"Wonder" Helps Speed Pattern Recognition

January 17, 2007

When assessing an environmental issue, it may be useful and appropriate to evaluate its links with human deaths and illnesses. Although this is difficult, and requires extensive professional analysis, you can begin to spot trends that raise questions worth asking by evaluating patterns of deaths and illness. Or you can use the patterns to confirm or challenge scientific research.

The data for doing this are increasingly available online to reporters and the general public. One of the latest developments is a recent upgrade of the US Centers for Disease Control and Prevention's WONDER data for underlying causes of death. WONDER stands for "Wide-ranging Online Data for Epidemiologic Research."

CDC has added a fifth year to the latest inventory (1999-2003), which helps smooth out the inevitable gyrations in number of deaths in any one location. That's particularly important for cities or counties with lower populations, since just a few deaths can dramatically change the death rate.

In addition, it now takes just minutes to map out the deaths in many ways. Mapping typically makes patterns vastly easier to decipher than using tables or other less-visual methods.

Data for hundreds of individual causes of death can be analyzed by age, race, and gender alone or in combinations for a broad spectrum of geographic settings, including counties, cities of numerous sizes, states, regions, and the country. To help make data comparable, you can "age-adjust" the results, which improves comparisons between, say, a college town and a retirement community.

In addition to maps, you can also quickly create tables, charts, and graphs, and export data.

Searching for mortality information begins here.

To start, you'll want to explore the site, so you begin to understand all the terms, as well as codes for states and causes of deaths. When you're ready, conduct a search. You'll get a table of results. Near the top of that page is a bar you can click to create a map or chart. With either choice, you'll be able to set up the final product in a number of ways. This also will take some experimentation to figure out what works best for your needs.

Keep in mind when using this data that weaknesses remain in the system. For instance, a doctor's selection of a specific code for a death can vary around the country. Also, for some causes of death, the total number of deaths is so small, particularly in a low-population city or county, that the data are considered unreliable, and labeled accordingly. In addition, it appears that deaths along the US-Mexico border may not be fully recorded in US records, since these areas often have very low rates, despite other indicators that suggest these populations aren't as healthy as the low rates indicate.

The full WONDER Web site is here. It contains many other resources for health data and information.