Scoping review of patients’ attitudes about their role and behaviours to ensure safe care at the direct care level

Abstract Background To improve harm prevention, patient engagement in safety at the direct care level is advocated. For patient safety to most effectively include patients, it is critical to reflect on existing evidence, to better position future research with implications for education and practice. Methods As part of a multi‐phase study, which included a qualitative descriptive study (Duhn & Medves, 2018), a scoping review about patient engagement in safety was conducted. The objective was to review papers about patients’ attitudes and behaviours concerning their involvement in ensuring their safe care. The databases searched included MEDLINE, CINAHL and EMBASE (year ending 2019). Results This review included 35 papers about “Patient Attitudes” and 125 papers about “Patient Behaviours”—indicative of growing global interest in this field. Several patterns emerged from the review, including that most investigators have focused on a particular dimension of harm prevention, such as asking about provider handwashing, and there is less known about patients’ opinions about their role in safety generally and how to actualize it in a way that is right for them. While patients may indicate favourable attitudes toward safety involvement generally, intention to act or actual behaviours may be quite different. Conclusion This review, given its multi‐focus across the continuum of care, is the first of its kind based on existing literature. It provides an important international “mapping” of the initiatives that are underway to engage patients in different elements of safety and their viewpoints, and identifies the gaps that remain.


| INTRODUC TI ON
The need to improve patient safety continues to be paramount. Patient safety is recognized as a serious global public health issue, with approximately one in 20 patients harmed while receiving medical care (primary, secondary and tertiary care settings). 1 When considering patient safety, it is important to understand the perspectives of patients and their families, and to recognize the need for more effective involvement of patients and families across the continuum of care. 2 As recipients of health-care services, patients and their families are a valuable resource and contribute to our understanding of safety issues. To advance our knowledge about patient and family involvement in promoting health-care safety, as distinct from their health generally, a multi-phase study was conducted that included a qualitative study of patients' perspectives about their knowledge, comfort level and behaviours in helping their safety while receiving health care. 3 This scoping review was phase 2, and the aim was to describe the literature about patients' and families' attitudes toward their role in health-care safety, as well as their reported behaviours in support of their safe care. A general multidimensional framework about patient and family engagement in health and health care underpinned this study. 4

| BACKG ROU N D
In preparation for the multi-phase study, a preliminary examination of the literature, which included findings from previous unpublished work, 5 provided an initial understanding of the evidence about patient perspectives regarding health-care safety, and views on their involvement in safety processes. The findings from this preparatory work, which helped determine the need for a scoping review, included the following concepts.

| Trust
The concept of trust in the health-care system and providers was also a common theme related to patient safety as perceived by patients and families. 6,9,11,13,[23][24][25] The overwhelming need of "safety" in ill hospitalized patients was tempered by the knowledge of being able to trust staff. 11 A group of 30 ambulatory patients receiving chemotherapy treatment recognized their own limitations in safety prevention, including deficits in knowledge, and therefore felt they had to trust the providers. 19 In another example, some parents, who believed their children had experienced a medical error, expressed regret at having trusted a health-care professional. 13

| Being trusted, seen and valued
Patients and their families view being trusted for their knowledge, respected and valued for their contribution as integral to safe patient care across hospital, outpatient and home settings. 8,13,16,17,[26][27][28] Reports of patients' experiences in primary care echoed the related theme of difficulties in interpersonal relationships with providers, particularly noting disrespect and insensitivity. 12 A study of 20 individuals with Parkinson's disease (or someone who spoke on their behalf-partner, wife, husband, daughter) and their experiences of medication errors most poignantly captures the lack of respect of patients' knowledge related to their medication management and the de-valuing of their insights-the cost of which can jeopardize a patient's safety. 27

| Patients as partners
Patients' beliefs about participating in patient safety in different ways have also been investigated. 19,24,[29][30][31][32][33][34] Positive attitudes about safety engagement have been identified, with qualifiers such as the type of action required by the patient, practitioners receiving this level of involvement by patients and their families, as well as the setting. 3 Tactics that patients (or family members of patients) use to protect themselves when interacting with the health-care system have been reported, including bringing family or friends to appointments and questioning unfamiliar medication. 8,9,18,25,[35][36][37] Patient involvement in ameliorating harm has also been documented. 19,21,[38][39][40] Generally, the most common approach to engage patients in health-care safety has taken the form of patient "tips" or strategies outlined in patient information or educational materials. There is evidence that some of these tips have been created with input from patients; however, this is not true of all recommendations, with a need for evidence-based content and evaluation. 41,42 Given the imperative of considering all approaches to reduce harm, as well as the importance of being patient-and family-centred, the preliminary work identified gaps and additional questions (including do patients intentionally think about their involvement in safety at the direct care level, and what does safety mean to them) necessitating and providing rationale to explore international evidence in a more fulsome, comprehensive way related to patient and family involvement in safety at the direct care level.

