HCUP e-News: the electronic newsletter of the Healthcare Cost and Utilization Project



September 2008
Issue #16


Contents

News and Announcements: Up-to-the minute information, including newly released databases and products
Recent Publications: A sample of materials, including peer-reviewed journal articles, utilizing HCUP data
Users´ Tech Tip: Answers to commonly asked questions when working with HCUP data
HCUP Calendar: Recent and upcoming conferences or meetings at which HCUP was/will be represented


NEWS AND ANNOUNCEMENTS

Updated HCUP Facts and Figures Report Released

AHRQ just released a new report, HCUP Facts and Figures: Statistics on Hospital-Based Care in the United States, 2006. This report presents data from the Nationwide Inpatient Sample (NIS) database on hospital care in 2006, as well as trends in care from 1993 to 2006. It is an update of the Facts and Figures 2005 report and also contains new special topics.

The HCUP Facts and Figures report showcases the wealth of statistics available from the HCUP NIS database. It features an overview of numerous hospital-related topics, including general characteristics of U.S. hospitals and the patients being treated; the most common diagnoses, conditions, and procedures associated with inpatient stays; the costs and charges associated with hospitalizations; and a special section on selected priority health conditions designated by the U.S. Department of Health and Human Services.

Highlights of hospital care in 2006 include the finding that six of the 20 most costly conditions associated with hospitalizations were related to the circulatory system. Though these stays accounted for 18 percent of all hospitalization costs—with three diagnoses (coronary artery disease, heart attack, and congestive heart failure) among the most costly in 2006—the growth in costs for these conditions has slowed dramatically since 2003. Examples of trend information presented in the report include findings that, between 1993 and 2006, the total number of Cesarean sections grew nearly 69 percent, while vaginal births remained steady. Maternal complications including hypertension, diabetes, and anemia increased for pregnant women, regardless of whether they were undergoing C-sections or vaginal deliveries, but were more common in women who had C-sections. Infant complications also increased during this period, including respiratory problems, "infant of a diabetic mother" syndrome, jaundice, and feeding problems, and were also more common in babies delivered via C-section than vaginally.

HCUP Facts and Figures: Statistics on Hospital-Based Care in the United States, 2006 is located on the Reports page of the HCUP-US Website. The 2005 version of the report is available on the same page. For more information, please visit the HCUP-US Website, or contact HCUP User Support at hcup@ahrq.gov.

New HCUP Statistical Briefs Are Now Available via the HCUP-US Website

The HCUP Statistical Briefs are short, focused reports on topics related to specific conditions, procedures, or populations. The Statistical Briefs are useful to a wide variety of audiences, including policy analysts, decision makers, media personnel, and others in need of summary facts and figures on current health care issues. Recently released Statistical Briefs include: Statistical Brief #55: Hospital Stays Resulting from Excessive Heat and Cold Exposure Due to Weather Conditions in U.S. Community Hospitals, 2005; Statistical Brief #56: Hospital Stays for Children, 2006; Statistical Brief #57: Meningitis-Related Hospitalizations in the United States, 2006; Statistical Brief #58: Hospital Stays Related to Asthma for Children, 2006; and Statistical Brief #59: The National Hospital Bill: The Most Expensive Conditions by Payer, 2006.

HCUP Statistical Briefs can be found in the Reports section of the HCUP-US Website.

2007 HCUP State Databases Released!

AHRQ is pleased to report that, since July, several HCUP State databases for the 2007 data year have been released. In the past, State data were typically available 12 to 18 months following the end of a calendar year. Thanks to new process improvements and strong relationships with HCUP State Partners, AHRQ has been able to begin to release 2007 databases in half that time.

This release includes the State Inpatient Databases (SID), State Ambulatory Surgery Databases (SASD), and State Emergency Department Databases (SEDD) for selected States. Researchers and policy makers can use these state-specific HCUP databases to investigate questions unique to one State, compare data from two or more States, conduct market area research or small area variation analyses, and identify State-specific trends in utilization, access, quality, charges, and outcomes. More State databases will be released throughout the year.

The following 2007 HCUP State databases are currently available:


As more 2007 State databases are released, the States will be listed on the HCUP-US Calendar of Database and Product Releases, and are also listed in the SID/SASD/SEDD Application Kit. These and other HCUP databases are available through the HCUP Central Distributor. More information can be found on the HCUP Central Distributor page of the HCUP-US Website.

