Background
Due to continuous migration, developed countries are becoming more diverse in their ethnic make-up, and the population of ethnic minority groups is increasing over time [
1]. Individuals from ethnic minority groups are known to be at higher risk of mental illness, to underutilise health services and to have lower participation in health research [
2-
4]. A systematic review of all published randomised controlled trials in panic disorder showed that only 44.7% reported ethnicity data for their included sample [
5]. Only 24% of US and Canadian studies have included Latino participants [
6], demonstrating a substantial underrepresentation of Latinos in clinical studies for OCD [
7]. Difficulties with recruiting research participants from ethnic minority groups is not solely restricted to mental health research; such problems have been encountered in range of other areas including research in nursing [
8], public health [
9] and cardiovascular disease [
10]. This low rate of participation among ethnic minorities reduces the generalisability of mental health research findings [
11] and impacts on the development of effective services and interventions [
9], leading to further widening of health inequalities [
12].
African-American adults have voiced their concerns about mistrust as the dominating barrier to research participation [
13]. However a review of research enrolment decisions of over 70,000 participants found ethnic minorities to be equally willing to participate in research [
14]. There is an immediate need to work towards developing appropriate strategies to improve participation among ethnic minorities in research [
12].
We published the only systematic review summarising barriers to the recruitment of ethnic minorities into mental health research [
15]. This review listed multiple barriers to recruitment and grouped the barriers into overarching themes, which we summarise in Table
1.
Table 1
Summary of Barriers to recruiting ethnic minorities to mental health research
10
Participant related barriers
|
Family/ community related
|
1. Explanatory models of illness | 18. Husbands’ influence |
2. Help-seeking/- ive attitude to psychotherapy | 19. Family perspectives |
3. Language spoken | 20. Stigma for family |
4. Religious beliefs and commitments |
Health services related
|
5. Trust | 21. Utilisation of mental health service |
6. Stigma | 22. Language of professional/intervention |
7. Gender | 23. Communication and cultural awareness between staff and participants |
8. Psychopathology/Substance Misuse |
9. Fear of being reported to immigration | 24. Staff personal attributes |
Practical Issues
|
Research process related
|
10. Lack of childcare | 25. Limited willingness and enthusiasm |
11. Transport provision | 26. Understanding the need for ethnic participation |
12. Financial constrains |
13. Culturally appropriate incentive | 27. Paucity of resources available |
14. Medical insurance | 28. Appropriateness of assessment tools |
15. Lack of time | 29. Non availability of translated materials |
16. Location of interview | 30. Lack of culturally competent staff |
17. Employment status | 31. Lack of culturally matched staff |
| 32. Under representation at recruitment Sites |
| 33. Understanding of consent process |
As a next step, we wanted to understand what recruitment strategies were recommended for overcoming barriers to recruiting ethnic minorities into mental health research. We extracted data from all included studies in our review [
15] with an aim to: (a) create a typology of recommended strategies to overcome barriers to research participation among ethnic minorities; and (b) match these recommended strategies to recruitment barriers. This will create a resource for researchers to assist with planning how to overcome recruitment barriers and increase confidence to enable the inclusion of more participants from ethnic minority groups.
Method
We revisited the nine [
16-
24] papers included in our published systematic review describing the recruitment of ethnic minorities into clinical studies (including trials and non-trials) in mental health. In the review, we combined search terms from three domains in Medline, Embase, PsycINFO and Cinahl: (1) participation or recruitment or retention; (2) ethnic minorities (exploded) or culture; (3) research or trial. Key words within the three domains were combined using “OR” and “AND”. We included English-language articles only. We did not aim to focus this review on any particular ethnic group or country of residence. Therefore no specific search terms for ethnic minorities or country were used. For details of the systematic review methodology please see the published paper [
15].
Data describing every strategy used to overcome recruitment barriers were extracted from each paper to create a typology of recommended strategies. Two authors (WW and AW) then matched these recruitment strategies to 33 recruitment barriers grouped into 5 broad categories that had been previously tabulated in our systematic review.
The matching of a recruitment strategy to a barrier is not mutually exclusive and any one strategy may be effective across several barriers, making some repetition within the review unavoidable.
Results
In our published systematic review the initial search identified 10,089 papers, of which 9 were included. All nine included papers originated from the United States. A brief description and the barriers described in the nine papers are provided in Tables
2 and
3 [
15].
