Barriers and determinants of asthma control in children and adolescents in Africa: a systematic review

Abstract

Objective

To identify reasons for poor asthma control in African children and adolescents.

Design

Systematic review

Data sources

PubMed, Scopus, CINHAL, PsycINFO, MEDLINE and Web of Science databases were systematically searched up to 31 May 2020. Hand searching was done on Sabinet, African Journal online and Google Scholar.

Eligibility criteria

Studies identifying barriers to asthma control, where asthma control was assessed by the validated Asthma Control Test/Child Asthma Control Test and/or Asthma Control Questionnaire were included.

Data extraction and synthesis

Two reviewers independently selected studies for inclusion with disagreements resolved by a research team discussion, including a third reviewer. Data were extracted using the Cochrane Effective Practice and Organization of Care data collection form. The quality of the included studies was assessed using the modified Newcastle-Ottawa quality assessment scale. Identified barriers were reported in a thematic narrative synthesis.

Primary outcomes

Poorly controlled asthma and associated factors.

Results

From 914 records, three studies conducted between 2014 and 2019 in Nigeria, Uganda and South Africa met the inclusion criteria. A total of 883 children aged 4–19 years were analysed. Older age, concurrent allergy and city-dwelling significantly impacted asthma control. Few children with asthma symptoms in the community had ever used inhaled corticosteroids (6.7%) and identified reasons included lack of asthma diagnosis (38.8%) and no prescribed treatment (47.6%).

Conclusion

Asthma control in African children is impacted by age, allergy, urbanisation and lack of access to asthma diagnosis and treatment. More studies focusing on identifying barriers to asthma control in Africa are needed.

PROSPERO

Registration no: CRD42020196755)

Strength and limitations of this study.

• This systematic review highlights the paucity of studies on barriers and determinants of asthma control in Africa.

• The sufficiently validated Asthma Control Test/Child Asthma Control Test was used to assess asthma outcomes and identify barriers to asthma control.

• Barriers to asthma control reported in this study contribute to, and match those described in the literature on paediatric asthma.

• A limitation of this study is that the heterogeneity of the studies precluded a meta-analysis.

Introduction

Asthma is a chronic non-communicable respiratory disease. According to the 2018 Global Asthma Report, asthma affects over 340 million people worldwide, the majority of whom reside in low-income and middle-income countries (LMICs).¹ In contrast to many high-income countries (HICs), the prevalence of asthma is steadily increasing in LMICs, particularly in Africa.¹ The latest systematic review on asthma prevalence in Africa shows that compared with 74 million in 1990, by 2010, asthma affected 119 million of the total population. Of concern, nearly half of these asthma cases were children under 15 years.² Countries with the highest childhood asthma prevalence in Africa, South Africa (20.3%), Congo (19.9%) and Ivory Coast (19.3 %), are also regions with increasing urbanisation rates.^{3 4} Factors associated with urbanisation including poverty, poor air quality and lifestyle and dietary changes may drive the rising asthma rate and impact asthma control.⁵ However, in this setting, access to asthma healthcare and diagnosis as well as asthma research and research infrastructure remains lacking.^6–9

The most commonly used validated tools for asthma control assessment are the composite score instruments: Asthma Control Test (ACT), Child Asthma Control Test (cACT) and the Asthma Control Questionnaire (ACQ).¹⁰ The ACT and ACQ provide a quantitative assessment of asthma control and have been designated as core measures by the National Institutes of Health (NIH) for clinical research and observational studies.^{10 11} ACT and ACQ are simple methods that can help quantify the impact of barriers on asthma control,¹² which may not be comparable between HICs and LMICs.¹³ This review was conducted to collate data on reported barriers to asthma control in children and adolescents in Africa.

Methods

The systematic review is registered with PROSPERO (registration no: CRD42020196755). We used the Population, Exposure, Comparator and Outcomes (PECO) acronym to aid with the systematic search. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting standards were followed.¹⁴ The Synthesis Without Meta-Analysis reporting items guideline was used in conjunction with the PRISMA.¹⁵

Search strategy

The following databases were searched: PubMed, Scopus, CINHAL, PsycINFO, MEDLINE and Web of Science. The search methodology for all the databases is provided in the supplementary material (online supplemental table 1). Hand searching of the following databases was also conducted: Sabinet, African Journal online and Google Scholar. Only scientific articles written in English with date restrictions from 1 January 2000 to 31 May 2020 were included.