Most 2006 HCUP State Databases Available

Several new 2006 HCUP statewide databases have been released and are available for purchase. Currently available 2006 databases include the following:


As more 2006 State databases are released, the States will be listed on the HCUP-US Calendar of Database and Product Releases, and are also listed in the SID/SASD/SEDD Application Kit. These and other HCUP databases are available through the HCUP Central Distributor. More information can be found on the HCUP Central Distributor page of the HCUP-US Website.

National ED Data—Coming to HCUP!

HCUP is pleased to announce the upcoming release of a unique and powerful database: the Nationwide Emergency Department Sample (NEDS). The NEDS will provide national estimates of emergency department (ED) encounters in the U.S. The NEDS can be used to create national estimates of ED visits, both treat-and-release and those that resulted in a hospital admission. It contains clinical and non-clinical information on all patients, regardless of payer—including those covered by Medicare, Medicaid, private insurance, and the uninsured.

Researchers and policymakers can use its more than 25 million records to investigate access to health care in a changing marketplace; identify trends in ED utilization, charges, and outcomes; and conduct market area research and variation analyses.

The 2006 NEDS is scheduled for release in early 2009. For more information, please contact the HCUP Central Distributor or HCUP User Support at hcup@ahrq.gov.

New HCUP Data Query Tool Under Development: Input Your Data—Output Your Website

AHRQ is developing a new web-based tool tentatively called EQUIP, or the Efficiency and Quality Improvement Portal. The tool will consist of a series of computer programs that will enable host users—such as state and local data organizations, chartered value exchanges, hospitals, or health plan providers—to input their own data and generate a data-driven website. EQUIP programs will analyze, summarize, and present the information in an evidence-driven format, ready for use by consumers and other decision-makers.

EQUIP is similar in concept to HCUP’s on-line query system, HCUPnet, except organizations will host the new tool on their own web server and populate it with their own hospital administrative data. Design options will allow EQUIP to be customized to match the organization’s interface and data interests.

Alpha testing is currently underway, and AHRQ anticipates the launch of a beta version in late January 2009. For more information, please contact HCUP User Support at hcup@ahrq.gov.

Cost-to-Charge Ratio Files for the 2006 NIS, KID, and SID Coming Soon

The Cost-to-Charge Ratio Files for use with the 2006 NIS, KID, and SID are expected to be released this October through the HCUP Central Distributor. HCUP databases include information on hospital charges, which reflect the total amount billed by the hospital per patient admission. However, these charges do not reflect the actual cost of providing patient care. The Cost-to-Charge Ratio Files allow HCUP data users to convert hospital charge data to cost estimates. Linking HCUP databases with the appropriate files will allow researchers to estimate an all-payer inpatient cost-to-charge ratio for nearly every hospital included in the 2006 NIS, KID, and SID.

To facilitate use of these files, user guides containing detailed descriptions of the files, instructions on how to link the files to the NIS, KID, and SID and other valuable documentation will be available on the HCUP-US Website. Once released, the Cost-to-Charge Ratio Files can be ordered free-of-charge by contacting the HCUP Central Distributor.

2006 KID Now Available

Produced every three years since the 1997 data year, the 2006 Kids' Inpatient Database (KID) was released this past June. The KID is the only dataset in the U.S. designed specifically to study hospital use, outcomes, and charges in the pediatric population. The KID includes all patients under age 21, regardless of payer.

The 2006 KID includes a sample of pediatric hospital stays from 3,739 hospitals in 38 states. The KID can be weighted to produce national and regional estimates, allowing researchers and policymakers to use the data to identify, track, and analyze pediatric health care issues in the U.S. The KID’s sample design and large size makes this database well-suited for research on both common and rare pediatric conditions.

The 2006 KID is available for purchase through the HCUP Central Distributor. More information can be found on the HCUP Central Distributor page on the HCUP-US Website. Additional information on the KID is available on the HCUP-US website at http://www.hcup-us.ahrq.gov/kidoverview.jsp.

2006 NIS Available for Purchase

The Agency for Healthcare Research and Quality (AHRQ) released the 2006 Nationwide Inpatient Sample (NIS) this past May. The NIS is the largest inpatient care database including all patients, regardless of payer—covering Medicare, Medicaid, privately insured, and uninsured patients.

The 2006 NIS is drawn from 38 States and includes eight million discharge records from more than 1,000 hospitals. NIS data can be weighted to produce national estimates, allowing researchers and policymakers to identify, track, and analyze national trends in health care utilization, access, charges, quality, and outcomes. The NIS is considered by health services researchers to be one of the most reliable and affordable databases for studying important health care topics.

