Strengths and limitations of this study

Extensive searching using the Population Intervention Comparison and Outcome standard questions, robust quality appraisal using the Newcastle-Ottawa Scale and pre-determined eligibility criteria, appropriate pooling model (weighted inverse variance random effects model), assessment of heterogeneity, sensitivity analysis, publication bias and subgroup analysis were employed.

Only studies reported in English language and that had defined good adherence as taking ≥80% of the prescribed doses of secondary antibiotics were included.

In this meta-analysis, there was high heterogeneity among the included studies.

Introduction

Rheumatic heart disease (RHD) is a medical condition in which the heart valves are permanently damaged leading to impaired cardiac function. It is the result of recurrent rheumatic fever which is an autoimmune-mediated disorder caused by group A beta haemolytic streptococcal bacteria. RHD is a critical public health issue causing significant morbidity and mortality.1 2 According to the Global Burden Disease report, there were about 33.4 million cases with RHD; and RHD was responsible for about 319 400 deaths per annum globally in 2015.3 It is also a cause of about 11.5 million disabilities worldwide. Furthermore, it is a great healthcare burden with an estimated annual expenditure of US$5400 billion.4

Fortunately, RHD is an easily preventable disease. The three levels of prevention known as primordial, primary and secondary levels can effectively prevent and reduce its incidence. The primordial and primary levels of prevention are used to prevent an acute rheumatic attack. But, secondary prevention is used to prevent recurrent rheumatic attack—a precondition for RHD. Secondary prevention is applied by regular administration of prophylactic antibiotics for relatively long duration either by injection or oral routes. Secondary antibiotic prophylaxis is a major determinant of cardiac outcome by preventing the occurrence of RHD or halting its progress.5–7

Secondary antibiotic prophylaxis cannot prevent RHD effectively unless the patients adhere to the regular administration of the prophylactic antibiotics.8 Even though, the exact level of adherence to prophylactic antibiotics for an effective prevention of recurrent rheumatic attack is not known, receiving 100% of the prophylactic antibiotic doses is a priority. But, in numerous research articles, the benchmark to define good adherence to secondary antibiotic prophylaxis among patients with RHD and/or acute rheumatic fever (RHD/ARF) is receiving at least 80% of the prescribed doses. Patients receiving less than 80% of their doses are defined as non-adherent; and they are considered as having a great risk of acquiring RHD.6 8

Worldwide, a number of studies have been conducted to assess the prevalence of adherence to secondary antibiotic prophylaxis and to identify the associated factors (reasons) for poor adherence among patients with RHD/ARF. However, results were highly inconsistent with a prevalence ranging from about 10%9 to 93%10 and relatively with different reasons or associated factors. Despite this inconsistency, as per the authors’ knowledge, there were no previous studies showing the global pooled adherence to secondary antibiotic prophylaxis. Assessing the pooled adherence to secondary antibiotic prophylaxis and reviewing the pooled associated factors/reasons for poor adherence can help to determine the global adherence level and to develop a reason/factor-oriented strategy targeted to increase the patients’ adherence which in turn is crucial for reducing the burden of RHD. Therefore, this systematic review and meta-analysis aimed to assess the pooled adherence to secondary antibiotic prophylaxis and to review the factors or reasons for poor adherence among patients with RHD/ARF from global evidence.

MethodsReporting

This systematic review and meta-analysis was presented according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guideline11 (Supplementary file 1).

Searching strategies

PubMed/Medline, Google Scholar, Cochrane Review and African Journals Online databases were searched to identify relevant research articles. Searching for grey literature from repositories and snowball searching were also employed to accommodate potentially related literatures. The comprehensive searching strategy was developed according to Population Intervention Comparison and Outcome standard questions (online supplemental table 1).

Eligibility criteria

After retrieving, articles were exported to the Endnote reference manager software V.7.0 to remove duplication. Two investigators (MB and FG) independently screened the selected articles by their titles and abstracts before the retrieval of full texts. Published and unpublished observational studies that had reported the prevalence of adherence (using an operational definition of adherent meant ≥80% adherence rate) to secondary antibiotic prophylaxis among patients with RHD/ARF; and published in English language from 1 January 2005 to 1 December 2022 were included. But, citations, research articles with no abstract and/or no accessible full texts, commentaries, editorials and anonymous reports were excluded. Finally, the full text of articles that met the inclusion criteria by title and abstract were deeply reviewed for their quality; and if they were in line with the objectives of this systematic review and meta-analysis; and those articles which fulfilled the inclusion criteria were included in the final analysis. Findings from relevant articles were summarised with narrative synthesis, and quantitative meta-analysis was then employed.

