Rahul Alam School of Nursing Midwifery & Social Work University of Manchester Jean McFarlane Building University Place Oxford Road Manchester M13 9PL UK E-mail: Rahul.Alam@manchester.ac.uk
Diabetes is a chronic condition requiring lifelong self-management. Patients are encouraged to access appropriate services to facilitate optimum management of diabetes. Although equitable access to healthcare in the United Kingdom is a legal right, not all groups and individuals in the community experience equity. Despite various equality laws and numerous efforts to minimise health inequalities related to access, particular community groups are more likely to experience inequitable access than others. The Bangladeshi community are one such community who experience some of the worst diabetes-related health outcomes in the United Kingdom. Little is known about their experiences and preferences in accessing diabetes healthcare information and services. Consequently, we undertook a scoping review of the literature by following the York Scoping Reviews Framework to identify the experiences and preferences of Bangladeshi patients and carers when gaining access to diabetes-related healthcare information and services. We identified eight articles and reported our results in relation to four domains of access: health service availability, health service utilisation, health service outcomes and the notion of equity. The review identified that language and literacy issues were the most common barriers hindering access to information and services. Patient knowledge regarding diabetes and its management was generally low, and friends and family were frequently being used as information sources and as informal interpreters. Additionally, there were feelings of isolation from mainstream information and services possibly resulting in the high prevalence of depression in the Bangladeshi community with women more affected than men. Social networks combined with religious and cultural beliefs as well as wider societal duties played a crucial role in accessing information and services for this population, and the identification of these issues merit further research and are possible avenues towards improved access to healthcare information and services for the Bangladeshi population.
• British Bangladeshis experience one of the highest rates of diabetes in the United Kingdom.
• Bangladeshis experience significant difficulties with literacy and communication issues in English which hinder access to healthcare.
• There is a high reliance on the utilisation of children as interpreters in the Bangladeshi community.
What this paper adds?
• Social networks including family and friends are an important resource in accessing diabetes-related healthcare information and services.
• Poor levels of diabetes knowledge exist amongst the Bangladeshi community.
• Religious fatalism and psychological distress were familiar themes and access to psychological services was uncommon.
Diabetes mellitus is characterised by insulin deficiency and insulin resistance, and an estimated 285 million people are thought to be affected worldwide (DUK 2010). In the United Kingdom, it affects more than 4% of the population, affecting more men than women. In 2008, 145 000 people were diagnosed with diabetes in the United Kingdom equating to approximately 400 new cases diagnosed every day (DUK 2010). It is estimated that a further half a million people remain undiagnosed (DH 2008a). Type 2 diabetes affects more than 90% of diabetes patients and is one of the major risk factors for cardiovascular disease, the single largest preventable cause of death in the United Kingdom (DH 2001, ONS 2004, Scarborough et al. 2010) and worldwide (Lip et al. 2007).
Various factors contribute to the increased risk of developing diabetes, many of which are linked to poor socioeconomic status and include smoking, unhealthy diets and low levels of physical activity which can lead to obesity which is one of the main independent risk factors for Type 2 diabetes (DUK 2010). Additionally, there are strong genetic influences; individuals are 75% more likely to develop diabetes if both parents have diabetes (DUK 2010). Ethnic background is also strongly correlated with developing Type 2 diabetes and prevalence is up to three times more common in people of Black African and Black Caribbean origin and up to six times more common in people of South Asian backgrounds (including Bangladeshi, Indian, Pakistani and Sri-Lankan) (DH 2001, Chowdhury et al. 2006, DUK 2010).
Given the extent of microvascular and macrovascular complications associated with Type 2 diabetes (DH 2001, DUK 2010) and the enormous burden it places on individuals, families and national economies (Ramachandran et al. 2010), it becomes of paramount importance to provide individuals (including those at high risk) with the skills and knowledge to prevent the development of diabetes (DH 2001, 2008a). Adopting healthy lifestyle changes which include changes to diet, increases in physical activity levels and reduction in weight can all help towards diabetes prevention (NICE 2010).
Additionally, patients living with diabetes are required to attend regular screening as well as monitor and understand blood glucose readings (Ismail et al. 2004). Responding to dangerous symptoms of hypoglycaemia and hyperglycaemia, carbohydrate counting, adhering to oral medication and titrating insulin dosage for those on insulin are crucial components towards effective self-management (Ismail et al. 2004, Hirsch et al. 2005, Abrahamson 2010). Consequently, providing diabetes patients with the skills and knowledge to effectively self-manage their condition becomes extremely important (DH 2001, 2008a).