The 2006 NIS and other HCUP databases are available through the HCUP Central Distributor. More information can be found on the HCUP Central Distributor page on the HCUP-US Website. Additional information on the NIS is available on the HCUP-US website at http://www.hcup-us.ahrq.gov/nisoverview.jsp.

Annual Joint HCUP and MEPS Training Workshop Sponsored by AHRQ to be Held in September

On September 24-25, AHRQ will sponsor a workshop for health services researchers to facilitate the use of two AHRQ resources: the Healthcare Cost and Utilization Project(HCUP) databases and the Medical Expenditure Panel Survey (MEPS) databases. This workshop will be held at AHRQ headquarters in Rockville, Maryland. Participants will receive information on the components and capabilities of the databases and will learn how to extract data for research purposes, working directly with HCUP data on a computer.

Notification of this workshop was sent via e-mail to the HCUP Mailing List and was announced on the HCUP Calendar of Events. Registration for the workshop is full, but it is expected that another joint HCUP and MEPS training workshop will be offered at AHRQ in September 2009.

HCUP to Host a Learning Session and Booth at APHA’s Annual Meeting

The 136th Annual American Public Health Association (APHA) Meeting & Exposition will be held October 25 – 29 in San Diego, CA. More than 13,000 physicians, administrators, nurses, educators, epidemiologists, researchers, and related health specialists are expected to attend this meeting, which will address a range of current and emerging health science and policy topics.

An HCUP Learning Institute (LI) will take place at the conference on Sunday, October 26, from 8:00 a.m.–11:30 a.m. This HCUP session will introduce health services and policy researchers to the HCUP databases and related resources that can enhance their research studies.

To attend the APHA-LI courses, a separate course registration fee is required. AHRQ has subsidized the registration fee for this course to $50. For an additional $50, participants can receive CME, CNE, or CHES continuing education credits for attending APHA-LI courses. Space for the LI is limited. Registration information is available on the APHA Website.

HCUP will also be sponsoring a booth at this meeting. Representatives will be available throughout the conference to provide visitors with an introduction to the HCUP databases, tools, and products, and to answer questions.

AHRQ’s 2nd Annual Conference a Success

AHRQ hosted its 2nd Annual Conference from September 7-10 at the Bethesda North Marriott Convention Center in Bethesda, Maryland. Attendees took part in sessions focusing on the meeting’s theme of "Promoting Quality...Partnering for Change" and learned how AHRQ's research is being implemented at all levels in health care delivery, including health information technology; comparative effectiveness; prevention and care management; patient safety and quality improvement; value-driven care; and health care innovation and emerging issues.

Seven sessions were held at the conference that drew on HCUP data, information, and tools, including three on how HCUP research helps inform national, state, and local policymakers and a lunch and learn about EQUIP, HCUP’s new web-based tool for quality and utilization reporting that is currently under development. In the mAHRQet Place Café, HCUP research posters were on display and an HCUP exhibit booth featured HCUP products, demonstrations of HCUPnet, and representatives that answered questions.

For more information on the AHRQ 2nd Annual Conference, please visit the meeting’s website at http://www.ahrq.gov/. We look forward to seeing you next year!

A complete listing of HCUP events are noted in the HCUP Calendar.


RECENT PUBLICATIONS

Lan TV, Nobuhara KK, Lee H, Farmer DL. Determination of risk factors for deep venous thrombosis in hospitalized children. J Pediatr Surg. 2008 Jun; 43(6): 1095-99.

This study uses the Kids’ Inpatient Database (KID) for years 1997, 2000, and 2003 to examine the trend and risk factors for lower extremity deep venous thrombosis (DVT) in pediatric patients between the ages of 1 and 17 years who had a hospitalization of four days or longer. Over 1.65 million pediatric discharges noted a DVT, a rate of 4.3 per 1,000 stays. The research determined that patients between the ages of 15 and 17 year were two times more likely to develop a DVT than younger ages, and there was a two percent increased chance with each day hospitalized. Certain underlying conditions also increased the likelihood of developing a DVT, including obesity, inflammatory bowel disease, hematologic malignancies, and surgical procedures, such as orthopedic operations and other types. Mortality for DVT was 2.5 percent—five times greater than the average mortality rate of their study group (0.5 percent). The article abstract is available via Pubmed.

Zhang W, Mueller KJ, Chen WL. Uninsured hospitalizations: Rural and urban differences. J Rural Health. 2008 Spring; 24(2):194-202.

