Review of Existing Data Sources to Assess the Impacts of the Affordable Care Act Medicaid Expansions: Working Paper. APPENDIX

04/01/2012

This appendix provides detailed information on each of the data sources included in the data scan. Appendix Table A-1a and Table A-1b summarize and compare the population surveys, Appendix Table A-2a and Table A-2b provide information on the provider surveys, and Appendix Table A-3 describes the administrative data sources.

Population Surveys

The American Community Survey (ACS), conducted annually by the U.S. Census Bureau, is a general household survey of the entire population (including persons living in group quarters) that replaced the decennial census long form. The ACS includes questions about demographic and socioeconomic characteristics, and a question on current health insurance coverage was added in 2008. This mandatory survey (persons are required to respond under law) samples from the National Master Address File and is conducted monthly by mail, by telephone, and in person. The ACS has a response rate of 98% and collects data from about 4.4 million people in 1.75 million households (the public use files include data from about 3 million and 1.2 million individuals and households, respectively). The Census Bureau releases summary reports and public use data files with state identifiers in the early fall each year, about 8-9 months after the end of the survey calendar year. Analysis using the full ACS file requires Census Bureau approval and must be performed in a Research Data Center.

The Current Population Survey (CPS) is a monthly survey of the civilian non-institutionalized population conducted by the U.S. Census Bureau. The primary purpose of the monthly survey is to collect data on labor force participation and unemployment. Data on income and health insurance are collected through the CPS ASEC in February through April of each year. The CPS ASEC asks about health insurance coverage for the prior calendar year and can be combined with information from the main CPS survey on other demographic and socioeconomic characteristics as well as with other information associated with health insurance coverage, such as firm size. The CPS, conducted in person and by telephone, uses an address-based census sample frame and surveyed about 205,000 individuals in 75,000 households in 2011. Nationally, the CPS achieved a response rate of 84% in 2011. Summary reports and public use data files with state identifiers, usually released in early fall, are available about 5-6 months after data are collected; with regard to timeliness of information, however, it is important to remember that the health insurance estimates refer to the previous calendar year.

The National Health Interview Survey (NHIS) is an in-person survey of the health of the civilian non-institutionalized population and is sponsored by the Centers for Disease Control and Prevention's (CDC) National Center for Health Statistics (NCHS). The NHIS, which uses an address-based census sample frame, has been conducted annually for more than 50 years. It asks about health insurance coverage, health care utilization and access, health conditions and behaviors, and general health status, as well as many demographic and socioeconomic characteristics. Within each household, separate household and family questionnaires are asked; in addition, one "sample adult" and one "sample child" are randomly selected from each family to complete the more in-depth sample adult and sample child questionnaires. With a national response rate of over 79% for the family questionnaire and 61% for the sample adult questionnaire in 2010, the NHIS public use file for 2010 includes information about nearly 90,000 individuals from the family questionnaire and more than 27,000 completed sample adult interviews. Summary reports, with state estimates for the 30 largest states, are released 6 months after data collection, as are public use data files (without state identifiers). Use of data files with state-level and other geographic identifiers requires NCHS approval, and the files must be accessed through a Research Data Center.

The Medical Expenditure Panel Survey--Household Component (MEPS-HC), sponsored by the Agency for Healthcare Research and Quality (AHRQ), is an in-person panel survey that includes several interviews over two full calendar years. Conducted since 1996, the MEPS-HC collects data on health status and conditions, health insurance coverage, access to and utilization of health care services, medical expenditures, and various demographic and socioeconomic characteristics. The MEPS-HC samples from a subsample of NHIS participants from the previous year and in 2009 had an overall response rate of about 57% for the full-year data file. For 2009, the MEPS-HC full-year data file includes nearly 37,000 individuals. Summary reports, with state estimates for the ten largest states, are released 6 months after data collection, along with public use data files (without state identifiers). Although the survey is not designed to produce state or local estimates, data files with state-level and other geographic identifiers can be accessed through an AHRQ or Census Bureau Research Data Center for approved projects.

