Background

Patient Reported Experience Measures (PREMs) are now among the key indicators of performance used to determine healthcare value [1]. PREMs produce local, service and system-level performance data that are essential to direct quality improvement and service development [1]. Inclusive PREMs that capture data from communities who have high healthcare utilisation and poor healthcare outcomes are therefore important to determine their experiences, and target for improvements [2]. Continued underrepresentation of people from ethnically diverse communities in PREMs data means that quality of care concerns from these communities are not identified and addressed.

People from ethnically diverse backgrounds, who speak a language other than a national language at home, have one or both parents born overseas, and/or have low proficiency in the national language/s, experience higher rates of healthcare-associated harm and preventable hospitalisations than the general population [3]. Understanding the experiences of people from ethnically diverse communities via PREMs provides an avenue to drive person-centric improvements to redress this inequity in service delivery and care outcomes. Yet limited accessibility of PREMs in terms of their structure, content and approaches to administration can prohibit their completion to improve care for people from ethnically diverse backgrounds, along with several other priority populations.

People from ethnically diverse backgrounds face specific barriers in accessing and completing PREMs that are subject to intra- and inter-group variation. Factors such as language proficiency, digital and health literacy [4], trust in government, culturally inappropriate content and limited resources to support participation create barriers for people from ethinically diverse backgrounds to participate in PREMs [5]. Widely used PREMs instruments are closed-item surveys, include technical and complex language and phrasing, and contain between 50–80 items [6, 7].

Targeted strategies and methods to increase uptake and completion of PREMs among ethnically diverse communities may contribute to reducing barriers to participation [8]. Synthesising evidence from existing studies that have captured patient-reported experiential data provides insight into the strategies and methods that have been used and may be effective in increasing participation of ethnically diverse communities in PREMs. This knowledge may inform population-based PREMs instruments and data collection approaches. Therefore, the aim of this review was to identify evidence in the peer-reviewed and grey literature of the strategies and methods employed in patient-reported experience measurement with people from ethnically diverse backgrounds to inform policy and practice.

Methods

A rapid evidence appraisal (REA) methodology was utilised to address the review objective because this project was undertaken to inform policy and practice for NSW Ministry of Health, Australia. REA is widely applied to answer policy-related questions that require expansive literature to be explored to answer a focused question within a limited timeframe [9]. REA rigorously follows established systematic review methodology to search and appraise existing evidence, limiting selected aspects of the review process to shorten the review timespan while still enabling the depth of current knowledge to be appraised [10]. In this REA, the search was limited to three electronic databases to enable a breadth of literature to be explored including grey material. The Centre for Evidence-Based Medicine (CEBM) guideline for REAs was followed [10].

To ensure a search strategy that was both sensitive and specific, comprehensive search strategies were developed by a medical information specialist for the electronic databases of published literature and for use with grey literature. The search strategy was applied to the following electronic databases in June 2022 by the medical information specialist: Medline; PsycINFO and CINAHL. A research team member (MPI) applied the same search terms to the websites of health agencies in Organisation for Economic Co-operation and Development (OECD) countries in which understanding and improving patient experience has been identified as a key outcome in relation to value-based care. In addition, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses—PRISMA statement—was used to guide the reporting of this REA [11].

Inclusion criteria

Articles were included if they met the following inclusion criteria:

  • Types of publication: Publications available in English, reporting original primary empirical or theoretical work, and published from the year 2000 onwards, which is contemporaneous with exploration of patient experience in health settings.

  • Types of settings: Any healthcare setting, including but not limited to public or private hospitals, day procedure centres, general practice or other primary/community care in OECD countries.

  • Types of study design: Conceptual, theoretical, quantitative, or qualitative studies of any research design.

  • Types of population: Health care consumers from ethnically diverse backgrounds who access health services were included; defined as born overseas or who have one or more parents born overseas in a country where English is not a national language, and/or who speak a language other than English at home; and/or who have low English language proficiency.

  • Interventions: Strategies or methods to increase uptake and/or completion of PREMs.