This study uses the 2002 Nationwide Inpatient Sample (NIS) to compare hospitalization patterns for uninsured patients in urban and rural settings. The findings showed that uninsured patients in rural areas were, on average, two years older and had shorter lengths of stay than those hospitalized in urban communities, and were more likely to live in areas with a median annual income of less than $35,000. Additionally, rural patients were more likely to be hospitalized for preventable conditions—1.5 times higher for adults and 1.7 times higher for children. The article abstract is available via Pubmed.

To access additional publications featuring HCUP data, please visit the Publications from the HCUP Databases page on the AHRQ Website.


USERS' TECH TIP

Weighting and the HCUP databases

A question the HCUP User Support team gets asked frequently is, "Is my HCUP data weighted?" Here are guidelines on how to know.

NIS
If you are using the full NIS (the ASCII file purchased through the HCUP Central Distributor), then the data you receive is not weighted. Unweighted, the NIS approximates a 20 percent sample of community hospitals in the U.S.

The Core NIS file includes a data element called DISCWT that can be used to produce national estimates. Applying the DISCWT variable is easy when working with a statistical software package like SAS, SPSS, or Stata. For example, in SAS the DISCWT variable can be applied via the WEIGHT statement in several SAS procedures, such as PROC FREQ and PROC MEANS. Once applied, the NIS data approximates all acute care discharges from non-rehabilitation, community hospitals in the United States. In 2006, this target universe contained 5,124 hospitals with nearly 40 million discharges.

Detailed information on the NIS sample design can be found in the report Introduction to the NIS, 2006 (http://www.hcup-us.ahrq.gov/db/nation/nis/2006NIS_INTRODUCTION.pdf).

KID
Like the NIS, the full KID file purchased through the HCUP Central Distributor is not weighted. In creating the KID, systematic random sampling is used to select 10 percent of uncomplicated in-hospital births and 80 percent of complicated in-hospital births and other pediatric cases from each frame hospital.

The Core KID file also includes the data element DISCWT that can be used to produce national estimates. Apply DISCWT to weight discharges in the Core file to obtain national estimates of pediatric discharges from all community hospitals in the U.S. For the 2006 KID, the target universe contained 3,739 hospitals.

Additional information on the weighting methodology used for the KID can be found in the Introduction to the Kids’ Inpatient Database (KID), 2006 report (http://www.hcup-us.ahrq.gov/db/nation/kid/KID_2006_Introduction.pdf)

State Databases
The SID, SASD, and SEDD contain the universe of the inpatient discharge abstracts for community hospitals in a given State. These state-level HCUP files are not based on a sample. No weighting needs to be applied to retrieve actual counts of discharges.

For doing State-level research, we do not recommend obtaining State estimates from the NIS and KID because of the sampling methodology used to create them. While both the NIS and KID include weights to allow researchers to generate national estimates from the raw counts, no weights are included for the calculation of State-level estimates. Always use the SID, SASD, and SEDD for State-level work.

For more information on the State Databases, please review the Introduction to the SID, Introduction to the SASD, and Introduction to the SEDD reports. Links to these documents are available on the HCUP-US Website http://www.hcup-us.ahrq.gov/databases.jsp under Database Documentation.

HCUPnet
If you are using HCUPnet (our free online query tool) to retrieve information on national inpatient hospital stays (based on the NIS or KID), the weighting has already been applied. The data output at the end of your query are national estimates. No additional work is necessary!

If you use HCUPnet to obtain State data (both inpatient and Emergency Department visits), the data you receive are actual counts of visits from that State’s discharge records.

Within the "State Statistics on All ED Visits" path, it is important to note that the aggregated ED data option does not provide a national estimate. In 2005, the aggregated option provides the total number of visits within the 23 SEDD participating states. These states comprise 51 percent of the U.S. population and 47 percent of ED visits in the U.S. National ED estimates will soon be available from HCUP with the release of the Nationwide Emergency Department Sample (NEDS) in December 2008.


HCUP CALENDAR

Upcoming Events:

September 24-25, 2008: Joint Medical Expenditures Panel Survey (MEPS) and Healthcare Cost and Utilization Project (HCUP) Workshop, AHRQ Headquarters, Rockville, MD

AHRQ staff will sponsor a hands-on workshop for health services researchers on two useful AHRQ resources: the Healthcare Cost and Utilization Project (HCUP) and the Medical Expenditure Panel Survey (MEPS). Participants will receive information on the components and capabilities of both HCUP and MEPS products and will learn how to extract data for research purposes. Pre-registration was required, and the Workshop is full. Information on the Workshop is on the MEPS Website.

