Frequently asked questions about the CIHI Hospital Frailty Risk Measure

  Back to Frailty among hospitalized seniors

1. What is the CIHI HFRM?

The CIHI Hospital Frailty Risk Measure (HFRM) is an acute care contextual measure developed by CIHI. It is a planning and resourcing tool that uses routinely collected administrative data to characterize the risk of frailty among seniors (age 65 and older) in hospitals across Canada. It can be used to ensure appropriate and targeted care for this acute care population, which has multiple and complex needs, in order to improve outcomes for seniors at risk of frailty. This measure was developed and validated through engagement with national and international clinical experts, and through stakeholders’ feedback. 

The main use of the CIHI HFRM is for overall health system planning and to support research and quality improvement. As a planning and resourcing tool, it can be used to help inform local health service needs (e.g., for hospital services planning, such as models of care for programming and staffing needs), discharge and end-of-life planning, and service planning between hospitals, home care and long-term care organizations. 

Please note that this measure is not a clinical assessment tool for individual patients, nor is it a health indicator to be used to compare performance across jurisdictions or organizations. Please see questions 4 and 7 for more information. 

2. How should the CIHI HFRM be used?

The CIHI HFRM is a good predictor of high use of hospital services and poor outcomes, such as in-hospital mortality, and risk of hospital readmission and institutionalization (e.g., admission to a long-term care facility).

This contextual measure aims to help health system administrators answer questions such as

  • Does my organization have the right infrastructure and resources in place to best serve the frail senior population?
  • What can be done once frail individuals are identified?

The uses of the CIHI HFRM can be generally categorized as follows: 

  • Provide information for service provision and overall health system planning (see question 1)
  • Link to patient-reported outcome measures and other data related to seniors
  • Provide information for commissioning and reimbursement decisions
  • Enable research and quality improvement1 

Key users include health system planners and decision-makers, care delivery managers and health researchers. 

3. Why is it important to measure frailty?

As the proportion of seniors in Canada grows, it is expected that an increasing number of people will become frail. Until now, there has been no standard for measuring and identifying patients at risk of frailty using routinely collected administrative data in acute care settings in Canada. The CIHI HFRM is designed to help jurisdictions understand the extent of frailty in their hospitalized population of seniors and is a good tool for overall health system planning that helps support quality improvement.

  • Complex medical/clinical needs that are common among frail seniors may require different approaches to care.
  • Understanding the extent of frailty in a hospital setting can guide the choice of care and staffing models and assist with determining what resources might be needed by hospitalized seniors at risk of frailty (compared with those who aren’t at risk). 
  • With the number of seniors living with frailty increasing, planning and resources are required to shift the focus to other health care settings beyond the hospital (e.g., home care, long-term care).
  • Quality of life and patient experience can be improved for the sickest patients, who often need frequent hospital services.
  • There are implications for discharge planning, end-of-life care and palliative care services.

4. Can the CIHI HFRM be used as a frailty clinical assessment tool for individual patients? 

No, the CIHI HFRM is not a clinical diagnostic tool for use with individual patients, nor can it be used for individual care planning. The current gold standard for frailty evaluation and management in clinical settings continues to be a comprehensive geriatric assessment.

It is important to note that this measure is solely for use in secondary health care and is not for use in clinical decision-making for individual patient-level care. It is designed to be used at an aggregate level by health system planners and administrators to better understand and plan for the needs of hospitalized seniors who are likely to be living with frailty. If the CIHI HFRM identifies an individual as being at risk of frailty, a direct clinical assessment would be required. 

5. How is frailty measured?

The CIHI HFRM follows a cumulative deficit approach (i.e., accumulation of deficits or frailty conditions is used to determine the individual’s risk of frailty).
There are 3 ways to contextualize frailty using the CIHI HFRM: 

  1. The continuous CIHI HFRM for each patient is calculated by dividing the total number of deficits by 36, which is the maximum number of deficits a patient can accumulate. The resulting value is a score between 0 and 1.
  2. 8 risk groups are used to represent a relative risk of frailty, ranging in severity from lowest (group 1) to highest (group 8), according to the number of deficits accumulated by a given patient. 
  3. The contextual measure Hospitalized Seniors (65+) at Risk of Frailty (%) reports the proportion of patients who accumulated 6 or more deficits (i.e., who belong to risk groups 4 to 8). This 6+ deficit cut-off was chosen as it has been demonstrated to include seniors at higher risk of poor outcomes, including mortality, more hospital readmissions and longer hospital stays.

