Hospital Harm
A lower rate for this indicator is desirable.
This indicator is expressed as the number of hospital discharges with at least one occurrence of harm per 100 discharges.
Unit of analysis: Hospital discharge
5 patient groups are hierarchically defined in the following order:
- Newborn patient group: Entry_code = N
- Obstetric patient group: Major clinical category (MCC) 13
- Pediatric patient group: Age younger than 18 years
- Surgical patient group: MCC partition code = I (intervention)
- Medical patient group: MCC partition code = D (diagnosis)
For further information on the methodology, please refer to the Hospital Harm Indicator General Methodology Notes (PDF).
Number of discharges from an acute care institution in a fiscal year
A subset of the denominator: discharges with at least one occurrence of harm identified during the hospital stay
Methodology
Name
Hospital Harm
Short/Other Names
Not applicable
Description
Hospital harm captured by this indicator is defined as the rate of acute care hospitalizations with at least one occurrence of unintended harm during a hospital stay that could have been potentially prevented by implementing known evidence-informed practices. This includes many types of harm at a system level (making it a big dot indicator). It also classifies harm into actionable clinical groups; therefore, improvement efforts in patient safety can be tracked for the overall measure and for each specific clinical group.
While not all instances of harm captured by this indicator can be prevented, adoption of evidence-informed practices can help to reduce the rate of harm.
Harm is captured only when it
- Is identified as having occurred after admission and within the same hospital stay;
- Requires treatment, alters treatment or prolongs the hospital stay; and
- Is one of the conditions from the 31 clinical groups in the Hospital Harm Framework (refer to the Hospital Harm Indicator General Methodology Notes (PDF).
The following are not captured:
- Near misses or incidents that did not reach the patient; and
- Reportable incidents or events that reached the patient and could potentially have caused harm or injury but did not.
For further details, please see the document Hospital Harm Indicator: Frequently Asked Questions (PDF).
Rationale
Patients expect hospital care to be safe, and for most people it is. Despite health professionals' focus on safety, a small proportion of patients experience some type of unintended harm as a result of the care they receive. Concern over patient safety during their hospital stay has grown steadily over the past decade, but understanding of strategies that prevent harm to patients has improved over this period, too.
Tracking and reporting harmful events is a vital first step to investigating, monitoring and understanding patient safety improvement efforts. Historically, reporting has been mostly voluntary and focused on particular risks such as infections. Until now, there has been no single measure that provides a broad perspective on patient safety in Canadian hospitals.
This indicator aims to provide a single estimate of the overall rate of hospital harm and to allow for tracking and monitoring of the rate over time.
Interpretation
A lower rate for this indicator is desirable.
HSP Framework Dimension
Health system outputs: Safe
Areas of Need
Getting Better
Targets/Benchmarks
Not applicable
Available Data Years
to (fiscal years)
Geographic Coverage
- All provinces/territories except Quebec
Reporting Level/Disaggregation
- National
Indicator Results
Update Frequency
Every year
Latest Results Update Date
Updates
Not applicable
Description
This indicator is expressed as the number of hospital discharges with at least one occurrence of harm per 100 discharges.
Unit of analysis: Hospital discharge
5 patient groups are hierarchically defined in the following order:
- Newborn patient group: Entry_code = N
- Obstetric patient group: Major clinical category (MCC) 13
- Pediatric patient group: Age younger than 18 years
- Surgical patient group: MCC partition code = I (intervention)
- Medical patient group: MCC partition code = D (diagnosis)
For further information on the methodology, please refer to the Hospital Harm Indicator General Methodology Notes (PDF).
Type of Measurement
Rate - Rate — per 100
Denominator
Description:
Number of discharges from an acute care institution in a fiscal year
Inclusions:
- Sex recorded as male or female
Exclusions:
- Discharges from Quebec acute care institutions
- Records with admission category of stillbirths and cadaveric donors (Admission Category Code = R or S)
- Discharges with invalid age
- Discharges with invalid admission or discharge dates
- Discharges with selected mental health diagnoses (i.e., most responsible diagnosis ICD-10-CA codes F10–F99). In Ontario, mental health discharges are submitted to the Ontario Mental Health Reporting System (OMHRS) and are therefore not in the Discharge Abstract Database (DAD). In order to create a standard hospital population, discharges with mental health disorders (with the exception of organic mental health disorders — ICD-10-CA codes F00–F09) were excluded from all provinces.
