DOCUMENT.TXT This file contains documentation for all of the files accompanying “Monitoring the Health Care Safety Net. Book I: A Data Book for Metropolitan Areas.” 1. INTRODUCTION The health care safety net—the Nation's system of providing health care to low-income and other vulnerable populations—was recently described as "intact but endangered." In response, the Agency for Healthcare Research and Quality and the Health Resources and Services Administration are leading a joint initiative to monitor it. The goal of this initiative is to help state and local policymakers, planners, and analysts monitor the status of their safety nets and the populations they serve. Strategies include providing baseline data and a set of tools that enable monitoring of the capacity and performance of local safety nets. This documentation and data accompanies two books. The first book, “Monitoring the Health Care Safety Net. Book I: A Data Book for Metropolitan Areas,” presents data on metropolitan areas from 30 states and the District of Columbia, including 354 counties and 171 cities in 90 metropolitan areas and 94 county residuals. The second book, “Monitoring the Health Care Safety Net. Book II: A Data Book for States and Counties,” shows data from all 1,818 counties in these states, including both metropolitan and non-metropolitan counties. Other than the Microsoft Excel spread sheets, the electronic data files combine the data displayed in these two data books. The following files are included. These files are also available at www.ahrq.gov/data/safetynet. 2. DOCUMENTATION FILES 2.a. DOCUMENT.TXT is the current file, which documents the contents of this disk and describes the data involved. The same information is also on the files DOCUMENT.DOC (formatted for Microsoft Word) and DOCUMENT.WPD (formatted for WordPerfect). 2.b. CODEBK.TXT (129 Kb) indicates the record layout (i.e., the start and end position of each variable) of the file. This codebook also includes listings of the variables in both alphabetical and positional order, as well as the following information for each variable: 8 character name, 40 character description, format (length), type (character or numeric), start and end positions, and unweighted and weighted frequencies. This information is also on the files CODEBK.DOC (formatted for Microsoft Word) and CODEBK.WPD (formatted for WordPerfect). 2.c. MEASURES.PDF (118 Kb) and MEASURES.HTML (96 Kb) have identical contents, and provide more detailed information on each measure, its definition, the data source used to derive it and any necessary methodologic information. In order to access PDF files such as MEASURES.PDF, Adobe® Acrobat® Reader™ software must be used. If you currently have this software installed on your computer, you can use it to view or print PDF files after you have downloaded them. Your browser may already be configured to use the Adobe® Acrobat® Reader™ as a "helper application," so that you can automatically view a PDF file when you select it. If you do not have the Adobe® software installed, your browser may give you the option of downloading and saving PDF files for later viewing and printing. In this case, you do not need to configure your browser, but you will still need to have the Adobe® software installed in order to access the downloaded PDF files. It is also highly recommended that PDF files be printed on a PostScript® printer, if one is available. Although PDF files may print legibly on other printers, a PostScript® printer will ensure a correct rendering of the document. If you do not have the Adobe® software installed, you can install it by downloading a free copy. Please note that the system requirements for this software are: 4 MB of RAM, 5 MB of hard disk space, plus 7 MB of additional temporary disk space available during installation. For instructions, go to the Adobe® site at http://www.adobe.com/prodindex/acrobat/readstep.html. Accessibility: Online conversion tools are available from Adobe® to help visually disabled users convert PDF documents into either HTML or ASCII text, which can then be read by a number of common screen reader programs. Go to: http://access.adobe.com/onlinetools.html. The folder named "images" contains information needed to display the file MEASURES.HTML. 3. DATA FILES 3.a. Microsoft Excel spreadsheets 3.a.1. MSA Tables1.XLS (4,149 Kb) contains spreadsheets formatted for use with Microsoft Excel 97-2002 & 5.0/95. These spreadsheets contain all of the data in “Monitoring the Health Care Safety Net. Book I: A Data Book for Metropolitan Areas” as they appear in print. 3.a.2. State Tables1.XLS (10,126 Kb) contains spreadsheets formatted for use with Microsoft Excel 97-2002 & 5.0/95. These spreadsheets contain all of the data in “Monitoring the Health Care Safety Net. Book II: A Data Book for States and Counties” as they appear in print. 3.b. Data files and programming statements for use with statistical software 3.b.1. SAFENET1.DAT (2,653 Kb) is a combined file containing all of the data from both books in an ASCII format to be read into any software package. It contains 2,183 records with 136 variables. This file has a logical record length of 1242 with an additional 2-byte carriage return/line feed at the end of each record. 