A systematic review of explanatory factors of barriers and facilitators to improving asthma management in South Asian children

The PICO and SPICE acronyms are established models for aiding systematic searches [10]. SPICE was used to delineate all elements of our research question.

The original aim was to source all articles including primary research, reviews, abstracts, conference proceedings and unpublished studies with a focus on minority ethnic community children in any country or South Asian children in minority or majority settings. Due to the vast literature, only full text primary research was included in the final analysis.

Studies were included if they met the following criteria:

An information scientist with specialist interest in ethnicity research derived the search strategy and searched the following databases and web resources supported by JW, LM and DB: MEDLINE, HMIC, EMBASE, ASSIA, Web of Science – SSCI and SCI, BNI, CINAHL, PsycINFO, OpenSIGLE, HTA, DARE, Scopus, Social Care Online, NHS Evidence, The Cochrane Library, Campbell Collaboration; CRD York, NICE, British Thoracic Society; American Thoracic Society; European Thoracic Society; Asthma UK. 5636 citations were returned.

Bibliographies of relevant articles, asthma management guidelines and previous systematic review were also hand searched by DB, LM and JW: 3735 citations were searched. Databases were searched for evidence from 1990 onwards with English-language translations available. Initial searches were conducted in 2009 and updated in May 2010.

Asthma management: relates to a process of treatment with the aim of achieving and maintaining asthma control.

Barriers: factors that impede the achievement and maintenance of asthma control and that include (but are not limited to) health beliefs, professional barriers, language and communication, prejudice or discrimination, delivery and organisation issues, issues around diagnosis, and treatment offered.

Facilitator: the inverse of a barrier (factors that assist achievement or maintenance of control).

South Asian: those of Indian, Pakistani or Bangladeshi descent.

Children: persons less than 18 years of age.

The search strategy included terms for “asthma”, “South Asian” (including terms specifying all major subgroups), and “children”. For example, the search strings used for Medline were:

GP, DB, LM and JW screened 9371 titles and abstracts against inclusion criteria: 9217 were excluded at this stage. A further 154 full papers were retrieved and scan-read where eligibility could not be determined by abstract alone: 108 papers were excluded at this stage. This left 46 potentially eligible articles for full text review and data extraction. Two reviewers assessed these papers independently, conflicts being resolved by discussion with the team. The study selection process is denoted in Figure 1.

GP extracted data using a piloted modified CASP worksheet [11] including: country of study; study aims; population studied, eligibility criteria and illness diagnosis; study design; ethical approval; sampling; data collection and analysis; presentation of the results; discussion of the findings; bias; value of the research; summary critique and limitations. DB conducted a reduced data extraction for all potentially eligible studies, and a full data extraction for 20% of the potentially relevant studies, with 95% agreement between GP and DB.

The Centre for Reviews and Dissemination (CRD) guidance emphasises the importance of using a structured approach to quality assessment when assessing qualitative studies for inclusion in reviews. However, it acknowledges the lack of consensus on the definition of poor quality [12]. Some argue that using rigid quality criteria may lead to the unnecessary exclusion of papers [13], and suggest that instead of excluding papers on quality grounds, a critique be included in the narrative to allow the reader to decide. This is the method used in this review.

The eligible studies tended to address very broad research questions, were conducted using qualitative and/or quantitative methods and were not presented in the format of controlled trials. The evidence reviewed is presented as a narrative report together with a critical reflection of the synthesis process used to assess its robustness [13‐15].

15 studies comprising of 25,755 children aged between 2 and 17 years (age range not specified in one paper), 18,483 parents/carers of children and 239 healthcare professionals working with children were included. 14 studies were quantitative (13 involving questionnaires) and 1 was qualitative. Outcomes measured included knowledge/understandings/beliefs about asthma (7 studies), symptom perception/understandings of ‘wheeze’ (3 studies), prescriptions and use of preventer medications (2 studies), impact of migration and ‘westernisation’ on asthma and atopy (2 studies). Two studies took place in Pakistan, five in India (South Asian majority) and eight in the UK (South Asian as minority).

For ease of reading the data have been grouped into themes, corresponding to a barrier or facilitator with evidence for any explanatory factors. These are denoted in Table 1.

None of the included studies specifically explored knowledge with UK resident South Asian families. Two studies did this in South Asian families resident in Pakistani and India [16, 17]. Both identify that parental knowledge was lacking but neither paper explored the reasons for this.