The Survey of Income and Program Participation (SIPP) is a panel survey of the civilian non-institutionalized population that has been conducted by the U.S. Census Bureau since 1984. Data on income and program participation, as well as the determinants of income and program participation, are collected in several waves over 4 years (however, some panels have been shorter because of budget constraints). The SIPP, which is an in-person and telephone survey, uses an address-based census sample frame to draw samples of households that are followed throughout the multiyear survey period. About 88,000 individuals responded to the sixth wave of the 2008 panel (conducted from May through August 2010). In Wave 1 of the 2008 panel, the national response rate was about 81%; about 16% of original respondents had been lost through Wave 6. Data are released about 9-13 months after collection. Although the SIPP is not designed to be representative within states, the public use data files do include state identifiers.

The Behavioral Risk Factor Surveillance System (BRFSS) is a state-based survey of the adult civilian non-institutionalized population that has been conducted annually since 1984. Sponsored by the CDC along with the 50 states and United States territories, the BRFSS inquires about health conditions, risk behaviors, preventive health practices, access to health care, and health insurance coverage. The BRFSS is a stratified random digit dial telephone survey of landline telephones (although the survey is experimenting with cell phones, these data are not yet included in public use files). In 2010, the BRFSS surveyed more than 450,000 adults across the 50 states and in United States territories. The response rate varies by state, with a median of about 36% in 2010 (ranging from 19% to 57%). The public use files, which include state identifiers, are released about 5 months after the end of the survey year.

The National Health and Nutrition Examination Survey (NHANES) is conducted by the CDC's NCHS using an address-based sample, with year-round data collection. NHANES has been conducted annually since 1999. This survey of the civilian non-institutionalized population collects information about population health, and it includes a medical examination component in addition to an in-person interview. NHANES oversamples several population groups, including Hispanic, Black, and low-income people, as well as those aged 60 and older. NHANES surveys about 5,000 people each year, with a response rate of about 77% in 2009-2010. Public use files are released biennially; the most recent file with 2009-2010 data was released in September 2011.

Finally, the National Survey on Drug Use and Health (NSDUH), conducted annually since 1990, is sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA). NSDUH, a survey of the civilian non-institutionalized population age 12 and older, focuses on issues related to drug and alcohol use and abuse or dependence as well as mental health. The survey is conducted in person, with a large self-administered component. The survey oversamples young people (ages 12-17 and 18-25), with more than 68,000 responses and a response rate of 66% in 2010. Public use data files for 2010 were released in December 2011. Although the public use data files do not include state identifiers, the survey is designed to support direct estimates for the eight largest states; SAMHSA uses small area estimation techniques to publish estimates for all states.

Provider Surveys

Hospitals

The National Hospital Ambulatory Medical Care Survey (NHAMCS), conducted by the CDC's NCHS annually since 1992, is a survey of ambulatory care services provided in hospital outpatient departments, hospital emergency rooms, and ambulatory surgery centers (freestanding ambulatory surgery centers were included beginning in 2010). Hospitals are selected to participate in the survey using a multistage sampling design based on geography. Data about participating facilities are collected through in-person interviews, and data on patient visits are collected through patient record abstraction. In 2009, the hospital response rate was 90%, and data were collected on about 35,000 emergency department visits and 33,500 outpatient department visits at 365 hospitals. Public use data files for 2009 were released in July 2011. The survey is not designed to produce state-level estimates and the public use data do not include state identifiers, but it is possible to use state data in a Research Data Center for an approved project.

The National Hospital Discharge Survey (NHDS), also conducted by the NCHS, focuses on inpatient hospitalizations. Its sample design is similar to the NHAMCS; data are collected either through hospital submission of Uniform Hospital Discharge Data Set files (also called the UB-04 administrative database) or through manual data abstraction. The NHDS has been conducted annually since 1965. In 2009, the survey achieved a hospital response rate of 79% and collected data on over 162,000 inpatient discharges from 205 hospitals. The 2009 public use data, released in April 2011, include geographic identifiers for census region but not state. The NHAMCS and NHDS are being integrated into a new National Hospital Care Survey, which will collect complete UB-04 data from participating facilities (rather than a sample, as in the past), as well as personal identifiers to allow linking of multiple records for an individual.