Results are aggregated at the hospital, regional and provincial/territorial levels. Measures 2 and 3 may be more meaningful and intuitive to understand for health system and hospital administrators, while measure 1 may be of more value in research settings. Health care organizations such as hospitals or health regions can get an overall picture of frailty within their organization using the Hospitalized Seniors (65+) at Risk of Frailty (%) contextual measure, or they may wish to use the 8 risk groups to target a smaller population at more severe risk of frailty within their organization. 

For information on the CIHI HFRM methodology, including the list of deficits, refer to CIHI’s Indicator Library.

6. How sensitive is this measure in detecting patients at risk of frailty?

The CIHI HFRM uses a very comprehensive methodology. The list of frailty-related condition categories (or deficits) and the corresponding International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada (ICD-10-CA) codes were established based on extensive consultation with experts in the field in Canada and the United Kingdom, as well as with CIHI’s Classifications specialists. 

While there are limitations to this measure (see question 7), the CIHI HFRM correlates well with clinical assessments of frailty across various populations. It also shows strong predictive power with frailty-related adverse outcomes such as mortality, hospital readmission, high use of hospital beds and admission to home care and long-term care.2  In particular, the predictive ability for these outcomes was incorporated into the development of the Hospitalized Seniors (65+) at Risk of Frailty (%) contextual measure and the underlying 8 frailty risk groups, which helps to mitigate some of the issues with under-capture of frailty conditions described in question 7. 

These measures provide a good reflection of the risk of frailty at the population level. For further information on the frailty conditions included in the CIHI HFRM and the development of the measure, refer to CIHI’s Indicator Library.

7. What are the limitations of this measure?

We suspect that many of the conditions of interest are not captured well in the administrative data on which the CIHI HFRM is based. 

  • Not all frailty-related conditions included in the CIHI HFRM are mandatory to code. There are jurisdictional differences in coding practices.
  • There is partial coverage of some administrative data sources used to identify frailty-related conditions. 
  • Certain conditions, such as functional and cognitive deficits, are not well documented with existing coding practices. 

As a result, the calculated number of deficits for each patient may be underestimated. Additionally, differences across jurisdictions in coding practices and data coverage lead to challenges in comparability across Canada. For this reason, the CIHI HFRM is a contextual measure and not an indicator for benchmarking performance. Direct comparisons between jurisdictions and organizations are discouraged unless data is known to be comparable. Where coding practices have been stable over time, users can monitor trends within their frail populations. 

The current measure provides a baseline to improve clinical documentation and to raise awareness on the topic of frailty in acute care and other clinical and community settings. To increase the capture of the frailty conditions in this measure, organizations and jurisdictions may mandate the capture of frailty related diagnoses. This could include abstracting data from relevant nursing notes, using non-mandatory codes — type (3) diagnoses — for the conditions included in the CIHI HFRM and continuing their efforts to improve the quality of physician documentation.

8. Can the CIHI HFRM be used in risk adjustment?

Where data is known to be comparable, the CIHI HFRM can be used as a variable when risk-adjusting health indicators. Since individuals living with frailty are more likely to experience poor health outcomes, this characteristic can be carefully considered and adjusted for in analytical products involving seniors. 

The CIHI HFRM can be used as a risk factor along with others such as sex, age and the Charlson Comorbidity Index in modelling a range of adverse outcomes. Adjusting for the CIHI HFRM will ensure comparability of performance indicators involving hospitalized seniors.

Due to differences in data coverage and coding practices, considerations should include the comparability of data between organizations within the analytical cohort. Data within a province/territory or health region may be comparable, and in this case could be used for risk adjustment. 
 

References

If you have a disability and would like CIHI information in a different format, visit our Accessibility page.