Update #6: 11/22/13
STATA Programming Statements have been added.
Update #5: 12/12/07
The variable FOODCT96 on the HC-012 data file should not be used because it is based on an invalid question in the survey instrument. This question asks "About how much did the family pay each month for food stamps?" when in fact, as of 1979, the food stamp program no longer maintained a purchase requirement. All responses are considered invalid.
Update #4: 08/03/01
On page C-76 of the documentation for the 1996 Full Year Consolidated Data File (HC-012), the sample sizes for MEPS families had been reported as 8,605 for cases with (WTFAMF96>0 & FMRS1231=1) and 8,652 for cases with WTFAMF96>0. These sample sizes have been updated to 8,586 and 8,655 respectively.
Update #3: 07/18/01
The version of the zipped SAS Transport HC-012 file, H12SSP.ZIP, which was added to this site on 6/1/01, was replaced on 7/2/01 in order to correct an error in that file.
Update #2: 06/06/01
Data for BIRTH96 and RSNINHOS were collected in different sections of the MEPS instrument. As a result, an inconsistency was noted between the number of births (BIRTH96) on HC-012 and the number of deliveries (RSNINHOS=4) and the number of births (RSNINHOS=5) on the Hospital Stays File (HC-010D). It is recommended that analysts use RSNINHOS to identify the number of births and/or deliveries that occurred in 1996.
Update #1: 04/12/01
AHRQ has released an update of the prescribed medicine variables previously released on the October 2000 version of the MEPS HC-012: 1996 Full Year Consolidated File. It was recently discovered that there is an inaccuracy in the number of times a household reported purchasing or otherwise obtaining a prescription drug in a particular round for a small percentage of household-reported medications. The variables that have been updated are TOTEXP96 -- TOTOSR96 and RXTOT96 -- RXOSR96. This inaccuracy is due to an instrument design flaw which caused interviewer error in isolated cases resulting in misreported large numbers of prescription refills for a medicine in a given round. In addition, prescribed medicine events in which a household respondent did not know/remember the number of times a certain prescribed medicine was purchased or otherwise obtained in a particular round were inadvertently excluded. After making additional adjustments for consistency with the 1997 MEPS prescription data editing, the cumulative effect of these corrections resulted in an 8.3 percent decline in 1996 MEPS prescribed medicine expenditures (from $71.2 to $65.3 billion) and an 11.9 percent decline in 1996 prescription drug utilization (from 2.116 to 1.865 billion prescriptions) when comparing the adjusted to the original release. Moreover, these adjustments result in a very modest reduction in the overall MEPS 1996 expenditure estimate of just 1.1 percent (from $554 billion to $548 billion).
Back
Email Updates