As such, patient empowerment and improving access to diabetes healthcare information on lifestyle issues, self-management, service availability, as well as improved access to diabetes services such as dietetic, podiatry, retinal screening, and psychological therapies, have been on the national agenda for over a decade (DH 2001, 2002, 2008a). Furthermore, the National Health Service (NHS) has made a pledge to ensure that healthcare services are appropriate to meet the needs of different groups of people in our society in order to reduce existing health inequalities (DH 2001, 2010). The most recent White paper by the Coalition Government on Equity and Excellence within the NHS states ‘Patients and carers will be able to access the information they want through a range of means, to ensure that no individual or section of the community is left out’ (DH 2010).
Where communities are not offered reasonable opportunities for equal access to and benefit from healthcare services, there are clinical consequences such as poorer health, higher mortality, reduced longevity, greater disease burden and consequently inequity (Marmot 2010). Whilst there have been commendable efforts by local and national government to reduce inequalities in access to healthcare services (Marmot 2010), there remains much disparity in the way some of the most vulnerable members of society, (including ethnic minority groups) gain access and negotiate their way through the healthcare system (Atkinson et al. 2001, Szczepura 2005, Randhawa 2007).
Although recent evidence suggests few ethnic inequalities in access to primary care services in relation to diabetes exist, inequalities in access to specialist secondary care services remain (Nazroo et al. 2009). Higher utilisation of primary healthcare services (which may be an indicator of greater need) (Nazroo 1997, Dixon-Woods et al. 2005, Sproston & Mindell 2006), subsequent lower rates of referral (Bhopal 2002, Gill et al. 2007, Fischbacher et al. 2009), lower usage of secondary (BHF 2007, Mindell et al. 2008) and tertiary healthcare services (Gill et al. 2007) have been reported for minority groups. Additionally, the evidence suggests that ethnic minority groups experience longer waiting times for appointments (DH 2004, Mead & Roland 2009), have fewer follow-up services (Scarborough et al. 2010), experience significant communication problems (DH 2008b), endure poorer health outcomes and are generally dissatisfied with healthcare services and the information received (DH 2008a,b). Unless these problems are addressed, these barriers are likely to increase the severity of illness and the rates of complications, placing additional demand on the NHS.
Although information alone is not sufficient for behaviour change and improved outcomes (Alberti et al. 2007), without appropriate access to healthcare information and services, desirable behaviour change is unlikely, leading to the poor uptake of services, poor self-care and ultimately poor clinical outcomes (DH 2004, Szczepura 2005, Knight et al. 2006). The provision of appropriate and accessible health education information has the potential to improve patient knowledge and awareness, thereby helping to prevent diabetes, promote self-management and improve long-term clinical and psychosocial outcomes (Nutbeam 2000, DH 2007).
However, providing relevant information and services in an appropriate manner and medium to patients with diabetes who possess poor communication and literacy skills is a complex and multifaceted problem (Greenhalgh et al. 2005, Grace et al. 2008). Whilst previous research studies have attempted to increase our knowledge in relation to the understandings, beliefs, attitudes and behaviours of Bangladeshi communities (Greenhalgh et al. 1998, 2005, Rhodes & Nocon 2003, Grace et al. 2008, Choudhury et al. 2009), there is little evidence on how the Bangladeshi community gain access to diabetes-related information and services and if those services meet the needs of the Bangladeshi population (Rhodes et al. 2003). It is therefore important to identify the Bangladeshi population’s preferences and experiences regarding access to information and services, including the content, complexity, delivery mode and settings. A greater understanding of the factors that can influence access, including identifying the barriers and facilitators to access, may lead to improved service delivery with the potential to improve the healthcare of patients.
We aimed to conduct a review of the literature to determine (i) the extent and nature of existing literature, (ii) to summarise the available evidence and (iii) to identify any gaps in the literature.
The University of York’s methodology for undertaking scoping reviews provided an appropriate framework to undertake such a review (Arksey & O’Malley 2005). The highly iterative process embedded in the methodology and its aim to capture all literature regardless of study design and quality (unlike most other approaches to mixed-methods synthesis) particularly suited this review.
We followed the five stages for undertaking a scoping review, which included (i) identifying the research question, (ii) identifying relevant studies, (iii) selecting appropriate studies, (iv) charting the data and (v) collating, summarising and reporting the results (Arksey & O’Malley 2005). Our scoping review did not include the optional consultation exercise (the sixth phase) (Arksey & O’Malley 2005). Furthermore, scoping review methodology does not aim to assess the quality of included studies, given the breadth of topics usually included and the potential diversity of study designs (Arksey & O’Malley 2005). However, we have provided summaries of quality assessment using the most appropriate tools developed by the NHS Public Health Research Unit (PHRU 2006) to enhance the review.
