Integrating equity into hospital incident reporting and patient concerns systems: study protocol for a mixed methods study

Abstract

Introduction

Preventable hospital patient harm events disproportionally affect certain patient populations. For some, harm extends beyond physical injury to include cultural, emotional or spiritual impacts. While these disparities are linked to socio-demographics (eg, race, education), they are driven by structural factors (eg, procedures and policies). Patient safety monitoring systems (eg, incident reporting, patient concerns) were not originally designed to identify equity-related harms and may inadvertently obscure or reinforce the injustices they should address. This study will examine how equity is currently considered within hospital incident reporting and patient concerns systems across Canada and will identify opportunities to strengthen these systems’ responsiveness to inequities in patient safety.

Methods and analysis

This 3-year exploratory sequential mixed-method study began in September 2024. Phase one involves qualitative interviews with patient safety and equity leads, patients/families/caregivers and leaders of innovative initiatives to explore current practices, gaps and innovations in how equity-related factors are identified and addressed within incident reporting and patient concerns systems. Findings will inform Phase 2, a modified Delphi process with patient safety and equity experts and persons with lived experience of equity-related harm events to refine and reach consensus on key equity-promoting features, considerations and recommendations for these systems. In Phase 3, consensus items will be used to develop a national cross-sectional survey assessing the extent to which equity is integrated into hospital incident reporting and patient concerns systems in Canada. A patient advisory committee will inform data collection, interpretation of findings and dissemination.

Ethics and dissemination

Ethics approval has been received for Phase 1, with subsequent approvals to be sought for later phases. Dissemination plans include peer-reviewed publications, presentations at international conferences and knowledge exchange activities to inform patient engagement, the design of incident reporting and patient concerns systems and policy development.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The sequential mixed methods approach will allow for the qualitative interviews in Phase 1 to inform a modified Delphi study in Phase 2, which will shape the development of a national survey in Phase 3.

• The study includes interview participants from three Canadian provinces, Ontario, British Columbia and Alberta, and international innovators in Phase 1; national and international experts for the Delphi study in Phase 2; and a Canada-wide survey sample in Phase 3.

• Challenges reaching consensus on key equity-focused features, considerations and recommendations in incident reporting and patient concerns systems in the modified Delphi study will be mitigated through multiple iterative rounds and structured prioritisation to support convergence.

• Although we have a robust survey recruitment strategy, there is a risk of a lower response rate than planned.

Introduction

With growing public and professional recognition of longstanding inequities in healthcare,1,3 the question is no longer whether these disparities exist, but how health systems should respond.^{4 5} Equity is defined as the ‘absence of unfair, avoidable or remediable differences among groups of people, whether those groups are defined socially, economically, demographically or by other dimensions of inequality’.⁶ Health equity is achieved when everyone can attain their full potential for health and well-being and is not disadvantaged from attaining it because of any socially determined circumstance.^{6 7} Equity is a fundamental dimension of healthcare quality.⁸ Yet the very patient safety monitoring approaches designed to protect patients, such as incident reporting and patient concerns systems, often fail to capture or address hospital harm events shaped by racism, socioeconomic exclusion or other inequities.9,13 Worse, these systems may inadvertently obscure or reinforce and perpetuate the injustices they should be surfacing. As healthcare organisations commit to equity as a core principle,¹⁴ a critical gap remains: how can we ensure that patient safety monitoring systems reflect and respond to the lived experiences of all patients? This multi-method study aims to address that question by exploring how equity can be more meaningfully integrated into incident reporting and patient concerns systems in Canada.

Equity, harm and safety

Equity-related variations in patient harm events are increasingly recognised yet remain poorly addressed in safety monitoring systems.9,13 In Canada, it is estimated that 1 in 17 hospital stays involves at least one harmful event, many of which are preventable.¹⁵ However, these events are not evenly distributed. Patients’ spoken language and education levels are associated with different likelihoods of harm.¹⁶ These patterns are driven by system-level factors (eg, provider bias), not by deficits in patients.⁵ Equity concerns extend beyond disparities in who experiences harm; they also relate to what types of harm are recognised. Traditionally, hospital-related harms have centred on physical harms (eg, harms from medication or surgical errors, healthcare-associated infections, patient falls).¹⁷ However, calls for equity in patient safety emphasise the need to broaden definitions of preventable harm to include psychological, cultural, spiritual and emotional harms, which are often rooted in systemic inequities such as racism, ableism and sexism.¹⁸

