Demonstrating linked PREMs, PROMs, clinical and costing data
June 18, 2024 — This report highlights the feasibility of linking health care data that is routinely collected from multiple sources at ºìÁì½í¹Ï±¨to facilitate the assessment of overall health system performance.
Bringing together measures from different sources of data can provide insight into how to achieve the best outcomes for patients at lower costs. This includes the correlations between linked patient-reported data, clinical outcomes (measured by need for readmission and/or revision surgery, and/or hospital harm incurred) and costs.
ºìÁì½í¹Ï±¨
It is important to collect patient-reported data to leverage the potential that this type of information can provide when linked with other health care data. This will ultimately improve patient-centred care and the patient experience. — Dr. Jason Werle, Orthopedic surgeon, Alberta
This analysis linked
- Patient-reported experience measures (PREMs) and clinical outcomes to see whether there is a relationship between patients’ hospital experience and positive clinical outcomes
- PREMs and patient-reported outcome measures (PROMs) to see whether patient experiences and patient-reported outcomes are related (i.e., whether good experiences in hospital result in a positive outcome)
- Costs and PROMs to see whether higher costs are correlated with better patient-reported outcomes
Correlations were explored to determine whether there is a relationship between the 2 measures of interest and the strength of this relationship. The top correlations from the analysis are presented in Table 1. A test of significance was used to determine whether the correlations found were statistically meaningful and not due to chance alone.
Table 1 Top correlations between clinical outcomes, PREMs, PROMs and costs, 2017–2018 to 2021–2022
PREMs | Clinical outcomes | Correlation |
---|---|---|
Communication With Doctors | Hospital Harm | -0.22* |
Involvement in Decision-Making and Treatment Options | 1-Year Revision | -0.20 |
Overall Hospital Experience | 30-Day Readmission | -0.19 |
Communication With Nurses | 30-Day Readmission | -0.14 |
Communication With Doctors | 30-Day Readmission | -0.13 |
PREMs | PROMs | Correlation |
---|---|---|
Overall Hospital Experience | 1-Year Satisfaction | 0.32* |
Communication With Nurses | 1-Year Satisfaction | 0.26* |
Communication With Doctors | 1-Year Satisfaction | 0.26* |
Involvement in Decision-Making and Treatment Options | 1-Year Satisfaction | 0.20* |
Information and Understanding When Leaving the Hospital | 1-Year Satisfaction | 0.20 |
Costs | PROMs | Correlation |
---|---|---|
Estimated Patient-Level Inpatient Costs | 1-Year Change in Functional Status | 0.08* |
Estimated Patient-Level Inpatient Costs | 1-Year Change in ºìÁì½í¹Ï±¨-Related Quality of Life (HRQL) | 0.11* |
Estimated Patient-Level Inpatient Costs | 1-Year Satisfaction | 0.03 |
Note
* Indicates an significant correlation (P<0.05).
Sources
Canadian Patient Cost Database, Canadian Patient Experiences Reporting System, Discharge Abstract Database, Hospital Morbidity Database, National Ambulatory Care Reporting System and patient-reported outcome measures data, Canadian Institute for ºìÁì½í¹Ï±¨ Information.
PREMs and clinical outcomes
When examining the relationship between patient experience measures and clinical outcomes, Communication With Doctors was correlated with decreased hospital harm.
PREMs and PROMs
Satisfaction with hip and knee surgical results after 1 year had a positive correlation with favourable responses for the patient experience measures (Overall Hospital Experience, Communication With Nurses, Communication With Doctors, and Involvement in Decision-Making and Treatment Options).
Costs and PROMs
There was a correlation between inpatient hospitalization costs and PROMs, including patient-reported change in functional status and change in quality of life after 1 year.
These findings demonstrate the need to continue to collect patient-reported data to fully leverage the potential of linked health care data. Though some correlations were statistically significant, the correlations observed were classified as weak (less than 0.39).Reference1 This suggests that additional factors may influence the relationship between measures. With larger data sets, more advanced analysis can be pursued such as accounting for non-linear relationships and multiple factors to better understand focus areas that can both improve patient outcomes and experiences and lower costs.
Related resources
References
1.
Schober P, Boer C, Schwarte LA. Correlation coefficients: Appropriate use and interpretation. Anesthesia and Analgesia. 2018.
How to cite:
Canadian Institute for ºìÁì½í¹Ï±¨ Information. Demonstrating linked PREMs, PROMs, clinical and costing data. Accessed April 15, 2025.

If you would like ºìÁì½í¹Ï±¨information in a different format, visit our Accessibility page.