Patient or public involvement

Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Outcome variable

The outcome variable for this meta-analysis was the pooled prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD/ARF.

Quality assessment

After removing duplicate studies and screening potentially relevant articles, two independent authors (MB and MA) appraised the quality of eligible articles using the Newcastle-Ottawa Scale (NOS) for observational studies as a quality appraisal tool. Disagreements between appraisers were solved by taking their mean scores. Interpretations for the score were based on the recommendations of the NOS tool for cross-sectional and cohort studies.12

Data extraction and statistical analysis

The data were extracted and cleaned using a Microsoft Excel worksheet; then they were exported to STATA V.11.0 statistical software for further analysis. SEs for the prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD/ARF were calculated using binomial distribution formula. Then, the overall prevalence of adherence in those patients was pooled based on the weighted inverse variance random-effects model at 95% CI. Results were presented by narrative synthesis tables and forest plots. Heterogeneity between included studies was assessed by the inverse variance (I2) with p values. The values of I2, 25%, 50% and 75% represent low, moderate and high heterogeneity, respectively. Publication bias was assessed by funnel plots and Egger’s regression test. Sensitivity analysis was also conducted to observe whether there was an influential study affecting the true value of the pooled prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD/ARF.

ResultsStudy selection and characteristics of the included studies

A total of 1904 studies were retrieved from the electronic database search. After screening and eligibility assessment, 33 studies9 10 13–43 were included in this meta-analysis (figure 1). All of the included articles were observational studies (online supplemental table 2). Three studies were both on oral and injectable prophylaxis; in which 20%, 5.5% and 1.6% of the participants were on oral prophylaxis.

Figure 1

PRISMA flow diagram showing the identification, screening, reasons for exclusion and number of included research articles in this systematic review and meta-analysis; 2022. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Adherence to secondary antibiotic prophylaxis (meta-analysis)

A total of 33 articles with a total sample size of 7158 patients were included in this meta-analysis. Among the included studies, the lowest prevalence of good adherence (≥80% of the prescribed dose) to secondary antibiotic prophylaxis among patients with RHD/ARF was 10.7%; while the highest prevalence was 92.9%. The pooled prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD/ARF was found to be 58.5% (95% CI: 48.2% to 68.7%; I2=99.2%; p<0.001; 33 articles) (figure 2).

Figure 2

Forest plot showing the prevalence of good adherence to secondary antibiotic prophylaxis among patients with rheumatic heart disease or rheumatic fever; 2022.

Heterogeneity

In this meta-analysis, the inverse variance (I2) was found to be 99.2% (figure 2); suggesting the presence of heterogeneity on the reported prevalence of adherence to secondary antibiotic prophylaxis among the included studies.

Sensitivity analysis

In this meta-analysis, a leave-one-out sensitivity analysis was conducted to examine if the pooled prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD/ARF has been greatly impacted by the result of a single study. But, all of the results of this sensitivity analysis were within the 95% CI limits of the pooled prevalence (48.2% to 68.7%); suggesting the absence of an influential study that potentially impacted the observed pooled prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD/ARF.

Publication bias

In this meta-analysis, the funnel plot showed symmetrical distribution (online supplemental figure 1), and the Egger’s regression test was found to be non-significant with a p value of 0.37; suggesting the absence of publication bias among the included studies.

Subgroup analysis

In this meta-analysis, subgroup analysis was conducted by study population category, income level of the study areas and study design category. Based on the study population, the included studies were categorised as studies conducted on adults, studies conducted on children and studies conducted on both adults and children. The subgroup analysis based on this category showed that the highest adherence to be observed among children with RHD/ARF (10 studies; pooled adherence=71.7% (95% CI: 52.7% to 90.6%; I2=99.4%; p<0.001)) (figure 3).

Figure 3

Forest plot showing the disparities of good adherence to secondary antibiotic prophylaxis among children and adult patients with rheumatic heart disease or rheumatic fever; 2022.

Regarding the income level category, the included studies were categorised as from high, middle and low income countries. Based on this subgroup analysis, the lowest adherence was observed among the studies conducted in high-income countries (pooled adherence=41.6% (95% CI: 30.4% to 52.8%; I2=97.2%; p<0.001); 11 studies (9 studies from Australia, 1 study from New Caledonia and 1 study from New Zealand)). Adherence was also found to be 69.1% (95% CI: 61.7% to 76.4%; I2=96.0%; p<0.001) in low-income countries (pooled from 14 studies (8 studies from Ethiopia, 4 studies from Uganda, 1 study from Gambia and 1 study from Rwanda)); and 64.1% (95% CI: 41.3% to 87.0%; I2=99.5%; p<0.001) in middle-income countries (pooled from 8 studies (4 studies from India, 1 study from Bangladesh, 1 study from Brazil, 1 study from Fiji and 1 study from Pakistan)) (online supplemental figures 2 and 3).

