Page semi-protected

Healthcare Cost and Utilization Project

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

HCUP Logo

The Healthcare Cost and Utilization Project (HCUP, pronounced "H-Cup") is a family of health care databases and related software tools and products from the United States that is developed through a Federal-State-Industry partnership and sponsored by the Agency for Healthcare Research and Quality (AHRQ).

General Information

HCUP provides access to health care databases for research and policy analysis, as well as tools and products to enhance the capabilities of the data. HCUP databases combine the data collection efforts of state data organizations, hospital associations, private data organizations, and the federal government to create a national information resource of patient-level health care data. state organizations that provide data to HCUP are called Partners. HCUP includes multiyear hospital care (inpatient, outpatient, and emergency department) data in the United States, with all-payer, encounter-level information beginning in 1988. These databases enable research on health research and policy issues at the national, state, and local market levels, including cost and quality of health services, medical practice patterns, access to health care, and outcomes of treatments. AHRQ has also developed a set of software tools to be used when evaluating hospital data. These software tools can be used with the HCUP databases and with other administrative databases. HCUP’s Supplemental Files are only for use with HCUP databases. HCUP databases have been used in various studies on a number of topics, such as breast cancer, depression, and multimorbidity, incidence and cost of injuries, role of socioeconomic status in patients leaving against medical advice, multiple chronic conditions and disparities in readmissions, and hospitalization costs for cystic fibrosis.

HCUP User Support Web Site (HCUP-US)

The HCUP User Support Web site is the main repository of information for HCUP. It is designed to answer HCUP-related questions; provide detailed information on HCUP databases, tools, and products; and offer technical assistance to HCUP users. HCUP’s tools, publications, documentation, news, services, HCUP Fast Stats, and HCUPnet (the online data query system) may all be accessed through HCUP-US. HCUP-US is located at https://www.hcup-us.ahrq.gov.

HCUP Overview Course

HCUP has developed an interactive online course that provides an overview of the features, capabilities, and potential uses of HCUP. The course is modular, so users can either move through the entire course or access the resources in which they are most interested. The On-line HCUP Overview Course can work both as an introduction to HCUP data and tools and a refresher for established users.

HCUP Online Tutorial Series

The HCUP Online Tutorial Series is a set of interactive training courses that provide HCUP data users with information about HCUP data and tools, and training on technical methods for conducting research with HCUP data. The online courses are modular, so users can move through an entire course or access the sections in which they are most interested. Topics include loading and checking HCUP data, understanding HCUP’s sampling design, calculating standard errors, producing national estimates, conducting multiyear analysis, and using the nationwide readmissions database.

HCUP Databases

HCUP databases bring together data from state data organizations, hospital associations, private data organizations, and the federal government to create an information resource of patient-level health care data. HCUP’s databases date back to 1988 data files. The databases contain encounter-level information for all payers compiled in a uniform format with privacy protections in place. Researchers and policymakers can use the records to identify, track, and analyze national trends in health care use, access, charges, quality, and outcomes. HCUP databases are released approximately 6 to 18 months after the end of a given calendar year, with state databases available earlier than the national dataset. For example, 2016 state data were available beginning in 2017, and nationwide data were available beginning in July 2018. Currently there are seven types of HCUP databases: four with national- and regional-level data and three with state- and local-level data.

National Databases

  • National Inpatient Sample (NIS) (formerly the Nationwide Inpatient Sample): Annual inpatient data from a stratified systematic sample of discharges from all hospitals in HCUP, equal to approximately 20 percent of all discharges in U.S. community hospitals, excluding rehabilitation and long-term acute-care hospitals. Data are available from 1988 forward, and a new database is released annually, approximately 18 months after the end of a calendar year. The NIS Overview and the NIS Database Documentation pages of the HCUP-US Web site contain additional information.
  • Kids’ Inpatient Database (KID): A nationwide sample of pediatric inpatient discharges. The KID was released every three years, from 1997 to 2012 and resumed release again in 2016.
  • Nationwide Emergency Department Sample (NEDS): A database of approximately 31 million records that yields national estimates of 143 million emergency department (ED) visits. The NEDS captures encounters where the patient is admitted for inpatient treatment, as well as those in which the patient is treated and released. The NEDS is released annually and is available from 2006 forward.
  • Nationwide Readmissions Database (NRD): The NRD is designed to support various types of analyses of national readmission rates for all payers and uninsured people. The NRD is released every year from 2010 forward.

