2009 NHAMCS MICRO-DATA FILE DOCUMENTATION PAGE 1 ABSTRACT This material provides documentation for users of the public use micro-data files of the 2009 National Hospital Ambulatory Medical Care Survey (NHAMCS). NHAMCS is a national probability sample survey of visits to hospital outpatient and emergency departments, conducted by the National Center for Health Statistics, Centers for Disease Control and Prevention. The survey is a component of the National Health Care Surveys, which measure health care utilization across a variety of health care providers. There are two micro-data files produced from NHAMCS, one for outpatient department records and one for emergency department records. Section I of this documentation, “Description of the National Hospital Ambulatory Medical Care Survey,” includes information on the scope of the survey, the sample, field activities, data collection procedures, medical coding procedures, and population estimates. Section II provides detailed descriptions of the contents of each file’s data record by location. Section III contains marginal data for selected items on each file. The appendixes contain sampling errors, instructions and definitions for completing the Patient Record forms, and lists of codes used in the survey. PAGE 2 2009 NHAMCS MICRO-DATA FILE DOCUMENTATION SUMMARY OF CHANGES FOR 2009 The 2009 NHAMCS Emergency Department and Outpatient Department public use micro-data files are, for the most part, similar to the 2008 files, but there are some important changes. These are described in more detail below and reflect changes to the survey instrument, the Patient Record form. Emergency Departments 1. New or Modified Items a. In item 1, Patient Information, there is a new checkbox item “Arrival by Ambulance.” This replaces the 2008 item, “Mode of Arrival.” b. In item 2, Triage, there are new items for: On Oxygen Glasgow Coma Scale (3-15) c. It should be noted that in item 1a. Date and Time of Visit, the “Seen by “ category was expanded from “Date and Time Seen by Physician” to “Date and Time Seen by MD/DO/PA/NP” (i.e., Doctor of Medicine, Doctor of Osteopathy, Physician Assistant, or Nurse Practitioner). Although actual “Seen by” times are not included on the public use file, these times are used in conjunction with arrival times to produce WAITTIME (waiting time to see the MD/DO/PA/NP) on the public use file. d. Data users should also note that, although Triage level (item 2b) is now asked using a 1-5 level scale on the Patient Record form, answers to this item have been edited in conjunction with responses from the hospital induction interview. Because some hospitals use different scales for triage, it was necessary to take that information into account when interpreting results in item 2b. Also, some emergency service areas (ESAs) do not use a nursing triage system. Missing values for item 2b for ESAs that do use nursing triage systems were imputed using a hot deck as described in Section 1. Results for the Triage item are presented on the public use file with the same categories used in previous years of NHAMCS: immediate, emergent, urgent, semi-urgent, and nonurgent. The variable name is now IMMEDR rather than IMMED to emphasize the changes. e. Pain Scale (item 2c), with values from 0-10, replaces the former categorical variable, Presenting Level of Pain. f. The Record Format item formerly referred to as PAYTYPE has been renamed PAYTYPER in order to emphasize the fact that it is a recoded item based on the collection of multiple expected sources of payment. This is explained in more detail below. IMPORTANT: Both the NAMCS and NHAMCS Patient Record Forms collected primary expected source of payment at ambulatory care visits from 1997 through 2004. This variable, called PAYTYPE, was included on the Public Use Files for those years. When the NAMCS and NHAMCS Patient Record Forms were modified to collect multiple expected sources of payment starting in 2005, all expected sources of payment were included on the Public Use Files. In addition, in order to provide a proxy for primary expected source of payment, a recoded variable, was included. This variable, called PAYTYPE for consistency with earlier years, used a hierarchical scheme to recode the multiple source data into primary source. As noted in the public use file documentation for each year from 2005 through 2007, the top of the hierarchy was Medicaid, followed by Medicare, Private Insurance, Worker’s Compensation, Self Pay, No Charge/Charity, Other and Unknown. This was the hierarchy chosen by branch staff for use in 2005 through 2007. Medicaid was listed first because many researchers are interested in studying care among patients who are poor enough to be on Medicaid; however, data users could always take the original expected pay source data on the public use file and recode it according to their own preferred hierarchy. 