- 2018–2019 data onward: Medical assistance in dying (MAID) (Discharge Disposition Code = 73)
Numerator
Description:
A subset of the denominator: discharges with at least one occurrence of harm identified during the hospital stay
Inclusions:
Harm is identified based on the International Statistical Classification of Diseases and Related Health Problems (ICD 10-CA)/Canadian Classification of Health Interventions (CCI) and the Canadian Coding Standards and is classified into 31 clinical groups under 4 categories of harm.
For detailed descriptions of the inclusions for clinical groups and categories of harm, please refer to the Hospital Harm Indicator General Methodology Notes (PDF).
Exclusions:
For detailed descriptions of the exclusions for clinical groups and categories of harm, please refer to the Hospital Harm Indicator General Methodology Notes (PDF).
Method of Adjustment
None
Adjustment Applied
Covariates used in risk adjustment:
None. At this time, only crude rate results are provided.
Geographic Assignment
Place of service
Data Sources
- DAD
Caveats and Limitations
The Hospital Harm indicator has the following limitations that may affect interpretation of results and comparison across organizations:
- Differences in processes, documentation and resources across hospitals may result in differences in their ability to capture data about harmful events, so hospitals with better documentation may have higher rates.
- The results are not risk-adjusted. Hospitals serve different patient populations and it is important to take this into account when comparing health system performance. Given the wide range of harmful events captured by this indicator, a risk-adjustment methodology to account for these differences was not developed.
- All occurrences of harm are considered to be of the same weight in terms of contribution to a hospital's overall rate, regardless of severity.
Trending Issues
In October 2016, CIHI released Measuring Patient Harm in Canadian Hospitals (PDF), a national-level report on hospital harm. Since the report was released, there have been changes to the methodology used.
References
Baker GR, Norton PG, Flintoft V, et al. The Canadian Adverse Events Study: The incidence of adverse events among hospital patients in Canada. Canadian Medical Association Journal. May 2004.
Canadian Institute for Health Information. Canadian Classification of Health Interventions (CCI). 2015.
Canadian Institute for Health Information. Canadian Coding Standards for ICD-10-CA and CCI, 2015. 2015.
Canadian Institute for Health Information. International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Canada (ICD-10-CA), 2015. 2015.
Canadian Institute for Health Information. Measuring Patient Harm in Canadian Hospitals. 2016.
Classen DC, Resar R, Griffin F, et al. Global Trigger Tool shows that adverse events in hospitals may be ten times greater than previously measured. Health Affairs. April 2011.
Hodgkinson MR, Dirnbauer NJ, Larmour I. Identification of adverse drug reactions using the ICD-10 Australian Modification clinical coding surveillance. Journal of Pharmacy Practice and Research. March 2009.
Jackson T, Duckett S, Shepheard J, Baxter K. Measurement of adverse events using "incidence flagged" diagnosis codes. Journal of Health Services Research & Policy. January 2006.
Layde PM, Meurer LN, Guse C, et al. Medical injury identification using hospital discharge data. In: Henriksen K, Battles JB, Marks ES, Lewin DI, eds. Advances in Patient Safety: From Research to Implementation (Volume 2: Concepts and Methodology). 2005.
Perla RJ, Hohmann SF, Annis K. Whole-patient measure of safety: Using administrative data to assess the probability of highly undesirable events during hospitalization. Journal for Healthcare Quality. September-October 2013.
U.S. Department of Health and Human Services, Agency for Healthcare Research and Quality. AHRQ Quality Indicators — Patient Safety Indicators: Software Documentation, Version 3.1 (PDF). 2007.
Zhan C, Miller MR. Administrative data based patient safety research: A critical review. Quality & Safety in Health Care. December 2003.
How to cite:
Canadian Institute for Health Information. Hospital Harm. Accessed April 25, 2025.
If you would like CIHI information in a different format, visit our Accessibility page.
Comments
Indicator results are also available in