Framework stage 1: Developing the research question
Scoping reviews aim to achieve in-depth and broad results as they attempt to identify all relevant literature regardless of study design and quality (Arksey & O’Malley 2005). One of its key strengths is redefining search terms as familiarity with the literature grows. Unlike a conventional systematic review, this iterative process allows for the inclusion and exclusion criteria to be developed post hoc (Arksey & O’Malley 2005, Brien et al. 2010). This approach subsequently re-defined our research question and objectives and enabled the inclusion of literature that included patients with co-morbidities, carers and significant others.
What are the experiences and preferences of Bangladeshi patients and carers in gaining access to diabetes-related health care information and services?
More specifically, the present review aims to address the following questions:
1 What are the key barriers and facilitators affecting access to diabetes healthcare information and services for Bangladeshis?
2 What are the preferred sources and forms of information for Bangladeshis?
3 What are the levels of knowledge regarding diabetes within the Bangladeshi community?
Framework stage 2: Identifying relevant studies
The literature was searched using the specified search terms listed in Table 1 in the following databases from their commencement to January 2010: The Allied and Complementary Medicine Database (AMED) (1985–September 2010), CAB Abstracts (1973–September 2010), CINAHL (1980–September 2010), EMBASE (1980–September 2010), Evidence Based Medicine (EBM) Reviews (1991–2010), GEOBASE (1994–September 2010), Ovid Medline (1950–September 2010), PsycINFO (1950–September 2010) and the Social Sciences Citation Index (1980–September 2010). Additionally, the NHS specialist library for ethnicity and health (http://www.evidence.nhs.uk) and evidence-based reviews including the Cochrane database (http://www.Cochrane.Org) and DARE (http://www.crd.york.ac.uk/crdweb) were searched with adapted search terms where applicable.
Table 1. Medical subject headings used during electronic database searches
[Bangladesh* or Bengal* or Bangal* or Sylhet* or Shylhet*] (mP = ti, hw, ab, it, sh, tn, ot, dm, mf, bm, tc, id) AND
[diabetes or mellitus or NIDDM or MODY or T2DM or type 2 or type 1 or gestational or insulin resistance or impaired glucose] (mP = ti, hw, ab, it, sh, tn, ot, dm, mf, bm, tc, id) AND
(All types of studies were eligible, and therefore, restrictions were not placed on study design)
Outcomes of interest
[access* or gain* or obtain* or receive* or Information or education or knowledge or health care or services or primary or secondary or tertiary or doctor or nurse or literacy or Illiterate or uneducated or untrained or linguistic or communication or written or oral or interpret* or translat* or advice or educat* or skill* or Instruct* or coach* or tutor* or train* or learn* or taught] (mP = ti, hw, ab, it, sh, tn, ot, dm, mf, bm, tc, id)
To complement the electronic database search, we identified three online journals: Ethnicity and Health, Diversity in Health and Social Care and Health and Social Care in the Community which were additionally searched using keywords as search filters. Reference lists of key articles related to accessing healthcare for minority groups were scanned for potentially relevant articles. We also used Google Scholar to undertake searches using keywords. The search results were imported into Endnote to store citations and track the abstract and article review process. The search strategy was revisited and updated in September 2010 and is summarised in Table 1.
Articles that discussed access issues related to diabetes information and services for the Bangladeshi population were eligible. Articles that included co-morbidities, patients and carers as well as articles published in any language and in any country were also eligible. There were no restrictions on the study design, participant’s age and generation or type of diabetes. Articles that included mixed populations were excluded because of the heterogeneity between different ethnic groups (Bhopal 2002).
Framework stage 3: Study selection
The eligibility criteria were consistently applied by one reviewer to all the abstracts retrieved from the searches. Where there was ambiguity or uncertainty from the abstract alone the full paper was obtained and scrutinised for its eligibility in the review. By using the iterative process for scoping reviews (Arksey & O’Malley 2005) and not restricting the review to type 2 diabetes and including patients with co-morbidities carers and significant others led to the identification of four further studies that contributed to the overall number of included studies.
From a total of 773 references, 704 references were judged not to be relevant and were excluded based on the abstract alone. Sixty-nine full text papers were obtained and read in full by one reviewer. From these, 12 papers describing eight studies were eligible and included in the review. Four papers were duplicate publications reporting on a particular angle of an already included study. The duplicate publications were scrutinised and were judged to add nothing further to their associated papers, and on that basis, the duplicate publications were excluded (Figure 1).
