Evaluating risk factors in disease outbreaks.

The investigation of disease outbreaks in herds and flocks includes the gathering of multiple types of discrete data on the characteristics of the disease and the potential risk factors involving agent, host, and environment. Descriptive statistics such as rates, ratios, and proportions offer the investigator tools for counting the disease occurrence, for communicating the results of the investigation to others, and for further analyzing the patterns of disease in order to identify the true risk factors. Because there are several alternative methods for calculating descriptive statistics, specific definitions of each statistic should be followed so that ambiguity and misinformation are avoided. The practitioner must choose one definition and apply it consistently in disease investigations. Consistency allows for comparison of descriptive statistics from one investigation to another, or from one time to another time within the same herd. Analytic statistics allow for a critical evaluation of the data represented in rates, ratios, and proportions. Statistical tests address the all-important question, "Could these observed associations between disease and potential risk factors have occurred by chance alone?" This question must be addressed so that the wisest intervention decisions can be made. The tools described in this article are useful in routine herd health practice situations as well as the investigation of disease outbreaks and impaired productivity. Descriptive statistics provide a system for counting disease so that the practitioner can evaluate progress in the individual herd. The ability of the practitioner to quantify changes in disease occurrence in the herd provides hard evidence for the producer of the impact of herd health approaches.