The American Hospital Association (AHA) Annual Survey has collected data from member and non-member hospitals on hospitals' capacity, services, utilization, personnel, and finances annually since 1946. The survey is sent to all United States hospitals each year, including all ownership types (e.g., federal, state, non-profit, for-profit) and all primary service types (e.g., general, specialty). Data are collected through online and mail responses. The 2010 survey response rate was approximately 76%, and the database for 2010 includes a total of 6,334 hospitals. Summary reports and data files are available for purchase. Currently, the data are released once each year (typically in October); however, in the future AHA plans to make new data available on a rolling basis to database subscribers.

The National Association of Public Hospitals and Health Systems (NAPH) conducts an annual survey of its members to collect data on capacity, utilization, patient characteristics, and financing at its member hospitals. The data are collected through an online survey; in 2009 the response rate was 95% (92 of 97 hospitals). The most recent data available are for 2009, released in December 2010. There is no public use file, but data for each participating hospital are published in an annual report.

Other Health Care Providers

The National Ambulatory Medical Care Survey (NAMCS) is an annual survey sponsored by CDC's NCHS. The survey collects data on ambulatory care services provided by office-based physicians; for participating CHCs, the survey collects data from both physician and non-physician practitioners. NAMCS has been conducted annually since 1989, and CHCs have been included since 2006. The sample for NAMCS is constructed using a multistage design that is based on geography and stratified by physician specialty. Data are collected through in-person interviews of physicians (62% response rate in 2009) and patient record abstraction. The 2009 public use data were released in May 2011, and the data file includes information on 32,281 patient office visits to 1,293 physicians. The survey is not designed to produce state-level estimates and the public use file does not include state identifiers, but it is possible to use the state information for an approved project in a Research Data Center; similarly, data on non-physician visits to CHCs are available only through a Research Data Center. Beginning in 2010, NAMCS has also conducted a supplemental mail survey of physicians, also stratified by geography and physician specialty. The supplemental survey has a larger sample size of physicians but a more limited scope; its focus is primarily on issues related to electronic medical record systems, but it also includes information on payer mix and whether a physician is accepting new patients, by payer.

The American Medical Association (AMA) Physician Masterfile is a continuously updated database with over 875,000 records on medical doctors, doctors of osteopathy, and medical residents and students. The database includes information on physician demographics and practice characteristics. Data are collected by mail, online, by telephone, and from secondary data sources. Annual reports and data files are available for purchase from the AMA and include state and substategeographic identifiers. (In addition, the ARF published by HRSA includes counts of physicians from the AMA Physician Masterfile aggregated to the county level.)

The National Survey of Substance Abuse Treatment Services (N-SSATS) is an annual survey of all known providers of substance abuse treatment in the United States, conducted by SAMHSA. The survey, conducted via mail, online, and by telephone, collects information about the characteristics of substance abuse treatment providers. In 2010, more than 13,000 providers participated in the survey, for a response rate of about 87%. The survey collects information about the number of clients being served at a specific point in time (the last weekday in March each year). The public data file for 2010, released in December 2011, includes national, regional, state, and substate geographic identifiers.

Administrative Data

The Healthcare Cost and Utilization Project (HCUP) is a database of inpatient discharges and emergency department and ambulatory surgery encounters assembled by AHRQ from state-level databases. The state databases are submitted to AHRQ voluntarily by states, hospital associations, and private data organizations. Currently, 44 states submit inpatient data, and 29 states submit emergency department and ambulatory surgery center data. AHRQ uses these data to construct a National Inpatient Sample (NIS) and National Emergency Department Sample (NEDS), representing a 20% sample of total visits. The most recent year available for the NIS and NEDS is 2009, released in May 2011 and October 2011, respectively. The national samples are not designed to be state representative, but complete state files are available separately (through either a central distributor or directly from the organization that submits data to AHRQ). The state databases are made available more quickly than the national inpatient and emergency department samples; the 2010 state databases were released in September 2011.