  • Outcomes: PREMs included any form of data “from patients on what happened to them in the course of care or treatment” were eligible for inclusion [12]. In this review, the focus was on experiences of a healthcare encounter or a service rather than general attitudes or perceptions of healthcare.

Exclusion criteria

Articles were excluded if they reported general beliefs or attitudes about healthcare rather than experiences of a care episode, along with those that did not meet the above criteria, or reported reviews, protocols, opinion, or editorial pieces.

Study identification and selection

Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia) was used for study screening and management. Two reviewers (MPI, UC) screened the titles and abstracts in Covidence against the eligibility criteria. Full-text documents were obtained for all potentially relevant articles. The eligibility criteria were then applied to the articles by three reviewers (AC, MPI, UC). Four team members then met to finalise the eligible articles for inclusion across the published and grey literature (MPI, RH, RM, EM).

Data extraction and synthesis

A narrative evidence synthesis was undertaken to address the project aim of collating established experience measurement approaches and the impact of these methods on participation of ethnically diverse individuals [13]. Separate data extraction tools were developed for full-text articles and the grey literature and each tool was used to extract relevant information using a data extraction form created in MS Excel. Evidence synthesis occurred in stages and was conducted using a team-based approach involving seven members of the research team (RH, MPI, AC, CA, UC, RM and EM). Following the tabulation of initial descriptions of the included studies, their approaches and techniques, and the resulting impact on participation (where reported), team members individually reviewed the included articles. The group met to discuss key findings, to explore commonalities in current approaches and techniques that have been successfully applied, and to identify any challenges and mitigation strategies adopted. Through this group discussion, initial themes were generated and used to describe the evidence available. Two research team members developed the results content and shared this content with the wider group to further refine the identified themes.

Results

Search results

The systematic database search retrieved 1992 articles. After removal of 80 duplicates, 1912 articles remained. A total of 1461 articles were excluded after title and abstract screening. The remaining 443 articles underwent full-text review, of which 357 were excluded. A total of 97 documents were included, composed of 86 peer-reviewed journals and 11 documents from grey literature. Figure 1 demonstrates the search and selection process. Descriptions of eligible studies and results were tabulated. Tables 1, 2 and 3 show summaries of the included quantitative, mixed methods and qualitative articles from the electronic data search respectively. Table 4 shows the summary of grey literature articles.

Fig. 1
figure 1

Prisma flow diagram CALD report

Full size image
Table 1 Summary of included studies: quantitative N = 27
Full size table
Table 2 Summary of included studies- Mixed methods N = 9
Full size table
Table 3 Summary of included studies- Qualitative (N = 50)
Full size table
Table 4 Summary of included grey articles (n = 11)
Full size table

Characteristics of included studies

Database search

Studies in the peer-reviewed literature (n = 86) originated from: the United States of America (US; n = 29), Australia (n = 16), the United Kingdom (UK; n = 13), Sweden (n = 6), Norway (n  = 5), The Netherlands (n = 4), Canada (n = 4), Germany (n = 3), Switzerland (n = 2), New Zealand (n = 1), Denmark (n = 1), Greece (n = 1), multiple European Union (EU) countries (n = 1). Studies were conducted in: hospitals (n = 38), community settings (n = 29), primary care (n = 14) and dental services (n = 2), or in more than one setting (n = 3). Patients’ experiences were explored in relation to: maternity (n = 21), integrated care and/or care coordination (n = 12), mental health (n = 7), general practice (GP) (n = 3), emergency care (n = 2), gynaecology (n = 2),, pharmacy services (n = 2), convalescence care (n = 1), end of life care (n = 2), dental care (n = 2), student directed clinic care (n = 1), paediatric care (n = 1), radiation therapy (n = 1) and hospital care for COVID-19 (n = 1).