This relationship between equity and harm is observed in varied clinical and geographical contexts. For instance, in a collaborative of 27 paediatric hospitals in the USA, the rate of serious harm experienced by patients varied by the patient and family’s race and ethnicity.¹⁹ In two UK maternity units, although there was no difference in the proportion of serious incidents based on patients’ ethnicity, there were differences in medical and pregnancy-related risk factors between the groups and in the root causes identified. Booking late for antenatal care occurred more frequently in women from ethnic minorities who were involved in high level incidents, and in the cases of serious incidents affecting these women, more frequently cited root causes related to staffing levels and workload and a failure to seek support from more experienced staff.²⁰ These findings underscore that harms are unevenly experienced, contributors to harms may vary, and that current approaches may overlook important equity-related drivers.

This paper describes a study designed to examine how hospital work systems conceptualise and respond to equity-related harms, drawing on insights from both social science and human factors. Further, it explores the potential for incident reporting and patient concerns systems to more systematically identify and address inequities in patients’ safety.

Incident reporting and patient concerns systems

Incident reporting and patient concerns systems are two of numerous monitoring systems (eg, administrative datasets, morbidity and mortality conferences and malpractice claims) that hospitals use to identify and respond to harm.²¹ These two systems differ in detection methods but overlap in goals to improve safety in alignment with institutional and societal commitments towards high-quality care. Incident reporting systems provide healthcare workers with a standardised process to report actual or potential harm.^{22 23} These reports are typically reviewed and analysed by healthcare leadership, internal quality and safety departments, or in some jurisdictions by external groups, to identify contributory factors and develop recommendations for improvement.²² Patient concerns systems, on the other hand, provide patients, family and caregivers a venue to formally raise concerns about care experiences, including safety events.^{24 25} Commonly managed by patient relations, risk management or legal teams, patient concern systems typically handle concerns on a case-by-case basis.²⁶ While the term ‘patient complaints’, rather than concerns, has been commonly used, it is critiqued for its negative connotations and characterisation of patients as ‘complainants’.^{9 27} We therefore use ‘patient concerns’ in our work. Although patient concerns systems were not created explicitly as a patient safety system, concerns submitted can be used as a trigger for further enquiry together with other quality and safety data.²⁶

We chose to focus on incident reporting and patient concerns systems because they are widely institutionalised, have regulatory dimensions and offer both provider and patient-initiated approaches. However, these monitoring systems also suffer from equity-related limitations that lead to differential detection of harm among patient groups defined by social factors such as race, spoken language and income.^{13 28 29} For example, in incident reporting systems, provider bias can affect whether a harm is recognised as reportable, leading to underreporting of harms experienced by some groups and overreporting in others.¹⁰ In addition, the processes commonly used to investigate and respond to safety events are often not attuned to structural inequities. Even when harms are recognised, reviews typically focus on clinical or procedural failures, overlooking how factors such as racism, ableism or bias may have contributed. Consequently, equity-related contributors are rarely reflected in the identification of root causes or in the recommendations that follow. For example, communication barriers may be central to an event, but their impact remains invisible if not meaningfully captured during analysis.¹¹ These gaps result in skewed data and an incomplete understanding of how different populations experience harms and the contributing factors. Patient concerns systems face similar limitations, as they are at risk of excluding those most affected due to barriers such as lack of awareness, inadequate language support and fears related to reporting.²⁹ As a result, these systemic shortcomings prevent the full range of patient experiences— especially those from marginalised or equity-deprived groups—from being heard, ultimately limiting the system’s ability to capture a comprehensive picture of patients’ safety.

Given the inequities that exist in patients’ experiences of harms, the limitations of patient safety systems whose purpose is to identify and respond to such harms, and the risks of these systems in perpetuating inequities, it is essential to increase our understanding about how hospitals can create safety systems explicitly designed to surface and respond to equity-related factors.