Hospitals that participate in Medicare are required to file annual Medicare cost reports; the primary purpose of these reports is for reconciliation and settlement of Medicare payments, but the reports also include information about facility characteristics, utilization, costs, and charges. Reports are due 150 days after the end of a hospital's fiscal year. The data are publicly available through a frequently updated public database. The database includes information at the hospital level.

The Uniform Data System (UDS) is a reporting system for Federally Qualified Health Centers (FQHCs) that receive funding from HRSA. The UDS report includes information on services available, utilization, staffing, patient demographics, finances, health outcomes, and quality measures. Beginning with calendar year 2011 (for reports that will be submitted in early 2012), FQHC lookalikes (centers that meet all of the eligibility requirements but do not receive federal funding) will also submit UDS reports. Reports are submitted in mid-February of each year for the previous calendar year. There are no public use data, but summary reports that aggregate the data to the state and national levels are publicly available.

For substance abuse treatment, the Treatment Episode Data Set (TEDS), sponsored by SAMHSA, collects patient-level information on admissions to substance abuse treatment facilities. The data are collected by states from facilities that receive state and federal funds to provide alcohol and drug treatment services. All states report a core data set, with a supplementary data set of Affordable Care Act's impacts, such as health insurance and income information, is included in the supplementary data set). Public data sets are made available annually and include state identifiers; the most recent available data file, for 2009, was released in June 2011.

TABLE A-1a. Population Surveys: ACS, CPS ASEC, NHIS, and MEPS
  ACS 2010 CPS ASEC 2011 NHIS 2010 MEPS 2009
Sponsor U.S. Census Bureau U.S. Census Bureau National Center for Health Statistics, Centers for Disease Control and Prevention Agency for Healthcare Research and Quality
Target population Entire U.S. population Civilian non-institutionalized population Civilian non-institutionalized population Civilian non-institutionalized population
Primary focus General household survey, replaced decennial census long form Labor force participation and unemployment Population health Health care access, utilization, and cost
Sample frame and design Address-based multistage sample stratified by geography Address-based multistage sample stratified by geography Address-based multistage sample stratified by geography. Oversamples Black, Hispanic, and Asian populations. Sample drawn from NHIS respondents with additional oversampling of low-income households. A new 2year panel is selected each year.
Data collection mode Mail, telephone, and in-person interviews In-person and telephone interviews In-person interviews In-person interviews with follow-up data collection from medical providers
Sample size (number of individuals) 4,368,578 204,983 89,976 total; 27,157 completed in-depth "sample adult" questionnaire 36,855
Response rate 97.5% 83.8% Family: 78.7%; sample adult: 60.8% 57.2%
Frequency and survey period Annually since 2000; conducted year-round. Health insurance questions were first asked in 2008. Annually since 1948; currently conducted February through April. Health insurance questions were added in the 1980s. Annually since 1957; conducted year-round. Annually since 1996; conducted year-round.
Data availability and timeliness Public use file includes a subset of responses--3,003,411 individuals in 2010. Data for 2010 were released in October 2011. Public use file from 2011 ASEC, with health insurance estimates for calendar year 2010, released in September 2011 Public use file from 2010 released in June 2011. Quarterly data available through Research Data Centers (January through June 2011 file was made available in December 2011). Full-year consolidated file for 2009 released in November 2011
Levels of geography in public use data National, state, and substate National, state, and some substate National and Census region National and Census region
Supports state estimates? Yes Yes Limited, and available for analysis only through Research Data Centers* Limited, and available for analysis only through Research Data Centers
Costs of acquiring/using data Moderate, because of large data file size Low Low for public use data; higher for analyses requiring Research Data Centers Relatively high due to longitudinal design and survey complexity; higher for analyses requiring Research Data Centers
SOURCES:
American Community Survey (ACS): U.S. Census Bureau, American Community Survey Accuracy of the Data (2010);State Health Access Data Assistance Center (SHADAC) tabulations from public use file.
Current Population Survey Annual Social and Economic Supplement (CPS ASEC): U.S. Census Bureau, Source and Accuracy of Estimates for Income, Poverty, and Health Insurance Coverage in the United States: 2010; SHADAC tabulations from public use file.
National Health Interview Survey (NHIS): National Center for Health Statistics, Data File Documentation, National Health Interview Survey, 2010.June 2011; Chris Moriarty, "The National Health Interview Survey: Changes Due to the Affordable Care Act, and Plans for a Future Online Analytic System," University of Minnesota seminar, October 19, 2011.
Medical Expenditure Panel Survey (MEPS): Agency for Healthcare Research and Quality, MEPS HC-129 2009 Full Year Consolidated Data File, November 2011.
*Beginning in 2011, increased sample size for the NHIS will enable reliable state estimates in about 25 states (previously 18-20 states). Future sample size expansions, dependent on funding availability, could enable estimates for 37-42 states.