| ME THODS
The scoping review was conducted to understand the breadth and depth of literature about patients' engagement in safe care. The intention of the review was to gather as much relevant literature about patients' attitudes toward having a role in preventing harm, as well as any safety behaviours they independently engage in or at the invitation of researchers or providers.
This scoping review was conducted using the search term "patient safety", but other iterations were also considered given patients/families may understand it in different ways. Additionally, the preparatory work was informative in identifying that researchers typically investigate components of safety (eg handwashing), and therefore, these elements were part of the search strategy. A search of the JBI (Joanna Briggs Institute) Evidence Synthesis Journal and Cochrane database indicated that no scoping review had been conducted on this topic. health-care providers]?

| Research design
Peters and colleagues indicate the appropriateness of a scoping review when "a body of literature…exhibits a large, complex or heterogeneous nature not amenable to a more precise systematic review". 43(p. 141) A generic methodology was taken to scope the literature on this topic, and this was overseen by the first author (LD) as part of her doctoral thesis, who was responsible for the management of data collection and analysis. Decisions were made primarily by the first author, and the thesis committee provided assistance if there were disagreements. Scoping reviews are unique in that they are typically about broader topics; often include many different study designs; and do not usually involve an assessment of the quality of studies that are included in the review. 44

| Study inclusion criteria
The PCC format for scoping reviews is an acronym for P = population; C = concept; and C = context. 43 The review was about a population which included all individuals engaged in the health-care system as patients, family members of patients or health-care consumers (no age criteria). The concepts were twofold, but both included the focus on safety actions as distinct from caring for one's health. First, it was the patients' and families' attitudes, opinions and beliefs about taking an active role to ensure safe care at the direct care level. Second, it was the patients' and family members' behaviours or actions that demonstrated active engagement and involvement to prevent harm at the direct care level. The safety behaviours or actions were either self-determined or directed as part of a study intervention. The review included evidence of whether patients and family members performed requested safety behaviours, or if they demonstrated self-determined strategies, to prevent harm while receiving health care. The context included all sectors of health care including acute care, ambulatory care and long-term care. Home settings (people's own home) were seen as a separate context and as such were excluded, unless they were included as part of multiple sites in a relevant paper. Direct care level was defined as any clinical care interaction, as based on the study framework. 4

| Types of literature
This review included systematic reviews, quantitative studies, qualitative studies, mixed-methods studies, scoping reviews, literature reviews, quality improvement projects, as well as opinion and discussion papers. Study or systematic review protocols were also included in order to give perspective on the interest in this topic and anticipated future results. It is acknowledged that some may not be actioned into full reviews but this does reflect current research activity. Studies published in abstract form (such as conference proceedings) were not included. Health Organization]) were searched focusing on the study objectives (see Appendix S1 for the site listing). In the searching of websites, given their design, it was most effective to search using general terms, such as "patient safety" or "patient beliefs on role in safety", or as based on the provided categories and headings. Note: The terms within each column were entered using 'OR', and the full sets of terms in each were combined across columns using "AND". "*exp" used with these terms to provide full scope of definition.

F I G U R E 1 PRISMA diagram
Records idenƟfied through database searching (n = 6,255)

| Procedure
The associated university research ethics board granted ethics approval for the multi-phase study, which included this scoping review (HSREB 6007637, NURS-299-12).

| Data collection and extraction
A standard approach was used to extract quantitative, qualitative and textual data from the literature about patients' and families' attitudes and behaviours related to their role in ensuring safe care. For all papers, general items, such as author(s), location of first author, title of document, name of organization or association (as relevant) and year of publication, were extracted. As relevant, data extraction also included, although not limited to, methodology, methods (eg interventions, sample, setting, data collection methods) and findings.

| Analysis
The analysis of the extracted data from the papers (formatted as tables) for this phase occurred in several ways. A descriptive approach was taken to determine the quantity of publications by type, author, location of author, year of publication and topic (ie attitudes versus behaviours; behaviours were further sub-divided into specific categories). Additionally, and as applicable, the results from the obtained literature were examined as a collective (published and unpublished papers together) using a content analysis approach for overall themes or patterns. This analysis was guided by the process as outlined by Miles, Huberman and Saldana, 45 including an initial coding process, followed by second cycle categorization into the larger groupings/themes. The results of this review are presented in narrative form. Study designs are reported as published by the author(s). as reflective of the search strategy for this review, is presented in Figure 1.

| RE SULTS
Of the 151 papers included in the final set of this review, there were 13 systematic reviews; one scoping review; six systematic review or study protocols; 53 quantitative studies; 29 qualitative studies; eight mixed-methods studies; eight literature reviews; 29 opinion or discussion papers; and five quality improvement projects.

| Patient attitudes
Three patterns emerged from the 35 papers about patients' and family members' attitudes and beliefs toward having an active role in promoting their safe care.

