Multiple Cause Mortality Analysis

"MultiCod" is the Vitalnet module for analyzing and disseminating multiple cause death data. MultiCod is a unique system, allowing better use of the wealth of information available on the death certificate.

Makes fuller use of the data - MultiCod analyzes all causes listed on the death certificate. This is especially useful for analyzing diabetes, heart disease, cancer, lung disease and other chronic diseases in an aging population. Most elderly people have multiple health problems (comorbidity). These chronic conditions are often listed as contributing causes on the death certificate, and are missed by underlying cause analysis.

Provides a new way of looking at the data - Multiple cause data give a different picture of the health of the community. Depending on the context, one or the other may be more useful, or you may use both.

Historically, multiple cause data have not been used much. We would suggest this is largely because of the great technical difficulty involved. Carrying out multiple cause analyses by hand is tedious and error-prone.

MultiCod makes analyzing multiple cause data simple, fast, and reliable. Publication-ready results are produced in a minute or two. Other systems for these kinds of analyses are few and far between. The only one we found is more limited in capability, and confusing to use.

A cost-effective, simple multiple cause method - You may ask why so few systems have the multiple cause capability. It's way too technically difficult, because: 1) Most systems do not analyze record level data; 2) The few that do analyze record level data are unable to handle the very complex logic to prevent double-counting the variable length record data.

We built a unique database architecture for multiple cause analysis. We invested the large effort needed to solve the many technical difficulties. You can spend millions trying to "reinvent the wheel", with little chance of success. Licensing Vitalnet is hassle-free, much less expensive, and gets you a working product in a few months.

Provides special multiple cause outcomes - MultiCod calculates death rates, age-adjusted rates, SMR, leading causes, YPLL and other statistics based on any mention on the death certificate. These results can be compared with those derived from underlying cause analysis.

MultiCod also produces the following statistics unique to multiple cause analysis: Mean causes per death (Higher means more comorbidity). Percent deaths with one cause (Higher means less comorbidity).

Multiple-cause analysis principles - MultiCod uses the analytical principles laid out in "Analytical Potential for Multiple Cause-of-Death Data", by Robert A. Israel, Harry M. Rosenberg, and Lester R. Curtin, published in American Journal of Epidemiology, 124: 161-179, 1986. In addition, MultiCod can be modified to provide new analytical capabilities requested by you.

Vitalnet multiple-cause analyses also apply to hospital discharge data (not shown).

Examples of ICD-9 multiple-cause analysis:

Leading causes - underlying cause analysis

Leading causes - multiple cause analysis

High "mean causes per death" (high comorbidity)

High "% deaths with one cause" (low comorbidity)

"% deaths with one cause" by age

"Causes listed" tabulation

Examples of ICD-10 multiple-cause analysis:

Leading causes - underlying cause analysis

Leading causes - multiple cause analysis

High "mean causes per death" (high comorbidity)

High "% deaths with one cause" (low comorbidity)

"% deaths with one cause" by age

"Causes listed" tabulation

There are striking differences between underlying and multiple cause analyses. Underlying cause greatly underestimates the impact from chronic conditions. For example, compare ICD-9 diabetes underlying cause deaths (4,888) with multiple cause deaths (13,355).

The example tables (shown above) just scratch the surface of the capabilities of multiple cause analysis. Each table can easily be customized in many different ways to explore the data in more detail.

Fill out contact form

Try VitalWeb with multiple cause data

---