Of the total 86 studies identified from the electronic database searches, 27 studies used quantitative methods via survey, 50 employed qualitative methods and nine employed mixed methods combining surveys and interviews. Of the 50 qualitative studies, 37 (74%) used a single method of qualitative data-collection: individual interviews (n = 26) [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39], focus-group discussions (FGD) (n = 10) [40,41,42,43,44,45,46,47,48,49] or group interview (n = 1) [50]. Thirteen of the 50 qualitative studies used two or more methods for data collection [51,52,53,54,55,56,57,58,59,60,61,62,63] combining individual interviews with FGDs (n = 9) [52, 53, 55,56,57,58,59, 63], with participant observation (n = 3) [54, 60, 62], with group interviews (n = 1) [61] or with FGDs and a site visit (n = 1) [40]. Almost all studies were with adults aged 18 and older, except for two studies that included samples aged 12–20 years old. Experiences were reported directly from patients (n = 85), or including both patients and carers/support persons (n = 8) aside from one study collecting end of life care experiences from the next of kin [64]. Of the studies that employed survey methods, 12 were cross sectional [65,66,67,68,69,70,71,72,73,74,75,76] and eight were experimental; four pre- and post-intervention studies [77,78,79,80], three randomised control trials (RCTs) [80,81,82], and one longitudinal study (1 and 3-months post intervention) [81]. Sample size ranged from 24 to 138,878 in the quantitative studies and 9 to 219 in the qualitative studies. Seven population-based studies had samples of > 1000 participants and included a subset of people from ethnically diverse backgrounds, identified via demographic survey items.

Grey literature

The grey literature search yielded 11 documents that originated from Australia (n = 4), New Zealand (n = 3), the UK (n = 3) and the US (n = 1). Of the four documents originating from Australia, one was aimed at a national level, developed by the Australian Commission on Safety and Quality in Health Care [83] and three were aimed at state level, developed by the NSW Agency for Clinical Innovation and Clinical Excellence Queensland [84,85,86]. The two documents from New Zealand were aimed at the national level, both developed by the Health Quality and Safety Commission New Zealand [87, 88]. In the UK, the identified documents were from the National Health Service (NHS; which includes NHS England, Scotland and Wales) [89, 90] and The Kings Fund [91]. The document from US was developed by The Beryl Institute [92]. Five of the grey literature documents were reports on patient experience surveys, of which three explored services and care in mental health settings [84, 89, 91] and two explored health care experiences system-wide [89, 93]. Three documents explored patient experience and satisfaction items, indicators, and survey instruments used for patient experience measurement [83, 87, 88]. Two resources were fact sheets about the importance of collecting patient experiences and instructions about how to complete them [85], and one was a video discussing the importance of the collecting diverse patient experiences [92]. One report described a survey instrument which was used to collect real-time data via a tracker device with specific discussion of responding to diversity via the support of carers, although ethnically diverse populations were not explicitly identified [86].

Review findings

Evidence from the 97 included papers was explored in relation to the review aim to identify the approaches and techniques employed in PREMs to improve participation amongst people from ethnically diverse backgrounds. Findings led to the development of four key themes: i) strategies for identifying and recruiting ethnically diverse communities to participate in PREMs, ii) approaches to creating data collection instruments and processes to support PREMs completion by people from ethnically diverse backgrounds, iii) the patient experience topics of relevance to ethnically diverse communities, and iv) the application of patient experience data from ethnically diverse communities in healthcare. Findings are presented in relation to each of these four areas.

