ºìÁì½í¹Ï±¨Hospital Frailty Risk Measure (HFRM)
The ºìÁì½í¹Ï±¨HFRM is a contextual measure, not a health indicator intended to benchmark performance.
The higher the value of the continuous ºìÁì½í¹Ï±¨HFRM, the more deficits a patient had accumulated and that were used to determine their risk of frailty.
The 8 risk groups represent frailty risk in terms of risk severity from lowest (group 1) to highest (group 8). The higher the number of the risk group, the more deficits that patient accumulated. Risk groups 4 to 8 are used to identify Hospitalized Seniors (65+) at Risk of Frailty (%).
As higher levels of frailty are linked to a number of adverse outcomes among seniors, when frailty severity increases, so does the risk for these adverse outcomes. Please refer to the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF) for details.
For more information on frailty, please refer to the Frailty web page.
Unit of Analysis: Patient
The ºìÁì½í¹Ï±¨HFRM follows a cumulative deficit approach (i.e., an accumulation of deficits is used to determine the individual’s risk of frailty).
The list of frailty deficits used for the ºìÁì½í¹Ï±¨HFRM includes 36 frailty condition categories, each of which corresponds to diagnosis codes from the ICD-10-CA. The list covers frailty-related deficits such as morbidity, function, sensory loss, cognition and mood.
An individual patient’s risk of frailty is calculated over a period of 2 years by counting the number of deficits, or frailty condition categories, looking back 2 years from their index discharge date, which is the most recent acute care inpatient discharge in the reporting year.
For more detailed information on how the condition categories were identified or how patient records were linked, please see the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF).
Results are presented in 3 different ways:
1) As a continuous ºìÁì½í¹Ï±¨HFRM score
To calculate the continuous score per patient, the total number of deficits for each patient is divided by 36, which is the maximum number of deficits a patient can theoretically accumulate. The resulting value is a number in a continuous range between 0 and 1.
Grouping of the ºìÁì½í¹Ï±¨HFRM into meaningful risk categories provides a better description of the seniors patient population, as opposed to a single continuous measure that ranges between 0 and 1.
Therefore, the following are also presented:
2) 8 risk groups
Patients are grouped into 8 categorical risk groups according to their total number of deficits, ranging in severity from lowest (group 1) to highest risk (group 8). The 8 risk groups are presented as a percentage of patients in each of the 8 groups.
3) Hospitalized Seniors (65+) at Risk of Frailty (%)
As a single measure of frailty risk, this is the proportion of patients in risk groups 4 to 8 divided by all patients included in the frailty cohort. This is the measure reported in the Your ºìÁì½í¹Ï±¨ System: In Depth tool.
Results are reported at the hospital, region and province/territory levels. Patients are included in the results for each hospital from which they were discharged. (Note: Patients are included only once in results for each hospital, region and province/territory.)
For results at the hospital level, place of service is used; for region and province/territory results, the patient’s place of residence is used.
For more information on the ºìÁì½í¹Ï±¨HFRM calculation, as well as on how patients were assigned to risk groups and flagged as being at risk of frailty, please refer to the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF).
Patients age 65 and older discharged from an acute care hospital
See Calculation: Description.
Methodology
Name
ºìÁì½í¹Ï±¨Hospital Frailty Risk Measure (HFRM)
Short/Other Names
Hospitalized Seniors (65+) at Risk of Frailty (%)
Description
The ºìÁì½í¹Ï±¨Hospital Frailty Risk Measure (HFRM) measures the risk of frailty among seniors (age 65 and older) in acute care.
The ºìÁì½í¹Ï±¨HFRM is an acute care contextual measure, not a health indicator intended to benchmark performance.
For more information on frailty, please refer to the Frailty web page, the ºìÁì½í¹Ï±¨HFRM FAQ and the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF)
Rationale
As the proportion of seniors (65+) in Canada grows, it is expected that an increasing number of people will become frail. Individuals living with frailty have an increased risk of hospitalization, hospital readmission, emergency department (ED) visits, requiring home care visits or transfers to a long-term care home, long hospital stays and in-hospital death. The risk of individual mortality may be better predicted by frailty than by chronological age.