 

TABLE A-1b. Population Surveys: SIPP, BRFSS, NHANES, and NSDUH
  SIPP 2008 Panel BRFSS 2010 NHANES 2009-2010 NSDUH 2010
Sponsor U.S. Census Bureau Centers for Disease Control and Prevention; states and territories National Center for Health Statistics, Centers for Disease Control and Prevention Substance Abuse and Mental Health Services Administration
Target population Civilian non-institutionalized population Civilian non-institutionalized population age 18 and older Civilian non-institutionalized population Civilian non-institutionalized population age 12 and older
Primary focus Longitudinal data on income and program participation Population health, risk factors, and health behaviors Population health and nutrition Drug and alcohol use and abuse/dependence; mental health
Sample frame and design Address-based multistage sample stratified by geography. Oversamples low-income populations. 2008 panel planned for 16 waves through 2013. Stratified sample of landline telephone numbers (specifics vary by state) Address-based multistage sample stratified by geography. Oversamples Hispanic, age 60+, Black, and low-income populations. Address-based multistage sample stratified by geography. Oversamples age 12-17 and age 18-25.
Data collection mode In-person and telephone interviews Telephone interviews In-person interviews and physical examinations In-person interview with self-administered component
Sample size (number of individuals) 88,150 451,075 About 5,000 each year; 2009-2010 data file includes 10,537 observations 68,487
Response rate 80.6% of eligible housing units in Wave 1 of 2008 panel; 16.2% of original respondents lost through Wave 6 Varies by state (19.3%-57.4%, median 35.8%) 77.4% of screened sample 66.3% (overall weighted response rate)
Frequency and survey period Multiyear panels since 1984; conducted year-round Annually since 1984; most states collect data year-round Annually since 1999; conducted year-round Annually since 1990; conducted year-round
Data availability and timeliness Some questions of interest are in topical modules that are not included in each wave. Core data released about 9 months after data collection; topical modules released about 13 months after data collection. Public use file for 2010 released in May 2011 Public use data released biennially; some data accessible only in NCHS Research Data Centers. 2009–2010 data released in September 2011. Public use files released annually; some data restricted. 2010 data released in December 2011.
Levels of geography in public use data National and state National and state National only National only
Supports state estimates? Not designed to produce state estimates Yes No Direct estimates for eight largest states; small area estimation techniques used to publish estimates for all states
Costs of acquiring/using data Relatively high because of longitudinal design and survey complexity Low Low for public use file; higher for restricted data Low
SOURCES:
Survey of Income and Program Participation (SIPP): U.S. Census Bureau, Source and Accuracy Statement for the Survey of Income and Program Participation 2008 Wave 1 to Wave 6 Public Use Files, July 2011; SIPP Users' Guide Sample Design and Interview Procedures, 2009. Sample size from State Health Access Data Assistance Center (SHADAC) tabulations of the Wave 6 public use file, number of interviewees in the first month of Wave 6.
Behavioral Risk Factor Surveillance System (BRFSS): Centers for Disease Control and Prevention (CDC), 2010 BRFSS Documentation Overview; CDC, Behavioral Risk Factor Surveillance System 2010 Summary Data Quality Report, version #1, revised May 2, 2011.
National Health and Nutrition Examination Survey (NHANES): CDC, NHANES 2009-2010 Public Data General Release File Documentation, http://www.cdc.gov/nchs/nhanes/nhanes2009-2010/generaldoc_f.htm, accessed December 14, 2011.
National Survey on Drug Use and Health (NSDUH): Substance Abuse and Mental Health Services Administration (SAMHSA), Results from the 2010 National Survey on Drug Use and Health: Summary of National Findings. NSDUH Series H-41, HHS Publication No. (SMA) 11-4658. Rockville, MD: SAMHSA, 2011; RTI International, 2010 National Survey on Drug Use and Health Public Use File Codebook, October 27, 2011.