Poor knowledge about asthma was identified amongst family physicians in Pakistan [18], [50% of physicians failed a 17 item questionnaire and 52% failed on management approach]. Physicians in India fared better with >75% of clinicians correctly answering questions on treatment importance, preventative medications use and steroid safety [19], however this study was biased as physicians could opt out of questionnaire completion. In both studies, newer graduates scored better than older graduates, suggesting that failure to refresh knowledge amongst older professionals may be explanatory. Assuming physicians as the primary information source for patients and families it is possible that poor physician knowledge translates to poor family knowledge [20].

Treatment non-compliance is a frequent explanation for poor outcomes in minority ethnic communities. The evidence however, supports two alternative explanations: under-diagnosis of asthma and under-prescription of preventer medications. One UK based prevalence study reported significant discrepancies between the percentage of children with diagnosed asthma and those with symptoms of asthma (labelled probable asthma) [21]. In contrast, another UK based prevalence study noted that inhaled corticosteroid [ICS] prescriptions were lower in South Asian compared to White children [OR 0.46, CI 0.28-0.77 in 1-year-olds, and 0.63, CI 0.44-0.91 in 2–4-year-olds], despite no evidence for asthma severity variation [3].

Finally, in a large UK based quantitative survey it was noted that Indian Subcontinent children with acute asthma were less likely to use the correct administration method or receive medications (OR 0.35, CI 0.22-0.57), compared to White and Other inner-city children (OR 0.82, CI 0.49-1.38) [22]. This suggests that under-prescription in South Asian children is related to the South Asian ethnicity and not to being a minority group. Whilst it appears unlikely that such under-treatment in the UK is related to medicine unavailability, exploration of such associations in both UK and South Asia in the literature remain sparse [3, 22].

Two studies in India and Pakistan reported that 48% and 61% of parents respectively did not accept their child’s asthma [17, 20]. Stigmatisation may be explanatory. In two studies based in India and another in Pakistan it was noted between 26% and 37% of parents believed asthma to be contagious [16, 17, 20]. In one, 50% of parents hesitated disclosing to others about their child’s asthma [20]. It is suggested that these findings reflect asthma associated social stigma impacting on treatment compliance.

This is reflected in a UK based focus group study which noted Bangladeshi participants suggesting that ‘other’ Bangladeshi people view asthma as contagious and that asthma associated stigma was present in Bangladesh but not in UK [23]. This was also noted in another UK study where Indian and Pakistani parents were the most likely to believe that their child did not have asthma compared to White parents [OR 4.04, CI 1.04-15.76] [24]. Although there appears to be an ethnicity-specific belief that asthma is contagious and stigmatising, the prevalence and impact of such beliefs may be lessening amongst minority (i.e. diasporic) South Asian population as a result of acculturation.

Several explanatory factors were identified for over-reliance on ED care. One may be inaccurate symptom assessment or underestimation of asthma severity. In an India-based questionnaire survey of Indians, the majority of patients were unable to assess their symptom severity with only 11.9% able to perceive warning signs of an acute attack [25]. A similar study undertaken in India compared the accuracy of parents and their children in recognising asthma severity symptoms and noted 62% and 60% accuracy respectively [26]. When peak flow measurements were performed, 92% of parents and 89% of children had underestimated asthma severity, most noticeably amongst those with severe asthma compared to mild/moderate disease [26].

Another factor may be beliefs concerning appropriate place of treatment. A UK-based study noted that South Asian parents were noted to more likely believe that ‘better help and support’ was available via hospitals rather than their GP [OR 4.27, CI 1.67 to 10.96] [24]. This was in contrast to another study of Bangladeshi mothers in the UK that noted that whilst GPs are recognised as the first step to treatment they would go to ED if their GP was inaccessible [23].

Communication encompasses many elements, one of which is language. Two papers support the existence of a language barrier in South Asian children asthma management. In a questionnaire study to identify parental understanding of wheeze in UK White and South Asian parents, 83.5% correctly identified wheeze as a whistling or squeaking noise. A correct answer however, was less likely if parents did not speak English as a first language [10]. This study however was hampered by low response rates and the administration of 3 study questionnaires within 5 years that may have led to ‘correct’ answer learning.

In another UK-based cohort study examining understandings of asthma and wheeze a large proportion of Bangladeshi mothers were noted to need help with translated questionnaires and suggested that language barriers may interfere with symptom reporting ability [27]. With the absence of language barriers being reported in data from South Asia we can conclude that language barriers reflect being in a minority position rather than an ethnicity-specific issue, although clearly language barriers may exist with specific ethnic groups within India, and language may be the defining feature of ethnic-difference that is associated with minority status among people of migrant origin in Europe and elsewhere.