| Lack of evidence about patient attitudes
The first pattern is the paucity of evidence specifically about pa-

| Patient belief may differ from their action
The third pattern relates to the second, specifically to those patients and consumers who believe they have a role in promoting safe care. While patients may indicate support and favourable attitudes toward safety involvement generally, intention to act or actual behaviours may be different. 49,60 The Theory of Planned Behaviour 71 has been used as a conceptual framework by several investigators when assessing patient attitudes about engagement in safety. 49,60,63,68 The premise of the theory is that an intention to act is influenced by the attitude toward that behaviour, the perceived influence of others and the perception of ease or difficulty of the behaviour. 71 Behavioural intent is predictive of behaviour. 71

| Patient behaviours
Four patterns emerged reflecting patients' and family members' behaviours and actions in promoting their safe care.

| Escalation in patient engagement research
The first is the increased interest in this topic in recent years, and the evolving range and diversity of how patients are being involved to support the safety of their care. The action of patients asking providers whether they have washed their hands has been a predominant focus of research in this field. 49

| Patients are engaged
The second pattern was the indication that patients are engaging in behaviours that promote safe care, either independently in their own ways 56,81,[90][91][92]  and may not even recognize the activities they do as safety measures. 51 Additional research is needed, but current evidence suggests that patients can and do perform actions to promote their safe care, and positive outcomes related to their involvement have been described. 97 Australians Hor and colleagues 98 believe that patients are already involved and that instead of asking whether patients ought to be involved with safety, we should be asking how we can make patient-provider collaborations for safety more effective.

| A focus on influencing and impeding factors to engagement
A third pattern was patients' and family members' behaviours about promoting safe care. Investigators have examined factors that influence or impede patients' participation in safety initiatives. 47,94 Factors that may support and enhance patient involvement in safety include the following: perceived risk, provider encouragement, perceived self-efficacy 47,73 and health-care setting 50 (eg primary care is seen as key in developing trusting relations with implications to safety 99 and is seen as more feasible to engage, 88 yet seen as paramount for family members in hospital settings 62 ). Issues that may negatively influence an individual's ability to engage in activities to support safe care include the following: severity of one's illness, perception of staff work pressure, lack of awareness of the benefit of their involvement, engaging in a task perceived as challenging authority (versus factual, information-sharing tasks) and belief that one's role should be passive. 47,49,94 There is recognition that there will be different motivators and inhibitors that will affect patients' behaviours related to participating in the safety aspects of their care, and while understanding the motivators and reasons for engagement was not a focus of this review, further unpacking influencing factors such as whether one's preference to engage is anxiety or fear-based that safe care could be compromised versus acting based on preference alone is warranted.

| The role of family
The final pattern was the role of family in advocating for and protecting the well-being of another. Parents whose children are requiring health care report and demonstrate a vigilance and protectiveness to ensure they are safe. 36,90,100,101 Of 130 parents of hospitalized children, 63% agreed or strongly agreed that they needed to safeguard their child against potential errors. 101   Finally, investigators have typically examined safety practices that they have identified, without consideration of patient preferences and without using a participatory action research approach. In most cases, the uptake of these behaviours has been variable and lacking consistency. Further, in many cases, an indirect finding has been that patients engage in strategies of their own. A more effective approach when considering patient engagement in patient safety may be to determine patients' understandings and preferences at the point of contact with the health-care system and strengthen and enhance those self-identified strategies. An efficient, user-friendly mechanism for determining this will be required. Vincent 107 and Spath 33 write of the complexity of involvement and the importance of discerning patient preferences, and it is offered that greater emphasis needs to be taken in this regard to most effectively involve patients and family members in ways that are right for them and in support of their safe care at the direct care level.

| Study strengths and limitations
It is offered that, as a component of a larger investigation, the strength of this study is its depth and breadth, particularly related to the focus on both attitudes and behaviours across the continuum of care, and cross-referenced with findings from the qualitative study phase. 3 To our understanding, no other review has been conducted with this comprehensive approach. It provides perspective internationally about initiatives and efforts that are underway to engage patients in different elements of safety, and illuminated the gaps that remain. It is acknowledged that a study limitation may include the oversight of papers not identified in the applied search strategy.

| CON CLUS ION
This review was about mapping what is known of patients' and families' attitudes regarding their role in safety at the direct care level, as well as their reported behaviours in support of their safe care. The review included 151 papers, and among the findings included patients' belief in having a role in safer care (although not for everyone), with indication of the need for further investigation in this regard, as well as degrees of variability in taking action.
The review also provided perspective of the rapidly evolving interest in this topic, particularly as it relates to behaviours generally, and more specifically about the involvement of patients in asking providers about handwashing, although involving patients in the research process and specifically in patient-identified engagement safety strategies is needed. We must appreciate that many patients are engaged while recognizing their own and system limitations, and better position ourselves as researchers, policymakers and providers in understanding and implementing approaches collaboratively as relevant and feasible. Queen's University, in conducting this study and manuscript preparation are acknowledged.

CO N FLI C T O F I NTE R E S T
None of the authors have any conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analysed in this study.