October 25-29, 2008: American Public Health Association (APHA) 136th Annual Meeting & Exposition, San Diego, CA

AHRQ staff will be presenting an HCUP Learning Institute on Sunday, October 26th from 8:30-11:30 A.M. at the 136th Annual American Public Health Association (APHA) Meeting & Exposition in San Diego, CA. The session will introduce health services and policy researchers to the HCUP databases and related resources that can enhance their research studies. A separate course registration fee is required. AHRQ has subsidized the registration fee for this course to $50. For an additional $50, participants can receive CME, CNE or CHES continuing education credits for attending APHA-LI courses. Please register early on the APHA Website. Space is limited. AHRQ will also be sponsoring on exhibit booth at the conference.

October 27-28, 2008: National Organization of Health Data Organizations (NAHDO) 23rd Annual Meeting, San Antonio, TX

HCUP will have a booth at NAHDO’s 23rd Annual Conference, to be held October 27-28 at the Westin Riverwalk in San Antonio, Texas. HCUP representatives will be available throughout the conference to provide visitors with an introduction to the HCUP databases, tools, and products. For more information and to register, please visit the NAHDO Website.

Recent Events:

September 10, 2008: Healthcare Workforce and Regionalization of Services. Session for National, State, and Local Policymakers, 2008 AHRQ Annual Conference, Bethesda, MD

A panel highlighted studies on important healthcare workforce and regionalization of services topics using recent research from HCUP to inform health policy. The studies showed the range of workforce specialty topics for which administrative data can provide information useful for policy decisions. The topics were:


The speakers were:
Rini Ratan, MD, Columbia University College of Physicians & Surgeons
Stephen C. Yang, MD, Johns Hopkins School of Medicine
Julie Anne Sosa, MA MD, Yale University School of Medicine
Matthew Cooperberg, MD, MPH, University of California, San Francisco
Laurent G. Glance, MD, University of Rochester School of Medicine

September 9, 2008: Using Administrative Data to Improve Public Health and Safety. Session for National, State, and Local Policymakers, 2008 AHRQ Annual Conference, Bethesda, MD

A panel highlighted studies on important public health and safety topics using recent research from HCUP to inform health policy. The studies are examples of the range of topics for which administrative data can provide information useful for health policy decisions. The topics were:


The speakers were:
Michael Jhung, MD, MPH, Division of Healthcare Quality Promotion, CDC
Marie Griffin, MD, MPH, Vanderbilt University Medical Center
Claudia Steiner, MD, MPH Center for Delivery, Organization and Markets, AHRQ
Anbesaw Selassie, DRPH, Medical University of South Carolina
Ted R. Miller, PhD, Pacific Institute for Research and Evaluation

June 25, 2008: Using HCUP for Pediatric Emergency Medicine Research – Emergency Medical Services for Children (EMSC) 2008 Annual Grantee Meeting, Bethesda, MD

HCUP presented a 90-minute workshop at the 2008 EMSC Annual Grantee Meeting in Bethesda, MD. This presentation provided an overview of HCUP, specializing in how it relates to pediatric emergency medicine research interests. The session also included an interactive demonstration of our free online querying tool, HCUPnet. Time was allocated to address participant questions.

June 22-25, 2008: Introduction to the Research Databases from the Agency for Healthcare Research and Quality (AHRQ) – American Society of Health Economists 2nd Biennial Meeting, Duke University, Durham, NC

AHRQ staff presented a half-day HCUP-MEPS introductory workshop on June 22 at the American Society of Health Economists (ASHE) 2nd Biannual Conference, at Duke University in Durham, North Carolina. A 90-minute introduction to HCUP databases, software, and related products, including HCUP tools like the Cost-to-Charge Ratio Files and Hospital Market Structure File, were presented. A demonstration of HCUPnet was also provided, and participants received a free CD-ROM containing valuable resources that expand on topics covered in the session. The second part of the presentation provided a general overview of the MEPS, its available data files, information about online data tools, and some examples of the type of research projects the MEPS data can support.

******************************************************

This publication and previous editions of HCUP e-News are available online at the HCUP-US Website.

HCUP e-News is distributed via the HCUP Mailing List.

For more information, or if you have any questions regarding HCUP e-News, please contact P. Hannah Davis, M.S., at hannah.davis@ahrq.hhs.gov, or Carol Stocks, RN, MHSA, at carol.stocks@ahrq.hhs.gov.