Review of Barriers to Engaging Black and Minority Ethnic Groups in Physical Activity in the United Kingdom

Introduction and Objective: The lower physical activity levels in Black and Minority Ethnic (BME) groups as compared with general population in the United Kingdom (UK) could relate to barriers to engaging these groups in physical activity. Hence, the aim to conduct a review to examine UK primary studies reporting barriers to engaging BME groups in physical activity. Method: This is a narrative review of literature from 1970 to 2008. The search looked for English literature from five bibliographic databases (MEDLINE, Embase, CINAHL, PsyINFO, Ethnicity and Health). Broad search terms ‘physical activity and minority’ were used and views from BME groups were considered in this review. Results & Conclusion: The search yielded 391 studies and 18 were finally included in the review. Our review identified 20 barriers clustered among four broad themes of: (a) perceived personal barriers; (b) socio-economic barriers; (c) cultural barriers; and (d) environmental barriers. Overcoming these barriers in these broad areas is important in development of sensitive multicultural health promotion addressing physical inequalities.


Introduction
Emerging evidence suggests that Black and Minority Ethnic (BME) groups have lower physical activity levels than their white counterparts and this may be one potential reason why BME groups are at higher risk of non-communicable diseases (Fischbacher et al. 2004;Haase et al. 2004;Health Survey for England 2004;Sport England 2006).Primary prevention of non-communicable diseases is paramount in public health in developed countries, and even more so for ethnic minorities.We use the United Kingdom (UK) as an example in this review as little is known about barriers to engaging in physically activity among BME groups in the UK and challenges of overcoming these barriers are critical in planning and implementing culturally sensitive interventions.
Physical inactivity remains a major risk factor for chronic diseases and obesity.It is clear that having inactive lifestyles has contributed to increasing rates of heart disease, stroke, diabetes, depression and cancer, and to poor mental health (Adams et al. 2007;Alessio 1997;Cooper & Brill 1988;Commandre et al. 2000;Santaularia 1995;Vuori 2001;Thompson 1994;Sproston & Mindell 2006;Warburton et al. 2006).Globally, 1.9 million deaths are attributable to physical inactivity (WHO 2002a) and yet the proportion of people meeting recommendations of physical activity is generally low.In public health practice, recommendations are: 30 minutes of moderate-intensity physical activity 5 days per week or 20 minutes of vigorous-intensity physical activity 3 days per week for adults.The World Health Organisation (WHO 2002b) and evidence-based reviews by the American College of Sports Medicine and American Heart Association (Haskel et al. 2007), Strong et al. (2005) and Department of Health, UK (2004) supported this assertion.Of interest are findings in BME groups not meeting recommended activity levels as compared to general population (Fischbacher et al. 2004;Haase et al. 2004;Health Survey for England 2004;Health Education Authority 2000).In the same way, research has shown that ethnic minorities stand a higher chance of having non-communicable diseases in UK (NRCEMH 2004;Fischbacher et al. 2004;Haase et al. 2004;Rudat 1994).For instance, the prevalence rate of diabetes in BME groups is three to four times higher in Scotland's BME population (NRCEMH 2004, NRCEMH/Diabetes UK Scotland 2006) and the risk of heart disease in this group is higher than the general population (Bhopal 2000;Gill et al. 2004).Stroke mortality rates in African-Caribbean are up to four times higher than the general population in England and Wales (Chaturvedi & Fuller 1996).
The BME groups are so diverse in terms of migration history, culture, language, religion and disease profiles.Defining 'ethnicity' is hard as the concept is multidimensional and used as a synonym for 'race' and there is no consensus on appropriate terms for use in the scientific study of health by ethnicity and race (Bhopal 2004).The term 'minority' may suggest a sociological group that does not constitute a politically dominant voting majority of the total population of a given society.They may be migrant, indigenous or landless nomadic communities.The term 'Black' is commonly referring to Africans and Afro-Caribbeans, but may also signify all non-white minority populations (Bhopal 2004), such as Indian, Pakistani, Bangladeshi and South Asian populations (Phillips 1998).
As physical activity inequalities may have resulted from barriers of participating in physical activity, commonly mentioned barriers in both BME groups and the general population are related to safety, availability, cost, personal barriers e.g.lack of time, health concerns, and lack of motivation.(Eyler et al.1998) and the social influence of peers (Keresztes et al. 2008).For BME groups 'restricting' factors such as culture, knowledge, strength of beliefs in health benefits of activity may be contributing (Haase et al. 2004).Increasing primary studies on physical activity issues among BME groups prompted this narrative review in order to give more dimension and insight into likely causes of physical activity inequalities.Overcoming these barriers is critical in developing and implementing interventions in multicultural environments.

Objectives
This review aims to undertake a narrative review of relevant literatures relating to barriers and influence involved in physical activity participation among BME groups.
Using a systematic approach will help to provide evidence of what is known about the dimension of barriers and influence to physical activity.This review brings together the findings of both quantitative and qualitative primary studies in the UK.

