Discussion
According to the World Bank classification (2021), there are 84 LMICs in sub-Saharan Africa, the Middle East and Latin America-Caribbean, with publications from 16 countries included in this systematic review. The majority of the studies identified in this review were from upper-middle-income countries, which is comparable to the trend seen in Asian LMICs.19 80 Therefore, the implementation of EHDI appears to be proportional to the country’s economy. While screening is being implemented, the relevant outcome of attaining early intervention is not yet known from these studies.14 17 81 Despite the extension of the age limit to 6 years for LMICs, many studies still included a larger age range for children to have better coverage. Therefore, this review also included older children beyond 6 years.
Hearing screening strategy
Hospital screening at birth, which included immunisation clinics, maternity units and well-baby care clinics, was the most commonly employed neonatal/infant hearing screening strategy.11 37 61 76 81 The contrast in the context of EHDI implementation in HICs must be considered before duplicating hearing screening programmes in LMICs to maintain sustainability and improve early intervention outcomes.9 10 For example, there is a considerable shortage of trained individuals in maternity and neonatal health, particularly in rural areas of LMICs7 8; births are often performed in remote primary clinics or home settings.82 83 To address this gap, the WHO developed criteria for screening tools and protocols customised to the country’s national, cultural and socioeconomic situations in 2021.10 In such contexts, community-based initiatives or a combination of hospital and community-based programmes are often recommended.84–87
Considering the burden of unidentified hearing loss, school-based screening has been increasingly employed for older children in some countries (India, Vietnam, Kenya, Brazil, Nigeria and South Africa). School-based hearing screening focuses only on children suspected of hearing loss (ie, poor academic performance and poor attention skills).62 Such targeted screening of children with poor academic performance was reported as a resource-saving method. Additionally, in some countries like South Africa and Kenya, mHealth-based screening was used effectively to screen infants as young as 2 years to older children up to 17 years. They engaged community health workers and trained volunteers as screening personnel, and whenever professional resources were limited, they advocated that these volunteers undertake diagnostic tests under supervision.37 39 44 59 75 These are some examples of finding alternative solutions specific to the region’s resources to achieve better outcomes.
Hearing screening methods
The majority of countries lacked a standardised process or sought to adopt the JCIH protocol.80 84 Only a few countries, including South Africa, Brazil, Jordan and Iran, adapted the JCIH protocol to develop a national-level hearing screening policy statement.85 88 EHDI, on the other hand, is not legislated as a national policy in any of these countries.84 89 A similar trend was observed in Asian LMICs,19 where most countries attempted to follow JCIH protocol but were not routinely available as a mandate. The lack of uniform methods across nations further limits the availability of data that can be used to assess EHDI results at the country level.9 15
Overall, there was a tendency to adhere to JCIH criteria for EHDI as several studies reported this as the reference benchmark that they adapted24 51 54 78 yet reported difficulty in adhering to it.35 78 While attempting to meet a benchmark (1, 3, 6 previously and now 1, 2, 3) designed by and for HICs is a good aspiration, such adherence nevertheless necessitates comparable resources and contexts. As a result, there is a simultaneous need to investigate context-specific techniques with impact assessments that can result in cost-effective methods suitable to LMICs regions. Targeted screenings are considered an alternative to universal screening when resources are limited27 55; however, the current study found that programme planners mostly attempted universal screening, with relatively few employing targeted screening. This decision is highly determined by the available resources, which in turn influences the overall efficiency, number of babies screened, cost outcomes and follow-up.85
Older child screening relied largely on subjective assessments, like in other Asian LMICs.19 They were not mandated; hence, there was no common protocol.86 While objective screening is preferred, short-term initiatives based on well-conducted pilot studies that use questionnaires, behavioural techniques and/or physiological markers could be adopted.7 A few studies, for example, employed questionnaires as screening techniques to identify children suspected of having hearing loss and only performed further screening tests on those identified through the questionnaire.60 65 74
Audiologists were most frequently involved in hearing screening of neonates and infants in these countries,90 like their Asian counterparts, while nurses routinely performed screening in HICs. Such a trend in LMICs must be investigated further to understand the rationale for using professionals who are scarce in these regions to do a basic screening and the implications for sustainability.