State Databases

  • State Inpatient Databases (SID): Databases from the universe of inpatient discharge abstracts from participating states, released annually. Data are available from 1995 forward. The SID are released on a rolling basis, as early as six months following the end of a calendar year.
  • State Ambulatory Surgery and Services Databases (SASD): Ambulatory surgery encounter abstracts from hospital-affiliated and sometimes freestanding ambulatory surgery sites in participating states. Data are available from 1997 forward. The SASD are released on a rolling basis, as early as six months following the end of a calendar year.
  • State Emergency Department Databases (SEDD): Hospital-affiliated emergency department data for visits in participating states that do not result in hospitalizations. Data are available from 1999 forward. The SID are released on a rolling basis, as early as six months following the end of a calendar year.

HCUP Tools and Software

HCUP provides a number of tools and software programs that can be applied to HCUP and other similar administrative databases.

HCUPnet

HCUPnet is an online query system that provides health care statistics and information from the HCUP national (NIS, NEDS, KID, and NRD) and state (SID, SASD, and SEDD) databases for those states that have agreed to participate. HCUPnet can be used for identifying, tracking, analyzing, and comparing statistics on hospital inpatient stays, emergency care, and ambulatory surgery, as well as obtaining measures of quality based on the AHRQ Quality Indicators. Select statistics are available at a national- and county-level. HCUPnet can also be used for trend analysis with health care data available from 1993 forward. HCUPnet also includes a feature called hospital readmissions that provides users with some statistics on hospital readmissions within 7 and 30 days of hospital discharge. Information on calculating readmissions for HCUPnet is available in the HCUP Methods Series report.

HCUP Fast Stats

HCUP Fast Stats is a web-based tool that provides HCUP-based statistics for health care information topics. The following topics are available:

  • State Trends in Hospital Use by Payer (formerly called Effect of Health Insurance Expansion on Hospital Use and Effect of Medicaid Expansion on Hospital Use) launched in July 2015, with data updates released quarterly starting October 2015. This topic includes statistics from up to 44 states on the number of hospital discharges by payer group (Medicare, Medicaid, private insurance, and uninsured) for categories of conditions (surgical, mental health, injury, maternal, and medical). Users can run state-by-state comparisons and analyze the effects of Medicaid expansion on hospital utilization levels and payment sources.
  • National Hospital Utilization and Costs was released in December 2015. This topic focuses on national statistics on inpatient stays: Trends, Most Common Diagnoses, and Most Common Operations.
  • State Trends in Emergency Department Visits by Payer was released in July 2016. These ED statistics are a supplement to the existing state-level inpatient stay trends by expected payer. Quarterly ED visit counts are presented from 2006-2016 for up to 32 states in a given year.
  • Opioid-Related Hospital Use was released in August 2017. This topic reports population-based rates of opioid-related hospital use by discharge quarter starting in 2008 for up to 45 states. Trends are available for inpatient stays and emergency department visits by expected payer (Medicare, Medicaid, private insurance, and uninsured). In June 2018, an interactive map for the topic was introduced that shows each state’s opioid-related inpatient or ED rate relative to the distribution across all states providing data in 2015.

HCUP Fast Stats will be updated regularly (quarterly or annually, as newer data become available).