Regarding the study design, the included studies were grouped as cross-sectional, retrospective and prospective in design. Lowest pooled adherence was observed among studies conducted with retrospective design (13 studies; pooled adherence=54.7%; 95% CI: 34.4% to 75.1%; I2=99.6%; p<0.001). Adherence from cross-sectional and prospective studies was found to be 61.3% ((95% CI: 51.0% to 71.6%; I2=97.2%; p<0.001; 13 studies) and 60.5% (95% CI: 42.4% to 78.6%; I2=98.7%; p<0.001; 7 studies), respectively (online supplemental figure 4).

Reasons or factors for poor adherence (systematic review)

Among the total 32 articles included in this meta-analysis, about 22 articles had reported reasons or factors for poor adherence to secondary antibiotic prophylaxis in patients with ARF/RHD. Those 22 articles were reviewed systematically and reported factors or reasons were extracted. After extraction, similar ideas were coded consistently to identify themes. Accordingly, thematic analysis was used to report the factors or reasons for poor adherence to secondary antibiotic prophylaxis in patients with ARF/RHD; and the following themes were identified.

Socioeconomic-related reasons or factors:

In the literatures reviewed, patients of older age were reported to have poor adherence to secondary antibiotic prophylaxis than younger patients. A lower educational status of the patient or lower educational status of the parents (for children) was also another variable reported to be a factor for poor adherence. Literatures also reported that patients living in rural areas had poor adherence to secondary antibiotic prophylaxis. Lack of money for buying medication, lack of money for transportation and lack of family (social) support were also reasons for poor adherence repeatedly reported in the literature.

Healthcare system and healthcare provider-related reasons or factors:

In the literature reviewed, healthcare inaccessibility, including long distances to health institution, shortage of prophylactic medication, long waiting times, long duration of secondary antibiotic prophylaxis, poor counselling or poor healthcare worker-patient communication and lack of knowledge or skill to inject prophylactic medications, was healthcare system or healthcare provider-related reasons for poor adherence to secondary antibiotic prophylaxis among patients with RHD/ARF.

Individual/patient-related factors or reasons:

In the literature reviewed, presence of comorbidities, longer duration with the RHD, having mild RHD (New York Heart Association (NYHA) class I and 2), schedule forgetting, feeling well or healthy while on prophylaxis, fear of injection pain and side-effects and lack of knowledge on the disease process and the prevention of RHD were patient-related factors or reasons for poor adherence to secondary antibiotic prophylaxis among patients with RHD/ARF.

Discussion

This systematic review and meta-analysis aimed to assess the prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD or rheumatic fever. It had also reviewed the factors or reasons for poor adherence. Good adherence was considered as taking at least 80% of the prescribed dose; and research articles reporting the prevalence of adherence based on this cut-off point were included. Accordingly, a total of 33 research articles with a total sample size of 7158 patients were included in the analysis.

Even if there is heterogeneity among the included articles, the pooled prevalence of good adherence to secondary antibiotic prophylaxis among patients with RHD/ARF was found to be 58.5% (95% CI: 48.2% to 68.7%; I2=99.2%; p<0.001). It is recommended that patients with RHD/ARF should take at least 80% of their prescribed secondary antibiotics for effective prevention of complications and death from the disease.6 44 But, according to this meta-analysis, only 58.5% of patients with RHD/ARF were taking ≥80% of their prescribed doses. This implies that about 41.5% of patients with RHD/ARF had poor adherence to secondary antibiotic prophylaxis; and were at higher risk of complications and death.

As per the subgroup analysis, children (age <18 years) were found to be more adherent with secondary antibiotic prophylaxis than adults. The pooled prevalence of adherence to secondary antibiotic prophylaxis among children with RHD/ARF was found to be 71.7% (95% CI: 52.7% to 90.6%; I2=99.4%; p<0.001; 10 studies). Studies conducted purely on adults are limited. One study conducted in Uganda reported that adherence to secondary antibiotic prophylaxis to be 30%.20 The pooled prevalence of adherence to secondary antibiotic prophylaxis among studies conducted on both children and adults was found to be 53.7% (95% CI: 41.6% to 65.7%, I2=99.0%, p<0.001; 22 studies); which was lower than the prevalence among studies conducted on children only. This might imply that children had higher adherence than adults which might be due to the reason that children are dependent on their parents for their healthcare and parents oversee adherence for their children. Another reason for the higher adherence in children might be shorter duration with the disease and on prophylaxis than adults.

