Survey Methods and Analytic Guidelines

Please note that complete documentation including survey methodology for all survey years can be downloaded for NAMCS and NHAMCS. Additional information on using NAMCS and NHAMCS data can also be found under Research Tools.

Survey Methods and Related Documentation

Survey Methods

Nonresponse Bias Analysis

Classification Systems Used with NAMCS and NHAMCS

  • International Classification of Diseases
  • National Drug Code Directory (used in NAMCS and NHAMCS prior to 2006)
  • Multum (used in NAMCS and NHAMCS starting in 2006)
  • Reason for Visit Classification, Drug Entry List, Generic Ingredient List: These classifications are updated on an annual basis and printed in the documentation for each survey year. In addition, text formats for reason for visit, drug entry, and generic ingredient codes are all available in downloadable SAS formats for data years 1992-2015, and SPSS syntax files and Stata do and dictionary files for 1992-2015. The SAS and SPSS formats also include text labels for the following codes: diagnosis, cause of injury, procedures, and therapeutic class of drug.

Survey Items over Time


Analytic Guidelines

Variance Estimation: Using complex sample design software such as SUDAAN, Stata, SPSS, R, SAS survey procedures, etc.

Currently, NAMCS public-use files for 1993-2015 and NHAMCS public-use files for 1992-2015 contain sample design variables in masked form. The initial release of the 2000-2015 public-use files includes masked sample design variables. We re-released the 1993-99 NAMCS and 1992-99 NHAMCS files to include these variables. For re-released files, file names and layouts are essentially the same, but the design variables were appended to each file according to a generic file layout that is available for downloading.

However, data users should note that, beginning with the 2002 public use files, two new masked design variables were added to the file, for use with statistical software that assumes a single stage of sampling. For the 2003 public use files and beyond, the decision was made to include only these two masked design variables for variance estimation. Therefore, data users who wish to combine years of data from 2003 and beyond with years prior to 2002 will need to create these two variables for each file prior to 2002. A technical paper, Using Ultimate Cluster Models with NAMCS and NHAMCS Public Use Files [PDF – 34 KB], gives instructions.

Additional data analysis resources can be found at Research Tools.