Engaging patients to improve quality of care: a systematic review

Patient engagement has become a cornerstone of quality of care [1‐6] and is a frequently stated goal for healthcare organizations. Traditionally, and most commonly, this engagement has focused on the relationship between patients and providers in making care decisions or how to improve patient efforts to manage their own care [7]. However, there are growing efforts to integrate patients in broader ways, including efforts to improve or redesign service delivery by incorporating patient experiences [8‐12]. These efforts are due in part to an increased recognition and acceptance that users of health services have a rightful role, the requisite expertise, and an important contribution in the design and delivery of services [4]. While the nature of patient engagement may vary from including patients as members of a board to time-limited consultation with patients on service redesign, its aims are consistent—to improve the quality of care [11, 13, 14].

Healthcare organizations have a long tradition of measuring the experience of patients, and health service “users” including families, caregivers, and clients, with their services. Yet, traditional satisfaction surveys often prove difficult to translate into improved service delivery [15, 16]. Indeed, research on patient engagement has pointed to the importance of augmenting traditional surveys and complaint processes, moving towards fuller engagement of patients in reviewing and improving the quality of service delivery in institutions and in the community [17‐25]. This recognition has been accompanied by a growth in the development of instruments to measure and improve the quality of care patients receive. Over the past two decades, assessments of quality of care from the patient perspective have shifted from patient satisfaction to patient experiences [26]. Increasing literature indicates that it is not only feasible to involve patients in the delivery or re-design of health care [9] but that such engagement can lead to reduced hospital admissions [27], improved effectiveness, efficiency and quality of health services [28‐31], improved quality of life, and enhanced quality and accountability of health services [9]. Frameworks of patient involvement have been developed that move from the traditional view of the patient as a passive recipient of a service to an integral member of teams re-designing health care [8, 11]. For example, one framework developed by Bate and Robert (2006) describes a continuum of patient involvement, which ranges from complaints, giving information, listening, and consulting towards experience-based co-design of services [8]. Low-level engagement, such as consulting, comprises largely unidirectional feedback (e.g., focus groups, surveys, interviews), whereas high-level engagement, like co-design, represents a partnership in the design or evaluation of services. A more recent framework developed by Carman et al. describes various levels of engaging patients and families in health and health care, from consultation or involvement to partnership and shared leadership in various activities including direct care, organizational design, and governance to policy-making [11]. Carman’s continuum of engagement was influenced by Arnstein’s formative “ladder of citizen participation,” a continuum of public participation in governance ranging from limited participation to a state of collaborative partnership in which citizens share leadership or control decisions [32].

Governments and health care institutions are urged by some experts to engage patients and other service users, including caregivers and relatives in more robust ways [8, 33] where patients are actively involved as partners or co-leads in organizational re-design and evaluation of health care delivery, as depicted by the red section in Carman’s framework (Fig. 1). Despite the substantive body of research on strategies to engage patients and their effects on patients and health services, the literature is dispersed and has not been recently synthesized into a coherent overview. If the benefits of engaging patients in the design or delivery of health care are to be realized at an organization or system level, then effective strategies and the contextual factors enabling their outcomes need to be identified so that learning can be generalized. We conducted a systematic review of international English language literature on strategies for actively engaging patients and families in improving or redesigning health care and the contextual factors influencing the outcomes of these efforts. The explicit questions that guided our review were:

There were several limitations to this review. Despite the large number of initial search results, there was only a small number of studies focused on involving patients in co-designing health service improvement. Therefore, despite our best attempts, the specificity of our search criteria was modest, a problem familiar to systematic reviews in health services research, which typically crosses many disciplinary boundaries [38]. Future searches would benefit from improved keywords or MeSH terms on the topic of patient engagement. In addition, studies characterized health service users and their involvement differently, ranging from user-centeredness, patient-centered care, and user involvement to patient involvement or participation. Indeed, “user” was a common term used in the UK, whereas other terms such as “patient” and “caregiver” are commonly used in the USA and Canada. These different conceptualizations might signify important distinctions, and the use of different terms, and the publication of these papers across many different journals, raises challenges in identifying and analyzing this literature. We addressed this limitation by using multiple terms and search strategies across multiple disciplinary databases that incorporated terms used in similar reviews. We deliberately sought out the terminology used in key articles to expand our search though may not have captured the entire breadth of terms, such as “consumer,” a popular term used in Australian health services research. We echo previous work that identified this “conceptual muddle” as “one of the greatest barriers to truly integrating patient involvement into health services, policy, and research” [79].

There was also significant variation in sample sizes and populations included in these engagement studies. Samples sizes ranged from 3 to 372 participants and included a variety of patients, families, caregivers, service users, health care providers, staff, board members, health care managers, administrators, and decision-makers. Many studies did not provide details on their sample. These variations illuminate the absence of a standard approach for designing and reporting engagement initiatives. This variation may also reflect the variety of journals in which this research is reported. Additional limitations include the variety of methods used and the limited evaluation of the engagement methods themselves. Where there was no explicit evaluation of engagement, other information including authors’ discussion of strengths and limitations was used to assess the effectiveness of engagement. However, this does not specifically comprise evaluation of the engagement process or its outcomes on care. Development of evaluative metrics and frameworks for the procedural and substantive outcomes of engagements appears warranted. A final important limitation is that our search ended in 2016, and therefore, these insights may differ in the future given the rapidly growing field of patient engagement. This is a limitation familiar to systematic reviews but a future review may be warranted.