The subgroup analysis based on the study design had revealed that lower adherence to be observed among articles done by retrospective design (13 studies; pooled adherence=54.7%; 95% CI: 34.4% to 75.1%; I2=99.6%; p<0.001). This might be due to the nature of the design; as retrospective follow-up studies use longer time period than cross-sectional designs; and longer follow-up period was reported to negatively affect adherence. Adherence in retrospective designs might also be affected by lack of secondary antibiotic prophylactic medication than in prospective and cross-sectional studies.

The subgroup analysis by study area income level had revealed surprising results. The lowest pooled adherence was among studies conducted in high-income countries (pooled adherence=41.6% (95% CI: 30.4% to 52.8%; I2=97.2%; p<0.001; 11 studies). On the other hand, highest adherence was observed among studies conducted in low-income countries (pooled adherence=69.1% (95% CI: 61.7% to 76.4%; I2=96.0%; p<0.001; 14 studies). This might be due to the differences in study design; as 9 out of 11 studies conducted in high-income countries were retrospective, and only 3 out of 14 studies conducted in low-income countries were retrospective, while about 8 out of 14 studies were cross-sectional in design (online supplemental table 2). But, authors of this systematic review and meta-analysis believe that this discrepancy needs further investigation.

This systematic review and meta-analysis also revealed the factors or reasons for poor adherence to secondary antibiotic prophylaxis among patients with RHD/ARF. Higher age was reported as a factor for poor adherence8 19; and this was supported by the subgroup analysis, as higher pooled adherence was observed in children (age <18 years) than adults. Living in rural areas was also a factor or reason for poor adherence.15 32 Rural residents have low healthcare access.21 They should travel long distance to access health institution13 15 22 30; and they may lack money for transportation.13 15 20 21 32 Furthermore, rural residents have lower educational and information access.19 22 32 So, they may not easily understand their disease process and the importance of taking their prophylaxis unless they gain thorough counselling from healthcare providers. Finally, they will end up with missing of their appointments; and will have poor adherence.

In the literatures reviewed, shortage of prophylactic medication,21 32 39 40 longer waiting time,21 22 insufficient counselling or poor healthcare worker-patient communication,21 30 32 lack of knowledge or skill to inject prophylactic medications21 were also the reported reasons for poor adherence to secondary antibiotic prophylaxis among patients with RHD/ARF. Longer duration with the disease and longer duration on prophylaxis8 15 39 were also reported as reasons for poor adherence. In fact, secondary antibiotic prophylaxis for RHD/ARF is administered for relatively longer periods, which might be tiresome for patients and lead to gradual economic loose (lack of money).

Presence of comorbidities like HIV/AIDS was also a reason for poor adherence; which might be due to multiple appointments and multiple drug regimens21 that might be tiresome for patients. Having mild RHD (NYHA class I and 2),19 schedule forgetting,30 feeling well or healthy while on prophylaxis,32 fear of injection pain and side-effects13 18 20 21 30 32 38 40 and lack of knowledge on the disease process and the prevention of RHD13 20 40 were also reported as factors or reasons for poor adherence to secondary antibiotic prophylaxis among patients with RHD/ARF.

This systematic review and meta-analysis used robust methodologies. Authors had used internationally qualified tools for evaluating the quality of included articles and they had employed test of heterogeneity, publication bias and subgroup analysis. Therefore, the results of this systematic review and meta-analysis might be used as input for designing novel strategies targeted at increasing adherence to secondary antibiotic prophylaxis among patients with RHD and/or ARF. But, this systematic review and meta-analysis might not be free from limitations. The discussion might be shallow; as there were no previous meta-analyses on the issue for comparison. Additionally, there was significantly higher heterogeneity among the included studies.

Conclusion and recommendations

The pooled prevalence of adherence to secondary antibiotic prophylaxis among patients with RHD/ARF was found to be low. About 41.5% of RHD/ARF patients were found to have poor adherence. This implies that more than two-fifths of RHD/ARF patients were at higher risk for complications and death from the disease. Healthcare inaccessibility including long distance from the health institution, schedule forgetting, poor counselling or poor healthcare worker-patient communication, lack of knowledge, lack of skill among healthcare workers and fear of injection were some of the easily modifiable factors or reasons for poor adherence. So, access to secondary antibiotic prophylaxis for rural residents should be increased through decentralisation of the follow-up care to the most accessible healthcare institution like health posts. Schedule reminding system should be designed. Targeted health education and thorough counselling should be provided for patientsRHD/ARF. Training on counselling and administration of secondary antibiotic prophylaxis should be given for healthcare workers.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. The datasets used and/or analysed during the current study are available within the manuscript.

Ethics statementsPatient consent for publication

Not applicable.

Acknowledgments

We would like to acknowledge authors of the included studies.