Human factors and social science theoretical framework

Patient safety and quality improvement scholars characterise inequity in healthcare as a systems-based problem, bringing attention to contributors of inequities beyond individual clinicians to wider organisational and societal structures that shape behaviours and practices.³⁰ This study combines human factors and social science approaches, each offering a different systems lens that can inform each other.³¹

Within patient safety, human factors emphasises the ‘work system’ and its interacting elements defined as person, tasks, tools/technologies, environment and organisation, as demonstrated by the Systems Engineering Initiative for Patient Safety (SEIPS) model.³² This systems approach brings attention to the importance of considering issues across the whole system and their relationships to each other, when aiming to improve processes and outcomes. For example, organisational practices such as failing to offer non-English education in infection prevention can increase hospital-acquired infection risks for patients with ‘emerging English proficiency’.³³ Yet, such equity-related factors are not routinely identified in incident reporting or patient concerns systems.¹¹ The SEIPS model will provide an analytical lens to guide system redesign to support the identification of equity considerations in incident reporting and patient concerns systems.

Social science theory about equity and patient safety will offer an additional and complementary analytical approach. This literature emphasises inequities arising from systemic discrimination biases that are embedded in tangible (eg, individuals, organisations, buildings) and intangible (eg, ideologies, goals, decision-making processes) system parts.^{34 35} An equity-focused systems orientation reveals, for example, the epistemic injustices in incident reporting whereby professionals, patients and families’ testimony in incident reporting is not always heard, understood, believed or valued due to power imbalances.^{36 37} Additionally, this literature invites examination of organisational processes affecting practices in patient safety reporting that provide a positioning for an equity lens. For instance, social science patient safety scholarship has demonstrated the range of factors that affect hospital healthcare providers’ decisions about whether or not to speak out about a safety concern.³⁸ Additionally, this research shows the ways in which narratives of patient safety are constructed through organisational processes such that check-boxes and pre-defined categorisations in incident report systems contribute to de-contextualised ‘narrow narratives’.³⁹ These findings provide essential background to exploring how to support identification and review of equity considerations in incident reporting.

A combined social science and human factors systems-based lens will allow for deeper understandings of how equity and harms are conceptualised within hospital contexts and how and why the ‘work system’ allows for, or misses, opportunities to integrate an equity lens into incident reporting and patient concerns systems.

We aim to examine how equity is addressed in hospital incident reporting and patient concerns systems across Canada, and to identify features, considerations and recommendations that could strengthen their responsiveness to inequities in patient safety. Specifically, the study objectives are to:

1. Elucidate and characterise features, considerations and recommendations for integrating equity into incident reporting and patient concerns systems.

2. Build consensus on equity-related features to be included in an online pan-Canadian equity integration assessment survey and an accompanying reference guide.

3. Generate national data on the prevalence of features and considerations for integrating equity in hospital incident reporting and patient concerns systems and deliver guidance to hospitals.

Methods and analysis

Preliminary work: critical interpretive synthesis of equity in safety monitoring

As preliminary work for this study, we conducted a critical interpretive synthesis⁴⁰ of the literature on equity in incident reporting and patient concerns systems. The review examined how equity is currently enacted in these systems and identified opportunities for more systematic integration. Through inductive analysis of 30 included articles, the synthesis generated four themes: (1) knowledge injustices in what patient harm events are recognised and addressed, (2) individual bias and systemic discrimination in reporting practices, (3) opportunities and limitations of data stratification and (4) alternate frameworks including restorative and human rights approaches, as ways to address inequities and humanise harm.^{41 42} These findings will inform the study qualitative interview guides and subsequent phases of analysis and tool development.

Methodological design

This 3-year study, launched in September 2024, will use an exploratory sequential mixed methods design consisting of three phases to achieve objectives 1-3.^{43 44} In a sequential design, each phase builds on the findings of the previous one to allow for in-depth exploration and methodological integration. Together, these phases will generate a comprehensive and actionable understanding of how equity can be meaningfully embedded in hospital incident reporting and patient concerns systems.