Quality Indicators (QIs)

The AHRQ Quality Indicators (QIs) are standardized, evidence-based measures of health care quality that use readily available hospital inpatient administrative data. AHRQ QIs can be used to highlight potential quality concerns, identify areas that need further study and investigation, and track clinical performance and outcomes over time. , , The AHRQ QIs consist of four modules measuring various aspects of quality:

  • Prevention Quality Indicators (PQIs) can be used with hospital inpatient discharge data to identify quality of care for “ambulatory care sensitive conditions.” These are conditions for which good outpatient care can potentially prevent the need for hospitalization or for which early intervention can prevent complications or more severe disease.
  • Inpatient Quality Indicators (IQIs) reflect quality of care inside hospitals, including inpatient mortality for medical conditions and utilization of procedures for which there are questions of overuse, underuse, and misuse.
  • Patient Safety Indicators (PSIs) also reflect quality of care inside hospitals but focus on potentially avoidable complications and adverse events following surgeries, procedures, and childbirth.
  • Pediatric Quality Indicators (PDIs) reflect quality of care inside hospitals and identify potentially avoidable hospitalizations among children.

Clinical Classifications Software (CCS)

The Clinical Classifications Software (CCS) provides a method for classifying diagnoses or procedures into clinically meaningful categories. These can be used for aggregate statistical reporting of a variety of topics, such as identifying populations for disease- or procedure-specific studies or developing statistical reports providing information (i.e., charges and length of stay) about relatively specific conditions. , Three versions of the CCS Software are available:

  • Beta CCS for ICD-10-CM/PCS is based on the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM)/Procedure Coding System, a uniform and standardized coding system. The Beta CCS for ICD-10-CM/PCS builds on the CCS for ICD-9-CM and attempts to map ICD-10-CM/PCS codes into the same categories. The ICD-10-CM/PCS has more than 69,800 diagnosis codes and 71,900 procedure codes collapsed into a smaller number of clinically meaningful categories.

The current CCS for ICD-10-CM/PCS has 285 mutually exclusive categories for diagnoses and 231 for procedures. For certain research interests, this smaller number can be more useful for presenting descriptive statistics than individual ICD-10-CM/PCS codes. Every effort was made to translate the CCS system into ICD-10-CM/PCS without making changes to the CCS assignments for diagnoses and procedures, but because of the new structure and expanded code availability this was not always possible. Because of the increased specificity of ICD-10-CM/PCS and the changes in the two code set structure, it was not possible to translate most multilevel categories to ICD-10-CM/PCS within the current structure – with the exception of the first- and second-level multilevel categories. The Beta CCS for ICD-10-CM/PCS will be updated annually.

  • Clinical Classifications Software (CCS) for ICD-9-CM is based on the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), a uniform and standardized coding system.

Since fiscal year 2008, CCS for ICD-9-CM includes categories from the Mental Health and Substance Abuse Clinical Classifications Software (CCS-MHSA). These categories replace the original CCS categories for mental health and substance abuse. Specifically, the CCS single-level software includes the CCS-MHSA general categories, and the CCS multilevel software includes the CCS-MHSA specific categories. The CCS for ICD-9-CM was updated annually starting January 1980 through September 30, 2015. ICD-9-CM codes were frozen in preparation for ICD-10-CM implementation and regular maintenance of the codes has been suspended.

  • Clinical Classifications Software (CCS) for Services and Procedures provides users with a method of classifying Current Procedural Terminology (CPT®) codes and Healthcare Common Procedure Coding System (HCPCS) codes into 244 clinically meaningful procedure categories. More than 9,000 CPT/HCPCS codes and 6,000 HCPCs codes are accounted for.

The CCS versions and their user guides are available for download from the HCUP-US Web site.