The search strategy was structured to include terms for ‘Child’, ‘Asthma’, ‘Barriers’, ‘Asthma Control Test’, ‘Africa’ and/or variations of these.

Selection of studies

Studies identified from searching electronic databases were combined and duplicates were removed. Two reviewers (REM, OK) independently screened references using a three-stage review of title and abstract, followed by a full-text review of included studies. The full text of potentially eligible studies was screened against the review criteria and potential articles identified. At each stage, disagreements were resolved by a team discussion with a third reviewer (RM).

Inclusion and exclusion criteria

The study’s focus was to identify barriers associated with poor asthma control in African children and adolescents with doctor-diagnosed/suspected asthma, where the validated ACT/cACT or ACQ tool was used to assess asthma control. The population included children between the ages of 6 and 18 years. Studies were included with broader age ranges if children aged 6–18 years were reported separately, or if >50% of the population were children within this age range.

Studies published from January 2000 to May 2020 were included to ensure the encompassing of all data since validation of the ACT and ACQ. Clinical trials assessing pharmaceutical treatment and diagnostic accuracy of tools were excluded. Grey literature from experts in the field, conference abstracts or unpublished material were also excluded (table 1).

Table 1.

Criteria for the search and rules devised to facilitate inclusion/exclusion criteria

Search strategy	Definition	Rules
Population	Children and adolescents between ages 6 and 18 years with a doctor diagnosis or a baseline prescription for asthma treatment or presumed diagnosis of asthma based on a history of recurrent wheeze.	Included: Studies with broader ranges of ages if children aged 6–18 were reported separately or if >50% of the population were children within this age range. Excluded: Studies in adults (>18 years).
Exposure	Environmental related factors Pollution (indoor and outdoor), environmental tobacco smoke, mould, biomass fuels, pets, physical exercise, sedentary lifestyle, antibiotic use, paracetamol use, industrial combustion, respiratory infections. Patient-related factors Attitudes, knowledge and perceptions, adherence, beliefs, inhaler technique, lifestyle, relationships, communication Healthcare and doctor-related factors Availability of treatment and healthcare facilities, doctor asthma knowledge, time spent on asthma education, availability of medications. Comorbidities Allergic rhinitis, obesity, obstructive sleep apnoea Gastro-oesophageal Reflux Disease	Included: Studies aiming to identify exposures that had a quantifiable impact on asthma control. Excluded: Clinical trials assessing pharmaceutical treatments. Studies assessing the diagnostic accuracy of tools. Studies assessing the validity of tools.
Comparison (if applicable)	Usual care in people of the same age with well-controlled asthma
Outcome	Asthma control measured using ACT/cACT and/or ACQ	Excluded: Studies using tools for measuring asthma control other than ACT/cACT and/or ACQ
Timeframe	20 years between January 2000 and May 2020	Excluded: Studies conducted before January 2000 and after May 2020
Setting	Africa	Excluded: Studies not done in Africa
Study	Cohort, case–control, cross-sectional	Included: Studies identifying exposures that impact asthma control as measured by cACT/ACT and/or ACQ

ACQ, Asthma Control Questionnaire; ACT, Asthma Control Test; cACT, Child Asthma Control Test.

Data extraction

The full texts of all studies found to be relevant and meeting the inclusion criteria were retained for data extraction and final synthesis. Data including study design, setting, population, authorship and statistical analysis was extracted using a standardised data extraction form modified from the Cochrane Effective Practice and Organization of Care data collection form.¹⁶ The authors were contacted where clarification was required, and data were missing. The selection process was summarised using a PRISMA flow diagram (figure 1).

Figure 1.

Study eligibility chart according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses criteria. ACQ, Asthma Control Questionnaire; ACT, Asthma Control Test.

Quality assessment

The included studies' quality was assessed using the modified Newcastle-Ottawa Scale for cohort, case studies and cross-sectional studies¹⁷ (online supplemental table 2).

Data analysis and synthesis

We anticipated that the population and statistical analysis heterogeneity of the studies would preclude a formal meta-analysis. We, therefore, grouped into themes asthma control barriers corresponding to literature; patient, environmental, healthcare/doctor-related factors and comorbidities.^{12 13} (online supplemental table 3). Statistical analyses were performed using MedCalc-Software, Ostend, Belgium; http://www.medcalc.org; 2018.¹⁸

Patient and public involvement

Patients and the public were not involved in the study design or conduction of the study.

