Subject Guides in the Health and Life Sciences

Resources in Vital and Health Statistics: Finding and Using Statistics

Identifying sources of relevant health-related statistics can be a challenge. Statistics are kept by groups as diverse as the World Health Organization, the U.S. Department of Energy, and Counting California. Unfortunately, due to the complexity of the information available, there is no single way to start looking. Outlined below are several points to consider when doing your search.

Is there a government or private agency that would be concerned about your search area?	All levels of government, from the United Nations down to individual cities, produce statistics in the course of fulfilling their individual missions. Often these statistics are then made available either in print form or on the Internet. There are also many private foundations and organizations that make information they collect and produce available to the public. If you can identify an appropriate agency, you can then search their web site and print publications, or contact the agency directly. Examples of this search method are: visiting the Centers for Disease Control site for current Anthrax statistics contacting the `Ahahui o na Kauka (Association of Native Hawaiian Physicians) to find the number of currently practicing Native Hawaiian physicians
Do you need very current (within the last year) or historical statistics?	The time frame of your research will also affect where you find your statistics. There is generally a lead time of at least one year before most statistics are published. If you need more current information, you can try looking in journal and newspaper articles or press releases. To do this, you will need to identify a topic that is relevant to the statistics you need and look for articles on that topic. Many statistics prior to 1960 are available only in print format, although you may be able to find these quoted as comparisons with more current research. Some examples of articles containing statistics like this are: a 2002 article in Diabetes Care on the current incidence of diabetes in Native Americans an article on the rates of cardiovascular disease from the late nineteenth century to the late twentieth century in the American Journal of Public Health
What type of statistics do you need? (Vital, demographic, health, etc.)	Vital statistics are records of births, marriages/divorces and deaths. Example: the number of women over 40 who gave birth in the Los Angeles area Demographics describe a specific population group; often this group is defined by geographic region. Example: the number of people who live in the Los Angeles area Health statistics, also called mortality and/or morbidity statistics, detail the incidence of certain diseases and conditions. Example: the number of deaths related to illegal drugs (note: this information may also be found both in vital statistics and in mortality and morbidity reports)
What level of statistics do you need?	Locating statistics on a national or international level is very different from locating statistics on a local level. Generally, it is a good idea to use a source as close to the area you are researching as possible. Examples of finding different levels of measles vaccination in children are: visiting UNICEF’s site to find the worldwide vaccination rate of children under age five (72% as of the end of 1999) visiting the Los Angeles County Department of Health Services Public Health site to find the vaccination rate of Los Angeles County kindergarten students (96% in 1999)

Using Statistics

Once you have found relevant statistics, it is important to consider the following ideas. In order to get complete information on the methods used in compiling statistics, it is often necessary to go to the original source, rather than use information quoted elsewhere.

Reliability of Source	It is important to know what group actually gathered the data you are going to use and why they did so. This is of particular concern when considering international statistics. It is also important to know what methods were used to gather the information you are using. One example of a problem in this area is if the survey was a written one in areas where much of the studied population cannot read or write. When comparing data gathered by multiple groups, it is also important to consider the differences in the study methodologies.
Time	Time is one of the chief variables in considering statistics. Even data gathered in a single year can be misleading. For example, if a nationwide study was done comparing the level of stress-related disorders on a state-by-state basis, it is likely that states surveyed after September 11, 2001 would show a higher incidence than they would have earlier in the year, even though all of the figures are from the same year. Also important to consider when data is presented for a year is whether that is a calendar year or a fiscal year. School years, for example, often go from summer to summer.
Geographic Location	Another consideration is what geographic area the statistics you are using cover. This is of particular importance when comparing statistics that have been gathered in different studies. In the United States, one of the biggest differences is whether information has been gathered on the census tract or zip code level. These areas rarely match up exactly, making it extremely difficult to relate data gathered in one way to data gathered in another.
Population	Try to get as specific a definition of the social group covered as possible. Even such seemingly simple concepts such as “adults in area x” can be deceptive. Does “adults” include individuals over 18? over 21? In many parts of the world this age can go as low as 15, or even 12. Comparing numbers from a study that includes anyone over 15 an adult to a study that considers anyone over 18 an adult can result in serious discrepancies.

Need More Help?

If you are a current UCLA student, faculty, or staff member and have any questions or would like to set up a personal consultation appointment with one of the Biomedical Library reference librarians, please e-mail us at biomed-ref@library.ucla.edu.