Developing an assessment methodology to identify patients at risk helps to ensure an appropriate care continuum and leads to improved measurement and assessment of health system performance, targeted care and better allocation of resources for seniors (65+).
Frailty indices and scales for acute care have been created using routinely collected data in both the United States and United Kingdom, including for the assessment of quality care and for planning of services. ºìÁì½í¹Ï±¨has developed this measure, using routinely collected administrative data, as a standard for measuring and identifying patients age 65 and older at risk of frailty in acute care settings in Canada.
For questions about the ºìÁì½í¹Ï±¨HFRM and its uses, please refer to the Frailty web page, the ºìÁì½í¹Ï±¨HFRM FAQ and the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF)
Interpretation
The ºìÁì½í¹Ï±¨HFRM is a contextual measure, not a health indicator intended to benchmark performance.
The higher the value of the continuous ºìÁì½í¹Ï±¨HFRM, the more deficits a patient had accumulated and that were used to determine their risk of frailty.
The 8 risk groups represent frailty risk in terms of risk severity from lowest (group 1) to highest (group 8). The higher the number of the risk group, the more deficits that patient accumulated. Risk groups 4 to 8 are used to identify Hospitalized Seniors (65+) at Risk of Frailty (%).
As higher levels of frailty are linked to a number of adverse outcomes among seniors, when frailty severity increases, so does the risk for these adverse outcomes. Please refer to the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF) for details.
For more information on frailty, please refer to the Frailty web page.
HSP Framework Dimension
ºìÁì½í¹Ï±¨ system inputs and characteristics: Efficient allocation of resources
Areas of Need
Not applicable
Targets/Benchmarks
Not applicable
Available Data Years
to (fiscal years)
Geographic Coverage
- All provinces/territories
Reporting Level/Disaggregation
- National
- Province/Territory
- Region
- Facility
Indicator Results
Web tool:
Update Frequency
Every year
Latest Results Update Date
Description
Unit of Analysis: Patient
The ºìÁì½í¹Ï±¨HFRM follows a cumulative deficit approach (i.e., an accumulation of deficits is used to determine the individual’s risk of frailty).
The list of frailty deficits used for the ºìÁì½í¹Ï±¨HFRM includes 36 frailty condition categories, each of which corresponds to diagnosis codes from the ICD-10-CA. The list covers frailty-related deficits such as morbidity, function, sensory loss, cognition and mood.
An individual patient’s risk of frailty is calculated over a period of 2 years by counting the number of deficits, or frailty condition categories, looking back 2 years from their index discharge date, which is the most recent acute care inpatient discharge in the reporting year.
For more detailed information on how the condition categories were identified or how patient records were linked, please see the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF).
Results are presented in 3 different ways:
1) As a continuous ºìÁì½í¹Ï±¨HFRM score
To calculate the continuous score per patient, the total number of deficits for each patient is divided by 36, which is the maximum number of deficits a patient can theoretically accumulate. The resulting value is a number in a continuous range between 0 and 1.
Grouping of the ºìÁì½í¹Ï±¨HFRM into meaningful risk categories provides a better description of the seniors patient population, as opposed to a single continuous measure that ranges between 0 and 1.
Therefore, the following are also presented:
2) 8 risk groups
Patients are grouped into 8 categorical risk groups according to their total number of deficits, ranging in severity from lowest (group 1) to highest risk (group 8). The 8 risk groups are presented as a percentage of patients in each of the 8 groups.
3) Hospitalized Seniors (65+) at Risk of Frailty (%)
As a single measure of frailty risk, this is the proportion of patients in risk groups 4 to 8 divided by all patients included in the frailty cohort. This is the measure reported in the Your ºìÁì½í¹Ï±¨ System: In Depth tool.
Results are reported at the hospital, region and province/territory levels. Patients are included in the results for each hospital from which they were discharged. (Note: Patients are included only once in results for each hospital, region and province/territory.)
For results at the hospital level, place of service is used; for region and province/territory results, the patient’s place of residence is used.
For more information on the ºìÁì½í¹Ï±¨HFRM calculation, as well as on how patients were assigned to risk groups and flagged as being at risk of frailty, please refer to the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF).
