have been conducted by individual researchers and organizations, most of these are not publicly available and are limited to very narrow topics, making them less useful for monitoring national child health issues over the long term. While these surveys or data collec-

tion systems are important sources of child health information, they are not comprehensive enough to allow for detailed examination of patterns, trends, and disparities in health.

The surveys are ordered from left to right according to the design and frequency of the survey. Organized from left to right are ongoing cross-sectional national surveys of children’s health, followed by shorter duration or paneled longitudinal design surveys, and then small one-time specialty surveys. For each survey, we summarize key aspects of the design, including the frequency, origin date, sponsor, research design, sample size, age-group focus, and respondent (Table B-1). We then review, in Tables B-2 through B-4, the child health measures and influences on child health. Each child health topic is coded with a somewhat arbitrary system that indicates whether the topic is measured “comprehensively” (defined as three or more questions), “adequately” (defined as one or two questions), through a biological or physical mechanism (e.g., blood test), using data from a birth certificate, or not at all. The variables examined in the tables are organized and correspond with the structure of the report. Note: Some questions included in the tables (e.g., MEPS) are asked only of specific age groups (e.g., ages 5-17).

The review identifies patterns of health topics that are commonly covered in national health surveys as well as gaps in which topics have rarely or never been addressed. We discuss the results with respect to whether the survey was a one-time endeavor or an ongoing initiative in order to analyze the consistency of data on specific topics. We also briefly discuss the measurement of race and ethnicity across surveys because of the national focus on eliminating racial and ethnic disparities in health.

Many of the national surveys ask parents to report on specific health conditions that are common among children and adolescents. The most common conditions are asthma, mental disorders (measured broadly), infectious diseases, pregnancy among teens, and child injuries. These conditions, however, were measured in no more than half of all the surveys. While the National Health Interview Survey (NHIS) and the National Health and Nutrition Examination Survey (NHANES) assess the greatest breadth of conditions, NHANES collects the most comprehensive and detailed information on specific conditions. Because NHANES collects biological samples from children (usually a blood test) to assess the presence of certain conditions, it also has the potential to collect things such as biomarkers.

Despite the comprehensiveness of NHIS and NHANES, there remain some gaps in the measurement of children’s health conditions across surveys. Among

the newer morbidities are mental health conditions such as Attention Deficit Disorder and Attention Deficit Hyperactivity Disorder (ADD/ADHD) and depression, and these conditions (though commonly diagnosed) are usually not included in national surveys of child health. Difficulty in identifying and measuring these conditions (beyond parent report of diagnosis) is likely to be a barrier to more frequent inclusion in large national surveys. Screening for the presence of a condition, such as depression or ADHD, often involves the use of 10 or more questions, which can take up a relatively large amount of scarce interview time. There are other nonsurvey data sources, however, that exist for monitoring child health conditions, including the annual National Ambulatory Medical Care Survey that collects data from medical offices about patient symptoms, diagnoses, and ambulatory care provided.

Nearly every national survey assesses some aspect of a child’s physical health. The majority of the measures currently in national surveys relate to aspects of physical functioning and impairments or deficits in mobility, ability to do usual activities, or more specific deficits in hearing, vision, or speech. Some more recent surveys have tended to adopt broader perspectives on children’s health, evaluating aspects of cognitive, emotional, and even social functioning. The broader perspective of what constitutes health is not routinely included in ongoing surveys but is rather the focus of more detailed and topic-specific one-time surveys.

The largest apparent gaps in the assessment of children’s health are in the evaluation of health potential (or rather more positive aspects of development and functioning). While many surveys asked about impairments in functioning, only a handful of one-time studies actually incorporated questions about positive developmental and functional trajectories, such as positive personal affect or self-sufficiency. Refined understandings of health and well-being recognize that health is more than merely the absence of illness, although this is yet to be reflected in ongoing national surveys of children’s health.

National surveys of child health have made tremendous steps in recognizing the importance of family determinants on child development. Nearly every survey measures race and ethnicity and some aspect of socioeconomic status (most commonly income and education) of the family. Many of the one-time special surveys have also recognized the contribution of family composition to children’s health and the threats to development that are potentially associated with family disruptions (such as divorce), parent health status, and aspects of parenting (such as discipline and providing rich learning environments). Of particular interest is the recent attention given to the role of child care (both formal and informal) in

national surveys. For example, the Early Childhood Longitudinal Study—Birth Cohort (ECLS-B) not only asks a series of questions to parents about their child care needs and beliefs, but also interviews the care providers and directors of the child care centers themselves. Interestingly, the ECLS-B also incorporates direct observation of child care practices to assess the quality of care that is provided.

While substantial progress has been made with the evaluation of family influences on health, few national surveys incorporate questions about community factors that may influence children’s health and development. Add Health is one of the few surveys to ask questions about community socioeconomic level, community unemployment rates, physical safety in the community (e.g., school and housing safety), and aspects of the social organization of the neighborhood (e.g., social cohesion, diversity, and social networks). Uniquely, Add Health further incorporates the use of direct observation of neighborhoods to note the safety of the neighborhood, housing adequacy, and other factors. While these community factors are increasingly known to influence children’s health, they are rarely included in ongoing national surveys.

National surveys of child health also frequently assess the influences of health systems on children’s health and development, most often incorporating measures of access to care and health service use. Ongoing surveys also seem to focus on childhood immunizations in order to support national efforts to monitor children’s immunization status. Health insurance is covered in all but three of the surveys.

Measures of quality of health care, particularly those examining more qualitative aspects of care (e.g., patient-provider interpersonal factors), are underrepresented in the ongoing national surveys. Coordination of care is nearly non-existent in the surveys, with the exception of the National Survey of Children with Special Health Care Needs, which examines experiences of children with special health care needs. Relatively little information is provided on the content of care, particularly the receipt of childhood preventive services other than immunizations. The National Survey of Early Childhood Health (NSECH) comprehensively asks parents about preventive services they have received, and uniquely ascertains whether the parents who did not receive the services would have found them helpful. This allows for some analyses of missed opportunities to provide needed preventive care. The NSECH is a one-time endeavor, but it may serve as a model for incorporation of preventive care questions into ongoing national surveys.

When considered together, this collection of national surveys covers a very large number of domains of children’s health and many family, community, and

medical influences on their health and development. When considered individually (because information from across surveys can rarely be combined), there remain substantial gaps in what is measured for children, particularly regarding positive functioning or health potential for children, family, or parenting processes, neighborhood and community influences, and aspects of health care quality. Moreover, with the exception of preventive care, there is almost no attempt to tailor measures to appropriate age groups or make them age-specific.

While recurrent surveys such as NHIS are well established and structured to routinely collect standard information about child health, their established nature makes them somewhat resistant to change other than through special supplements. The NHANES is, perhaps, one exception because it already collects biological samples from children, and these could be used to easily screen for additional biomarkers. Despite these gaps in measurement, a number of one-time surveys have successfully collected data on measures at the forefront of children’s health issues. Collection of these data not only demonstrates that these issues can be successfully measured, but also serves as a testing ground for the validity and reliability of the measures, easing their transition into other ongoing national surveys.

TABLES B-1 THROUGH B-4 FOLLOW