2009 NHAMCS MICRO-DATA FILE DOCUMENTATION PAGE 3 In our analyses of emergency department trends from 1997 through 2007 using the PAYTYPE variable, we observed a distinct discontinuity between 2004, the last year that primary expected source of payment was collected, and 2005, when multiple sources of payment were recoded into a hierarchical variable. Our best interpretation of what we have observed results from the fact that the Medicaid-dominant hierarchy is inconsistent with insurance industry practices. Centers for Medicare & Medicaid Services (CMS) treats payment for services for persons who are beneficiaries of both Medicare and Medicaid (dual eligibles) by assigning Medicare as the primary source of payment. We recommend that when analyzing trends across these two time periods, researchers should keep in mind that two different methods were used to collect payment data, and that they may wish to recode PAYTPE in the 2005-2007 NAMCS and NHAMCS public use files to assign Medicare as the primary source of payment for dual eligibles. For 2008, NCHS recoded the PAYTYPE variable to more accurately reflect CMS practices, with Medicare at the top of the PAYTYPE hierarchy. This hierarchy has been continued in 2009. As an additional step, the variable name is being changed from PAYTYPE to PAYTYPER on the upcoming 2009 NAMCS and NHAMCS Public Use Files, to ensure that data users understand that this item is not the same as the primary expected source of payment data collected prior to 2005. In the near future, this name change will also be made to the 2008 NAMCS and NHAMCS Public Use File documentation for consistency with 2009. It is planned that the 2005-2007 NAMCS and NHAMCS Public Use Files will eventually be re-released to include a Medicare-dominant hierarchy for each year, in order to achieve a consistent version of PAYTYPER from 2005-2009. One other caveat regarding expected sources of payment and the hierarchical recode to a primary expected source of payment is this: Although those with both Medicare and private insurance are coded in PAYTYPER as having a primary expected payment source of Medicare, users should be advised that private insurance will be the primary payer for some individuals who are at least 65 years old, work full-time, and have employer-sponsored health insurance. g. In item 6, Provider’s Diagnosis for this Visit, a second section has been added, “Does patient have”, which includes checkboxes for 5 chronic conditions (cerebrovascular disease/history of stroke, congestive heart failure, condition requiring dialysis, HIV, and diabetes. h. In item 7, Diagnostic/Screening Services, there is a new checkbox for HIV test. j. In item 8, Procedures, there are new checkboxes for: Suturing/Staples Pelvic exam Central line j. In item 10, Providers, there are now checkboxes for: Consulting physician Mental health provider k. In item 12, Visit Disposition, there are new checkboxes for: Transfer to psychiatric hospital The previous checkbox for “Admit to observation unit” has been replaced by two more specific categories: Admit to observation unit, then hospitalized Admit to observation unit, then discharged PAGE 4 2009 NHAMCS MICRO-DATA FILE DOCUMENTATION l. In item 13, Hospital Admission, part a, “Admitted to:”, please note that the order of checkboxes #4 (mental health or detox unit) and #5 (cardiac catheterization lab) was switched on the Patient Record Form, relative to the order used in 2007 and 2008. The 2009 order was also used in 2010 and 2011. Prior to 2007, a smaller set of checkboxes was used (critical care unit, OR/cath lab, other bed/unit, unknown). Please keep these changes in mind when combining data across years or trending. m. In item 13, Hospital Admission, there is a new sub-item for Admitting Physician. Time data items have been modified to collect date and time bed was requested for hospital admission and date and time patient actually left the emergency department. n. Item 14, Observation Unit Stay asks about date of observation unit discharge. 2. Deleted Items a. In item 1, Patient Information, the categories for the sub-item on patient residence have been modified. The categories of “Other institution” and “Other residence” have been combined into a single “Other category. b. In item 1, Patient Information, the former “Mode of Arrival” sub-item has been deleted and replaced with “Arrival by Ambulance.” (See above.) c. In item 7, Diagnostic and Screening Services, the MRI sub-categories of “Head” and “Other than Head” have been removed. d. In item 8, Procedures, deleted checkbox categories include “Laceration repair,” “wound debridement,” and “NG tube/gastric suction.” e. In item 10, Providers, the category, “On-call attending physician/Fellow/Resident” has been deleted. f. In item 11, Visit Disposition, the category, “Refer to Social Services” has been deleted. Outpatient Departments 1. New or Modified Items a. In item 7, Diagnostic /Screening Services, there are new checkboxes for: Foot examination Retinal examination HIV test b.