3.b.2 SAFENET1.SD7 (2,721 Kb) is a combined file containing all of the data from both books, formatted for use with SAS version 8. See the file SNSAS1.TXT for information applicable to SAS users of the data. 3.b.3. SAFENET1.SSP (2,621 Kb) is a combined file containing all of the data from both books in SAS Transport format. This file can be used with earlier versions of the SAS software. See the file SNSAS1.TXT for information applicable to SAS users of the data. 3.b.4. SNSAS1.TXT (41 Kb) is an ASCII file that contains information for SAS users of the SAFENET1.DAT file. This includes information on the SAS Transport version of the data, as well as SAS statements for converting the ASCII data file into a SAS data set (in lieu of using the SAS Transport file) and for creating and using SAS formats. This file also includes helpful information for inexperienced SAS users, such as sample SAS program code. 3.b.5 SAFENET1.SAV (2,688 Kb) is a combined file containing all of the data from both books, formatted for use with SPSS for Windows. 3.b.6 SAFENET1.DTA (2,380 Kb) is a combined file containing all of the data from both books, formatted for use with Stata version 7. 4. VARIABLES Detailed information on the variables included in the data files can be found in MEASURES.PDF and MEASURES.HTML. These two files have identical contents, and contain information on the measures, their definitions, and their data sources. This information, along with additional details about data sources and geographic areas, is printed in the technical appendix at the back of “Monitoring the Health Care Safety Net. Book I: A Data Book for Metropolitan Areas.” This technical appendix is also available electronically at www.ahrq.gov/data/safetynet. The ASCII (SAFENET1.DAT), SAS (SAFENET1.SD7 and SAFENET1.SSP), Stata (SAFENET1.DTA) and SPSS (SAFENET1.SAV) versions of the data file merge information from both the MSA and state Excel tables into a single file. These files include a number of variables to help analysts link data across geographic areas; these variables are documented in this section. There are five types of observations in these data files, corresponding to the five types of geographic areas included in the books: states, Metropolitan Statistical Areas/Primary Metropolitan Statistical Areas (MSAs/PMSAs), counties, cities, and county residuals. MSAs/PMSAs are generally referred to throughout the products as MSAs, although the variable containing the MSA names clearly indicates whether it is an MSA or a PMSA. A county residual is the part of a county outside of the city. For example, the county residual of Maricopa County, Arizona is the part of Maricopa County that lies outside the city of Phoenix. For additional information on geographic areas, please see “Appendix A: Technical Information,” in “Monitoring the Health Care Safety Net. Book I: A Data Book for Metropolitan Areas.” This is available in the print version of the book or electronically at www.ahrq.gov/data/safetynet. 4.a. Variables indicating place names 4.a.1. STATE This is the state name associated with each record. For MSAs located in more than one state, the state listed first in the MSA name is listed in STATE. STATE is valid for all records. 4.a.2. MSA This is the name of the MSA/PMSA in which the MSAs, counties, cities and county residuals are located. MSA is set to -222 for states and non-MSA counties. 4.a.3. COUNTY This is the name of the county in which the counties, cities and county residuals are located. COUNTY is set to -222 for states and MSAs. 40 Virginia independent cities (18 in metropolitan areas, 22 in non-metropolitan areas), Baltimore city and St. Louis city are treated as counties. 4.a.4. CITY This is the city name and is valid only for city records. CITY is -222 for states, MSAs, counties and county residuals. New York City and Washington, DC are treated as city records. They do not contain county information. 4.a.5. RESIDUAL This is the county residual name, which is valid only for county residual records. An example is ‘Residual – Maricopa County,’ which is the part of Maricopa County, Arizona that is outside the city of Phoenix. RESIDUAL is set to -222 for state, MSA, county and city records. 4.b. Variables describing the type of observation 4.b.1. OBSTYPE This variable is a flag for observation type and is set to 1 for states, 2 for MSA/PMSAs, 3 for counties, 4 for cities and 5 for county residuals. 4.b.2. METRO This is a flag variable. METRO is set to 1 for all MSA and MSA-component records. These include all MSA records, all counties in MSAs, all cities, and all county residuals. There are no non-metropolitan cities or county residuals included in the data set. METRO is set to -222 for state records. METRO is set to 0 for all counties that are not in MSAs/PMSAs. 4.c. FIPS codes Federal Information Processing Standards (FIPS) codes uniquely identify states, MSAs, counties, and cities. For analysts working with another data set that contains FIPS codes, these variables will enable the appropriate merging of these data onto other data sets. FIPS codes are developed by the National Institute of Standards and Technology (NIST), and additional information is available at http://www.itl.nist.gov/fipspubs/. 