  • i. Strategies for identifying and recruiting ethnically diverse communities

Gathering patient-reported experiences of people from ethnically diverse backgrounds is contingent upon identifying ethnically diverse communities and engaging these communities in PREMs. Evidence from the peer-reviewed literature indicated that patient experience data were often captured from individuals or families whilst in a clinical setting. In 49 studies, PREMs were captured while individuals were visiting, being admitted to, or discharged from, hospitals [14,15,16,17, 21, 23, 25, 29, 31, 33, 36, 37, 44, 49, 51, 52, 54, 60, 62, 63, 65, 68, 73, 76, 81, 82, 94,95,96] or while individuals were visiting primary care clinics [18,19,20, 26, 30, 38, 42, 43, 54, 58, 59, 69, 70, 75, 78, 97,98,99,100,101,102]. Thirty-one further studies reported PREMs being gathered via community and/or online environments. For example, ethnically diverse respondents were often identified through community organizations or via attending social events using flyers and social media [72, 80, 98, 99, 102, 103]. Other methods for identifying ethnically diverse communities included using administrative data. Given the limited sociocultural data available regarding ethnically diverse communities, these methods were based on identifying ethnically diverse communities from lower socio-economic groups such as via the Index of Multiple Deprivation (IMD) in England to identify eligible people based on their postcode and derived diversity [104], or via birth registration records from the Office of National Statistics. Identified individuals were then mailed leaflets in multiple languages to seek ethnically diverse respondents [22]. One study sought experiences of refugees aged 18–25 years using mental health services, which used multiple modalities for recruitment (e.g., project flyers on noticeboards, mail-outs, and presentations to professionals and youth from refugee backgrounds).a multi-faceted approach [35].

Qualitative studies often used additional sampling techniques to try to access relevant ethnically diverse participants, which were often directed to a particular community or group of communities who had access to a specific service or were experiencing a specific health condition. Twelve studies used purposive sampling to reach their target population [14, 19, 25, 28, 29, 31, 32, 35, 45, 47, 53, 57], including stratified, purposive sampling to represent the dominant cultural and language groups in their community. Snowball sampling was reported in five studies [24, 27, 32, 47, 57] and convenience sampling was reported in two studies [27, 28]. Community leaders were used to identify potential respondents in two studies via referrals [24, 47].

Bicultural workers were often engaged in the process of identifying and recruiting ethnically diverse communities. Twenty-nine studies reported the involvement of multicultural and/or multilingual team members or community leaders. Bicultural workers included researchers, multicultural health workers, community networks, formal and informal interpreters. Bicultural workers contributed to projects at a range of stages. In the establishment of projects to aid identification and recruitment of ethnically diverse participants. Seven studies discussed the involvement of multicultural and bi/multilingual staff members in facilitating recruitment [19, 33, 44, 48, 58, 61, 63]. The strategies identified involved contacting eligible ethnically diverse participants via telephone or speaking to potential participants in person to explain the study purpose and to obtain informed consent. Materials were frequently created in ethnically diverse community languages required to aid recruitment of communities, with bicultural workers also supporting the recruitment process. Seventeen studies developed documents in community languages and/or engaged bilingual personnel in the consent and recruitment process to support language translation of study information [14, 15, 19, 20, 33, 36, 42, 44, 48, 51, 53, 55, 57,58,59, 61, 63]. Ten studies provided patient-information sheets and consent documents in relevant community languages [14, 15, 20, 36, 42, 51, 55, 57, 59, 63]. Bicultural staff also aided in developing interview and focus group discussion guides [19, 33, 44, 48, 58, 61, 63]. In three studies, bicultural staff were trained in preparation to contribute to study processes; and three studies discussed provision of training to bilingual research personnel [42, 46, 63]. The training was in relation to support the personnel’s involvement in data-collection [42, 46] and one study trained staff for recruitment and data collection [63]. Six studies reported providing cash incentives to patients [69, 70, 77, 80, 98, 103]. The grey literature provided little detail about how patients from ethnically diverse backgrounds might be identified or recruited to complete PREMs.

  • ii. Creating data collection instruments and processes to support completion

Several strategies were used to support the process of data collection of patient experience. Twenty-five of the 36 studies included surveys that had been translated into multiple languages to aid completion [65, 67,68,69,70,71,72,73,74,75,76,77,78,79, 81, 94, 97,98,99,100, 102, 103, 105,106,107]. Whilst using translated surveys was considered to enable a wider population to complete the surveys, no direct evidence of the impact of using translated surveys on the quality and quantity of data collected was reported.