Results

Search results

There were 914 articles identified: 863 articles through electronic database searching (EBSCO host=27, PubMed=136, Web of Science=97, Scopus=603) and an additional 51 articles through hand searching (Google scholar=23, Sabinet=12, AJOL=16). The total number of articles found after duplicates were removed was 498. Of the 498 articles screened, 484 were excluded as they were not appropriate or did not relate to the study. The remaining 14 full articles were assessed for eligibility, and 11 articles were excluded for the following reasons: wrong age group=2, did not use ACT/ACQ=2, not original research=2, assessed impact rather than barriers of poor asthma control=5. Three studies met the inclusion criteria (figure 1).

Characteristics of the studies

All three studies conducted in Nigeria, South Africa and Uganda^19–21 were cross-sectional; two hospital-based and one community-based. The sample size was smaller for hospital-based studies with 207 and 115 participants in Nigeria¹⁹ and South Africa,²⁰ respectively, compared with the community-based study of 561 participants in Uganda.²¹ Publication dates ranged from 2014 to 2019. The ages of participants ranged from 4 to 19 years. Asthma diagnosis was based on doctor diagnosis^{19 20} guided by the Global Initiative on Asthma (GINA),¹⁹ and symptom screening by the International Study of Asthma and Allergies in Childhood questionnaire.²¹ One study adjusted for age, gender and concurrent allergy,²¹ while the rest did not report adjusting for potential confounders, reducing their quality score^{19 20} (online supplemental table 2). To recruit participants, two of the hospital-based studies used consecutive enrolment from a group of children attending asthma clinic.^{19 20} The community-based study derived participants from a large case–control study investigating risk factors of asthma in school going children^{21 22} (table 2).

Table 2.

Characteristics of included studies

Author ref	Study type	Setting	Year of publication	Country of origin	Sample size	Age ranges (years)	Asthma definition	Asthma control definition	Recruitment	Exposures	Quality score	Reviewers comment
Ayuk et al19	Cross-sectional	Hospital	2018	Nigeria	207	4–18	Doctor diagnosis, GINA	ACT/cACT >19 controlled <19 uncontrolled	Consecutive enrolment for 1 year from a group of children attending the asthma clinic	Family size, socioeconomic status, urban vs rural dwelling, allergy status (by ISAAC), Triggers (particulate and non-particulate)	7/10	Author contacted for further information on participant numbers.
Garba et al20	Cross-sectional	Hospital	2014	South Africa	115	5–18	Doctor diagnosis	ACT/cACT =25 (ACT)/27 (cACT) total control >19 well-controlled ≤19 uncontrolled 16–19 somewhat controlled <16 Poorly controlled	Consecutive enrolment for 4 months from a group of children attending the asthma clinic	Presence of a smoker at home, presence of pets, cockroaches and use of biomass fuel, the child’s sleeping environment (dust, carpets and soft toys in the bedroom). Compliance with medications and inhaler technique. Allergy status (by clinical examination)	5/10	Author contacted for further information on recruitment strategy, data analysis and participant numbers.
Mpairwe et al21	Cross-sectional	Community School	2019	Uganda	561	5–17	Screening ISAAC questionnaire	ACT/cACT >19 Well controlled 15–19 partly controlled <15 Poorly controlled	Recruitment from children with self-reported breathing problems at schools in an urban area	Age, sex, regular physical exercise as recommended by WHO, area of residence in first 5 years of life (rural, town or city), concurrent allergy, antimalarials	10/10	Describes participants as derived from a large case–control* study to investigate risk factors of asthma.

*Mpairwe et al.²².

ACT, Asthma Control Test; cACT, Child Asthma Control Test; GINA, Global Initiative for Asthma; ISAAC, International Survey on Asthma and Atopy in Children; WHO, world health organisation.

Assessment of asthma control

All the studies measured asthma control using ACT and cACT. Scores were based on the cut-off point of >19 for controlled asthma and ≤19 for uncontrolled asthma. The prevalence of uncontrolled asthma in the Nigeria, South Africa and Uganda was 30.9%, 44.3% and 44.5% respectively.