Patient and public involvement

The principal applicants will jointly lead the project in collaboration with team co-applicants, a patient advisory committee (PAC) and an expert panel. The PAC consists of a diverse group of individuals with lived experiences of healthcare inequities, preventable hospital harms and experiences with incident reporting and/or patient concerns systems. In pursuing diversity, we will be sensitive to the complex interplay of factors such as race, ethnicity, language, socioeconomic status, disability, gender, rural/remote geography and other social identities relevant to equity in healthcare. To ensure patient/public perspectives are structurally embedded and have meaningful influences, the PAC will be engaged at the ‘collaborate’ level, as defined by the Canadian Institutes of Health Research Patient Engagement Framework.⁴⁵ This means that members will participate in shared decision-making on elements such as recruitment strategies, interpretation of findings and dissemination plans. In addition, two seats on the expert panel are designated for PAC members to ensure consistent representation in study governance. To guide the planning and conduct of patient and public involvement in this project, we will draw on the Learning Together Evaluation Framework for Patient and Public Engagement in Research.⁴⁶ We will use its guiding principles as a reference point for designing our engagement activities and ensuring they are meaningful, inclusive and aligned with the project’s equity goals. Financial compensation will be provided to PAC members.

Phase 1: qualitative interview-based study

Phase 1 will involve qualitative interviews⁴⁷ to explore current practices and gaps in how equity-related factors are understood, operationalised and addressed within hospital incident reporting and patient concerns systems. These interviews will also allow for insights into innovative approaches to incorporating equity into identifying and responding to harms.

Sampling and recruitment

This phase of the study will include three participant groups: (1) hospital and healthcare system quality and safety and/or equity specialists; (2) patients, families and caregivers; (3) leads for innovative initiatives for integrating equity into incident reporting and patient concerns systems or new systems for identifying and responding to equity-related harms. The first two groups will be recruited from three provinces in Canada—Alberta, British Columbia and Ontario. The three chosen provinces exhibit variability in the organisation of incident reporting and patient concerns systems and policies impacting these systems as well as the organisation of healthcare systems, that will allow us to explore the ways in which these contextual factors influence equity and hospital safety monitoring practices. The third group will be recruited from Canada and internationally.

For participant group 1 (target n=30–60), we will recruit from 2 to 3 hospitals or health systems (depending on the organisation of patient safety monitoring systems and the healthcare system) from each province, using purposive maximum variation sampling⁴⁸ for variability in geographical context, population and hospital/health region size. We will electronically distribute study information with interview invitations to individuals with positions in incident reporting and patient concerns, and individuals who hold equity-focused leadership roles (eg, patient safety or patient relations lead, risk manager, equity lead).

For participant group 2 (target n=20–25), we will recruit patients, family members or caregivers with experiences of equity-related hospital harms and have or have not been involved with incident reporting and patient concerns systems. Variability in their system experiences will allow for the exploration of multiple reasons for engaging and not engaging with these systems. We will use purposive maximum variation sampling⁴⁸ to identify individuals who experience different forms of inequity (eg, related to the complex interplay of factors such as race, religion, gender, sex and disability) in relation to their experiences of hospital harm and perspectives from residing in the three identified provinces. We will recruit through provincial and national patient safety organisations and direction from PAC members.

For participant group 3 (target n=15–20), we will recruit individuals in Canada and internationally who are leading innovative initiatives for integrating equity into incident reporting and patient concerns systems, or who are developing new equity-sensitive systems to identify and respond to harms. These individuals will be identified through criterion sampling using: (1) research team connections with national and international health and patient safety organisations and networks, (2) searches of authors of publications included in the critical interpretive synthesis and (3) snowball sampling⁴⁸ by asking participants to suggest others who meet our sampling criteria.

Participant recruitment for all three groups will be guided by the principle of information power, which shifts the focus from the number of participants to the power of information the sample holds.⁴⁹ We will expand our sampling if greater variability arises within each group.

Data collection and analysis

This phase will use semi-structured interviews to collect qualitative data (online supplemental file).⁴⁷ Interviews will be conducted by experienced researchers, including one interviewer with lived experience of a serious patient-safety event. The research team will draw on trauma and violence-informed principles and engage in reflexive practices, including post-interview memos and periodic peer debriefs. These practices will support the creation of a supportive and respectful environment for participants, particularly during patient interviews. Furthermore, they will enable researchers to reflect on how one’s own knowledge and experiences are influencing how they are collecting and responding to the data.^{50 51} Interview guides will be developed for each of the three sample groups, aiming to understand current hospital and health system practices, the lived experiences of patients/families/caregivers, and innovative initiatives and approaches. The guides are informed by the critical interpretive synthesis, the PAC and research team members with lived experience. Interview participants will be informed about the confidentiality of their participation and that all details will be anonymised. Following informed consent, interviews will be audio-recorded and transcribed.