The quality of studies were assessed by one researcher using the Critical Appraisal Skills Programme (CASP) tools developed by the Public Health Resource Unit, England (2006). Six qualitative studies were assessed using the qualitative CASP tool (Table 2) and the randomised controlled trial (RCT) was assessed using the CASP RCT tool (Table 3). For the purposes of illustrating the quality assessment process in Tables 2 and 3, only abridged versions of the CASP questions are provided. It was not possible to assess the quality of the pragmatic diabetes self-management education class during Ramadan (Chowdhury et al. 2003) as this was a process evaluation. The only CASP evaluation tool available was deemed inappropriate as this was an economic evaluation tool and the paper in question possessed no economic components. Despite this and considering that formal quality assessment are not typical for scoping reviews, the article was judged to contribute towards the overall aims of this review, and hence, we justify its inclusion.
Table 2. Quality criteria and results of qualitative studies (n = 6)*
2. Was this a randomised controlled trial and was it appropriately so?
3. Were participants appropriately allocated to intervention and control groups?
4. Were participants, staff and study personnel ‘blind’ to participants’ study group?
5. Were all the participants who entered the trial accounted for at its conclusion?
6. Were the participants in all groups followed up and data collected in the same way?
7. Did the study have enough participants to minimise the play of chance?
8. Were the results presented in a meaningful way with the main results being discussed?
9. Are the results precise enough to make a decision?
10. Were all important outcomes considered so the results can be applied?
Framework stage 4: Charting the data
The fourth step in the review process involves reading and re-reading the text to identify and ‘chart’ key emerging themes that support and contradict emerging ideas from all the papers included in the review (Ritchie & Spencer 1994, Pope et al. 2000).
A standard data extraction sheet was developed and refined as familiarity with the included papers grew to capture all relevant aspects of our research aims and questions. The template included descriptive characteristics of included studies such as author(s), country, setting, study design and more detailed accounts of individual study aims, patient experiences and preferences regarding access, barriers and facilitators to access, levels of diabetes knowledge, the study’s main findings and conclusions. We then utilised the conceptual framework of ‘access’ (availability, utilisation, outcomes and equity of services) (Gulliford et al. 2002) to chart our extracted data using Excel.
Framework stage 5: Collating, summarising and reporting the results
Stage 5 of the scoping review aims to provide a coherent structure to the literature so that it can be contextualized by the reader. However, unlike other scoping reviews (Weeks & Strudsholm 2008, Brien et al. 2010, Manthorpe et al. 2010), our review did not reveal a large body of evidence, despite the relatively comprehensive and broad search strategy used. It is likely that the dearth of evidence for this population resulted in the low number of papers retrieved. Nonetheless, we aimed to provide a narrative account of the included studies in two ways. First, we reported the results in relation to the four thematic components of access (Gulliford et al. 2002), and second, we used this opportunity to identify gaps in research knowledge.
There is a clear lack of research on access to health care for Bangladeshi patients with diabetes. Given that the few included studies varied considerably in their aims and designs, it was not possible to analyse the studies by study design. We provide a brief description of the studies below, followed by a narrative summary of the key themes emerging from the literature in relation to the various domains of access (Gulliford et al. 2001, 2002). Details of the eight included studies are provided in Table 4.
Understanding the beliefs of diabetes in the UK Bangladeshi population
Friends and family were important sources of advice on diet and exercise. However, the majority of participants had poor knowledge of the causes of diabetes and its prevention and were passive about their own self-management. Most patients claimed to understand information provided by their HCP regarding diet either because they spoke English or their HCP was Bengali. Some relied on children interpreters
Age range: 26–67
4 men and 10 women
One to one structured interviews undertaken at patient homes with diabetes patients who were invited to attend a peer education programme
Recruitment was through word of mouth, posters and announcements at the local mosque. Time constraints restricted the number of participants involved
The majority felt that diabetes education is important (although only 3 had attended education programmes), but some felt that current information was not useful, and some felt that they had to be content with the information provided by HCP as they had no other alternative. On occasions this led to patients accessing the help of children causing them to feel bad and guilty
Access to information and experiences were also influenced by location of practices and the availability of Bengali-speaking doctors
To understand lay beliefs and attitudes including that of religious teachings and professional perceptions in relation to diabetes prevention
80 participants without diabetes
Participants, particularly the first generation, identified poor fluency in English as a major barrier to accessing and understanding basic health information. Although knowledge of diabetes was found to be generally high, information was primarily obtained through the experience of diabetes in a relative or friend. There was a heavy reliance on friends and family to interpret health information for them
Mean age: 35 years
37 men and 43 women
29 Islamic scholars and religious leaders
Mean age: 35 years
14 men and 15 women
28 healthcare professionals
Healthcare professionals faced substantial barriers when attempting to communicate basic lifestyle information, including; time pressures, difficulties with interpreters and limited knowledge of the patient’s culture
Mean age: 41
3 men and 25 women
Focus groups and semi-structured interviews set in Tower Hamlets.
