- Owner / manager
- Study and sample characteristics
- Major foci
- Special supplements and resources
- Links to other datasets
- Papers published
- Dataset accessibility and cost
- Help desk
Key web links
Home Page
www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm
NHAMCS Survey Instruments
https://www.cdc.gov/nchs/ahcd/ahcd_survey_instruments.htm
List of survey items 1973 – 2006:
https://www.cdc.gov/nchs/ahcd/datasets_documentation_related.htm
Ambulatory Care Drug Database System
http://www.cdc.gov/nchs/ahcd/ahcd_database.htm
Dataset Summary
NAMCS and NHAMCS comprise nationally representative surveys of outpatient and emergency department visits in the United States from 1973 through the present. Each year, physicians in community-based office practices, hospital-associated office practices, and emergency rooms are asked to complete forms about outpatient visits by individual patients. Altogether, both surveys collect data on approximately 80,000 patient visits per year. The surveys are clustered and weighted in such a way that results from the research databases can be easily extrapolated to produce nationally representative estimates. Data is available for downloading off the web at no cost. All records are completely de-identified for protected health information. Participating physicians (and their patients) vary from year to year, so there is no longitudinal followup, although the serial cross-sectional nature of the survey allows for tracking of trends over time.
Expert comments
NAMCS and NHAMCS are wonderful resources. The data are easy to access and use, and the website provides highly useful documentation about how to program Stata, SAS, and SPSS to adjust for survey clustering, stratification, and weighting. These datasets are ideally suited to evaluate practices that occur at individual office visits; for example, the prevalence of reasons for office visits, characteristics of patients receiving care (and clinicians providing care) in different outpatient settings, interventions offered at individual visits (such as medication prescribing, diagnostic test-ordering), and so forth. Both surveys have also been used to evaluate chronic diseases and their treatments, for example, evaluating chronic NSAID or statin use in older patients. However, both surveys are less ideally suited to evaluating the prevalence and treatment of chronic diseases, since the main focus of the survey is on the content of the individual visit and chronic diseases or medications may be more likely to not be recorded on the survey. It is also important to note that the majority of surveys are filled out by office staff and NHCS representatives based on clinic notes, rather than the physicians themselves. NAMCES was validated against direct observation (Gilchrist Med Care 2004) and found to be most accurate for procedures and examinations; behavioral counseling was underreported and visit duration overestimated compared with direct observation.
Dataset Details
Dataset owner / manager
National Center for Health Statistics
Study and sample characteristics
Nationally representative sample of visits to non-federally employed office-based physicians who are primarily engaged in direct patient care (NAMCS) and to emergency departments and outpatient departments of general and short-stay hospitals, exclusive of Federal, military, and VA hospitals (NHAMCS). Each year, physicians from each of these settings are recruited to complete data forms for a representative sample of patient visits. Sampling is conducted using a multi-stage stratified probability approach, and visit weights and clustering variables are available to convert survey data to nationally representative estimates.
Data are available annually from 1973 to 1981, in 1985, and annually since 1989.
https://www.cdc.gov/nchs/ahcd/datasets_documentation_related.htm
See also the “Scope and sample design” and “Downloadable documentation” links from the home page: http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm
Major foci
Data collection forms vary slightly from year to year. However, the overall format of the surveys have remained consistent, and include the following types of variables. Individual years may have additional information not included below:
- Clinician and institutional characteristics including clinician specialty, national region, type of office practice, practice / hospital ownership, and more
- Patient characteristics including basic demographic information and insurance status
- Reason for the visit and diagnoses related to the visit
- Diagnostic and screening services and non-medication based treatments ordered or provided
- Medications (including immunizations and other office-administered drugs) prescribed at the visit or continued from previous use
- Visit disposition
Survey instruments: https://www.cdc.gov/nchs/ahcd/ahcd_survey_instruments.htm
A list of survey items from 1973-2012 are available as a series of links from(look mid-way down the page):
https://www.cdc.gov/nchs/ahcd/ahcd_survey_instruments.htm
Additional, restricted data is available at special request from the Research Data Center, including:
- Physician / hospital demographic and practice characteristics
- Geographic variables (including state, county, and zip codes)
Census variables (e.g. percent of population in patient’s zip code with high school diploma, percent who are non-English speaking, zip-linked median household income, etc.)
For more information, see: “Additional NAMCS and NHAMCS Data Available at the Research Data Center” on the NAMCS / NHAMCS home page: https://www.cdc.gov/nchs/ahcd/rdc_data.htm
Special supplements and resources
The Ambulatory Care Drug Database System provides information about medications listed in NAMCS and NHAMCS.
http://www.cdc.gov/nchs/ahcd/ahcd_database.htm
In recent years, both NAMCS and NHAMCS have included a cervical cancer screening supplement. See downloadable files at:
https://www.cdc.gov/nchs/ahcd/datasets_documentation_related.htm
Papers published
Click here for a PubMed search for articles using this dataset.
A complete list of NAMCS and NHAMCS publications can be found at: http://www.cdc.gov/nchs/ahcd/ahcd_products.htm
Examples of papers published using NAMCS and NHAMCS include:
Trends in prostate-specific antigen testing from 1995 through 2004.
Farwell WR, Linder JA, Jha AK.
Arch Intern Med. 2007 Dec 10;167(22):2497-502.
Medication use leading to emergency department visits for adverse drug events in older adults.
Budnitz DS, Shehab N, Kegler SR, Richards CL.
Ann Intern Med. 2007 Dec 4;147(11):755-65.
Preventive health examinations and preventive gynecological examinations in the United States.
Mehrotra A, Zaslavsky AM, Ayanian JZ.
Arch Intern Med. 2007 Sep 24;167(17):1876-83.
The treatment of smoking by US physicians during ambulatory visits: 1994 2003.
Thorndike AN, Regan S, Rigotti NA.
Am J Public Health. 2007 Oct;97(10):1878-83.
Electronic health record use and the quality of ambulatory care in the United States.
Linder JA, Ma J, Bates DW, Middleton B, Stafford RS.
Arch Intern Med. 2007 Jul 9;167(13):1400-5.
Dataset accessibility and cost
Data is free and available for downloading off the NAMCS / NHAMCS website. All data are de-identified.
http://www.cdc.gov/nchs/ahcd/ahcd_questionnaires.htm
Additional data is available via an application process with the Research Data Center. Fees are charged for this service and vary by the type of data request. For more information, see links to “Additional NAMCS and NHAMCS Data Available at the Research Data Center” on the NAMCS / NHAMCS home page. http://www.cdc.gov/nchs/ahcd/ahcd_questionnaires.htm
Help Desk
The general venue for help requests is cdcinfo@cdc.gov, or call Susan Schappert at (301)-458-4480
Downloadable documentation is excellent and is available via web links from the home page.
SAS, Stata, and SPSS input statements, variable labels, and other data resources are available via web links from the home page.
For information on conducting trend analyses, see:
http://www.cdc.gov/nchs/ahcd/trend_analysis.htm
Request a consultation (SGIM members only)
Members of SGIM may request a one-time consultation with an expert in this dataset, for example, to explore research ideas or to troubleshoot a problem or vexing question. Please click here for guidelines and the request process.