Table 2
Description of included papers
| San Francisco, USA, Hospital | Latino | Depression |
Thompson et al. 1996 [ 17] | Detroit, USA, Two hospitals | African American | Schizophrenia or mood disorder |
| Mexico City, Mexico: Hospital, community health care centre | Latino | Postpartum depression |
Washington DC, USA: Hospital, community health care centre |
| San Francisco Bay, USA, Community clinic | African American | Depression or anxiety |
| Cleveland, USA, Community | African American | Depression |
Gallagher-Thompson et al. 2004 [ 21] | San Francisco, USA, Community | Latino | Dementia |
| New York City, USA, Neighbourhood health centre | Asian American | Depression |
| South Eastern USA, Aacademic medical centres | African American | Schizophrenia & Schizoaffective disorder |
Loue & Sajatovic. 2008 [ 24] | San Diego and North East Ohio, USA | Latino | Schizophrenia, bipolar disorder or major depression |
Table 3
Summary of proposed strategies
Participant Related Barriers
| Explanatory models of illness | | | | | | | √ | | |
Help-seeking/-ve attitude to psychotherapy | | | | | | | √ | | |
Language spoken | √ | | √ | | √ | √ | | | √ |
Religious beliefs and commitments | | | | √ | | | √ | | |
Lack of trust | | | | | √ | | | √ | √ |
Stigma | | | | √ | | | | | √ |
Gender issues | √ | | | | | | | | |
Psychopathology/Substance Misuse | | | | | | | | | |
Fear of being reported to immigration | | | √ | | | | | | |
Practical Issues
| Lack of Childcare | √ | | | | | | √ | | |
Transport Provision | √ | | | | √ | √ | √ | | |
Financial constrains | | √ | | | √ | √ | √ | | |
Culturally inappropriate incentives | √ | | | | | | √ | | √ |
Medical insurance | √ | √ | | | | | | | |
Lack of time | | | | √ | | | | | √ |
Location of interview | | | | √ | | | | | √ |
Employment status | | | | | | | √ | | |
Family/Community Related
| Husbands’ Influence | √ | | | | | | | | |
Family perspectives | √ | | | | | | √ | | |
Stigma for family | | | | √ | | √ | | | |
Under utilisation of mental health service | | | | | | √ | | | |
Language of professional/intervention | √ | | √ | | | | | | |
Lack of communication and cultural awareness between staff and participants | | | √ | | | | √ | | |
Staff personal attributes | √ | | √ | | | √ | | √ | |
Research Process
| Limited willingness and enthusiasm | √ | | √ | | | √ | √ | | |
Understanding the need for ethnic participation | | | | | | | √ | | |
Paucity of resources available | | | | | | | | | |
Appropriateness of assessment tools | | √ | | | | √ | | | |
Non availability of translated materials | √ | | | | | | √ | | √ |
Lack of culturally competent staff | | √ | √ | √ | | √ | | | √ |
Lack of culturally matched staff | | √ | | √ | | | | | √ |
Under representation at recruitment sites | √ | √ | √ | | | | √ | √ | √ |
Understanding of consent process | | √ | √ | | | | √ | | |
Explanatory models of illness
Help-seeking/negative attitude to psychotherapy
Psychopathology/substance misuse*
Addressing this barrier was beyond the scope of any of the research teams; this needs addressing at the level of health service provision.
Fear of being reported to immigration
Practical issues
Culturally inappropriate incentive
Underutilization of mental health services
Language of professional/intervention
Lack of communication and cultural awareness between staff and participants
Staff personal attributes
Research process
Limited willingness and enthusiasm of researchers
Understanding the need for ethnic minority participation
Paucity of resources available
No direct solution to this issue was made in the included papers.
Appropriateness of assessment tools
Non-availability of translated materials
Lack of culturally competent staff
Lack of culturally matched staff
Under-representation of ethnic minorities at recruitment sites
Understanding of the consent process
Discussion
Our previously published systematic review on barriers to ethnic minority recruitment in mental health provided us with an opportunity to look in depth into the interplay of various barriers to recruitment. It became evident that there is significant overlap between categories and they are often interlinked. This may suggest that multi-component strategies are needed to overcome these barriers.
