Categorical Regression Analysis (CatReg)
CatReg complements EPA’s existing Benchmark Dose Software (BMDS) by greatly enhancing a risk assessor’s ability to determine whether data from separate toxicological or epidemiological studies can be pooled into a single dose-response-time meta-analysis.
CatReg can be used to support the analysis of health effect studies and other types of toxicity data used to support toxicity assessments developed by EPA and other interested stakeholders.
Introduced by EPA in 2006, CatReg was available only as an application written in and run using the R statistics package.
CatReg 3.0 still requires R be installed and relies on the same statistical algorithms. However, CatReg 3.0 offers a Windows-based interface that makes specifying variables, filters, stratification, clustering, hypotheses, and applying the algorithms an easier and more efficient experience.
Subscribe to the BMDS newsletter to stay informed on CatReg availability, software updates, and training opportunities.
- Purpose of CatReg
- Special CatReg Features
- Advantages of CatReg Compared to Other Dose-Response Tools
Purpose of CatReg
CatReg was developed to support risk assessors, toxicologists, and health scientists in conducting exposure-response analyses, most often for controlled animal experiments. CatReg reads data from ordinary comma-separated data files and outputs both plots and text-based results.
CatReg can be used to perform categorical regression analyses on toxicity data after effects have been assigned to ordinal severity categories (e.g., no effect, adverse effect, severe effect) and associated with up to two independent variables corresponding to the exposure conditions (e.g., concentration and duration) under which the effects occurred. CatReg calculates the probabilities of the different severity categories over the continuum of the variables describing exposure conditions. The categorization of observed responses allows expression of dichotomous, continuous, and descriptive data in terms of effect severity and supports the analysis of data from single studies or a combination of similar studies.
There are many potential applications of the CatReg program in the analysis of health effects studies and other types of data. It was originally developed to support toxicity assessments for acute inhalation exposures, but modifications have been made to the software (e.g., the ability to estimate background responses and derive upper and lower bound confidence limits on BMD estimates) that facilitate its broader use for the evaluation of chronic studies and for its use in the derivation of traditional EPA Integrated Risk Information System (IRIS) reference values.
Special CatReg Features
- Stratifying the analysis by user-specified covariates (e.g., species, sex, etc.)
- Choosing among several basic forms of the exposure-response curve
- Using effects assigned to a range of severity categories, rather than a single category
- Using cluster-correlated data
- Using aggregate data
- Filtering data to analyze subsets of data and perform sensitivity analyses
- User-friendly hypothesis testing interface that can be used to determine whether data associated with variables such as gender, species, genetics, and other factors that can influence a dose-response-time relationship should be pooled together or analyzed separately for estimating the model parameters (e.g., intercept, dose, and time model coefficients)
For screenshots and more information on CatReg, download and view the CatReg User Guide.
Advantages of CatReg Compared to Other Dose-Response Tools
CatReg’s advantages over other dose-response analysis tools include its abilities to:
- Use multiple independent variables to explain the response
- Predict exposures related to various levels of effect severity
- Combine multiple studies in a single analysis
- Test hypotheses
Use Multiple Independent Variables
Most often, the independent variables modeled in in CatReg are exposure (or dose) and duration of exposure. The latter allows for estimating risk at different exposure durations—even time points not included in the dataset being modeled.
Evaluate Severity Levels and Severity Categories
Using categorical regression to evaluate different levels of severity is of particular use for developing emergency planning guidelines, which are defined in terms of the exposure limits needed to protect people from increasingly severe effects (e.g., protection from mild effects,. severe effects, and death).
Combine Multiple Studies in a Single Analysis
The meta-analytical use of categorical regression can be invaluable when individual studies provide only minimal useful dose-response data. Classifying toxic effects into severity categories provides a way to put different endpoints on a common scale for analysis and also allows the use of all types of data in the analysis. For example, individual animal studies investigating liver lesions and serum biochemistry can be combined with human studies reporting preclinical and clinical liver effects to provide an overall estimation of the dose-response for liver toxicity.
Finally, the CatReg 3.0 implementation of categorical regression offers valuable and practical hypothesis testing tools, combined with a user-friendly interface, to assist the user in determining whether data associated with variables that can sometimes influence a dose-response (e.g., sex, species, diet, age, genetic differences) should be evaluated separately or combined together into a more robust, data-rich multiple-study meta-analysis.