Chronic Condition Indicator

The Chronic Condition Indicator (CCI) facilitates health services research on diagnoses using administrative data. Two versions of the CCI software are available, CCI for ICD-9-CM and Beta CCI for ICD-10-CM. The CCI tools categorize ICD-9-CM/ICD-10-CM diagnoses codes into two classifications: chronic or not chronic. A chronic condition is defined as a condition that lasts 12 months or longer and meets one or both of the following tests: (a) it places limitations on self-care, independent living, and social interactions; and (b) it results in the need for ongoing intervention with medical products, services, and special equipment. The identification of chronic conditions is based on all 5-digit ICD-9-CM or 7-digit ICD-10-CM codes. External cause of injury codes are not classified because all injuries are assumed to not be chronic. Currently, there are approximately 14,000 diagnosis codes in version ICD-9-CM and 69,800 diagnosis codes in version ICD-10-CM. The tool also assigns diagnosis codes into one of 18 body system indicator categories, allowing users to create indicators listing which specific body systems are affected by a chronic condition. The body system indicator is based on the chapters of the ICD-9-CM/ICD-10-CM codebooks. This indicator may be useful as a means of counting the number of body systems affected by chronic conditions. Alternatively, the Clinical Classification Software (CCS) may be used in conjunction with the CCI to obtain a count of the number of relatively discrete chronic conditions. The ICD-9-CM CCI was updated annually and is valid for codes from January 1, 1980, through September 20, 2015. ICD-9-CM codes were frozen in preparation for ICD-10-CM implementation and regular maintenance of the codes has been suspended. The ICD-10-CM CCI is updated annually and is valid for codes from October 1, 2015, forward.

Elixhauser Comorbidity Software

Elixhauser Comorbidity Software assigns variables that identify comorbidities in hospital discharge records using ICD-9-CM or ICD-10-CM diagnosis coding and has been used in various analyses. Two versions of the Elixhauser Comorbidity Software are available: Beta Elixhauser Comorbidity Software for ICD-10-CM and Elixhauser Comorbidity Software for ICD-9-CM. The Beta Elixhauser Comorbidity Software for ICD-10-CM consists of two SAS computer programs for personal computers. Although the programs are written in SAS, they are distributed in ASCII so that they can be readily adapted to other programming languages. The first program, Creation of Format Library for Elixhauser Comorbidity Groups, generates a SAS format library that maps diagnosis codes into comorbidity indicators. Additional formats are created to exclude conditions that may be complications or that may be related to the principal diagnosis. The second SAS program, Creation of Elixhauser Comorbidity Variables, applies these formats to a data set containing administrative data and then creates the 29 comorbidity variables. The Elixhauser Comorbidity Software for ICD-9-CM (Version 3.7) contains a third SAS program, Creation of Elixhauser Comorbidity Index Scores, that applies the weights and creates the two indices for the Elixhauser Comorbidity Software ‒ one for in-hospital mortality and one for readmission. The Elixhauser Comorbidity Software for ICD-9-CM is based on ICD-9-CM and Medicare Severity Diagnosis-Related Group (MS-DRG) codes and valid through September 30, 2015. The Elixhauser Software for ICD-9-CM was updated annually from January 1, 1980, through September 30, 2015. The ICD-9-CM codes were frozen in preparation for ICD-10 implementation and regular maintenance of the codes has been suspended. The Beta Elixhauser Comorbidity Software for ICD-10-CM is updated annually and based on the ICD-10-CM and MS-DRG codes that are valid through September 30 of the designated fiscal year after October 1, 2015. The Elixhauser Comorbidity Software is available for download on the HCUP-US Web site.

Procedure Classes

Procedure Classes facilitate research on hospital services using administrative data by identifying whether an ICD-9-CM or ICD-10-CM procedure is (a) diagnostic or therapeutic, and (b) minor or major in terms of invasiveness and/or resource use. There are two versions of Procedure Classes tools, Procedure Classes for ICD-9-CM and Beta Procedure Classes for ICD-10-PCS. The Procedure Classes can be used to categorize procedure codes into one of four broad categories:

  • Minor Diagnostic: Nonoperating room procedures that are diagnostic (e.g., B244ZZZ, Ultrasonography of Right Heart)
  • Minor Therapeutic: Nonoperating room procedures that are therapeutic (e.g., 02HQ33Z, Insertion of Infusion Device into Right Pulmonary Artery, Percutaneous Approach
  • Major Diagnostic: Procedures that are considered valid operating room procedures by the MS-DRG grouper and that are performed for diagnostic reasons (e.g., 02BV0ZX, Excision of Superior Vena Cava, Open Approach, Diagnostic)
  • Major Therapeutic: Procedures that are considered valid operating room procedures by the MS-DRG grouper and that are performed for therapeutic reasons (e.g., 0210093, Bypass Coronary Artery, One Site from Coronary Artery with Autologous Venous Tissue, Open Approach).

The Procedure Classes for ICD-9-CM were updated annually from January 1, 1980, through September 30, 2015. The ICD-9-CM codes were frozen in preparation for ICD-10 implementation and regular maintenance of the codes has been suspended. The Beta Procedure Classes for ICD-10-PCS are updated annually and valid for codes from October 1, 2015, forward. Procedure Classes are available for download from the HCUP-US Web site.

Utilization Flags

Utilization Flags combine information from Uniform Billing (UB-04) revenue codes and ICD-9-CM or ICD-10-PCS procedure codes to create flags—or indicators—of utilization of services rendered in health care settings such as hospitals, emergency departments, and ambulatory surgery centers. There are two types of Utilization Flags, Utilization Flags for ICD-9-CM and Beta Utilization Flags for ICD-10-CM/PCS. The Utilization Flags can be used to study a broad range of services, including simple diagnostic tests and resource-intense procedures, such as use of intensive care units. They can also be used to more reliably examine utilization of diagnostic and therapeutic services. The Utilization Flags for ICD-9-CM were updated annually from January 1, 2003, through September 30, 2015. The ICD-9-CM codes were frozen in preparation for ICD-10 implementation and regular maintenance of the codes has been suspended. The Beta Utilization Flags for ICD-10-CM/PCS are updated annually and valid for codes from October 1, 2015, forward. The Utilization Flags are available for download from the HCUP-US Web site.

Surgery Flags

Surgery Flag Software consists of a SAS program and two files that include information about the classification of procedures into the broad and narrow definitions of surgeries in ICD-9-CM or CPT-based inpatient and ambulatory surgery data. Three versions of the Surgery Flag Software are available. The initial release in September 2014 is valid for ICD-9-CM codes through September 2013 and CPT codes through December 2013. A second version was released in June 2015. A third version, focusing on CPT only, was released in April 2017. This version brought the Surgery Flag software up to date for CPT codes through 2017. The software assignments are validated by certified coding specialists. The Surgery Flag Software identifies two types of surgical categories: NARROW and BROAD. NARROW surgery is based on a narrow, targeted, and restrictive definition and includes invasive surgical procedures. An invasive therapeutic surgical procedure involves incision, excision, manipulation, or suturing of tissue that penetrates or breaks the skin; typically requires use of an operating room; and requires regional anesthesia, general anesthesia, or sedation to control pain. BROAD surgery includes procedures that fall under the NARROW category but adds less invasive therapeutic surgeries and diagnostic procedures often performed in surgical settings. Users must agree to a license agreement with the American Medical Association to use the Surgery Flags before accessing the software.

HCUP Supplemental Files

The HCUP Supplemental Files augment applicable HCUP databases with additional data elements or analytically useful information that was not available when the HCUP databases were originally released. They cannot be used with other administrative databases.