Religious leaders and lay participants felt that faith could be one mechanism by which information could be conveyed. Many first-generation participants held fatalistic views, and there were strong cultural norms (especially for women) which prevented the adoption of healthy lifestyles. Women were also more likely to have poor mental well-being and ill-health. Poor availability of local services and practical and material constraints restricted the utilisation of wider services. This made adopting a healthy lifestyle difficult for many
Recruitment of lay participants was through community centres, mosques and GP practices with information being provided in written English and Bengali as well as orally in Sylheti
To develop a complex intervention for diabetes education for ethnic minority groups who do not speak English
8–42 regular participants over an 18-month period based in a community hospital setting.
A weekly story-telling group revealed that the group of Bangladeshi women preferred an informal environment to discuss and learn topics. Conversations seemed chaotic with multiple conversations going on at the same time and often the HCPs stepped in to resolve disputes
(Patient characteristics not specified other than elderly Bangladeshi women)
The group was established through word of mouth
A narrative written by a bilingual health advocate describes how a Bangladeshi first-generation woman was experiencing significant difficulties and learning about the onset of diabetes led to depression. She was heavily reliant on wider family members as she did not know what to do (including administering insulin). The example reports how the link worker’s were able to bridge the communication and cultural gaps by monitoring her and informing the HCPs of her progress.
The main objective for the participants was to know what to do about diabetes and most would attend to check their blood glucose readings (despite being provided their own devices). The informal group provided a forum to exchange information and advice. The authors report an example of a participant’s GP dismissing claims that she had diabetes despite having very high blood glucose readings during the sessions
There was evidence that mean glucose concentrations of long-term attenders decreased over time. A focus group evaluation conducted 1 year from the beginning revealed that the ‘story telling group’ was well received and appreciated by the participants
To determine the effective-ness of a culturally adapted lay-led self-management programme
Randomised Control Trial
476 participants with diabetes and chronic disease comprising of cardiovascular musculoskeletal and respiratory conditions
Information during recruitment and during the intervention was provided through video cassettes and lay-led expert patients able to speak in Bengali, thereby overcoming language barriers
Patients receiving the education intervention showed improvements in self-efficacy, improvements in self-management behaviours and reduced HADS depression scores. There were no improvements in any other outcome measures including communication with GPs and frequency of visits to GPs and clinical measures were not collected
238 participants (66% diabetes)
Mean age: 48.9
105 men and 133 women
Follow-up work revealed that social and religious barriers hindered attendance. Wider societal duties such as caring, cooking and attending healthcare appointments took precedence
238 participants (71% diabetes)
Mean age: 48.0
99 men and 139 women
The cost of the programme was £123 per participant
The programmes were delivered by accredited Bangladeshi lay tutors living with chronic conditions in GP practices or community centres
Recruitment included writing to patients on the disease register in English followed by telephone calls by Sylheti speakers
To identify the health and social care needs of informal carers of Bangladeshi relatives living with chronic conditions
20 informal carers
Bangladeshi carers and patients faced numerous barriers when accessing health and social care services. The lack of knowledge regarding the availability of services was most prominent resulting in heavy reliance on close family and community networks and resources, especially through the use of word of mouth. There was a resistance to accepting formal care with fears that they would not meet cultural and religious needs
Age range: 16–62 years
4 men and 16 women
Nine cared for patients with diabetes, six for gastrointestinal problems, five for cardiac problems including hypertension, three for lung cancer and one for respiratory ailments.
The doctor was viewed as the most important source of information though the information was not always understood leading to the use of older children as interpreters and translators. This led to adverse effects on children such as damaging their education. Language barriers were common and the burden of caring was mainly on women causing significant suffering
One-to-one interviews with carers
Recruitment was through purposive and snowball sampling with information being provided in written English and Bengali as well as oral explanations
To provide an educational class on diabetes self-management during Ramadan
120 patients attended the group educational class (patient characteristics not specified).
