FAQ Data Descriptions Demographics 

Data Set I contains information on recipient demographics, geography, eligibility and program enrollment, chronic conditions, service use, and the costs associated with the services. (Please see our data dictionary for descriptions of columns included in the recipient data.)

Demographic data contains information on age, gender, race, and ethnicity. The data sets do not include dates of birth, but do include death dates for recipients who have died in the experience period. Ages are presented in ‘age bands’ (where age is in integers, rounded down) condensed from the ages of recipients as of the anchor date or last eligibility date: aged 0 (less than one year), 1 to 18 years, 19 to 20 years old, 21 to 44 years old, 45 to 64 years old, and 65+ years old.

Geographic data is included two ways: as zip codes, and as counties. The data set includes the five-digit zip code on file as of the anchor date of the experience period (December 31, 2010) or a recipient’s last eligibility date (if their eligibility ended prior to December 31, 2010). HFS does not clean zip codes, which means null and incorrectly reported zip codes may be found in the data set. County information is recorded at beginning of the recipients most recent eligibility span in Medicaid and reflects the county of the public aid office where the individual’s eligibility was processed. When an office serves several (rural) counties, we have attempted to increase data specificity by assigning counties based on recipient zip codes. Recipients do not necessarily go to the office in their county or change offices when they move. County attribution should therefore be considered imperfect.

The anchor date refers to the last day of the experience period, December 31, 2010. Over the course of a year recipients can enroll and disenroll from Medicaid, change program status, qualify for waiver programs, change their address, be born, die, and undergo other changes. This transience can make data analysis difficult. We have summarized most enrollment data by anchor or the nearest available date, such as the last date enrolled prior to the anchor date or the first day of the eligibility span closest to the anchor date. Our data dictionary defines the applicable date for each data field.

Some zip codes contain a small population of people; combining the zip code with all the other information on a small recipient pool can theoretically make certain recipient records identifiable. This is in violation of Health Insurance Portability and Accountability Act (HIPAA) regulations for a limited data set. (For further description, please see our Data Use Agreement, Section C of Attachment D of the Solicitation (pdf).) The use of banded ages otherwise limits the potential for recipient identification.

The data set’s experience period is one year: January 1, 2010, to December 31, 2010. Additional years of data (covering the same information) may be available in subsequent data releases; your feedback based on the shortcomings of this data set may also be used to improve upon the information currently offered. However, we do not anticipate that additional years of data can be released until after the deadline for the current Solicitation. Therefore we expect prospective partners to submit proposals based upon the current data release. Partners are free to submit proposals containing supplementary data from other sources as they see fit and/or their caveats on data limitations.