Just over half of the surveys used (20/36; 56%) were previously validated, English-language instrument(s) [64, 65, 67, 71, 72, 74,75,76,77, 79, 94, 98, 101, 103,104,105,106, 108, 109]. Of which, seven were adapted based on pilot testing with a subset of the target population [65, 67, 72, 76, 94, 106, 109]. Changes that resulted from pilot testing of existing instruments were: modifications to the response options to respond to diversity [72, 109], increasing the appropriateness of the tool content for ethnically diverse communities [76, 106], and reducing the complexity and reading level [75]. Three studies reported using shorter versions of a survey to increase completion rates [75, 94, 97], with response rates ranging from 38 to 60% in these studies.

A further 10 studies created purpose-built PREMs for ethnically diverse populations [66, 68,69,70, 73, 78, 80, 81, 96, 100]. Of these, five PREMs were developed drawing items together from a range of existing validated surveys [66, 69, 70, 80, 100], and the remainder were developed as new surveys [69, 70, 96]. Six of the studies that developed new surveys reported pilot testing the surveys with the target population to ensure appropriateness and comprehension [68,69,70, 73, 78, 96]. Three studies reported conducting reliability or validity analyses [68, 75, 76].

Five documents from the grey literature discussed adaptations to PREMs data collection instruments to aid completion by diverse respondents. Two documents provided insight about PREMs administration with a variety of existing mainstream PREMs in Australia i.e. Victoria Patient Satisfaction Monitor, Picker survey, National Healthcare Agreement, Queensland patient survey, Press Ganey and Hospital Consumer Assessment Healthcare Providers and Systems (HCAPS). These documents proposed the use of a range of data collection methods via health services, mail, online, and telephone interviewing to respond to diverse population needs [83, 84]. Similar approaches to diverse data collection modes were identified by the Health Quality & Safety Commission New Zealand [87]. The Health Quality & Safety Commission New Zealand created additional questions toconsider the assessment of culturally safe care through patient experience measurement [88]. A comprehensive approach was undertaken to develop culturally safe and appropriate questions, which involved exploring available frameworks, designing questions and cognitive pre-testing of questions to check for relevance, importance, and clarity with a diverse group of patients. The resulting additional patient survey questions captured experiences of the following aspects of care: listening, engaging in shared decision making, kindness and comfort, respect, and recognition of personal needs (cultural, spiritual and individual). Other grey literature documents identified additional modalities for PREMs, including a patient experience tracker; a computer-based tool to collect real-time patient experience data [86], the use of computer-assisted personal interviews (CAPI) [93], and offering PREMs in multiple languages [84, 87, 89].

In addition to creating or adapting survey instruments, flexibility in the process of data collection was a feature of studies aiming to capture PREMs from ethnically diverse communities. In survey studies, PREMs were administered via face-to-face, verbal surveys (N = 10) [67, 69,70,71, 74, 78,79,80, 94, 101, 105], via telephone verbal surveys (N = 2) [96, 109], paper-based surveys (N = 9) [64,65,66, 74, 82, 100, 105, 107, 109], and online (N = 2) [70, 74]. The mode of data collection and location appeared to be associated with the completion rates in these studies. Response rates were reported in 16 survey studies, which ranged from 20 to 95%. Generally, surveys with higher response rates (> 70%) were completed on-site in hospitals/clinic waiting areas and/or completed as verbal surveys conducted face-to-face. The survey with the highest response rate of 95% included only three items [99]. The lowest response rates (< 30%) were for paper-based, postal surveys. [66, 109, 110]. Interview studies adapted data collection methods to increase uptake among ethnically diverse communities by conducting interviews in an individual’s preferred language (n = 7) [65, 80, 81, 95, 96, 98, 103] and in a location convenient for them, often going into communities (n = 4) [65, 95, 96, 103]. Studies also offered participants the choice of completing [15, 16, 26, 63] individual interviews or focus groups [63], or of interviews being completed face-to-face or via telephone [15]. Participants were also offered the option to conduct individual interviews or interviews in pairs [26].