On the other hand, screening for older children was done by community health workers at schools or camps. This differs from the practices observed in Asian LMICs and HICs, where audiologists, school teachers or nurses conduct school screening.90 Task shifting to community health workers has been advocated as an appropriate technique in low-resource settings with the scarcity of hearing healthcare experts.39 59 Some investigations have shown that task shifting can also be used to perform automated diagnostic pure tone audiometry on older children, where otolaryngologists monitor and analyse the results.31 66 67
Diagnostic methods
A few studies on neonates and babies found significantly high referral rates. The reasons were attributed to various reasons such as the short screening time after birth (<6 hours), noisy screening site and unsuitable equipment.33 34 This is significantly higher than the criteria set by JCIH (<4%). Similarly, in older children, noisy screening site, reliance on behavioural response and prevalent ear pathologies like wax impact and OME43 50 59 were the causes for higher refer rates. While HICs have achieved satisfied benchmarks, LMICs including the Asian study still report similar high referral rates as a contributing limitation.91 92
While programmes that successfully track the progress of identified children have been documented in HICs and Asian LMICs, they are scarce in non-Asian LMICs. One significant challenge noted in several studies was a failure to follow-up for second screening and diagnostic testing.16 24 34 47 60 63 86 87 The reasons were a lack of financial support to obtain these services, transportation to the testing site and parental understanding of the importance of early detection and rehabilitation. Another factor was the excessive wait times for diagnostic testing sessions and insufficient follow-up strategies. Some studies attribute the loss of follow-up to the high rate of infant mortality in their regions. The lack of effective follow-up and monitoring systems to complete all the stages of the programme seems to be a significant challenge.85
The goal of an EHDI programme is to lower the age of identification so that intervention can begin within the critical period.85 Some analyses of EHDI programmes in HICs suggest that the age of identification is around 5 weeks.18 80 However, the benefit of a hearing screening programme in lowering the age of identification in LMICs is unknown.16 19 87
Diagnostic ABR alone was used to estimate thresholds in neonates/infants, while subjective assessment with otoscopy was performed in older children as young as 2 years old. While WHO (2021) and JCIH (2019) recommend a test battery that includes ABR/ASSR, tympanometry, auditory reflex testing, otoscopy and a medical examination, this has not always been feasible in LMICs.93 This needs additional considerations as it is beyond the affordability of the screening programme.
While audiologists performed diagnostic testing on neonates/infants, capacity limitations were overcome in these countries by training community health workers and nurses to perform subjective tests such as pure tone audiometry to diagnose older children.
The prevalence was estimated regionwise as per the World Bank classification and was similar to prevalence rates reported by WHO.10 These data, however, should be viewed with caution owing to limited studies, lack of information on prevalence based on hearing thresholds, and small sample size. Another major limitation is the lack of data on the type and degree of hearing loss. The variations in hearing loss classifications also make it difficult to associate the criteria with standards given in HICs. So, the prevalence rates also varied considerably across studies, as reported previously as well.9
Intervention methods
In HICs, follow-up for intervention was usually within 3 months of identification, and the maximum age of intervention was 13.5 months.80 However, such information could not be gathered from LMICs as aspects of the intervention were reported in very few studies.26 27 40 41 43 58 61 66 67 69 71 73 Information on children who availed of interventions, including HAs, CIs and aural rehabilitation, as well as the age of intervention, was not available in the studies. The outcome of a hearing screening programme is complete only when the child receives appropriate rehabilitation or treatment.13 Due to the lack of accessible resources to support children and their families with hearing loss, the number of children receiving these interventions is lower.14 16 55 Hence, it is unclear if the expected outcomes/goals of EHDI are met in these regions. This is similar to many other studies that quote the lack of treatment-related information as one of the major limitations of the screening programmes.16 80 86 87
Strengths and limitations
Overall, this systematic review is the first known effort to understand the outcomes of hearing screening programmes in LMICs, including sub-Saharan Africa, the Middle East, Latin America and the Caribbean. This study adheres to all the required guidelines for a systematic review (PRISMA, CASP and RoB). However, there are some limitations to consider, such as the fact that we did not eliminate any articles based on the RoB assessment and that, due to data heterogeneity, we could only perform a narrative synthesis rather than a meta-analysis. Another limitation is that we did not restrict the review to include only studies that identify permanent hearing loss. Hence, the results should be viewed with caution as it includes all types and degrees of hearing loss. Furthermore, publication bias is probable as not all hearing screening projects in LMICs may have published their data in English, given the diversity of native languages in these countries.