Treatment non-adherence is a well-known barrier to asthma management. In the UK, it was reported that South Asian parents were less likely to give regular inhaled corticosteroids (ICS), more likely than White parents to believe that medicines were addictive [OR 3.89, CI 1.47-10.27] and that medicine does more harm than good [OR 3.19, CI 1.22-8.34] [24]. Similar results were noted in South India that reported that amongst 100 children requiring acute asthma treatment (i.e. nebulisers) at least once, only 12 children used inhalers at home. Additionally, parents reported that aerosol therapy was addictive and continuous medication use during symptom free intervals would impair the child’s ability to grow out of asthma with 47% of parents not aware that inhalers could be administered at home [17]. This supports the notion that knowledge and beliefs are closely linked and that misconceptions around medication side effects may underlie non-adherence.

Medication format may also be an explanatory factor. 82% of Pakistani parents in Pakistan felt that inhalers were superior to oral treatment, despite 77% currently using oral treatments [16]. In contrast, a questionnaire study of parental attitudes towards western and complementary medicine in India noted only 43% of adult and children were prescribed inhalers and only 48% of children chose to use inhalers over oral medications [25]. It suggests that reluctance to use inhalers relates to stigmatisation, with unproven examples of girls in particular wanting to hide that they had asthma as it may impact upon their marriage prospects.

Complementary medicine is widely practiced in South Asian countries. In a Pakistan-based study, Hazir et al. [16] noted that whilst 98% of Pakistani parents preferred going to the doctor for treatment, 9% also visited a Hakim [traditional practitioner] and 11% a homeopath. Lai et al. [20] reported that 65% of Indian parents sought treatment outside the hospital whilst Singh et al. [25] noted that 79% of Indian parents used alternative therapies to manage their child’s asthma, including ayurvedic treatments, homeopathy and yoga. Finally, in a UK based study, it was reported that Bangladeshi mothers held generally positive attitudes towards the medical profession and western medicines whilst generally accepting western medications as means of controlling asthma symptoms [23].

Whilst complementary therapy use is sometimes suggested as a barrier to asthma management or an explanation for non-adherence to western medications it is not clear whether alternative remedies are used in conjunction with western treatments, or instead of them. Nor is it clear why complementary therapies are used. Though easy to presume that such remedies are traditional, it is possible that parents turn to complementary therapies to fill perceived gaps in western medications. From the evidence, data that links barriers to explanatory factors and management behaviours is lacking.

It was noted in a UK based study that whilst White British mothers were concerned about preservatives and food allergies and Black Caribbean mothers advocated junk food avoidance and the importance of a healthy diet, Bangladeshi mothers identified specific foods to be avoided by a child with asthma [bananas, cold milk, ice-cream] and commented that “doctors here don’t even mention restricting foods”. Additionally, specific foods and drinks were reported as avoidable triggers in 89% of parents in a Pakistani study [16], 68% of parents in an Indian study [17] and an unspecified percentage in a second Indian study [20].

Furthermore, 40% of GPs in Delhi [19] believed that drinking milk increased mucus production in asthmatic children, and that asthmatic children should not eat bananas [64%], chilled food [24%], dairy products [53%] sour food [35%], nor drink chilled liquids [41%]. These beliefs were closely correlated to length of time in practice, with more experienced (i.e. older) GPs being more likely to hold these beliefs than newer GPs. Thus dietary modification, over and above healthy eating or food allergy avoidance, appears to be a factor specific to South Asian families, not a reflection of minority status.

Very little was identified relating to facilitators to achieving asthma control in South Asian children. South Asian families appear to have a highly positive attitude towards healthcare professionals: Smeeton et al. [24] found that Indian and Pakistani parents were the most likely to accept/believe that their child had asthma [OR 4.04, CI 1.04-15.76] and Lai et al. [20] reported that Indian parents of asthmatic children viewed their doctors as the major and most reliable source of information. As the prevailing attitude amongst South Asians is to have a strong degree of trust in medical professionals, this should be sensitively explored and utilised to assist families in gaining control over asthma.