To analyse interviews, we will use a thematic analysis⁵⁰ approach informed by both human factors and social science theories as described previously, and by building on the themes from the critical interpretive synthesis. These analyses will continue developing an evolving list of equity-oriented features, considerations and recommendations for each of the two monitoring systems (incident reporting and patient concerns) and approaches to innovation. The constant comparison method⁵² will guide data analysis as we move back and forth within and between data collected across the different participant groups, the three provinces, the two monitoring systems and innovative approaches. Consistent with the constant comparative method, themes of equity features, considerations and recommendations will be modified iteratively in response to the data, resulting in re-specification of identified themes and new themes being developed. We will attend to differences and overlaps of findings as they relate to the two monitoring systems, for example, to more precisely describe how biases may influence reporting across both systems but in system-specific ways, or what kinds of equity-related harms are present or absent across the two systems.

Phase 1: main deliverable

A comprehensive list of features, considerations and recommendations for integrating equity into incident reporting and patient concerns systems.

Phase 2: modified Delphi study

Findings from Phase 1 will inform Phase 2, a modified Delphi process^{53 54} with expert panellists experienced in domains of patient safety and equity and/or lived experiences of equity-related hospital harms to refine and reach consensus on key equity-focused features, considerations and recommendations in incident reporting and patient concerns systems.

Sampling and recruitment

The aim is to recruit a panel of 11–30 expert members purposively sampled to reflect a diverse range of perspectives on equity and patient safety monitoring systems.^{53 55} The inclusion criteria will be one or more of the following: (1) experience working in hospital patient safety and/or equity, (2) lived experiences of inequity-related hospital harms and/or incident reporting or patient concerns systems and (3) experience studying in the fields of patient safety and/or equity. We will recruit national and international representatives, with a majority representing the first two groups—people actively involved in patient safety practice and patients/family/caregivers. The study expert panel, PAC and research team will suggest potential panellists. Invitations to take part in the study will be sent by email. We will use snowball recruitment by asking potential panellists to suggest others who they consider suitable to participate. We will then invite identified potential panellists, provided they meet the inclusion criteria.

Data collection and analysis

We will create a questionnaire about equity-related features, considerations and recommendations based on findings in Phase 1 and distribute it online in English and French to Delphi participants. Consistent with Berenholtz et al⁵⁶ guidance for producing practical tools that are helpful to practitioners, we will frame each identified feature or consideration as a question (eg, Do institutional policies acknowledge inequities as a source of harm?). We will generate a list of system-based recommendations based on findings from Phase 1 (eg, for policies, tool redesigns, staff training). Participants will rate the importance of features and considerations using 5-point scales for relevance, feasibility and impact, as accepted criteria for priority setting in health services research.⁵⁷ Panellists will also have the option to provide comments to support their decision or volunteer changes to the question.

The modified Delphi will include three rounds. In round 1, items judged by at least 70% of participants^{58 59} to be important (a score 4 or 5) will be considered a consensus and will be automatically included in the equity integration assessment survey and reference guide. If consensus is not obtained in round 1 for all items, that is, judged by 70% of participants, round 2 will include items that did not meet that threshold. For round 2, items will be modified where possible based on round 1 feedback and include new items suggested by participants. In round 2, respondents will be shown round 1 results, their own answers given back and asked if they want to change their vote (with option to provide rationale). A research team and PAC member will co-lead a virtual meeting after each round to share and discuss findings and support consensus. There will be two meeting times to accommodate time differences, and both meetings will be recorded for those who are not available to attend. A final virtual meeting (round 3) is planned to resolve any items that do not reach consensus in rounds 1 and 2, as has been done in other studies.⁵⁴ Final items identified as important (reached by consensus) will be included in the equity integration assessment survey and reference guide.

Phase 2: main deliverable

A prioritised, consensus-based subset of equity-related features and recommendations from Phase 1 to be included in the equity integration assessment survey and reference guide.

