To ensure the continued relevance of its Health System Performance indicators, the Canadian Institute for Health Information (CIHI) regularly reviews indicator methodologies and updates them as required. Methodology updates are made every 3 years to coincide with changes to standards in classification systems (ICD-10-CA and CCI), clinical administrative databases (DAD and NACRS) and case-mix methodologies.

Results using the updated methodologies will be available in the fall 2019 releases of the Your Health System (YHS) web tool and the Health Indicators e-Publication.

In the upcoming releases, the methodologies used for the following indicators will be updated (please refer to Summary of methodology changes below for the full list of changes):

  • 5 Readmission indicators
  • Hip Fracture Surgery Within 48 Hours
  • Hospital Deaths (HSMR)
  • Hospital Deaths Following Major Surgery
  • In-Hospital Sepsis
  • Obstetric Trauma

Indicator results for 5 years calculated using the updated methodologies will be publicly available in YHS and the Health Indicators e-Publication on November 28. Prior to the public release, results will be available in YHS: Insight (CIHI’s private eReporting tool) on September 19. Technical notes reflecting these methodology changes will be available in the Indicator Library on November 1.

Methodology changes made to selected indicators

  1. All indicators listed above
    • Updated risk-adjustment models: Regression coefficients used for risk-adjustment were updated using combined data from 3 fiscal years (2015–2016, 2016–2017 and 2017–2018).
    • Updates to Charlson Index: Diagnosis codes and diagnosis types used to derive the Charlson Index were updated to reflect the most recent Canadian Coding Standards. The updated algorithm is available upon request.
    • New methodology for trend assessment: Trend-over-time assessment in YHS: In Depth is now determined using logistic regression instead of a series of z-tests. This is to improve the sensitivity of trend tests in identifying improvement/weakening in performance.
    • Exclusion of MAID: Records with the discharge disposition medical assistance in dying (MAID) were excluded to comply with legislation in certain jurisdictions.
  2. Readmission indicators
    • Age cut-off: The age cut-off for 3 readmission indicators was changed to align with the definition of the pediatric patient population:
      • The age cut-off for the Patients 19 and Younger Readmitted to Hospital indicator was changed to age 17 and younger; the indicator name was changed to Pediatric Patients Readmitted to Hospital.
      • The age cut-off for the medical and surgical readmission indicators was changed to 18 and older.
    • Top 80 CMG list: The list of case-mix groups (CMGs) included in the risk-adjustment models (where applicable) was updated using combined data from 3 fiscal years (2015–2016, 2016–2017 and 2017–2018).
    • Please refer to the Summary of methodology changes below for the full list of changes.
  3. Hospital Deaths (HSMR)
    • Performance assessment methodology: Hospital-level results are now compared with the peer group average to align with other clinical indicators.
    • Top 80 diagnosis list: The list of diagnosis groups that account for 80% of in-hospital deaths was updated using combined data from 3 fiscal years (2015–2016, 2016–2017 and 2017–2018).
    • Exclusions removed:
      • Exclusion of records with a most responsible diagnosis of brain death was removed due to changes in the Canadian Coding Standards.
      • Exclusion of patients who sign out against medical advice or who do not return from a pass was removed in response to stakeholder feedback.
  4. Hospital Deaths Following Major Surgery
    • Age cut-off: The age cut-off was changed to 18 and older to align with the definition of the pediatric patient population.
    • Top 80 CMG list: The list of CMGs included in the risk-adjustment model was updated using combined data from 3 fiscal years (2015–2016, 2016–2017 and 2017–2018).
  5. Obstetric Trauma
  6. Other methodology changes
    • Peer group assignment: Hospital peer group assignments in YHS were updated using combined data from 3 fiscal years (2015–2016, 2016–2017 and 2017–2018).

i. International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada, and Canadian Classification of Health Interventions.

ii. Discharge Abstract Database and National Ambulatory Care Reporting System.

 

Summary of methodology changes

Indicator name Case selection Risk factors

All Patients Readmitted to Hospital 

  • Include sign-out against medical advice and not return from pass (previously excluded)
  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Chemotherapy exclusion: Use diagnosis code Z51.1 with diagnosis type M, 1, C, W, X or Y (only type M was used previously)
  • Updated coefficients
  • Updated top 80 CMG list

Pediatric Patients Readmitted to Hospital (previously named “Patients 19 and Younger Readmitted to Hospital”)

  • Age cut-off: 17 and younger (previously 19 and younger)
  • Include sign-out against medical advice and not return from pass (previously excluded)
  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Chemotherapy exclusion: Use diagnosis code Z51.1 with diagnosis type M, 1, C, W, X or Y (only type M was used previously)
  • Updated coefficients
  • Updated top 80 CMG list

Medical Patients Readmitted to Hospital

  • Age cut-off: 18 and older (previously 20 and older)
  • Include sign-out against medical advice and not return from pass (previously excluded)
  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Chemotherapy exclusion: Use diagnosis code Z51.1 with diagnosis type M, 1, C, W, X or Y (only type M was used previously)
  • Updated coefficients
  • Updated top 80 CMG list
  • Updated Charlson Index
  • Reference age group: 18 to 44 (previously 20 to 44)

Surgical Patients Readmitted to Hospital 

  • Age cut-off: 18 and older (previously 20 and older)
  • Include sign-out against medical advice and not return from pass (previously excluded)
  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Chemotherapy exclusion: Use diagnosis code Z51.1 with diagnosis type M, 1, C, W, X or Y (only type M was used previously)
  • Updated coefficients
  • Updated top 80 CMG list
  • Updated Charlson Index
  • Reference age group: 18 to 44 (previously 20 to 44)

Obstetric Patients Readmitted to Hospital 

  • Include sign-out against medical advice and not return from pass (previously excluded)
  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Chemotherapy exclusion: Use diagnosis code Z51.1 with diagnosis type M, 1, C, W, X or Y (only type M was used previously)
  • Updated coefficients

Hip Fracture Surgery Within 48 Hours

  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Updated coefficients

Hospital Deaths (HSMR) 

  • Updated top 80 most responsible diagnosis list:
    • Add 3 diagnosis groups: A49, I33, J10
    • Remove 3 diagnosis groups: C85, E86, I24
  • Remove brain death exclusion
  • Include sign-out against medical advice and not return from pass (previously excluded)
  • Updated coefficients
  • Updated Charlson Index

Hospital Deaths Following Major Surgery

  • Age cut-off: 18 and older (previously 20 and older)
  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Updated coefficients
  • Updated top 80 CMG list
  • Updated Charlson Index

In-Hospital Sepsis 

  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • Updated coefficients
  • Updated Charlson Index
  • New ICD-10-CA code for immunocompromised state: B48.5 

Obstetric Trauma 

  • Exclude MAID (discharge_disposition = 73) from 2018–2019 data onward
  • New ICD-10-CA codes for third-degree tear: 

O70.211
O70.221
O70.231
O70.281
O70.291

  • Updated coefficients

Supporting methodology documents

  1. Indicator Library: General Methodology Notes — Clinical Indicators, November 2019 (PDF)
  2. Model Specifications — Clinical Indicators, November 2019 (PDF)