AHRQ--Agency for Healthcare Research and Quality: Advancing Excellence in Health Care

Using AHRQ Quality Indicators

  • Why do the draft ICD-10 specifications for the PSIs seem to classify cases as surgical that were previously classified as medical in ICD-9-CM PSI specifications, thereby increasing the denominator for “perioperative” and “postoperative” PSIs? — New!

    AHRQ uses the list of major operating room (OR) procedures that is developed and maintained by the Centers for Medicare & Medicaid Services (see Draft ICD-10 MS-DRG v28 definitions, Appendix E at https://www.cms.gov/icd10manual/fullcode_cms/P0033.html ). The Procedure Code/MS-DRG Index is a reference source of all ICD-9-CM procedure codes that affect MS-DRG assignment, the MDCs and MS-DRGs to which they are assigned, and a description of the surgical categories. Some procedures that were previously classified as a non-OR procedure in the ICD-9-CM MS-DRG list are currently classified as an OR procedure in the draft ICD-10 MS-DRG v28. For example, codes such as percutaneous insertion of a catheter device in an “upper artery” for blood pressure monitoring (ICD-10-PCS code = 03HY32Z) and diagnostic paracentesis by percutaneous drainage of the peritoneal cavity with no device (ICD-10-PCS code = 0W9G3ZX) are currently considered an OR procedure in CMS’ draft ICD-10 list. As a result, patients who had these procedures are counted in the denominators for “perioperative” and “postoperative” events such as PSI 04 (Death Rate among Surgical Inpatients with Serious Treatable Conditions), PSI 08 (Postoperative Hip Fracture Rate), PSI 09 (Perioperative Hemorrhage or Hematoma Rate), PSI 10 (Postoperative Physiologic and Metabolic Derangement Rate), PSI 11 (Postoperative Respiratory Failure Rate), PSI 12 (Perioperative Pulmonary Embolism or Deep Vein Thrombosis Rate), PSI 13 (Postoperative Sepsis Rate), PDI 08 (Perioperative Hemorrhage or Hematoma Rate), PDI 09 (Postoperative Respiratory Failure Rate), and PDI 10 (Postoperative Sepsis Rate). Composite measures that include these indicators are also affected.

    AHRQ is aware of CMS’ classification change, and addressing the change as it relates to the PSIs. In the next version of the ICD-10 software, AHRQ is excluding these procedures from the denominator, in addition to a few other procedures that were previously excluded in the ICD-9-CM PSI technical specifications (e.g., allogenic bone marrow transplants). Please be aware that the next version of the ICD-10-CM/PCS QI software (forthcoming Spring/Summer 2016) will include significant changes relative to the draft specifications and software that are currently posted for public comment and evaluation. AHRQ welcomes input from the user community about the beta version of the ICD-10-CM/PCS software currently posted on the AHRQ QI website.

  • Which modules are impacted by v5.0.3 changes? — New!

    Three modules include corrections in the updated v5.0.3 software. The PDI module includes corrections to risk-adjustment coefficients that impact expected, risk-adjusted, and upper and lower risk adjusted confidence limits rates for seven area-level PDIs (PDI14, 15, 16, 18, 90, 91 and 92). In addition, the PQI, IQI and PDI modules also include corrections to signal variance calculations that impact smoothed rates and the smoothed standard error, which calculates the confidence levels for the smoothed rates, for all PQIs and area-level IQIs and PDIs. Finally, the PQI module includes corrections to the reference population rates used to calculate risk-adjusted PQI rates.

    The PSI module is not impacted by these changes.

  • Which specific QIs will be affected by this change? — New!

    The PSI module is not impacted by these changes. Observed rates also are not impacted by these changes for any affected modules. Below, the specific changes to each module are listed.

    • The expected, risk-adjusted, smoothed rates and upper and lower confidence intervals of the following PDIs were corrected as part of this software update:
      • PDI14 Asthma Admission Rate
      • PDI15 Diabetes Short-Term Complications Admission Rate
      • PDI16 Gastroenteritis Admission Rate
      • PDI18 Urinary Tract Infection Admission Rate
      • PDI90 Pediatric Quality Overall Composite
      • PDI91 Pediatric Quality Acute Composite
      • PDI92 Pediatric Quality Chronic Composite
    • The smoothed rates and smoothed standard error, which is used to calculate the smoothed confidence intervals, for the following area-level PDIs were corrected as part of this software update, specifically
      • PDI14 Asthma Admission Rate
      • PDI15 Diabetes Short-Term Complications Admission Rate
      • PDI16 Gastroenteritis Admission Rate
      • PDI18 Urinary Tract Infection Admission Rate
    • The smoothed rates and smoothed standard error, which is used to calculate the smoothed confidence intervals, of all area-level IQIs were corrected as part of this software update, specifically
      • IQI26 Coronary Artery Bypass Graft
      • IQI27 Percutaneous Coronary Intervention (PCI) Rate
      • IQI28 Hysterectomy Rate
      • IQI29 Laminectomy or Spinal Fusion Rate
    • The risk adjusted rates, smoothed rates and smoothed standard error, which is used to calculate the smoothed confidence intervals, of all PQIs were corrected as part of this software update
  • What aspects of the QIs contain discrepancies in the March 2015 v5.0 software? — New!

