Efficacy and Effectiveness of Community-based Newborn Hearing Screening

Efficacy and effectiveness of community-based newborn hearing screening in developing countries: a systematic review protocol of quantitative studies

BACKGROUND

In children up to age 15, disabling hearing impairment is defined as a permanent hearing loss with a severity of above 30 decibels (WHO, 2014). In newborns, hearing loss remarkably affects receptive and expressive language and, consequently, social development (Olusanya and Newton, 2007). Genetic and environmental factors including intrauterine infections, prematurity or exposure to constant noise can cause neonatal hearing impairment (Wroblewska-Seniuk et al., 2016).

Burden of disease

0.5 to 5 out of 1000 neonates are born with hearing impairment worldwide (WHO, 2014). In developing countries, this number has been estimated as 2-4 per 1000 livebirths (Olusanya and Newton, 2007). The alarming burden of newborn hearing impairment in resource-limited nations is explained by the higher burden of risk factors for this impairment, such as congenital infections, consanguineous births, natal trauma and postnatal infections (Olusanya and Newton, 2007). The impact of hearing impairment in low- and middle-income countries (LMIC) is even more disabling, hampering children optimal development and leading to life-long educational and economic consequences (Olusanya and Newton, 2007). Similarly, disabled children are more likely to encounter social discrimination and stigmatisation (McPherson, 2012).

Newborn hearing screening

Primary interventions to prevent hearing impairment are challenging to implement in LMIC (Olusanya, Wirz, and Luxon, 2008). Newborn hearing screening (NHS) is an effective approach in early detecting hearing loss in newborns and improving the prospects of affected neonates (NHS, 2019). Different protocols for NHS have been defined in terms of the targeted population and the implemented tests. In relation to the population, universal (UNHS) and targeted NHS (TNHS) for those newborns at-risk, are the two mainly used screening protocols worldwide (Huang et al., 2012). Regarding the type of physiological tests used to evaluate hearing impairment, some of them are (i) distortion product otoacoustic emissions (DPOAE) (ii) otoacoustic emissions test (OAE) and (ii) automated auditory brainstem responses (AABRs) (Friderichs, Swanepoel and Hall, 2012; WHO, 2009). However, the most frequently applied and cost-effective protocol is a two-stage screening consisting of first OAE and then AABRs test for those neonates failing the first phase of the protocol (NHS, 2019; Olusanya, Wirz and Luxon, 2008).

Get Help With Your Nursing Essay

If you need assistance with writing your nursing essay, our professional nursing essay writing service is here to help!

Nursing Essay Writing Service

In developed nations, NHS is usually offered to neonates before hospital discharge. However, in LMIC most women give birth outside hospital services, making hospital-based approaches not feasible in this context (Montagu et al., 2011; Olusanya and Okolo, 2006). It has been suggested that a more effective approach would be a community-based programme such as linking screening to immunisation clinics, doing it at home by a health visitor or in primary healthcare clinics (McPherson, 2012; Olusanya and Akinyemi, 2009; Olusanya and Okolo, 2006; Swanepoel, Hugo and Louw, 2006).

Early intervention for hearing impairment

Interventions for hearing loss include hearing aids, speech and language therapy, sign language education and psychological support (Nelson, Bougatsos and Nygren, 2008; WHO, 2004). The early diagnosis of hearing impairment in neonates accelerates intervention initiation and, therefore, improves children’s language and educational prognosis (Kennedy et al., 2006). It has been suggested that impairment detection in infants aged 6 months or less have better prognosis than those with delayed detection (Firoozbakht et al., 2014).

Rationale for this review

Screening effectiveness requires that every step of the programme works adequately to provide an optimal follow-up in order to deliver appropriate interventions (Uus and Bamford, 2006). Several reviews have systematically summarised the effectiveness of NHS compared to no screening (Wolff et al., 2009), the effectiveness of UNHS compared to TNHS (Nelson, Bougatsos and Nygren, 2008; Thompson et al., 2001; Grill et al., 2005), the overall follow-up rate in NHS (Ravi et al., 2016) and its cost-effectiveness (Colgan et al., 2012). However, these reviews only considered hospital-based programmes with a global focus instead of poor settings. This review aims to systematically review the literature on efficacy and effectiveness of community-based newborn hearing screening programmes in LMIC.