4.c.1. UNIQFIPS This is a unique FIPS ID, which consists of 15 fields. The fields are state FIPS (2 bytes) + MSA FIPS (4 bytes) + county FIPS (3 bytes) + city FIPS (5 bytes) + X or Y (1 byte, X if county residual, Y if others). 4.c.1.a. State records have UNIQFIPS=State FIPS (2 bytes) + YYYYYYYYYYYYY (13 bytes). 4.c.1.b. MSA records have UNIQFIPS= State FIPS (2 bytes) + MSA FIPS (4 bytes) + YYYYYYYYY (9 bytes). 4.c.1.c. MSA county records have UNIQFIPS= State FIPS (2 bytes) + MSA FIPS (4 bytes) + county FIPS (3 bytes) + YYYYYY (6 bytes). Five counties under New York City have special values for UNIQFIPS. These counties are Bronx County (005), Kings County (047), New York County (061), Queens County (081) and Richmond County (085). For example, Bronx County has UNIQFIPS=36560000551000Y. 4.c.1.d. Non-MSA county records have UNIQFIPS=State FIPS (2 bytes) + YYYY (4 bytes) + county FIPS (3 bytes) + YYYYYY (6 bytes). 4.c.1.e. City records have UNIQFIPS=State FIPS (2 bytes) + MSA FIPS (4 bytes) + county FIPS (3 bytes) + city FIPS (5 bytes) + Y. New York City and Washington, DC have special UNIQFIPS because they do not have a county FIPS. UNIQFIPS is 365600YYY5100Y for New York City and 118840YYY50000Y for Washington, DC 4.c.1.f. County residual records have UNIQFIPS=State FIPS (2 bytes) + MSA FIPS (4 bytes) + county FIPS (3 bytes) + YYYYY (5 bytes) + X. 4.c.1.g. San Francisco County’s FIPS code is used for the ‘San Francisco City and County’ record. The same method is used for ‘Denver City and County,’ as well as ‘Philadelphia City and County.’ Dade County’s FIPS is used for Miami-Dade County. 4.c.2. STFIPS This is the state FIPS code associated with each record. For MSAs located in more than one state, the state listed first in the MSA name is listed in STFIPS. STFIPS is valid for all records. 4.c.3. MSAFIPS This is the MSA FIPS code for all MSAs/PMSAs and MSA counties, a well as all cities and county residuals. MSAFIPS is -222 for states and non-MSA counties. 4.c.4. CONTFIPS This is the county FIPS code for counties, cities, and county residuals. CONTFIPS is -222 for states and MSAs. 4.c.5. CITYFIPS This is city FIPS code, which has valid values for cities only. CITYFIPS is -222 for states, MSAs, counties and county residuals. 4.c.6. RESFIPS This is a constructed code, which has valid values for county residual records only. It is set to county FIPS + X. RESFIPS is -222 for all other records. 4.d. Variables indicating position/sequence within this data set 4.d.1 STATEID This is a unique sequence ID based on an alphabetical sort of state names. The range is 1-31, with all records having a value. 4.d.2. MSAID This is a unique sequence ID for MSAs based on the order in MSA data book. The range is 1-90 for MSAs, MSA counties, cities and county residuals. MSAID is -222 for states and non-MSA counties. 4.d.3. COUNTYID This is a unique sequence ID based on an alphabetical sort of abbreviated (2 letter) state names and county names. The range is 1-1,818 for counties, cities and county residuals. COUNTYID is -222 for state and MSA records. 4.d.4. CITYID This is a unique sequence ID for cities based on the order in MSA data book. The range is 1-151 for cities. CITYID is -222 for states, MSAs, counties and county residuals. 18 Virginia independent cities, Baltimore, St. Louis, Philadelphia, Denver, and San Francisco are all treated as counties rather than cities. 4.d.5. RESID This is a unique sequence ID for county residuals based on the order in MSA data book. The range is 1-94 for county residuals. RESID is -222 for states, MSAs, cites and counties. 5. Missing values The following missing values codes are used: -111 N/A indicates that data should be available for this specific area, but are missing. -222 LEVEL NOT AVAILABLE indicates that data are not available at this area level. For example, estimates of the percent of the population that is uninsured can be made only at the state and MSA levels with our data, so counties, cities, and county residuals are coded -222. -333 NO DATA indicates that data were not collected or processed for this specific area. -444 NO HOSPITAL indicates that there is no hospital in the county. -555 ONE HOSPITAL indicates that there is only one hospital in the county, so there is no meaningful analysis of this measure. -666 SUPPRESSED indicates that data have been suppressed to protect the identity of a single facility or to protect personal privacy (cell size less than 5 individuals). -777 LOW POPULATION indicates that data are not presented because the small population size of the area prevents statistically meaningful analysis. 6. Accuracy of data and contact information Every effort has been made to ensure that the existing data used for these books has been processed accurately. However, errors may still exist, either due to an oversight in the data processing or problems with the original data. Questions or feedback concerning the data may be directed to SAFENET@AHRQ.GOV.