Having developed a broad classification of barriers in the initial review, we next revisited the original papers to extract data on proposed solutions to these barriers. We made an attempt to find solutions related to each barrier and (as mentioned earlier), we found that solutions to these barriers were again overlapping, interlinked and often multifaceted. It is possible that, as the barriers are overlapping, that a strategy to overcome one barrier can in fact make a positive impact on others. However, it would be a mistake to assume that only a few recruitment strategies are needed to increase recruitment rates.
As a specific example of this, the introduction of culturally sensitive methods at the time of recruitment, without matching cultural adaptations in the intervention itself may reduce the impact achieved. Participants may consent to join the study, but if they find the intervention clashing with their cultural norms, their chances of continuing in the study would diminish.
Based on our own research experience and findings of these two reviews, it is important that researchers plan strategies to overcome these barriers at the proposal writing stage. Once the project has begun, providing extra human resources and meeting additional costs may be beyond the financial and staffing capabilities of the project. Initiating culturally sensitive strategies at the start of the study is not enough; the team should regularly monitor recruitment rates and if possible quantify the cost effectiveness of each strategy on enrolment into the study. This will help in identifying ineffective strategies and help direct funds into strategies that are effective [
25,
26].
In examining the barriers and the proposed solutions, it becomes evident that the solutions can be categorised based upon human and financial costs. We would like to propose the following:-
a)
Solutions that are not culturally specific and may also apply to majority ethnic (i.e. ‘white’) participants: This means that the proposed solutions are basically adopting good practice and apply to the general population but a cultural emphasis is needed when recruiting ethnic minority participants. Strategies like community engagement, provision of child care, transport and incentives will fall under this category.
b)
Solutions requiring additional resources specifically for ethnic minorities: Various types of solutions, such as procurement of extra materials, translations, adaptations and provision of multilingual staff will need extra financial input.
What we have documented in this paper is the accounts of authors about how they devised various strategies to improve recruitment. Authors have generally not been able to quantify the positive impact of these strategies on recruitment in a rigorous way that would demonstrate clearly their advantages over routine methods. There have been a number of published trials conducted to evaluate recruitment strategies [
27], but none have reported the effects of interventions for ethnic minority participants. A key next step to further the science of recruitment in this area is the formal testing of these strategies, using appropriate experimental or quasi-experimental methods. The optimal way of doing this is to conduct nested trials of recruitment interventions in ongoing randomised clinical trials, where some patients in the trial are randomly assigned to an ethnically sensitive method of recruitment, and some receive a standard model. In this way, the effectiveness and cost effectiveness of the various strategies may be formally tested [
28]. There may be some barriers to the use of such nested trials – for example, ethical concerns may arise if culturally appropriate recruitment is to be randomised. Other forms of evaluation may be required at times [
29]. However, demonstrating the impact of these strategies will be important in encouraging their uptake among research teams and funding agencies, especially where the strategies are associated with significant costs.
Another important point worth considering is that we need to further increase cultural sensitivity of research in ethnic minorities outside the US. It is a fact that ethnic minority groups are found in almost all developed countries, but their countries of origin vary considerably: in the US we mainly come across people of African-Caribbean and Spanish descent; in the UK the majority are South Asians; while in Europe there is a large representation from North Africa. There will be some commonalities in barriers to recruitment and strategies to overcome them, but there are clear cultural differences between these groups. In addition, there is a wide variation in health systems and a number of barriers are related to the peculiarities of the health service, highlighting the need for solutions to be tailored according to the structure of the health service under study.
Limitations
The main limitation of this review is that we have only given verbatim description of various strategies to overcome ethnic recruitment barriers as described by the original authors. Quantitative impact of these strategies on recruitment rates and the extra costs to implement these strategies were not fully described by the authors.
Conclusions
It can be a challenge to engage ethnic minorities in research; challenges that may be confounded in the context of mental health issues. Investigators need to employ a range of strategies to overcome these barriers, and our proposed typology of strategies provides some guidance. The strategies in this paper may not just be relevant for mental health patients alone and further evaluation of their impact in research across disciplines and in lower and middle income countries is also required.
Acknowledgement
Adwoa Hughes-Morley is funded by the National Institute for Health Research (NIHR), through a Doctoral Research Fellowship (Award Reference number: DRF-2012-05-128). This article presents independent research funded by the NIHR. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (
http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
WW and AW conceived and designed the study. WW and AW extracted data. All authors contributed to matching of strategies. WW and PB drafted the initial manuscript. AW, AHM and GA contributed to subsequent revisions. All authors read and approved the manuscript.