Literature Search
Five electronic bibliographic databases were searched from 1950 to April 2008, including MEDLINE, EMBASE, CINAHL, PsyINFO, Ethnicity and Health including government websites; NHS Scotland Library, Health Technology Assessments (HTA), National Institute of Health and Clinical Excellence (NICE).The review used combination of MeSH terms, free text word and truncated terms of 'physical activity', 'ethnic group', and 'barriers' were used in search strategy.The MEDLINE search strategy was adapted for searching other databases.Reference and library source searching plus hand-searches was also performed to identify relevant studies.A total of 391 articles and reports were identified.

Study Selection
Studies are included if they contained reported and separate BME groups results related to barriers or influence of physical activity in the UK.The sifting of literature performed in stages against inclusion criteria was initially based on 'title', and then 'abstract' that resulted in 71 potentially relevant references.The retrieved full texts of 71 references were similarly assessed to yield a total of 13 papers.Figure 1 summarises the sifting process with inclusion and exclusion criteria.Five papers were identified from hand-searched literature to make total of 18 papers that make up this systematic review.
A data extraction form was developed and piloted for the purpose of this review.Full data extraction was performed by one reviewer and checked by a second reviewer.The qualities of studies were assessed by nine-item checklist constructed specifically for this review (copy available from first author).The quality criteria were informed by several sources (Higgins 2006;Thomas et al. 2003;Spencer et al. 2003;NHS Centre for Reviews andDissemination 1996, NHS Health Scotland 2008;Greenhalgh 1997).Discrepancies between reviewers were resolved through discussion.

Description of Studies
Findings are reported as narrative summary and in tabular form.The search strategy produced 391 articles.Out of these, 18 met the inclusion criteria of reporting separately the barriers of physical activity among BME Groups in the UK.The study designs of the included studies were of various types: nine studies used solely qualitative method, three were quantitative, and the remaining six studies used mixed method (Table 1).The qualitative studies employed methodologies that range from case studies to semi-structured interviews, and in-depth interviews to focus groups discussions.There were three intervention studies identified (Snape & Binks 2008;Jolly et al. 2007;Carroll et al. 2002), one evaluated the intervention programmes by way of a RCT (Randomised Controlled Trial) (Jolly et al. 2007), while others gave case studies reports of the interventions.Sample sizes ranged from six to 175 in the qualitative studies, and from 25 to 4452 in the quantitative studies.All studies considered perspectives of BME groups although three papers included views from teachers (Khunti et al. 2007), programme staff (Snape & Binks 2008), GPs (General Practitioners), Health Authorities and leisure centres (Carroll et al. 2002).
It is the aim of this review to consider only UK studies, we found that all papers were conducted in England except one (Lawton et al. 2006) which was conducted in Scotland.We also found that five of the studies expressed solely women's views and it is important to note that all these come from South Asian populations.One paper took its perspectives solely among men (Naeem 2001) and the rest included male and female.In the majority of studies the participants are adults above 18 years, only two studies explored the views of schoolchildren under the age of 15 (Khunti et al. 2007;Mahoney 1992).
The sample frames for included studies were drawn from variety of ethnic groups, settings and gender.Ethnicity was described by all the studies with 10 out of 18 papers restricted to South Asians mainly Indians, Pakistan, and Bangladesh (McKenna & Ludwig 2008;Sriskantharajah & Kai 2006;Carroll et al. 2002;Lawton et al. 2006;Khanam & Costarelli 2008;Farooqi et al. 2000;Snape & Binks 2008;Burney 2006;Khunti et al. 2007).Six studies combined mixed populations of South Asians, Whites and/or Afro-Caribbean (Greenhalgh et al. 1998;Mahoney 1992;Jolly 2007;Darr et al. 2008;Health Education Authority 1999;Rai & Finch 1997) while one study each focused on Afro-Caribbean (Molokhia & Oakeshott 2000) and West Africans (Elam et al. 2001).In methodological quality, the studies had clearly defined purpose and gave adequate descriptions of the sampling and justification for data collection in general, but details for sample validation and assessment of generalisability varied widely.There was limited evidence of triangulation with other sources and limitations of studies made clear by authors.Details of the studies' aims, design, samples and assessment of qualities are summarised in Table I and IV.

Physical Activity Barriers
Our review identified 20 barriers clustered among three broad themes of perceived personal, socio-cultural and environmental barriers.With no attempt to quantify results in this review, the range of barrier outcomes is appreciated (see Table III).Whilst some barriers are distinct, some are inter-related for example lack of time may be due to work or family commitment.Lack of time was the most reported barrier (n = 8) but further probing for reasons was poorly reported (Tables II and III).
Identifying literatures from databases have been difficulty as majority are qualitative studies which tend to use creative titles that provide no information on barriers of physical activity among BME groups.The included studies can be broadly classified into a) those that focus on physical activity alone (n=8), and b) physical activity and other lifestyles (n=10).The groups included in the studies included patients with diabetes, cardiovascular heart disease (CHD), osteoporosis, and obesity.Studies were generally poorly reported and inconsistent questions asked to identify physical activity barriers.The physical activity preferences of BME groups were poorly explored by included studies.