Cost-to-Charge Ratio Files (CCR)

The Cost-to-Charge Ratio (CCR) Files are hospital-level files designed to convert the hospital total charge data to cost estimates when merged with data elements exclusively in the HCUP NIS, KID, NRD, and SID. The HCUP databases are limited to information on total hospital charges, which reflect the amount billed to the payer per patient encounter. Total charges do not reflect the actual cost of providing care or the payment received by the hospital for services provided. This total charge data can be converted into cost estimates using the CCR Files, which include hospital-wide values of the all-payer inpatient cost-to-charge ratio for nearly every hospital in the participating NIS, KID, NRD, and SID. CCR files can be used for various types of cost analyses. Cost information was obtained from the hospital accounting reports collected by the Centers for Medicare & Medicaid Services (CMS). Researchers and policy makers can use the converted cost estimates to examine a variety of topics, including use and cost of hospital services, health care cost inflation, and how the cost experiences of a given hospital or health plan compare with national or state trends. The Cost-to-Charge Ratio Files are updated annually and available for the HUCP inpatient databases beginning with 2001 data.

Hospital Market Structure (HMS) Files

The Hospital Market Structure (HMS) Files are hospital-level files designed to supplement the data elements in the NIS, KID, and SID databases. The HMS Files contain various measures of hospital market competition. These aggregate measures are meant to broadly characterize the intensity of competition that hospitals may be facing under various definitions of market area. Hospital market definitions were based on hospital locations, and in some cases, patient ZIP Codes. Hospital locations were obtained from the American Hospital Association (AHA) Annual Survey Database, Area Resource File (ARF), HCUP Historical Urban/Rural – County (HURC) file, and ArcView GIS. Patient ZIP Codes were obtained from the SID. Users can merge the data elements on the HMS Files to the corresponding NIS, KID, or SID hospitals by the hospital identification number (HOSPID). Using the merged data elements, users can include hospital market structure measures in analyses. Hospital market structure measures are generally useful for performing empirical analyses that examine the effects of hospital competition on the cost, access, and quality of hospital services. They are most useful to analysts as a secondary control variable (e.g., for assessing whether a statistical relationship exists between two variables when hospital market structure is controlled). The Hospital Market Structure Files are updated every three years. The HCUP Hospital Market Structure Files are currently available for 1997, 2000, 2003, 2006, and 2009.

HCUP Supplemental Variables for Revisit Analyses

The HCUP Supplemental Variables for Revisit Analyses allow users to track sequential visits for a patient within a state and across facilities and hospitals settings (inpatient, emergency department, and ambulatory surgery) while adhering to strict privacy guidelines. The available clinical information can determine if these sequential visits are unrelated, an expected follow-up, complications from a previous treatment, or an unexpected revisit or rehospitalization. Users must merge the supplemental files with the corresponding SID, SASD, or SEDD for any analysis. Data are available from 2003-2008 in ASCII format. Beginning with 2009 data, the revisit variables are included in the Core file of the HCUP State Databases when possible.

NIS and KID Trend Files

The NIS-Trends and KID-Trends files are available to help researchers conduct longitudinal analyses. They are discharge-level files that provide researchers with the trend weights and, in the case of the NIS-Trends, data elements that are consistently defined across data years.

American Hospital Association (AHA) Linkage Files

The American Hospital Association (AHA) Linkage Files are hospital-level files that contain a small number of data elements that allow researchers to supplement the HCUP State Databases with information from the AHA Annual Survey Databases (Health Forum, LLC © 2012). The files are designed to support richer empirical analysis where hospital characteristics may be important factors. Linkage is only possible in states that allow the release of hospital identifiers and are unique by state and year. The HCUP AHA Linkage Files for the SID, SASD, and SEDD are available starting in 2006 from the HCUP-US Web site. Nationwide Inpatient Sample (NIS) Hospital Ownership Files The NIS Hospital Ownership Files are hospital-level files designed to facilitate analysis of the NIS by hospital ownership categories. These HCUP supplemental files allow the user to identify in the 1998-2007 NIS the following three types of hospitals: government, nonfederal; private, nonprofit; and private, investor owned.

HCUP News and Reports

HCUP produces material to report new findings based on HCUP data and to announce HCUP news.