Bicultural workers were often used to support the data collection process in addition to identifying and recruiting ethnically diverse populations. Bicultural workers supported data collection (n = 24) [14, 15, 20, 21, 24, 25, 31, 33, 34, 37, 42,43,44, 46,47,48, 50,51,52,53, 56, 58, 61, 63] and data analysis (n = 12) [14, 15, 24, 31, 33, 43, 44, 47, 52, 53, 55, 56]. Most (86%; 43/50) of the qualitative studies conducted data collection in the respondents’ preferred community language via bicultural workers or interpreters [14, 15, 17,18,19,20,21,22, 24,25,26,27,28,29,30,31,32,33,34,35, 37,38,39, 41,42,43,44,45,46,47,48,49,50, 53,54,55,56, 58,59,60,61,62,63]. In addition to supporting language and cultural needs, the need for flexible and supportive data collection processes to enable ethnically diverse communities to engage in PREMs was reported in several studies. Strategies included scheduling focus groups at a range of times, including evenings or weekends [43, 45], providing transport for participants to and from the data collection site [48], and providing childcare facilities [42]. The opportunity to include a support person/s was also identified as helpful in the data collection process and comprehension of materials. Seven studies reported enabling a suppport person, such as a family member, to be present during interviews or focus groups [16, 20, 23, 53, 55, 59, 63], and four survey studies invited the involvement of a support person to assist respondents to complete PREMs [66, 67, 102, 109].

Support for the process of collecting PREMs among participants from diverse communities, in the grey literature, was outlined in four resources. One organisation provided fact sheets translated into multiple language for patients about how to share their experiences [85]. While another resource was a video with easy-to-follow animations that encouraged patients from ethnically diverse backgrounds to participate in PREMs [89]. The collection of patient experiences through an underpinning lens of equity and bias was identified in two resources, which included a brief video for staff members to address and promote an understanding patient experience through the lens of equity and bias [88, 92].

  • iii. Patient experience topics of relevance for ethnically diverse communities

The included studies showed that most PREMs captured standard healthcare experiences commonly featured in mainstream population surveys, however aspects specific to ethnically diverse populations were notable. These were: healthcare navigation and access [64, 67,68,69,70, 72, 75, 78, 96, 97, 99,100,101,102,103, 105, 106, 109], care experiences of models and processes of care designed to be culturally relevant [65, 73, 74, 77, 79, 80, 94, 95, 98, 107, 110], cultural competence of healthcare teams and services [74, 78, 96, 97], experiences of patient-provider communication [64, 69, 70, 72,73,74,75,76, 82, 94, 100, 104, 106, 107, 109, 110], and of respect in the context of patient and clinician interactions [73, 74, 80, 101, 109, 110].

Health system navigation and access are topics of central focus in studies exploring migrant experiences of healthcare, but evidence of the impact of health system navigation and access problems experienced by migrants were apparent within studies of a range of services. Key examples are shown in relation to reliance on emergency departments for the treatment and management of chronic conditions and cancer due to lack of knowledge and understanding about how to access the health services via primary and outpatient care [68, 70]. Poorer experiences of models and processes of care that have been established for the mainstream population were also notable among migrant populations, with evidence of a lack of shared understanding of patient education materials, processes to coordinate care, and the actions expected of patients in their own care in service areas such as maternity and cancer care [63, 65, 67]. Poorer experiences of care were often reflected in low scores on patient satisfaction measures [67, 68].

Topics of importance further differed based on key features of ethnically diverse communities, such as ethnicity, country of origin and English language proficiency. For example, communities in which English language proficiency was low identified challenges in health system navigation and communication with healthcare teams more so than for groups with good English language proficiency [82]. In communities with low English proficiency, a language concordant health professional was identified as increasing satisfaction with care experiences [41, 97]. Communities that were well established within the host country also indicated fewer problems in understanding and accessing the healthcare system [82]. Cultural competence in relation to specific healthcare issues, such as the experience of cancer and of maternity care, in which expectations may vary based on cultural and religious background, emerged as important to consider in evaluating experiences of care [16, 23, 41, 51, 82]. Low healthcare engagement of migrant communities and opportunities to develop skills that enable self-management were also identified as factors contributing to poor healthcare experiences in these areas of care [23, 41].