Discussion
This review revealed a limited literature on physical activity among BME groups, and even less investigating the barriers to engaging in physical activities among the same UK population.Most studies explored the views of South Asian population as compared with other UK minorities.All studies conducted in England with exception of only one which was conducted in Scotland (Lawton et al. 2006).The information on methodology and results provided by most studies varied widely in terms of how it was presented and the level of details reported.This may be due to strict word-length restriction by some academic journals.
The difference in physical activity levels between BME groups and general population revealed by Fischbacher and colleagues may contribute to increase risks of non-communicable diseases such as diabetes, hypertension (Fischbacher et al. 2004).The strength of this review lies in providing better insight on the range of barriers that are consistently identified in several different studies related to BME groups.Knowledge of barriers to engaging in physical activity among this population may prove useful in the development of physical activity interventions for BME groups.The findings here are consistent with a review that focused on children in the general population (Brunton et al. 2003).
We identified inter-related personal, socio-cultural and environmental barriers as classified by several other studies (Andajani-Sutjahjo et al. 2004;Booth et al. 1997;Browson et al. 2001;Cleland et al. 2008).For example, personal barriers relate to lack of motivation; socio-cultural barriers are associated with lack of family support, religion, language barriers; and environmental barriers have to do with lack of information, access, safety, cost and time.Of interest is the cultural influence on physical activity in which participants negatively perceive exercise.In this case there were beliefs that exercise connotes 'illness', as informed by lay concept of exercise in Bangladesh's local language 'beyam' (Greenhalgh et al. 1998); the belief that women are not allowed to walk alone in the evening (Khanam & Costarelli 2008); and also the perception of exercise as sexual connotation since it makes you look attractive for seduction purpose (Snape & Binks 2008).The attitudes of participants towards physical activity were affected by other cultural factors such as language barriers, religion and cultural dress codes, unavailability of women only sessions at sport facilities.Ruland and Moore (2001), Wilcox and colleagues (1999) and Burke and colleagues (2006) have all emphasized the influence of ethnic background on physical activity preferences.Our review pointed out lack of evidence as related to activity preferences.Understanding BME groups' preferences is crucial alongside evidence of barriers to physical activity in developing strategies of effective Physical Activity interventions.
This review falls short of assessing effectiveness of interventions used to promote physical activity among BME groups because only one study evaluated intervention promoting physical activity adherence (Jolly et al. 2007).The majority of studies were epidemiological and exploratory focusing on beliefs and attitudes towards healthy lifestyles and risks of diseases.
Challenges faced in searching for studies exploring the views of BME in UK may reflect either paucity of literature in this field or failures of search strategy to capture literature since creative titles not including keywords being used mostly.Hence, we cannot exclude the possibility of some studies being missed out in this review.A detailed quality assessment was not conducted in this review due to heterogeneity of study designs and the lack of consensus in the literature on quality criteria in use for mixed methods and qualitative studies (O'Cathain et al. 2008;Hannes 2008).As indicators of quality criteria in the field of Ethnicity and Health, we looked into justification for designs and sampling, ethical issues, language interpretation and triangulation of results with other sources.This is as informed by sources (Higgins 2006;NHS Health Scotland 2008;Thomas et al. 2003;Spencer I. 2003; University of York NHS Centre for Reviews and Dissemination 1996; Greenhalgh 1997.There is, however, a great need for development of standardised tools for assessing mixed methods.
The results of most of the studies may not be generalisable, this is more so because failure or inadequacy of the studies to include control groups, hence the result might have been over-interpreted.

Conclusion
This review has given more insights to potential causes of what is currently known in this field, namely that BME groups do have lower physical activity levels.Evidence of personal barriers, socio economic, cultural and environmental barriers that influence physical activities have been described in this paper.Overcoming these barriers in these broad areas is essential in development of sensitive multicultural health promotion addressing physical inequalities.Future research needs to focus on different contexts of physical activity preferences among BME groups as such evidence alongside the identified barriers to engaging in physical activity could help develop effective physical activity interventions targeted at BME groups.The paucity of literature in this field necessitates future research to take opportunity of including ethnic minorities in all research in UK with separate analysis, thus having foundation for more evidence and external validity of studies.

Figure 1 .
Figure 1.Flow diagram: Searching and sifting process

Table 2 .
Findings of Studies' Barrier outcomes

Table 3 .
Classification of BarriersGlobal Journal of Health Science

Table 4 .
Quality Criteria of Studies