    The original release of the v5.0 software (March 2015) contained discrepancies in the coefficients used to calculate expected, risk-adjusted and smoothed PDI rates. In addition, it contained discrepancies in signal variance calculations used to calculate smoothed rates and smoothed standard error, which is used to calculate the smoothed confidence intervals, for all PQIs and area-level PDIs and IQIs. Finally, an interim release v5.0.2 contained incorrect reference population rates for the PQI module. The updated software (v5.0.3) contains corrected coefficients, signal variance calculations and reference population rates and produces the correct risk-adjusted, expected and smoothed PDI rates and standard errors (used to calculate confidence intervals). Observed PDI, IQI and PQI rates are correct in both the original v5.0 and updated v5.0.3 software; no changes in these rates are expected. All rates for provider-level IQIs (reported by hospital) are correct. All rates for the PSI module are also correct.

  • How will these changes affect my rate? — New!

    Provider-level rates (those reported by hospital) are not impacted by this change. Only area-level rates are impacted, specifically for the PQIs, which are all reported for geographic areas such as county, and area-level PDIs and IQIs. The updated parameters files for the issue correct an overstatement of the signal noise estimates. The initial release resulted in an underweighting of the reference population rate when the smoothed rate is calculated. Users with a limited number of discharges will see their smoothed rate move closer to the reference population rate. The larger the number of discharges, the less likely the user will see an impact. The updated PQI reference population rates will result in higher risk-adjusted PQI rates than were seen before the correction. For example, rates calculated for PQI 13 with the new version of the software will be approximately 10 times higher than what they were before the correction.

  • I used the original v5.0 (March 2015) software that contained discrepancies. What do I need to do? — New!

    If you do not use the PQIs, or area-level PDIs or IQIs, you do not need to do anything. If you have, or plan to, use the PQIs, IQIs or PDIs in the future, you should download and install the updated 5.0.3 software (available for SAS and Windows). See FAQs below for instructions for how to do that

  • I am running v5.0 (March 2015) software. What changes should I expect to see after installing and running the updated v5.0.3 software? — New!

    Users re-running their data using the updated software should expect to see a change in the expected, risk-adjusted, and smoothed rates for PDI14, PDI15, PDI16, PDI18, PDI90, PDI91 and PDI92. They should also expect to see a change in smoothed rates and the smoothed standard error, which is used to calculate the smoothed confidence intervals, for all PQIs and area-level IQIs and PDIs. Finally, they should expect see a change in risk-adjusted PQI rates

  • What is new in the AHRQ QI v5.0 software?

    The AHRQ QI software is updated on an annual basis to reflect changes in the AHRQ QI technical specifications. New software versions and updated technical specifications are released simultaneously. Routine annual updates include yearly International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM), Medicare severity-Diagnostic Related Groups (MS-DRG), Major Diagnostic Categories (MDC), updated version of the 3MTM APR-DRG grouper, new Census population files and newly derived risk adjustment parameters.

    Noteworthy upgrades to the AHRQ QI v5.0 include updates to risk-adjustment models for all of the AHRQ QI using 2012 reference population data, use of more current regression coefficients, and modifications to the technical specifications for 34 IQIs, 17 PQIs, 26 PSIs, and 24 PDIs. The technical specification modifications involve a combination of changes to numerator and denominator specifications, in addition to certain ICD/E-codes and DRG codes being added or removed. The v5.0 update also includes removal of the Prediction Module (PM) that estimates the prevalence of conditions as present on admission. Going forward, the user of the AHRQ QI v5.0 software will simply specify whether or not the input data includes POA data.

    For complete information about all of the changes in the AHRQ QI v5.0, please refer to the software Release Notes on the SAS QI and WinQI software pages.

  • Can my organization use the AHRQ QIs and, if so, what data do I need?

    AHRQ provides the software to any interested organization for use with their own hospital discharge data. The Quality Indicators software is designed to read hospital administrative discharge data that generally conforms to HCUP specifications, in which each hospitalization is reported on a single record.

    • The software expects that the DRG or ICD-9 code on any given discharge record is valid for the fiscal year of the discharge date. The software is designed to be backwards compatible with previous fiscal year versions.
    • The AHRQ QI software is intended to be used with data that cover an entire patient population (e.g., all discharges from a hospital in a year) or that were sampled from a patient population using simple random sample. The software ONLY accepts three common data formats: Text (comma separated values), Microsoft Access®, and Microsoft Excel®. Two key formatting issues are that each row of data represents a separate discharge record and each column of data represents a single variable for all discharges.
  • Can I have multiple versions of the AHRQ QI software installed on my computer?

    No, once you download an updated version of the software, the previous version will be removed.

  • How do I interpret the output of the AHRQ QI software or specific cases?