Care coordination emerged as a critical topic for ethnically diverse communities, but was connected with the quality and availability of interpreter services and access to suitable health service information [63, 69]. Topics of importance to ethnically diverse communities may therefore include experiences of care coordination, particularly in relation to chronic conditions, maternity care and cancer, but require consideration of the features of additional support that are needed to create a positive experience. For example, experiences may be impacted by whether interpreters available for all aspects of care in which they were required, easy to access by the patient and of a good quality such that the patient and professional generated a shared understanding of the situation and next steps.

Communication and respect in the patient-professional dyad was a notable topic for ethnically diverse communities in their experience of care. Evidence from studies of chronic disease management demonstrated that the quality and clarity of communication with healthcare professionals (or lack of) was a key contributor to their healthcare experiences [53]. Studies that evaluated culturally tailored interventions, such as a family-based obesity intervention, further demonstrated that when people had culturally appropriate communication they had more positive experiences of care [18]. A series of language-specific focus groups revealed key indicators of a positive patient experience for ethnically diverse communities were compassionate, kind and respectful treatment, and the negotiated involvement of their family in health care decision making [44]. Lack of culturally competent behaviour among health professionals was perceived as leaving patients feeling powerless in their own care [44].

  • iv. Applications of PREMs from ethnically diverse communities in healthcare

Patients’ experiences were gathered and reported predominantly to understand experiences of services of new programs and interventions targeted to ethnically diverse communities (69%; 59/86 studies). Five of these studies gathered PREMs seeking to improve capability in intercultural communication among healthcare staff [65, 73, 105], in person-centred care and shared decision making that is responsive to patient diversity [73, 104] and for managing patient distress [66]. In these studies, PREMs data was used to create educational materials and programs for staff training and development. Beyond these five studies, the evidence generated from the reviewed literature was not described in terms of its application. While not being directly applied to create change, 33 studies did provide recommendations for future health system planning to improve experiences among the ethnically diverse community based on their PREMs results [14, 16, 17, 19, 24, 25, 28, 30, 34,35,36,37,38,39, 41, 42, 44, 46, 47, 49, 51,52,53, 57, 58, 61, 65, 67, 72, 80, 82, 104, 105]. Areas for action included improving patient-professional communication, understanding of culturally respectful behaviours [16, 23, 51,52,53], access to interpreters [16] and personalised and culturally relevant patient education [20, 23, 37, 40, 44, 46, 51, 53, 63]. The grey literature predominantly provided PREMs to be used for evaluating service delivery rather than reporting applications of PREMs with ethnically diverse communities to change care [83, 84, 89, 93].

Based on the review findings, a preliminary maturity model (Fig. 2) was produced to guide short- medium- and long-term action, which may have application to other diverse population groups.

Fig. 2
figure 2

Maturity model for diversity in PREMs

Full size image

Discussion

A substantial literature demonstrated the methods and approaches used to collect patient-reported experiences among ethnically diverse communities, with 97 papers included in the review. Data extraction and synthesis revealed five broad strategies employed throughout the process of PREMs development and collection that sought to increase uptake and completion of PREMs with a range of communities in community and healthcare settings. These five strategies were: 1) ensuring the project design was responsive to culturally and linguistically diverse needs when establishing the project team and processes, 2) targeting resources and communication about the purpose, process and application of PREMs, 3) embedding flexibility into processes of collecting PREMs, 4) creating inclusive data collection processes relative to the community needs and preferences, and 5) recognising the direct and indirect costs with reimbursement and remuneration.

Central to most identified strategies, was collaboration with bicultural/multicultural staff and community members who provided critical support in one or more stages of PREM design and collection. Bi/multi-cultural staff and/or community members contributed to creating and communicating information about the purpose of PREMs, supporting translation to ensure high quality, advising on suitability of the content and its complexity with reference to community needs and pilot testing materials. Engaging bi/multicultural team and community members to elicit patient-reported experiences offers a number of benefits, but requires consideration and planning at the outset of PREMs work to ensure mutual learning, empowerment, and adequate resourcing [111, 112]. Cross-cultural research highlights a range of considerations when working over cross-cultural boundaries [112].

