BIOSTATISTICS

Biostatistics

Biostatistics



Introduction

The relatively young discipline of statistics evolved during the last century to become an important aspect of all the sciences. The term statistics now usually includes both descriptive and analytical domains. Those analytical methods that have been found to be particularly useful and prevalent in the design and analysis of medical and public health research studies have been termed biostatistics. The use of these methods will be discussed later in the section titled 'Basic analytical statistics (biostatistics) concepts.'

While most readers are unlikely to be practicing biostatisticians, most probably they will be expected to understand and interpret correctly journal articles summarizing research projects relevant to public health practice. Most such articles use statistical techniques in their summaries and analyses (perusal of any recent research journal in medicine or public health verifies this assertion). We review the rationale and interpretation of basic statistical summaries and biostatistical analyses. (Details on many specific statistical methodologies may be found in the several textbooks listed in the section titled 'Further reading.')

Descriptive Statistics in Public Health

Many nations and all U.S. states mandate that certain data are routinely collected on births, deaths, and other key measures of population health. Under federal law, the U.S. National Center for Health Statistics (NCHS) is required to collect, compile, and disseminate a large number of such health measures. As a result, several important health statistics programs have been developed by NCHS in recent decades to better estimate and understand the status and trends of the health of the American population and several subpopulations. These programs include not only the standard vital statistics (NCHS, 2006) such as births, low birth weights, and deaths, but also national surveys of morbidity and health practices. Naturally the collection of such voluminous amounts of data, by NCHS and other public health organizations, requires suitable methods for summarization, depiction, and interpretation. Collectively labeled descriptive statistics, there are a wide variety of these methods including rates, proportions, and other summary measures.

Several summary descriptive statistics have become the bedrock for assessing population health. As such, they can also provide quantified goals for national and state health programs (U.S. Department of Health and Human Services, 2000 U.S. Department of Health and Human Services, Healthy People 2010: Understanding and Improving Health (2nd edn.), U.S. Government Printing Office, Washington, DC (2000).U.S. Dept. of Health and Human Services, 2000). For example, the infant mortality rate is widely perceived as the single most important statistic by which to assess the overall health status of large populations. This is the rate at which infants less than one year of age die. In the United States the infant mortality rate for the entire country was 26 per 1000 live births in 1960, which decreased to 6.9 per 1000 live births in 2000, clearly documenting significant progress in maternal and infant health.

Mortality statistics, by age, gender, cause of death, and state are one of the routine, ongoing reports that appear annually in NCHS publications (Heron and Smith, 2006). To account for the aging population, age-adjusted mortality ...