    Documentation (SAS QI or WinQI) is available to help users with interpretation of software output. Unfortunately the AHRQ Quality Indicators technical support team does not provide support with interpretation of AHRQ QI results, individual cases or advice on how to apply codes to a specific case.

  • Is benchmark information available? How is the AHRQ QI benchmark information created?

    Yes. New benchmark tables have been created for SAS QI v5.0 software.

    The Inpatient Quality Indicator v5.0 Benchmark Data Tables, Pediatric Quality Indicator v5.0 Benchmark Data Tables, Prevention Quality Indicator v5.0 Benchmark Data Tables, and Patient Safety Indicator v5.0 Benchmark Data Tables documents provide benchmark values, including average volumes, provider rates, and population rates across an aggregation of discharges across an aggregation of SID state files, depending on the type of indicator. In addition, the Technical Specifications document for each indicator lists all of the International Classification of Diseases, Clinical Modification (ICD-9-CM) and Medicare severity-Diagnostic.

    Benchmark Data Tables are based on the discharges from an aggregation of the 2012 Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID) from 45 States. The resulting file included more than 30 million discharges from community hospitals that are not designated as rehabilitation or acute long-term care facilities. Users can refer to these tables to determine if their rates approximate the population rate and how their case-mix compares to the population rate.

  • What if I have concerns about the coding approach taken to assign the ICD-9-CM codes?

    The AHRQ QI support line is available to answer your questions about the QI definitions and receive your feedback on those definitions; however, the QI support line does not provide consultation on appropriate use of the codes. If you have questions about individual cases or general coding guidelines please follow either of these approaches:

    1. If you have questions or disagreements about coding of individual cases, a special quarterly publication called Coding Clinic for ICD-9-CM provides interpretation and coding guidance on the proper use of ICD-9-CM. This quarterly resource is published by the American Hospital Association. However, the four Cooperating Parties for ICD-9-CM [i.e., American Hospital Association (AHA), American Health Information Management Association (AHIMA), Centers for Medicare and Medicaid Services (CMS) and National Center for Health Statistics (NCHS)] have final approval of the content of this official publication. The Editorial Advisory Board for Coding Clinic for ICD-9-CM is a voluntary advisory board. Any interested stakeholder may submit a case for coding adjudication to the American Hospital Association's Central Office on ICD-9-CM, following the procedure described here:

      Note: As of Jan, 2014, the AHA Central Office shifted its attention to ICD-10-CM/PCS and no longer accepts nor responds to requests for ICD-9-CM coding advice. You can submit a question about ICD-10-CM/PCS using the online form available at: http://www.codingclinicadvisor.com

    If you have questions or disagreements about general coding guidelines, the CMS and the NCHS are the two departments within the U.S. Federal Government's Department of Health and Human Services (DHHS). They are in charge of the ICD-9-CM Coordination and Maintenance Committee. They provide the official guidelines for coding and reporting using the International Classification of Diseases. These guidelines have been approved by the four organizations that make up the Cooperating Parties for the ICD-9-CM (i.e., AHA, AHIMA, CMS and NCHS). NOTE: Beginning with the September 2011 ICD-9-CM Coordination and Maintenance Committee meeting only proposals for new codes/revisions for a new disease/new technology procedure will be accepted for ICD-9-CM. All other proposals must be submitted as changes to ICD-10-CM/PCS using the required submission process.

  • What is the status of the AHRQ QI conversion to ICD-10-CM/PCS?

    The conversion from ICD-9-CM to ICD-10-CM/PCS requires re-specification of the AHRQ QI and development of new software to compute them. The AHRQ QIs will be updated for ICD-10-CM/PCS on a timeline that is consistent with the Centers for Medicare and Medicaid Services (CMS) implementation. There is currently a report, entitled “ICD-9-CM to ICD-10-CM/PCS Conversion of AHRQ Quality Indicators” available on the QI website, which details some of the considerations that are part of the conversion.

    The ICD-9-CM Coordination and Maintenance Committee implemented a partial freeze of the ICD-9-CM and ICD-10 (ICD-10-CM and ICD-10-PCS) codes prior to the implementation of ICD-10. On April 1, 2014, the Protecting Access to Medicare Act of 2014 (PAMA) (Pub. L. No. 113-93) was enacted, which said that the Secretary may not adopt ICD-10 prior to October 1, 2015. Accordingly, the U.S. Department of Health and Human Services issued a final rule on August 4, 2014 that changed the compliance date for ICD-10 from October 1, 2014 to October 1, 2015. The final rule also requires HIPAA covered entities to continue to use ICD-9-CM through September 30, 2015.

    Meanwhile, AHRQ’s QI software redesign will occur over a number of software releases and will include beta testing. AHRQ has provided free ICD-10 Alpha version of the SAS software for organizations to test data that they have converted to ICD-10, to help assess any impacts that may have occurred due to that data conversion. The Alpha version of the ICD-10 software, released on Feb. 27, 2015, and accompanying information can be found here.

Updated February 1, 2016

[ ]