 

TABLE A-2a. Provider Surveys--Hospitals
  Acute Care Hospitals Safety Net Hospitals
NHAMCS 2009* NHDS 2009* AHA Annual Survey 2010 NAPH 2009
Sponsor National Center for Health Statistics, Centers for Disease Control and Prevention National Center for Health Statistics, Centers for Disease Control and Prevention American Hospital Association National Association of Public Hospitals and Health Systems
Primary focus Ambulatory care services provided in hospital emergency and outpatient departments and ambulatory surgery centers Inpatient hospital utilization Hospital characteristics, utilization, staffing, and finances Characteristics of public safety net hospitals
Sample frame and design Multistage sample design to select geography units, hospitals within selected geographies, units within hospitals, and patient visits within selected units. Excludes federal hospitals, hospital units of institutions, and hospitals with fewer than six staffed beds. Multistage sample design to select geography units, hospitals within geography units, and inpatient discharges within hospitals Survey of all U.S. hospitals NAPH members
Data collection mode In-person interviews of facilities and patient record abstraction Acquisition of Uniform Hospital Discharge Data Set files or manual data abstraction Online and mail Online
Sample size 34,942 emergency department visits and 33,551 outpatient department visits at 365 hospitals 162,151 discharges from 205 hospitals 6,334 hospitals 92 hospitals
Response rate 89.7% (hospital) 79% (hospital) 76% 94.8%
Frequency and survey period Annually since 1992; data collected year-round Annually since 1965; data collected year-round Annually since 1946; survey is conducted during the first half of the calendar year (data collected for a hospital's most recently completed fiscal year) Annually
Data availability and timeliness 2009 public use data released in July 2011 2009 public use data released in April 2011 2010 survey data released in October 2011. Beginning in 2012, subscribers will be able to access new data on a rolling basis. 2009 survey results published December 2010; no public use data
Level of geography in public use data National only National and census region National, state, local, and individual hospitals No public use data, but report includes national totals and individual hospitals
Supports state estimates? No, but possible to use state data in Research Data Centers No Yes No
Cost of acquiring/using data Low Low Moderate. Approximately $8,000 per year to purchase data; requires licensing agreement. Not available
SOURCES:
National Hospital Ambulatory Care Medical Survey (NHAMCS): National Center for Health Statistics (NCHS), 2009 NHAMCS Micro-data File Documentation.
National Hospital Discharge Survey (NHDS): NCHS, National Hospital Discharge Survey 2009 Public Use Data File Documentation, March 2011; NCHS, "NAMCS and NHAMCS Restricted Data Available at the NCHS Research Data Center," http://www.cdc.gov/nchs/data/ahcd/Availability_of_NAMCS_and_NHAMCS_Restricted_Data.pdf.
American Hospital Association (AHA): AHA Survey Database Fiscal Year 2010 Public File Layout and Code Descriptions; personal communication with Sara Beazley, senior information specialist at AHA, January 2012.
National Association of Public Hospitals and Health Systems (NAPH): NAPH, America's Public Hospitals and Health Systems, 2009, December 2010.
*NHAMCS and NHDS are being integrated into a new National Hospital Care Survey.