Diet may also be protective. In a UK based prevalence survey, Carey et al. [28] reported that for South Asian children, a traditional Asian diet was inversely associated to airway hyper-reactivity risk in a dose response fashion: ORs for hyperreactivity with an exclusively Asian, mostly Asian, or a mixed diet relative to an English diet were 0.31 [95% CI 0.15 to 0.62], 0.88 [0.56 to 1.37], and 0.99 [0.65 to 1.49]. This suggests that South Asian dietary habits may be protective with changing diets associated with second/third generation families and acculturalisation potentially accounting for some variation in asthma outcomes.

A lack of explanations underlying reported barriers was identified. Where offered, they tend to be theorised rather than proven. This reflects the nature of quantitative research that is inherently limited by specific questioning. Whilst qualitative research can be time consuming and costly, particularly if bi-lingual facilitators are used, it remains the best method of identifying underlying explanations for barriers.

Stigmatisation prevalence and its impact on asthma management in South Asians remain poorly understood in the UK, particularly in multi-generation families, extended families and in the context of easy international travel and communication.

The impact of diet on asthma remains underexplored despite its relevance to the South Asian community, given the high prevalence of beliefs on dietary restrictions in children with asthma and suggestions that the South Asian diet is protective against developing asthma.

Healthcare professional bias also requires further investigation. In a review of impact of ethnicity on asthma management [29], it is suggested that subconscious bias and stereotyping affects care quality. Ethnographic research involving direct observation and event questioning as they unfold is likely to be needed to compare and contrast clinical management and expectations in children of different ethnic backgrounds.

Additionally, impact of out-dated professional knowledge and behaviours on patient knowledge and behaviours requires further research. This was identified almost 30 years ago when UK Asian women had antenatal care provided by GPs with insufficient obstetric knowledge [30] and suffered significantly increased perinatal mortality and morbidity. It is suggested that British South Asian families gravitate towards South Asian GPs for both cultural and linguistic reasons. Given the evidence reported in this review, they may be lacking in current asthma knowledge. It can therefore be argued that asthma knowledge amongst GPs should be explored.

Whilst we found several studies investigating ethnic differences in prevalence of asthma and medication use, no clear association between a barrier, explanation and management behaviour was identified. Without linkage, there is considerable risk of presuming explanations or impacts on management that do not apply. This has implications for resource use efficiency: tailoring interventions to minority ethnic communities has been shown to be effective in asthma management [31]. It would be inappropriate however, to tailor interventions to all ethnic groups if barriers are not ethnic specific, but rather reflect being a minority group.

Defining ethnicity is challenging. The problem lies in how ethnicity is defined – geographically, linguistically, and historically or another way. In UK studies, various methods were used in defining ‘origin’; mother’s self-identification as Indian, Pakistani or Bangladeshi [3, 27, 32], 2001 UK census category self-selection [25, 28], ethnic group classification by spoken home language, fieldworkers subjective assessment of ethnicity [23], no definition beyond ‘Asian patients’ [22], or no description of ethnicity definition [24]. Using pre-set categories by country of origin is simple, but takes no account of ethnic (linguistic or religious/cultural) variations within countries; and challenging for older generations who may identify with one country but were born in another.

UK census category use allows for generalisation and ease of ethnic group analysis, but is limited by fixed categories changeable with each census. Language use is a pragmatic solution, particularly when work is conducted in multiple languages. Conversely, subjective assessment by a fieldworker is fraught with potential pitfalls. It is argued that allowing ethnicity self-definition without pre-set categories is most effective however this may present significant challenges in analysis and applicability [33]. We join others in recommending that ethnicity be clearly explained in all ethnic health variation research by specifying the way in which ethnicity has been defined and operationalised, so that readers can determine applicability to their own populations [34].

Given the large number of barriers and explanations that may impact upon asthma management, clinicians may be daunted by the prospect of addressing these barriers in everyday practice. We do not suggest that clinicians ask for an exhaustive social history nor possess an encyclopaedic knowledge of all religions and cultures. Rather, clinicians should take a holistic view which will allow them to see children as existing within their family, social and cultural setting. This encompasses cultural competency; awareness and discussion of barriers and ethnic-specific explanations that exist and their impact on their patient’s management and self-management

Our findings highlight the importance of cultural and community beliefs. For example, dietary restriction is not considered in the widely used British Thoracic Society [BTS] guidelines on asthma management, yet beliefs and practices concerning food and asthma are prevalent amongst UK South Asian families and the possible stigmatisation of asthma in South Asian communities may also present a barrier to timely presentation and diagnosis. At the same time, a lack of culturally competent practice and adequate communication support in the NHS may provide significant system level barriers to effective asthma management in South Asian children.