Prescribed drugs pills

These MEPS summary tables provide statistics on total expenditures, total purchases, and number of persons with purchases for prescription medicines or therapeutic class groups.

Use the options below to select a statistic of interest, data years, and grouping variable (therapeutic class or generic drug name). After you select the available options, the table will automatically be updated. The table can be sorted by drug name or therapeutic class name, or data values for each year by clicking on the column header.

To view data as a plot, select up to 10 rows by clicking in the table. The "Code" tab displays R and SAS code needed to replicate the data shown in the table. The generated table, plot, and codes can be downloaded with the download button under each tab. To view standard errors in the table or 95% confidence intervals in the plot, select the "Show standard errors" checkbox.

Select years
Updating, please wait...

To activate the 'plot' tab, select up to 10 rows by clicking in the table below.

Loading data...

Please choose items to plot by selecting rows under the Table tab.

No non-missing data available for selections. Please try again.
Download code

To run the code, first download and unzip the required public use data files from the MEPS data files page, and save them to your computer. More information on downloading and analyzing MEPS data in R, SAS, and Stata can be found at the AHRQ GitHub site External Link. Note that some standard error estimates may differ between R and SAS, since SAS doesn't support any options to adjust for lonely PSUs.

The following code can be used to generate the selected estimates, where the SAS transport data files (.ssp) have been saved to the folder 'C:\MEPS'. For trend estimates, example code is shown for the most recent year selected:



This tool is provided as a convenience. It is the responsibility of the user to review results for statistical significance and overall reasonableness.

About the data

The MEPS Household Component collects data on all members of sample households who are in-scope for the survey. These data can be used to produce nationally representative estimates of medical conditions, health status, use of medical care services, charges and payments, access to care, experience with care, health insurance coverage, income, and employment. The target population represented in the tables and figures is persons in the U.S. civilian non-institutionalized population for all or part of the year.

Suggested citation