As health systems move towards value-based care, PREMs are increasingly important as one of the key indicators to demonstrate that health services are improving what matters to patients [113]. Almost half of the Australian population have ethnically diverse backgrounds with one or more parents born overseas, making PREMs participation from these communities essential to provide representative data [114]. Recent evidence indicates that ethnicity is one factor that may be assocaited with different patient experiences [115]. This review demonstrates that there are methods that are commonly and successfully used to increase responses from ethnic minority groups. When considering the strategies that have been used to greatest effect, it is apparent that additional resource is required at all stages of PREMs collection to increase community engagement and responses. Requirements for public health services to achieve a representative sample of respondents are lacking along with the tools and resources to enable this. Despite strong traction towards value-based care, without these implementation supports, communities will continue to lack appropriate PREMs tools and the support needed to engage in their completion.

Implications

In the context of existing PREM instruments and processes, the maturity model resulting from this review has several implications for progressing PREMs engagement among ethnically diverse populations. Firstly, the model highlights the need to be able to recognise diversity in population groups. Limited and incomplete socio-cultural data is a barrier to this in many health systems worldwide [2]. The Australian context provides an example of how the need to capture information on diversity has advanced at a system level. In 2022, in addition their minimum core set of data variables (country of birth; main language other than English spoken at home; proficiency in spoken English, and Indigenous status), the Australian Bureau of Statistics (ABS) recommended collection of: country of birth of father and mother, religious affiliation, and year of arrival in Australia to increase the ease of identification of key features of diverse communities and factors that may impact on health and social well-being. Ensuring comprehensive routine capture of socio-cultural data via standardised data variables is a central system requirement to support better understanding of and provision for societal needs [107]. Access to these data may help to identify and facilitate comparison of PREMs results between services, cohorts and over time to track improvement where comparable measures are used.

With the target population identified, the second stage of the model identifies the need to respond to diversity via flexible and supported data collection methods. A wide literature exists on inclusive and cross-cultural survey content and methods, which demonstrates key aspects of survey behaviour that may be targeted to increase access and minimise bias in uptake, completion and outcomes from PREMs with ethnically diverse communities [108, 109]. Finally, the model depicts the need for a shift in the conceptualisation and design of PREMs that ensures measurement tools and processes are created to include diversity, with collaborative and community-partnership approaches championed. Aligning with the premise of person-centric healthcare, it is proposed that by developing PREMs topics, instruments and methods with community members, resulting tools and processes are likely to better suit community needs and preferences, leading to greater participation.

Limitations

As an REA, the search scope was limited to three electronic databases and relevant material may have been omitted from the review. Although a wide range of search terms were developed in consultation with a medical librarian who is experienced in constituting and executing literature searches, the concept of experience is broad and relevant material, which may have also constrained the sensitivity of the search process. The inclusion of grey literature enabled this review to identify current approaches used in health systems for PREM collection and the extent to which these have sought to respond to diversity.

Conclusions

A large literature has reported experiences of healthcare among ethnically diverse populations that demonstrate approaches that can be used to increase participation in PREMs collection among diverse communities. Five strategies have been predominantly used to increase PREMs participation, most often reported in research studies. Strategies include being responsive to community needs, targeting recruitment resources and communication, embedding flexibility in data collection processes, creating inclusive data collection processes, recognising the costs of participation and providing remuneration. Whilst strategies are available that appear to be effective in encouraging minority ethnic groups to participate in PREMs, there was limited evidence of their application in surveying conducted by public health system organisations. Public health systems must dedicate resource to enable the engagement of bi/multicultural health staff and community leaders to design and support the administration of PREMs if they are to ensure representative participation in PREMs across communities. In the longer-term, by working in collaboration with their communities, systems and services may co-create targeted PREMs that enable uptake and completion among communities.