Patient-related factors

Age

Two studies assessed the impact of age on asthma control. The large community-based study showed that older age (13–17 years) was significantly associated with poorer asthma control (adjusted regression coefficient (95% confidence interval), p-value: (−1.07 (−1.20 to –0.94), p<0.0001).²¹ The exception was a small clinic cohort of moderate quality, which showed no association.²⁰

Gender

Two of the studies^{20 21} that examined gender showed no significant association with asthma control.

Asthma medication use

Two studies^{20 21} examined the use and compliance of asthma medication. The study among school-going children²¹ showed that the majority (73%) had never used inhaled asthma medications. Additionally, regular use of inhaled asthma medication in the last 12 months was inadequate for salbutamol (18.1%) and corticosteroid (6.7%) even though the majority (55.8%) had a doctor diagnosis of asthma. Although not significant, in the same cohort, children with poorly controlled asthma preferred regular use of (salbutamol and prednisone) tablets rather than inhaled salbutamol and corticosteroids.²¹ Conversely, in the cohort of children attending asthma clinic,²⁰ good adherence to medications was seen in 82.6% of patients. In these doctor-diagnosed children, asthma control was significantly associated with good adherence to medication, where 37.9% and 62.1% of patients had uncontrolled asthma and controlled asthma, respectively (χ²=0.217, p=0.002).²⁰

Ethnicity

There was no significant association between asthma control and ethnicity (χ²=3.22, p=0.359) in Black-African, Caucasian, mixed-ethnicity and Indian participants in South Africa.²⁰

Environmental related factors

Two studies conducted in Uganda²¹ and Nigeria¹⁹ examined the effects of rural versus urban domicile on asthma control. The school-based Ugandan cohort showed that city residence in early life was associated with poor asthma control (−1.99 (−3.69 to –0.29), p=0.02).²¹ In contrast the clinic-based cohort in Nigeria showed, although without significance, that within the rural community, more children with current allergies had better control of their asthma (85.7%) when compared with their urban counterparts (66.7%). Interestingly, the children who lived in rural areas without concurrent allergy had poorly controlled asthma (50.0%) compared with their urban counterparts (28.3%), Fisher’s exact test=2.076, p=0.17, although this too was not significant.¹⁹

All three included studies considered the presence of asthma triggers in their participants' environments, but only the South African study examined these triggers in relation to asthma control. Common triggers included dust, cold air, physical exercise, fumes or air pollution, pollen, pets, smoking and biomass fuels (figure 2). In the South African cohort, home circumstances including dust, cockroach, carpet, pets, toys in bed and smoking were not found to be associated with asthma control.²⁰ The use of biomass fuel was uncommon in South Africa (6.1%) compared with Nigeria (22.1%) and was not found to be significantly associated with asthma control (χ²=6.202, p=0.185).^{19 20}

Figure 2.

Prevalence of asthma triggers among study participants across African studies using the Asthma Control Test/Childhood Asthma Control Test to identify asthma control barriers. ETS, environmental tobacco smoke.

Healthcare and doctor-related factors

Only the field-based study in Uganda, reported the impact of healthcare-seeking behaviour on asthma control. In 553 children who reported treating their asthma in the last year, 26.8% reported having ever used inhaled asthma medications, and a similar proportion, 29.7% reported having ever used herbal remedies for asthma management. On enquiry about previous asthma assessments and follow-up, 73 (13.2 %) visited a health facility to monitor their asthma, 45 (8.2%) children had ever had a lung function test; 2 (0.4%) had ever used a peak flow metre as an asthma monitoring tool at home, and only 3 (0.5%) had a personal written asthma action plan.²¹ The reason for having never used inhaled asthma medication was investigated in 405 children and included inhaled asthma medications had never been prescribed for them (47.6%), never been diagnosed (38.8%), high cost of inhalers (4.5%), fear of side effects of inhalers (4.5%), alternative treatment with salbutamol or steroid tablets (1.4%) and non-medicinal treatment, that is, wrapping up in warm clothes and resting.²¹

Comorbidities

All three studies assessed children for allergic rhinitis, but only two^{19 21} in relation to asthma control. In the larger powered community-based study,²¹ children with concurrent allergic rhinitis were more likely to have lower asthma control scores (−1.33 (−2.28 to –0.38), p=0.006), whereas no significant association was found between atopy and asthma control in the small cohort clinic-based study.¹⁹ However, in the latter study, children with current allergy had more emergency hospital visits due to asthma exacerbations (χ²=10.09 (df: 1) p=0.002; Spearman’s R=0.22, p=0.001).¹⁹

Discussion

Older age, concurrent allergic rhinitis and early life urban residence are barriers similar to HICs and significantly impact asthma control in African children. Access to healthcare and appropriate asthma medication remains limited, with a minority of children with asthma symptoms ever having used inhaled corticosteroids (ICS).