Phase 3: cross-sectional survey

In Phase 3, items that reached consensus from the modified Delphi study will be used to develop a national cross-sectional survey to assess the current state of equity integration in incident reporting and patient concerns systems.

Sampling and recruitment

We plan to distribute the survey to all hospitals across Canada using comprehensive lists sourced from provincial/territorial health ministries, the Public Health Agency of Canada, the Guide to Canadian Healthcare Facilities,⁶⁰ hospital websites and Google Maps. To ensure surveys reach the most appropriate individuals, we will use multiple strategies:

• Institutional champions and direct contacts: Healthcare Excellence Canada and our research team members, who have established relationships with provincial quality and hospital organisations, will assist in identifying and directing surveys to suitable contacts (eg., Patient Safety Officers, Quality Directors).

• Professional networks: We will reach out to hospital leadership associations (eg, provincial health authority networks, nursing leadership councils) for guidance in routing the survey to the most relevant staff members.

• Adaptive follow-up: If initial emails are unresponsive, we will follow up via phone or email with central administration offices to identify the correct point of contact.

We anticipate an overall response rate of 330-415 hospitals (54-68% of all Canadian hospitals)^{61 62} based on past Canadian healthcare surveys.^{63 64}

Data collection and analysis

We will develop a survey to assess hospitals’ integration of equity into incident reporting and patient concerns systems using rigorous survey design.⁶¹ Questions will be compiled from equity-related features, considerations and recommendations identified in Phases 1 and 2, with input from the expert panel and PAC. Participants will be asked to indicate which of the two safety monitoring systems they work with and will then be provided with appropriate survey questions available in English and French. The invitation email to the survey will ask potential respondents to reach out to the research team about any survey accessibility considerations that the team can support.

The equity integration assessment survey will be tested before distribution to improve its content validity and reliability. To test the survey, we will use a convenience sample (n=15–20) of hospital workers with experience distributed across the two monitoring systems from three participating hospitals. Respondents will be asked to rate the relevance of each item on a 4-point Likert scale (1=not relevant, 2=somewhat relevant, 3=quite relevant, 4=very relevant) and a content validity index will be computed.⁶⁵

The final survey will be delivered to respondents online with a secure application,⁶⁶ and using the Dillman approach⁶⁷ with a reminder at 2 weeks, and a third reminder after another 2 weeks. The survey will remain open for 1 month after the last email reminder. The response rate will be monitored monthly and if after 2 months the target response rate has not been reached, recruitment will be extended up to 6 months to support targeted recruitment efforts. We will target provinces with lower response rates to attempt a consistent response rate across all provinces.

Hospital demographics will be summarised using descriptive statistics and compared with those of the national sample from the Canadian Institute of Health Information using t-tests. Differences in survey answers will be evaluated using linear regression models based on hospital size (number of beds), affiliation (academic vs community) and province, using a stepwise selection procedure. Once we have compiled these ratings from all hospitals, we will be able to identify opportunities to improve hospital integration of equity into incident reporting and patient concerns systems and highlight gaps where strategies for integration of equity have not yet been developed.

Phase 3: main deliverable

A Canada-wide comprehensive data source about how equity is currently integrated into incident reporting and patient concerns systems, and a reference guide to support healthcare systems in integrating equity features, considerations and recommendations into these safety monitoring systems. To support ongoing learning and adaptation, we will explore designing the guide as an online, updatable resource that can evolve over time in response to emerging evidence.

Ethics and dissemination

Ethics

Research ethics approval for Phase 1 of this study has been received from the University of Toronto and other hospital and healthcare authorities requiring research ethics board approvals for collecting data in their organisations. Research ethics approvals will be obtained from the University of Toronto for research Phases 2 and 3.

Dissemination

Dissemination plans include publications in peer-reviewed quality and safety journals, presentations at international conferences and knowledge-exchange activities with partners and audiences across care settings to optimise patient engagement and inform hospital practices and policy development; where feasible and appropriate, outputs will be open access and accompanied by a plain-language summary. We will partner with the PAC as co-dissemination leads: committee members will review patient-facing materials for clarity and relevance, help identify trusted channels (patient organisations, caregiver groups, community venues, online forums) and, where desired, co-present at public sessions. PAC contributions to public products will be acknowledged, with co-authorship or co-presentation offered.