The educational classes were well received and appreciated by all the participants (outcome measures were not reported). In addition to the classes approximately 5–7 calls were received each day and 5–7 visits were made by patients to the drop-in sessions
A 3-hour class involving Bengali diabetes link worker a senior diabetes specialist nurse and a GP. Additionally, a telephone help line service was provided for urgent advice as well as a drop-in session
In addition to the HCPs, advice was offered by religious leaders re-enforcing health messages and dispelling myths with regards to fasting obligations and health. However, patients were unwilling to attend outpatient clinics or have blood tests carried out during the month of Ramadan due to strong religious beliefs. The authors report to overcoming this by not booking outpatient clinics over this time period
Recruitment was through written invitations, and advertisements with posters at GP’s surgeries, diabetes clinics, local mosques and community centres
The authors also suggest that in coming years fasting will become longer as a result of the Islamic lunar calendar when Ramadan begins 10 days earlier each year. The summer months will be particularly challenging for diabetes patients and the authors highlight how the Bangladeshi population may need additional help in coming years
An investigation into access from the point of view of Bangladeshis
12 Bangladeshi first-generation patients with type 2 diabetes
Many patients accessed their GP’s for information and services but felt that their needs were neglected by their HCPs and the patients were often resigned to a fatalistic acceptance of the minimal information and services they were provided with. Patients were largely unaware of the care they might expect or of the full range of services available. There were views that the information provided was not in appropriate formats and that HCPs had little knowledge and understanding of patient needs
Age range: 43–75 years
4 men and 8 women
One-to-one semi-structured interviews, carried out by a Sylheti speaker in patient’s homes
Recruitment involved purposive sampling from disease registers followed by sending out letters and telephone calls by a Sylheti speaking interviewer
The presence of Bengali-speaking doctors reduced communication barriers for some, but there were still reports of poor levels of diabetes knowledge and wide spread dissatisfaction with the care they received, which included service level barriers such as appointment systems and poor availability of alternative sources of information
Friends/family and community networks were seen as an important source of information and children (of working, college and school age) were frequently used as interpreters causing significant disruption to patients and their children
Women were particularly disadvantaged through their heavy reliance on family members, fear, lack of social support and poor access and familiarity with pubic transport
To investigate the health beliefs and folk models of diabetes in British Bangladeshis
40 patients with diabetes
Lay sources of information was the most popular source of information with 17 of the 18 participants in the structured vignette agreeing that this was the best way to learn about diabetes. However, overall knowledge regarding diabetes (management and preventative care) appeared to be poor. Participants reported other practical difficulties in living with diabetes including language barriers when speaking to HCPs, the use of children as interpreters and the inability to understand written leaflets, the lack of availability of local facilities. Women were reported to be particularly affected as they rarely left their homes
Age range: 21–80
23 men and 17 women
Qualitative study using narratives, semi-structured interviews, focus group, structured vignettes and pile sorting exercises
Recruitment included sending letters followed up by telephone calls, opportunistically through the practice reception as well as manually searching through patient case notes
Fatalistic attitudes were common amongst the group, and many informants reported the presence of physical or psychological stress
The majority of participants viewed the doctor as an important authoritative and knowledgeable professional, and nurses were sometimes viewed as a source of information and advice
Studies highlighted how healthcare professionals also found it difficult to communicate basic lifestyle information to Bangladeshis (Rhodes et al. 2003, Grace et al. 2008). Their limited understanding of cultural aspects of lifestyle as well as the difficulties with interpreted consultations and the lack of availability of time compounded the problems (Rhodes et al. 2003, Grace et al. 2008).
Low car ownership, low social economic status (Rhodes et al. 2003, Grace et al. 2008), limited familiarity in using public transport services (Choudhury et al. 2009), fear of travelling beyond their locality because of street crime and poor fluency in English (especially for women) (Greenhalgh et al. 1998, Grace et al. 2008) were barriers to access. One study reported how Bangladeshi patients were being referred back to their GP from specialist care as a result of not being able to keep clinic appointments owing to practical rather than clinical reasons (Rhodes et al. 2003).
Availability of Bengali speaking healthcare professionals
Limited literacy and communication difficulties affected utilisation
Poor communication skills and poor fluency in English hampered the ability of patients to utilise services to the best of their ability. There was a general lack of awareness and knowledge of the services and support available to patients and carers, and this led to limited utilisation of services such as diabetes education programmes (Choudhury et al. 2009), support services from nurses and health visitors (Merrell et al. 2005), and dietetic, retinal screening and podiatry services (Rhodes et al. 2003). Language barriers also delayed utilisation of services (sometimes for years) owing to absent family members, particularly for women who required interpreters and chaperones and who were more reliant on wider family members to communicate with healthcare professionals (Greenhalgh et al. 1998, 2005, Rhodes et al. 2003, Merrell et al. 2005, Grace et al. 2008, Choudhury et al. 2009).
