Overview of the ICD-11 Quality and Safety TAG AHRQ 2012 Annual Conference

1. Overview of the ICD-11Quality and Safety TAGAHRQ 2012 Annual Conference Patrick S. Romano, MD MPH UC Davis School of Medicine On behalf of William Ghali, MD, MPH (U. of Calgary) Harold Pincus, MD (Columbia U.)

2. Topic Advisory Groups • “Organize the work for the revision of ICD in a specified domain based on procedural, terminologic and taxonomic guidelines of the revision process” • “Advise WHO in all steps leading to the revision of the topic sections of ICD” • Most TAGs work “vertically” in relation to existing ICD-10 chapters (e.g., mental health, dermatology)

3. Safety and Quality • A horizontal theme that applies across all chapters (not just in the external causes/injury domain) • An ICD-11 ‘use case’ (analogous to coding mortality to track causes of death, coding morbidity for payment) • Measurement using ICD supports surveillance, trending, benchmarking, research, quality improvement

4. WHO Revision Steering Group ICD Revision Organizational Structure Gastroenterology WG Cardiovascular WG Health Informatics and Modelling TAG (HIM TAG) Internal Medicine TAG Hepatology & Pancreatobiliary WG Paediatrics TAG iCAT Software Team Nephrology WG Dentistry TAG Dermatology TAG Endocrinology WG Rare Diseases TAG Rheumatology WG External Causes and Injuries TAG Morbidity TAG Haematology WG GURM Respiratory WG Mortality TAG Mental Health TAG Working Groups Musculoskeletal TAG Functioning TAG Neoplasms TAG Quality & Safety TAG Genito-Urinary, Reproductive and Maternal TAG Neurology TAG Ophthalmology TAG Cross-Sectional TAGs Traditional Medicine TAG

5. The need for a Q&S TAG • ICD-10 challenged the safety/quality use case (b/c existing tools were for ICD-9) • Iezzoni’s Complication Screening Program • AHRQ Quality Indicators/OECD • Other classification initiatives • NQF Serious Reportable Events • International Classification for Patient Safety • AHRQ Common Formats • Best to inform ICD-11 writing through a safety/quality lens, building on ICD-9 and 10

6. TAG Work Plan • Co-chaired by William Ghali (U. of Calgary) and Harold Pincus (Columbia U.) • Membership includes experts from N. America, Europe, Asia, Australia • Funding from CIHI and AHRQ Conference Grant program secured • Face-to-face meetings: • June 2010 • Feb 24-25 and Nov 15-16, 2011 • April 3-4 and Nov 7-8, 2012

7. TAG Work Plan • Focus of work in 5 general areas: • Develop an inventory of existing Q&S indicators, and consider potential new indicators, to ensure applicability in ICD-11 • Review concepts of ICPS for applicability to ICD-11 • Restructure the coding scheme for health care related injuries in chapters 19/20 • Review and refine volume 2 coding rules for ICD-11 • Review and critique ICD-11 alpha/beta drafts • Develop strategy for field testing • Collaboration with External Causes and Injury TAG, Morbidity TAG, Revision Steering Group

8. Specific recommendations to date • Main condition coding • “Reason for admission” versus “resource consumption” • Former approach more consistent with Q&S use case • Latter approach requires complex decision logic that may not have sufficient inter-observer reliability, compromising comparability across providers and countries • Diagnosis timing indicator • Crucial to flag complications of hospital care • Simpler classification scheme (2 or 3 categories) favored • Suffix/prefix modifier within code (related to concepts of “personal history of,” “family history of,” etc.)

9. Specific recommendations to date • ICPS (not fully developed) • Ideally, all of the ‘incident type’ categories from ICPS should map to codes in ICD-11 (what about “near misses”?) • This requires a “virtual chapter” structure • Number of diagnosis fields must be sufficient to support reporting of all significant complications that arise during an episode of care, in addition to significant risk factors • Empirical work using data from OECD

10. Specific recommendations to date • Clustering would allow post-coordination of diagnostic concepts that cannot be clearly linked in ICD-10 (or are linked using combination codes) • The cause of injury (e.g., a substance/drug, a procedure/intervention, or a device) • The mechanism or mode of injury (i.e., the mechanism by which the factor caused injury, as currently defined in Chapter 20) • The actual injury or condition that resulted (i.e., anatomic or physiologic abnormalities, as currently defined in Chapters 1-19) • Could accommodate “place,” “activity at time of injury”

11. Example problem • ICD-9/CM: 998.2, “Accidental puncture or laceration during a procedure” • ICD-10: T81.2, as above “NEC” • ICD-10-CM: Many chapter-specific codes, e.g., I97.5, “Accidental puncture and laceration of a circulatory system organ or structure during a procedure” • I97.51, “…during a circulatory system procedure” • I97.52, “… during other procedure”

12. Example of clustering in ICD-11 • Concept 1: Injury or harm • Injury of heart with haemopericardium (S26.0) • Concept 2: Mechanism or mode of injury or harm • Mechanical cut or puncture • Concept 3: Cause of injury or harm • …during endovascular surgical procedure • Suffix for diagnosis timing • 7th digit clustering flag on all related diagnoses

13. Other considerations • “Complication of other aspect of care” (e.g., physiotherapy) • “Observation for suspected condition” (with no actual harm) • Rationalization of substance-related concepts in a single chapter • Mortality use case • Backward compatibility with ICD-10 (and country-specific versions)

15. Methodological work • Overview (Ghali et al.) • Defining main condition or “principal diagnosis” for international comparability (Quan et al.) • Number of diagnosis fields (Drősler et al.) • Defining diagnosis timing (“present on admission” for international comparability (Sundararajan et al.)

16. Quality and Safety TAG • An exciting mandate – May 2015! • Real potential for ICD enhancement for quality measurement and data extraction • Dialogue/interaction with all other TAGs • Develop international consensus on data standards and methods • Harmonize data definitions • Linkages with WHO, OECD, AHRQ, CIHI

17. Questions and discussion • Please e-mail us: • wghali@ucalgary.ca • pincush@nyspi.columbia.edu • psromano@ucdavis.edu • Questions for your consideration: • How could ICD be improved in ICD-11 to facilitate measurement of quality and safety? • How can we make ICD easier to use but also more useful for research and QI?