Improving patient safety and quality of care: Applying an equity lens to hospital harm

October 17, 2024 — Measurement is fundamental to health care quality improvement. Quality improvement efforts need to be guided by data and focused on care that is safe, timely, effective, efficient, equitable and patient-centred.Reference1 The U.S. Institute for Healthcare Improvement, a global leader in quality improvement, has identified the importance of equity in quality and patient safety improvement.Reference2

CIHI developed the Hospital Harm indicator to support patient safety measurement and improvement in hospitals. This indicator measures harmful events that occur during an acute care hospitalization that could have been prevented through evidence-based practice. In Canada, the average cost of a hospitalization for a patient who had a harmful eventFootnotei during their stay was more than 4 times the cost of a hospital stay without harm ($42,558 versus $9,072).Footnote,.ii

In this analysis, we apply an equity lens to the Hospital Harm indicator to measure variations in harm across patient population groups. Education and language stratifiers were analyzed by linking the 2016 Census to the pooled Discharge Abstract Database (DAD) files (2016–2017 to 2018–2019) that were available in Statistics Canada’s Virtual Data Lab at the time of the analysis. Age, recorded sex or gender, neighbourhood income and patient place of residence were analyzed using DAD files (2014–2015 to 2022–2023). Health systems and hospitals can replicate this approach and use this information to prioritize quality improvement efforts, reduce equity gaps, and improve quality and safety for all populations.

Patients who do not speak English or French are 30% more likely to experience harmful events compared with patients who speak one of those languages. This represents 1.5 additional cases of harm per 100 hospitalizations compared with the rate for those who speak English or French.

Similar results were found in an analysis of patient safety events among home care recipients who were admitted to hospital in Ontario, Canada.Reference3 Reference4 The authors suggested that higher rates of harm for this population can potentially be explained by communication challenges that arise when the patient and provider speak different languages.Reference5 Communication challenges may lead to poor patient understanding, improper use of medications, and improper tests and procedures.Reference6

The measurement of various aspects of patient care, such as patient satisfaction and safety, enables health care providers and organizations to pinpoint where improvements are needed. By systematically collecting and analyzing data, health care organizations can make informed decisions to enhance patient care, reduce errors and improve overall outcomes. Equity-stratified patient safety data takes this a step further by recognizing that health care outcomes can vary significantly across different demographic groups, such as race, ethnicity, gender and socio-economic status. — Maya Sinno, Associate Vice President, Acute Care Clinical Transformation and Growth, Mackenzie Health

Patients who have less than high school education are 20% more likely to experience harmful events compared with patients who have high school education or more. This is about 1 additional case of harm per 100 hospitalizations compared with the rate for patients who have high school education or more.

To our knowledge, this is the first analysis of patient educational attainment and risk of hospital harm in Canada. Evidence from the United States has found a similar association between harm and a patient’s level of education.Reference7 Reference8 Low education is associated with low health literacy,Reference9 and health literacy is in turn associated with the ability to navigate the health care system.Reference10 Health information can be complex and may include unfamiliar medical terms, which may make it more difficult to understand for people with lower levels of education.

Applying an equity lens to patient safety indicators like Hospital Harm helps uncover how social determinants of health, like education and languages spoken, are part of the picture and need to be considered as part of efforts to improve outcomes. This helps generate discussion that can inform quality improvement and additional analysis. — Lisa Corscadden, Senior Director, System Reporting and Population Health Analytics, British Columbia Ministry of Health

Older patients are at increased risk of experiencing a patient safety event during a hospital stay. Patients age 85 and older are 5 times more likely to experience patient safety events compared with patients younger than 20. This represents about 8.5 additional cases of harm per 100 hospitalizations for patients age 85 and older. See the data tables for more information.