Effects of tailored interventions on service utilisation
Four interventions provided tailored educational programmes to Bangladeshis. Of the two structured interventions, rates of utilisation were low. For the Expert Patient Programme, approximately 50% of those who registered, attended (Choudhury et al. 2009). Similarly, 51% of intervention participants attended three or more sessions of the Chronic Disease Self-Management Programme (Griffiths et al. 2005). The less structured components of the Expert Patient Programme were reported to be preferred by participants (Choudhury et al. 2009).
Rates of service utilisation for the remaining two interventions are more difficult to interpret as they had no pre-determined targets. However, the Ramadan diabetes education classes were utilised approximately 300 times over a 3-month period (through a mixture of classes, ad-hoc phone access and drop-in access) (Chowdhury et al. 2003). The action research programme, which was also less structured, grew from eight to 42 regular attenders over an 18-month period (Greenhalgh et al. 2005).
Cultural and religious influences affected service utilisation
Religious events such as Ramadan also affected service utilisation. During this time, patients were less likely to attend outpatient clinics and less willing to have their blood tests taken (Chowdhury et al. 2003) or have their dietary needs met in hospital settings (Merrell et al. 2005). Similarly, Griffiths et al. (2005) report that social and spiritual barriers prevented attendance at their intervention. Some studies utilised religious leaders in their work to address religious misconceptions held by participants. Chowdhury et al. (2003) utilised religious leaders to re-iterate the need to abstain from fasting for vulnerable groups such as those with erratic blood glucose control and pregnant women. Grace et al. (2008) reported how religious scholars disapproved of religious fatalism and attributed this to misinterpretations of Islamic teachings. Choudhury et al. (2009) appointed a religious leader to be the peer educator in their programme (for the purposes of trust and credibility from within the community).
The present review highlights some important experiences and preferences (including barriers and facilitators) in relation to accessing information and services by the Bangladeshi population. Various factors affected this population’s access to healthcare information and services from availability, utilisation, outcomes and in relation to the equitable provision of services. It is likely that barriers in one domain of access such as the availability of information services (or lack of) had an effect on other domains of access such as utilisation, outcomes and equity. Consequently the design and delivery of healthcare services must consider all aspects of access simultaneously to ensure equitable access.
One of the major recurring themes was the poor levels of literacy and communication difficulties which appear to have compounded the effects of poor access for this community. The poor availability of linguistically competent services is likely to have limited the ability of patients to understand healthcare information, communicate effectively with healthcare professionals and have limited knowledge of wider healthcare services that are available. This may have led to a greater reliance and mobilisation of personal resources such as drawing on family and friends and wider community networks for information, advice and support. These findings are consistent with a previous review under the broader remit of disadvantaged groups which included South Asian minorities (Dixon-Woods et al. 2005), and the present review supports this assertion specifically for the Bangladeshi community. However, although language and literacy barriers can be overcome through family, friends and community networks, the exchange of advice and information may not be accurate and there are dangers of undue burden, pressure and breach of privacy, which have been well documented (Cohen et al. 1999, Rhodes & Nocon 2003, Gerrish et al. 2004). Children were frequently reported to act as interpreters during healthcare appointments; however, it is not within the scope of this review to discuss the advantages and disadvantages of using formal and informal interpreters. The studies provided little information on the nature of children as interpreters such as their gender, age and occupation which may explain why particular informal interpreters were often chosen or preferred. Similarly, studies contained both low literate and illiterate patients and carers, most likely with varying degrees of literacy and communication skills. Consequently, the findings may only be generalisable to a broad spectrum of low literate Bangladeshi patients. Nonetheless, despite the acknowledgement of language and communication barriers for many years (Hawthorne 1994, Gerrish et al. 2004, Dixon-Woods et al. 2005), the findings of this review suggest that the problems associated with this important domain of access remain to be addressed. Short of deploying more Bengali/Sylheti speaking doctors, nurses and other healthcare professionals nationally or ensuring that English is made compulsory for all citizens accessing healthcare services, it is clear that an intermediary position that is both acceptable to the patients and cost-effective for the NHS is adopted. Although the evidence on how language and literacy issues can be best overcome for the Bangladeshi population is scarce, there is some evidence to suggest that link workers/support workers may be effective in conveying key health information (Curtis et al. 2003, Raghavan et al. 2009). Work with other minority groups have suggested for similar approaches in order to minimise the language and cultural disparities that exist (Khanchandani & Gillam 1999, Lawton et al. 2006a). However, with no nationally set criteria for staff training and in the absence of long-term evaluations of such schemes (Lawton et al. 2006a), their feasibility and effectiveness remain largely unknown and merit further research.