  • The HCUP eNews summarizes activities of the HCUP project quarterly.
  • The HCUP Mailing List sends e-mail updates on news, product releases, events, and the quarterly eNews to interested subscribers.
  • HCUP Statistical Briefs provide health care statistics for various topics based on HCUP databases.
  • HCUP Infographics show data from the HCUP Statistical Brief series. Topics have included inpatient vs. outpatient surgeries in U.S. hospitals, neonatal hospital stays related to substance use, and characteristics of hospital stays involving malnutrition.
  • HCUP Methods Series Reports offer methodological information on the HCUP databases and software tools.
  • HCUP Projections Reports use longitudinal HCUP data to project national and regional estimates on health care priorities.

See also

References

  1. Zoorob RJ, Salemi JL, Mejia de Grubb MC, et al. A nationwide study of breast cancer, depression, and multimorbidity among hospitalized women and men in the United States. Breast Cancer Res Treat 2018 Nov 21. [Epub ahead of print]
  2. Zonfrillo MR, Spicer RS, Lawrence BA, et al. Incidence and costs of injuries to children and adults in the United States. Inj Epidemiol 2018 Oct 8;5(1):37.
  3. Yuan S, Ashmore S, Chaudhary KR, et al. The role of socioeconomic status in individuals that leave against medical advice. S D Med 2018 May;71(5):214-219.
  4. Basu J, Hanchate A, Koroukian S. Multiple chronic conditions and disparities in 30-day hospital readmissions among nonelderly adults. J Ambul Care Manage 2018 Oct/Dec;41(4):262-273.
  5. Vadagam P, Kamal KM. Hospitalization costs of cystic fibrosis in the United States: a retrospective analysis. Hosp Pract (1995) 2018 Oct;46(4):203-213. Epub 2018 Aug 9.
  6. Bath J, Dombrovskiy VY, Vogel TR. Impact of Patient Safety Indicators on readmission after abdominal aortic surgery. J Vasc Nurs 2018 Dec;36(4):189-195. Epub 2018 Oct 2.
  7. Nguyen MC, Moffatt-Bruce SD, Van Buren A. Daily review of AHRQ patient safety indicators has important impact on value-based purchasing, reimbursement, and performance scores. Surgery 2018 Mar;163(3):542-546. Epub 2017 Dec 21.
  8. Engineer LD, Winters BD, Weston CM, et al. Hospital characteristics and the Agency for Healthcare Research and Quality Inpatient Quality Indicators: a systematic review. SMJ Healthc Qual 2016 Sep-Oct;38(5):304-313.
  9. Al-Qurayshi Z, Baker SM, Garstka M, et al. Post-operative infections: trends in distribution, risk factors, and clinical and economic burdens. Surg Infect (Larchmt) 2018 Oct;19(7):717-722. Epub 2018 Sep 5.
  10. Chan L, Chauhan K, Poojary P, et al. National estimates of 30-day unplanned readmissions of patients on maintenance hemodialysis. Clin J Am Soc Nephrol 2017 Oct 6;12(10):1652-1662. Epub 2017 Sep 28.
  11. Gardner J, Sexton KW, Taylor J, et al. Defining severe traumatic brain injury readmission rates and reasons in a rural state. Trauma Surg Acute Care Open 2018 Sep 8;3(1):e000186. eCollection 2018.
  12. Moore BJ, White S, Washington R, et al. Identifying increased risk of readmission and in-hospital mortality using hospital administrative data: the AHRQ Elixhauser Comorbidity Index. Med Care 2017 Jul;55(7):698-705.
  13. Harris CR, Osterberg EC, Sanford T, et al. National variation in urethroplasty cost and predictors of extreme cost: a cost analysis with policy implications. Urology 2016 Aug;94:246-54. Epub 2016 Apr 20.
  14. Cerullo M, Chen SY, Dillhoff M, et al. Variation in markup of general surgical procedures by hospital market concentration. Am J Surg 2018 Apr;215(4):549-556. Epub 2017 Oct 23.