 

TABLE A-2b. Provider Surveys--Physicians, Community Health Centers, and Substance Abuse Treatment Facilities
  Physicians Community Health Centers Substance Abuse
Treatment Facilities
NAMCS 2009 AMA Physician Masterfile NAMCS 2009 N-SSATS 2010
Sponsor National Center for Health Statistics, Centers for Disease Control and Prevention American Medical Association National Center for Health Statistics, Centers for Disease Control and Prevention Substance Abuse and Mental Health Services Administration
Primary focus Ambulatory care services provided by office-based physicians Physician demographics and practice characteristics Ambulatory care services provided at CHCs by office-based physicians and non-physician practitioners Facilities and providers treating substance abuse disorders
Sample frame and design Multistage sample design to select geography units, physicians within selected geographies stratified by specialty, and patient visits within physician practices. Includes non-federal office-based physicians who are primarily engaged in direct patient care. AMA database of medical doctors, doctors of osteopathy, residents, and medical students (includes AMA members and non-members) List of physicians and non-physician practitioners at 104 selected CHCs; selection of up to three providers per CHC All known providers of substance abuse treatment in the U.S. are included in the survey
Data collection mode In-person interviews of physicians and patient record abstraction. Supplemental mail survey on EMR/EHR systems. Mail, online, telephone, and secondary data sources In-person interviews of physicians and patient record abstraction. Supplemental mail survey on EMR/EHR systems. Mail, online, and telephone
Sample size 32,281 patient office visits to 1,293 physicians* Approximately 875,000 records in database 3,590 visits to physicians 13,339 facilities
Response rate 62.4% (physicians) N/A 74.7% (physicians) 86.7%
Frequency and survey period Annually since 1989; data collected year-round Established in 1906; continually updated CHCs included in NAMCS since 2006; data collected year-round Annually since 1995 (except 2001); client counts are at a point in time (last weekday in March). Survey conducted between March and October.
Data availability and timeliness 2009 public use data released in May 2011 Continually updated 2009 public use data released in May 2011. Includes physicians at CHCs but not non-physician practitioners--full CHC file available for use only at Research Data Centers. 2010 public use data released in December 2011
Level of geography in public use data National and census region National, census region, state, and substate National and census region National, regional, state, and substate
Supports state estimates? No, but possible to use state data in Research Data Centers. Supplemental mail survey supports state estimates.* Yes No, but possible to use state data in Research Data Centers. Supplemental mail survey supports state estimates. Yes
Cost of acquiring/using data Low Available for purchase through several licensed vendors; cost varies Low Low
SOURCES:
NAMCS--National Center for Health Statistics, 2009 NAMCS Micro-data File Documentation. NCHS, "NAMCS and NHAMCS Restricted Data Available at the NCHS Research Data Center," http://www.cdc.gov/nchs/data/ahcd/Availability_of_NAMCS_and_NHAMCS_Restricted_Data.pdf; personal communication with Sandra Decker, December 2011; CDC web page for NAMCS participants, http://www.cdc.gov/nchs/ahcd/namcs_participant.htm. American Medical Association (AMA): Medical Marketing Service, Inc. data file layout. National Survey of Substance Abuse Treatment Services (N-SSATS): Inter-University Con+B14sortium for Political and Social Research, "National Survey of Substance Abuse Treatment Services (N-SSATS), 2010"; Substance Abuse and Mental Health Data Archive, "National Survey of Substance Abuse Treatment Services (N-SSATS), 2010: Codebook."

*The National Ambulatory Medical Care Survey (NAMCS) sample size has been substantially increased in recent years, to about 4,700 office-based providers in 2011; further sample size increases are planned for 2012. The NAMCS supplemental mail survey is primarily about EMR/EHR systems, but also includes questions about payer mix and whether a provider is accepting new patients. Beginning with 2010 data, the mail supplement supports state estimates, and it has a 64% response rate. The supplemental mail survey is stratified by geography and physician specialty. EMR/EHR = electronic medical record/electronic health record; N/A = not applicable.