Despite these limitations, our study revealed key insights into the factors that influence the ability of health care organizations and decision-makers to create opportunities for engagement that are not provided in individual studies, which cross disciplines and geographical boundaries. We found that successful patient engagement resulted in culture change within the organization, meaningful collaboration and mutual learning, and shared or neutralized power, which tended to arise in settings where co-design is used. Optimal engagement often includes some of the following strategies: use of deliberative spaces to share experiences, external facilitation, broadening power and control to include users in all aspects of the process, flexible approaches for involving users, user training, clarity of roles and objectives, providing feedback, leadership by local champions and securing institutional and/or executive level commitment, and sponsorship from local authority by way of dedicated resources and on-going contact with management and executives. Leadership is key, but there may be a potential temporal trend in leadership actions; top-down approaches to patient engagement tended to be reported in earlier studies [40, 63] whereas more clinician or community-driven initiatives emerged from more recent studies [42, 77]. Another important factor is the timing of engagement. If the engagement occurred after a decision had been made, the success (or even function) of the engagement became highly questionable from the patient’s perspective. Taken together, this analysis suggests that co-design methods supported by executive sponsorship or driven by local champions that use externally facilitated, deliberative, experience-based discourse with trained users can promote successful patient engagement and outcomes.

Mental health settings emerged as a frequent venue for patient engagement in our review. The earliest reports in our review [61, 63, 80] are in this setting, suggesting that the therapeutic approaches, the nature of the population, or the orientation of mental health services might encourage greater patient participation in this area. Indeed, enabling service user involvement in care planning is a key principle of contemporary mental health guidance in the UK [81] and a potentially effective method of improving the culture and responsiveness of mental health services in light of a service history founded on aspects of containment and compulsion, and the stigmatization of those using mental health services [82]. Many of the co-design engagement activities that led to staff and organizational changes such as improved collaboration and mutual learning [42, 47, 76, 77], sharing or neutralizing power among patients and providers or staff [52], developing new competencies, and negotiating for service changes [39, 59] also occurred in mental health. While patient engagement is now occurring in many settings, the experiences in mental health settings serve as important examples of effective patient engagement.

Ultimately, the effectiveness of any patient engagement should be judged by its impact on patient care. There is a growing body of literature that indicates that engaging patients can lead to improved effectiveness, efficiency, quality of care [28‐31], health outcomes, and cost-effective health service utilization [27, 83, 84]. The outcomes reported in our review spanned beyond improved care to include enhanced governance and informed policies and organizational planning, which illustrates the breadth of quality of care initiatives that might be sought through patient engagement. However, drawing causal associations between engaging patients in health services improvement and health outcomes is difficult. Furthermore, it remains unclear whether these improvements translate into sustained or improved quality of care beyond local settings at a system level. Indeed, one study found a lack of evidence that patient involvement leads to the implementation of patient-centered care [85]. Some evaluative tools are emerging [86], yet more studies are needed that assess the conditions on which these tools and strategies can sustain the quality of care systemically.

Our review builds upon previous reviews in this field by providing insight into the associations between quality improvement methods and the varying system-level outcomes they yield. Indeed, our review echoes previous research indicating that patient engagement can lead to a multiplicity of health services outcomes with sufficient role definition, training, and alignment of patient-provider expectations but that the quality of the reporting has been poor and the full impact of patient engagement is not fully understood [87‐89]. Previous reviews have been limited to specific countries [87], care settings (e.g., mental health [89]), hospitals [90], or study design (e.g., qualitative studies [88]). In this way, our review provides a comprehensive perspective of optimal strategies used internationally, across care settings and using multiple methodologies to engage patients, caregivers, and relatives in quality of care improvement initiatives. Our review also provides novel insights into how the level of engagement influences the outcomes, namely, discrete products (e.g., development of tools and documents) largely derived from low-level engagement (consultative unidirectional feedback), whereas care process or structural outcomes (e.g., improved governance, care or services) mainly derived from high-level engagement (co-design or partnership strategies). If the benefits of engaging patients in the design or delivery of health care are to be realized at an organization or system level, then effective strategies and the contextual factors enabling their outcomes need to be identified so that learning can be generalized. Importantly, our review provides guidance on the effective strategies and contextual factors that enable patient engagement including techniques to enhance the design, recruitment, involvement, and leadership action, and those aimed to create a receptive context.

Future research would benefit from greater consistency in the conceptual, methodological, and evaluative frameworks employed. Greater emphasis is also needed on a procedural evaluation that assesses group composition, group cohesion or collaboration, equality of the participation, and the level of deliberation/reasoning. Such assessments are being developed in the deliberative democracy field [91] and could be informative in patient engagement initiatives. The limited evaluation of patients’ experiences is particularly ironic given the intent of these services to be patient-centered. Additional evaluative metrics should be developed to examine patients’ experiences. Finally, since it is difficult to draw causal relationships between patient engagement and health outcomes, future research should incorporate longitudinal measures and approaches to explore the impact of patient co-design on quality of care.

Several practice implications also emerge and reflect factors linked to the success of quality improvement initiatives more generally. Senior leadership support is critical to success since it increases the likelihood that the relevant decision-makers will implement the findings, and dedicated resources may encourage staff commitment to these efforts.

Despite the substantive body of research on strategies to engage patients and their effects on patients and health services, the literature is varied and dispersed. This study provides a comprehensive review of the strategies used to engage patients in service planning and design, identifies the outcomes, and contextual factors shaping optimal patient engagement to improve quality of care. Patient engagement can inform education, tools, planning, and policy (discrete products) as well as enhance service delivery and governance (care process or structural outcomes). The level of engagement appears to influence the outcomes of service redesign; discrete products are largely derived from low-level (consultative to co-design) engagement, whereas care process or structural outcomes mainly derived from high-level (co-design) engagement. Further evidence is needed to understand patients’ experiences of the engagement process and whether these outcomes translate into improved quality of care.