Age, pre-existing conditions and frailty (i.e., reduced function and health in older adults) are well-known risk factors for health care–associated harm.Reference11 Reference12 Reference13 Age-associated hearing loss and cognitive decline may also contribute to communication errors, which have the potential to result in harm.Reference14

The following strategies support effective communication with and care outcomes for patients who don’t speak English or French, patients with lower education and/or older patients: 

• Use the teach-back method when relaying information to patients (i.e., have patients repeat instructions in their own words to ensure they understand)Reference15
• Promote clear, effective patient–provider communication that considers patient health literacyReference6 Reference16
• Use qualified interpreters where possibleReference16
• Establish policies that acknowledge health interpretation as a “medically necessary” serviceReference17
• Raise organizational awareness about the impact of health literacy and English/French proficiency on patient safetyReference18
• Empower staff with strategies that facilitate patient safety reporting among patients with low English or French proficiency and/or lower educationReference15
• Emphasize patient-centred communication by tailoring communication to distinct patient groupsReference14

The Hospital Harm indicator provides health care leaders with an overall picture of patient safety in their organization, with the ability to drill down further to identify priority areas for safety improvement. 

To demonstrate, we analyzed inequalities in hospital harm by patient sex, age, neighbourhood income and place of residence (urban versus rural/remote geography), further stratified by patient type (medical or surgical) and the 4 categories of hospital harm.Footnoteiii The data tables provide detailed results from this analysis at the national level for 2014–2015 to 2022–2023. 

For example, while the overall risk of experiencing hospital harm increases with age, this pattern varies by the category of hospital harm. Patients age 85 and older are at highest risk of health care– or medication-associated harms, while patients age 65 to 74 are at highest risk of procedure-associated harms. 

This data clearly shows that the system does not provide the same safe, reliable, high-quality care to all patients. Drilling down on this data will inform approaches to improve equity in safe, high-quality care — for example, by improving communications with older patients, those who have limited language skills in English and French, and patients with lower education levels. — Dr. Ross Baker, Researcher and Professor, Institute for Health Policy, Management and Evaluation, University of Toronto

Our analysis explored equity in patient safety in Canada, which is an emerging area of focus for health systems. Analysis of administrative data complements other patient safety measures and tools that may also incorporate an equity lens, such as

• Morbidity and mortality reviews
• Error or incident reporting systems
• Chart or electronic medical record reviews
• Direct observation of patient care
• Clinical surveillance

To improve overall patient safety outcomes and reduce inequalities among patient populations in hospitals, consider the following:

• Review patient safety culture using valid and reliable assessment tools that incorporate equity, such as a health equity checklistReference19
• Integrate patient socio-demographic dataFootnoteiv — including language, education and racialized group — in patient safety measurement
• Implement tailored patient safety interventions that address the underlying causes of patient safety inequalitiesReference20
• Increase the diversity of lived experiences and expertise among patient safety teamsReference20
• Examine patient safety systems for biasesReference20

Integrating patient socio-demographic dataFootnoteiv — including language, education and racialized group — with patient safety measurement is key to identifying whether certain patient populations are experiencing a disproportionate burden of harm. Socio-demographic data is equally important for monitoring the effectiveness of an intervention for target populations as part of quality improvement initiatives. The goal is to improve overall patient safety outcomes and reduce inequalities among patient populations.

Patient safety improvement resources:

• TeamSTEPPS Canada Essentials Course (Healthcare Excellence Canada)
• Patient Safety Essentials (Healthcare Excellence Canada)
• A Guide to Patient Safety Improvement (Healthcare Excellence Canada)
• Rethinking Patient Safety Discussion Guide (Healthcare Excellence Canada)
• Effective Governance for Quality and Patient Safety (Healthcare Excellence Canada)
• Equity, Diversity and Inclusion Virtual Learning Exchange (Healthcare Excellence Canada)
• Patient Safety Essentials Toolkit (Institute for Healthcare Improvement)
• Improvement Tools (PSNet, Agency for Healthcare Research and Quality)

• Hospital Harm Project 
• Measuring Health Inequalities: A Toolkit
• Health equity and population health

Back to Equity and patient safety: An analysis of hospital harm