 

TABLE A-3. Administrative Data Sources
  Acute Care Hospitals Community Health Centers Substance Abuse
Treatment Facilities
HCUP 2009, 2010 Medicare Cost Reports UDS TEDS
Sponsor Agency for Healthcare Research and Quality Centers for Medicare and Medicaid Services Bureau of Primary Health Care, Health Resources and Services Administration Substance Abuse and Mental Health Services Administration
Primary focus Inpatient discharges; emergency department and ambulatory surgery encounters Information needed for reconciliation and settlement of Medicare payments; also includes facility characteristics, utilization, costs, and charges CHC patient demographics, services provided, staffing, clinical indicators, utilization, and financial data Admissions to substance abuse treatment facilities
Data collection State databases are submitted voluntarily by states, hospital associations, and private data organizations. NIS and NEDS are created from the state databases, representing a 20% sample of community hospitals and emergency departments. Medicare-certified providers submit annual cost reports to a fiscal intermediary; due 150 days after the end of the reporting year Reports submitted by grantees under section 330 of the Public Health Service Act, including CHCs, migrant health centers, Health Care for the Homeless grantees, public housing primary care grantees. FQHC look-alikes added for CY2011 reports. Data collected by states from facilities receiving state and federal funds to provide alcohol and drug treatment services. Supplementary data set including health insurance and income data is not reported by all states.
Frequency Annually. NIS since 1988; NEDS since 2006. Historical data vary by state. Annually; data available beginning with 1996 hospital fiscal years Annually, due by February 15 each year. States encouraged to report monthly. Public data sets are annual, beginning in 1992.
Data availability and timeliness 2009 NIS released in May 2011. 2009 NEDS released in October 2011. 2010 state databases released in September 2011. Data files updated frequently No public use data, but national and state summary reports are available 2009 data file released in June 2011
Level of geography in public data files National, census region, state; hospital for some states. Number of states participating varies by data type: inpatient (44), emergency department (29), and ambulatory surgery center (29). Individual facilities N/A National, census region, state, and some substate
Supports state estimates? National samples not designed to be state representative; complete state files available separately Yes Yes Yes, for reporting facilities; however, scope is incomplete
Cost of acquiring/using data High. Data files are very large, and purchasing state databases can be very costly. Data use agreement and training required. High, because of size and complexity of data files Not publicly available Moderate, due to large file size
SOURCES:
Healthcare Cost and Utilization Project (HCUP): Agency for Healthcare Research and Quality (AHRQ), "Introduction to the HCUP Nationwide Inpatient Sample (NIS) 2009," May 2011; AHRQ, "Introduction to the HCUP Nationwide Emergency Department Sample (NEDS) 2009," September 2011; AHRQ, "Fact Sheet: Databases and Related Tools from the Healthcare Cost and Utilization Project (HCUP)," revised March 2011.
Medicare Cost Reports: Centers for Medicare and Medicaid Services (CMS), Provider Reimbursement Manual; CMS, "Cost Reports: General Information," http://www.cms.gov/CostReports/Downloads/CRGeneralInfo.pdf, accessed December 17, 2011.
Uniform Data Set (UDS): Bureau of Primary Health Care, "BPHC Uniform Data System Manual," November 23, 2011.
Treatment Episode Data Set (TEDS): Substance Abuse and Mental Health Services Administration (SAMHSA), "TEDS--Treatment Episode Data Set," http://wwwdasis.samhsa.gov/webt/information.htm, accessed December 17, 2011; SAMHSA, Office of Applied Studies, "Treatment Episode Data Set--Admissions (TEDSA)--Concatenated, 1992 to 2009," ICPSR Study No. 25221; SAMHSA, Office of Applied Studies, "Treatment Episode Data Set--Admissions (TEDS-A)--Concatenated, 1992 to 2009, Part 4: 2005 to 2009 Codebook"; SAMHSA, "Treatment Episode Data Set (TEDS) State Instruction Manual," February 2010.
CHC = community health center; FQHC = federally qualified health center; NEDS = National Emergency Department Sample; NIS = National Inpatient Sample.

View full report

Preview
Download

"ACAImpact.pdf" (pdf, 1.37Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®