Sample Surveys in the Health Sciences

Abstract

The use of sample surveys in biostatistics arises from the need to answer questions about health issues in the community. The earliest uses of survey methodology in health research in the United States were motivated more by policy issues than by epidemiologic questions and paralleled the use of surveys by the US Census Bureau and other Federal agencies to address social and economic concerns. More recently, it has been recognized that research into the causes and outcomes of disease can benefit from studies in the entire community, rather than just in the formal health care system, particularly in countries such as the United States, where access to health care is not uniform. The Affordable Care Act, passed in 2010 with the goals of increasing access to and reducing the costs of health care in the United States, has motivated both government and nongovernment studies to assess its impact. A number of practical concerns about conducting studies in the community then lead to the use of sample surveys, rather than complete enumeration. In this article, we review the history of sample surveys in the health sciences, present an overview of the major elements and features in the design of sample surveys and analysis of data from such studies, and discuss major methodological advances especially in the last decade. Finally, we provide a large set of references that might be of use to the reader wishing to delve more deeply into specific areas.