Type of Measurement
Percentage or proportion
Denominator
Description:
Patients age 65 and older discharged from an acute care hospital
Inclusions:
- All inpatient care discharges (Facility Type Code = 1)
- Age at index discharge 65 years and older
Exclusions:
- Records with an invalid health card number (HCN)
- Records with an invalid code for province/territory issuing HCN
- Records with an invalid discharge date
- Records with admission category of cadaveric donor or stillbirth (Admission Category Code = R or S)
- Records with delivery (ICD-10-CA: O10–O16, O21–O29, O30–O37, O40–O46, O48, O60–O69, O70–O75, O85–O89, O90–O92, O95, O98, O99 with a sixth digit of 1 or 2; or Z37 recorded in any diagnosis field) or abortion (ICD-10-CA: O04)
- Records with medical assistance in dying (MAID) (Discharge Disposition Code = 73)
The above applies only to the identification of the reporting cohort for a given fiscal year; it does not apply to other data sources used to extract additional medical records through linkage for identification of frailty conditions.
For more information on inclusion/exclusion criteria, please refer to the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF).
Numerator
Description:
See Calculation: Description.
Inclusions:
The list of frailty deficits used for the ºìÁì½í¹Ï±¨HFRM includes 36 frailty condition categories, each of which corresponds to diagnosis codes from the ICD-10-CA.
For the full list of condition categories, corresponding ICD-10-CA codes and descriptions, please refer to the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF).
All diagnoses and conditions (i.e., all diagnosis type codes, not just the most responsible diagnosis) that are present on a patient’s record, regardless of record type, are included in flagging the frailty conditions.
Exclusions:
Not applicable
Geographic Assignment
Place of residence or service
Data Sources
- DAD
- HMDB
- NACRS
Caveats and Limitations
The differences in processes, documentation and resources across hospitals may result in differences in their ability to capture data about frailty conditions, so hospitals with better documentation may have higher frailty scores.
There are jurisdictional differences in the coverage of data sources used in the identification of frailty conditions, and the patient linkage standard cannot be applied across all jurisdictions consistently.
Direct comparisons between organizations or provinces/territories are discouraged (i.e., it is not recommended to compare ºìÁì½í¹Ï±¨HFRM results across different facilities and different jurisdictions unless data is known to be comparable).
The type and number of deficits that can be identified using health administrative data is somewhat limited.
When a certain frailty deficit was not triggered, it is unclear whether it was looked for but not present in a patient or was not recorded because it was either overlooked or undocumented.
The quality of the underlying clinical data can affect the results.
The conditions covered by the ºìÁì½í¹Ï±¨HFRM focus more on diseases and less on functional outcomes, as functional and cognitive deficits are not well documented with existing coding practices.
For details on caveats and limitations, please refer to the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF).
Trending Issues
Trending may be impacted by changes in coding practices within jurisdictions over time.
References
Muscedere J. . The Journal of Frailty and Aging. 2020.
Rockwood K, et al. . Canadian Medical Association Journal. 2005.
Kim DH, et al. . The Journals of Gerontology: Series A. 2017.
Gilbert T, et al. . The Lancet. 2018.
Soong J, et al. . BMJ Open. 2015.
Clegg A, et al. . The Lancet. 2013.
Mitnitski AB, et al. . Mechanisms of Ageing and Development. 2002.
Canadian Institute for ºìÁì½í¹Ï±¨ Information. ºìÁì½í¹Ï±¨Hospital Frailty Risk Measure (HFRM): November 2023 — Methodology Notes (PDF). 2023.
How to cite:
Canadian Institute for ºìÁì½í¹Ï±¨ Information. ºìÁì½í¹Ï±¨Hospital Frailty Risk Measure (HFRM). Accessed April 6, 2025.
If you would like ºìÁì½í¹Ï±¨information in a different format, visit our Accessibility page.
Comments
More information on the ºìÁì½í¹Ï±¨HFRM is available on the Frailty web page, the ºìÁì½í¹Ï±¨HFRM FAQ and the ºìÁì½í¹Ï±¨HFRM methodology notes (PDF)