Older age

Mpairwe et al²² found adolescents in Uganda have inadequate asthma control and outcomes. Similarly, the age group 12–17 years was more predictive of exacerbations than other age groups in a European cohort study using the General Practice Research Database.²³ One reason for this can be explained by adolescent studies that show poor adherence compared with other age groups.²⁴ Social stigma, forgetfulness and poor understanding of medication play a significant role in adherence and warrant further exploration.^{25 26}

Concurrent allergic rhinitis

The Ugandan and Nigerian studies found that children with allergic rhinitis (AR) had less well-controlled asthma and were more likely to be hospitalised. Similarly, in a large UK retrospective cohort of 9522 children with asthma, the presence of AR significantly increased the likelihood of physician visits and more than doubled the likelihood of hospitalisation. Furthermore, drug use and costs were significantly higher among children with asthma and concurrent AR.²⁷ Active search and recognition of AR when assessing children remain critical in comprehensive asthma management.

Rural versus urban residence

Studies in Africa show a decreasing gradient in asthma prevalence between urban and rural areas.^{28 29} In this context, biomass fuel exposure remains a significant contributor to inflammatory lung diseases, including asthma and chronic obstructive pulmonary disease.^{30 31} Few studies in Africa have compared asthma control between rural and urban areas.19 21 32 33 Urban residence was significantly associated with poorly controlled asthma in Uganda, where asthma risk among schoolchildren²¹ was three times higher in children who in early life resided in cities rather than rural areas.²² Similarly, rural to urban migration appears to be an important determinant of the increasing prevalence of wheeze among schoolgoing children in Latin American cities.^{34 35} Increasing asthma rates in urban settings could be related to over-crowding, reduction of exercise, poorer air quality and changes in lifestyle and diets.

Access to diagnosis and healthcare

Six out of 10 children attending healthcare institutions have good asthma control, while a similar number of undiagnosed children in the community have poorly controlled asthma.^19–21 Even after a diagnosis of asthma, ICS use is limited in communities^{21 36} compared with clinic patients²⁰ who once diagnosed, have significantly better asthma control. The preference of tablets (salbutamol and corticosteroids) over ICS may largely be explained by their quick relief and ease of administration combined with underlying suboptimal knowledge and asthma medications cost.³⁶ Furthermore, traditional healers remain integral to medical care in communities due to local cultural practices and beliefs. There is a need to communicate asthma management strategies to communities in a culturally sensitive manner.^{32 37} Triggers including dust, air pollution, pollen, pets and smoking common across the globe, indicate the feasibility of a global checklist and the necessity of avoidance education.³⁸

Strengths and limitations

We may not have identified all significant barriers that impact asthma control as other asthma control tools, that is, GINA and National Asthma Education Programme, were excluded because they are not as sufficiently validated as the ACT and ACQ.¹⁰ Nevertheless, we identified variables in each group classification for poor asthma control in current literature.¹³ Our wide-ranging search strategy found no non-English articles requiring exclusion. The studies' heterogeneity in terms of outcome analysis and population precluded a meta-analysis; therefore, we reported all the factors within the emerging themes.

Implications for clinical practice, healthcare systems and policymakers

Strategies that improve medication access, including initiatives like the WHO Essential Medicines List, low-cost equipment like plastic spacers³⁹ and implementing culturally appropriate educational programmes for healthcare workers and the public, remain vital.^{40 41}

Implications for future research

Studies beyond healthcare institutions that include communities in identifying barriers and their impact on asthma control are needed in African children.

Conclusion

Asthma control barriers requiring focus in Africa are lack of accurate diagnosis, limited access to inhaled therapy, lack of asthma knowledge and poor air quality. Better education and advocacy through community-based public interventions are needed to improve African children’s asthma control and outcomes.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Not applicable.