In addition to communication difficulties, the utilisation of services also appeared to be influenced by strong cultural and religious beliefs. The issue of ‘religious fatalism’ was present in the review and resonates with previous work in diabetes for Pakistani and Indian populations (Lawton et al. 2006a,b, Ludwig et al. 2011) and for the Bangladeshi population in relation to maternal health (Noor & Rousham 2007), heart disease (Netto et al. 2007) and healthcare delivery (Abbott & Riga 2007). Nonetheless, the present review does highlight the potential for community organisations and religious scholars in promoting self-care behaviours for patients with diabetes. Similarly, others have suggested for the potential involvement of community/religious organisations in health promotion and service delivery for the Bangladeshi population (Abbott & Riga 2007, Netto et al. 2007) and other populations (Dessio et al. 2004, Yu 2010). However, to date there has been little debate regarding the appropriateness of such specialised services (Jabbour & Fouad 2004) and there is little evidence of successful implementation, impact and feasibility which all merit further research.
It is important to note that whilst one study reported adequate levels of knowledge amongst the community (Grace et al. 2008), the majority reported poor and dangerously low levels of diabetes-related knowledge. The findings of improved knowledge may be due to repeated exposure of diabetes interventions in the same geographical area. Of the eight diabetes studies conducted with this population in the United Kingdom, this was the fifth study undertaken in the East End of London (Grace et al. 2008). The poor levels of knowledge amongst the remaining papers (Greenhalgh et al. 1998, 2005, Rhodes et al. 2003, Griffiths et al. 2005, Choudhury et al. 2009) could be partially explained by poor access to appropriate information sources and poor communication between Bangladeshis and healthcare professionals. However, given that some participants utilised family interpreters and some had access to Bengali speaking doctors, it is probable that external factors other than that of communication affected patient understanding and subsequent knowledge levels. Indeed, some studies highlighted a preference towards a participatory learning style which involved blood glucose testing for example (Chowdhury et al. 2003, Greenhalgh et al. 2005, Choudhury et al. 2009). It is not within the remit of this paper to discuss adult learning styles for low literate populations; however, it has been suggested that the simple exchange of didactic information may not be the most effective method for the acquisition of knowledge for this population (Greenhalgh et al. 2005).
Our findings also alluded to the presence of psychological distress in the Bangladeshi community (particularly amongst women who seemed to be more socially isolated than men). Poor awareness of services, poor knowledge of symptoms and poor access to mental health services have previously been illustrated for this population in the United Kingdom (Parvin et al. 2004). More recently, a study conducted in rural Bangladesh reports over 30% of women with diabetes to have symptoms of major depression with a higher proportion of women being affected than men (Asghar et al. 2007). Whilst the influence of psychological well-being to patients living with type 2 diabetes has long been recognised (Greenhalgh 1997, Ismail et al. 2004, DH 2008a), access to psychological services for the Bangladeshi population in this review was uncommon. The reasons for this were not explored; however, there is a further need to identify the reasons for such poor service utilisation for this aspect of their health.
It appears that the Bangladeshi community continue to experience poor access in relation to diabetes. This review identifies various findings and gaps in our knowledge which may help towards improving access to diabetes healthcare information and services for the Bangladeshi community. However, given the dearth of evidence because of the limited number of papers retrieved in this review, the diversity of their aims and their moderate quality, these findings need to be interpreted with caution. It is clear that further research (both qualitative and quantitative) are required on the various domains of access to determine the most fruitful approaches towards ensuring the equitable and appropriate delivery of healthcare information and services. Unless attempts are made to improve the current situation, it is likely that the Bangladeshi population will continue to receive variable standards of care.
Strength and weaknesses
One of the strengths of this review was that all the studies were conducted in reasonably homogenous Bangladeshi populations in the United Kingdom, thereby removing confounding factors associated with different healthcare systems. Publication bias was addressed by removing multiple publications pertaining to the same study and by searching through a wide range of sources. The scoping review methodology provided an iterative and flexible strategy to incorporate relevant articles, and this was particularly useful, given the scarcity of evidence for the Bangladeshi population. Furthermore, the four thematic components of access (Gulliford et al. 2002) provided this review with a robust framework for data extraction and analysis. We were flexible in our interpretation of the framework and this enabled us to categorise emerging themes that overlapped, accurately and objectively.
Weaknesses include the failure of the majority of papers to provide detailed information on access issues, participant characteristics such as assessed literacy levels and characteristics of children interpreters, the limited number of studies retrieved and on occasions, the small sample size of studies. Additionally, the inability to assess the quality of one paper as well as data extraction and analysis being conducted by one researcher owing to time constraints are weaknesses of this review.
Sources of funding
Funding to conduct this research was provided by the Greater Manchester Collaboration for Leadership in Applied Health Research and Care (CLAHRC), Health Care Services Research Theme.