METHODS

Protocol preparation

This systematic review protocol was developed following the recommended items in the PRISMA-P (Preferred Reporting Items for Systematic review and Meta-Analysis Protocols) 2015 checklist (Moher et al., 2015). This systematic review will be conducted from June 2019 to December 2021.

Eligibility criteria

Only quantitative studies will be included in this systematic review. However, no other criteria for study design will be applied to maximise the inclusion of studies. Inclusion criteria will be restricted to studies conducted in LMIC. Countries will be classified by income following the World Bank classifications (The World Bank, 2019). No language or time restrictions will be applied in this systematic review. Inclusion criteria will be based on the following PICO:

Participants: Neonates and infants (aged up to 2 years). No restriction on gender will be applied.

Interventions: Universal or targeted community-based NHS programmes consisting of DPOAE , OAE, AABRs or both OAE and AABRs will be included. Community-based programmes will include screening at immunisation clinics, primary healthcare clinics, private obstetric clinics, maternity centres or home visits. Studies on programmes based on hospital-based approaches or on qualitative / behavioural diagnosis test will be excluded.

Comparators:No applicable.

Outcome: (i) To evaluate efficacy screening coverage, first- and second-stage referral, overall follow-up and age at testing will be evaluated. (ii) To evaluate effectiveness age at diagnosis, sensitivity and specificity will be assessed.

Information sources

Several electronic databases will be consulted to identify relevant studies including Medline (Ovid), Embase, PsycINFO, CINAHL, Web of Science, Global Health and Cochrane databases. Other sources will include reference lists of all identified eligible studies, grey literature, google scholar, professionals in this area and registers of not published studies. Finally, governmental or public health reports from LMIC will be also considered.

Search strategy

A similar search strategy will be applied for each database, an example for Medline (Ovid) is given in Appendix 1. Broad terms related to the PICOs of interest will be searched and truncation will be used to comprise more studies around this topic. Studies in any language will be included and a professional translator will be contracted for researches in languages other than English. Otherwise, they will be excluded from the review. Boolean operators such as ‘OR’ and ‘AND’ will be used.

Study selection

Electronic papers identified from the indicated information sources will be exported into EndNote software to manage records and references. Screening of title and/or abstracts will be independently done by two researchers to evaluate if identified records are eligible according to the previously specified inclusion criteria (Figure 1). If disagreement arises a third reviewer will arbitrate it to reach a consensus. Retrieved studies after initial screening will be fully reviewed to confirm that eligibility criteria are met.

Data collection process

Data will be extracted in duplicate by two independent reviewers. If disagreement arises, a third researcher will arbitrate the situation to reach a consensus. A data extraction form will be constructed and piloted on a couple of studies. The data collection form will include the following items: language, first author, year of publication, country of data collection, aims, study design, participants (number, gender, mean age), targeted population (UNHS or TNHS), testing protocol (DPOAE, OAE, AABRs or both OAE and AABRs), screening performance (screening coverage, age at testing, first- and second-stage referral rate), screening effectiveness (sensitivity, specificity and age at diagnosis) and methodological quality. Authors will be contacted if further information is needed.

Outcomes and prioritization

For screening efficacy, the primary outcomes will be:

-          Screening coverage

-          First- and second stage referral rate

-          Overall follow-up rate

-          Age at testing

Since age at testing is essential in determining the infant’s prospects, it will also be defined as an outcome.

To assess effectiveness, the primary outcome will be age at diagnosis,  because it has been proved that detection in infants aged 6 months or less improves prognosis. Anticipating data missingness for these variables, sensitivity and specificity will be defined as secondary outcomes.

Quality assessment

Methodological quality and potential risk of bias for individual researches will be evaluated in duplicate and independently by two researchers following the Centre for Reviews and Dissemination (CRD) guidance (NHS Centre for Reviews and Dissemination, 2001). If included studies present deficiencies they will be classified as major or minor deficiencies. Items such as confounding, sample size, individuals flow transparency or comparability of baseline characteristics will be evaluated. The QUADAS tool will be applied to evaluate the quality of diagnostic precision in included researches (Whiting et al., 2003).

Data synthesis

Conceptual, methodological and statistical heterogeneity across studies is anticipated, considering study design, participants (universal versus targeted screening), interventions (DPOAE, OAE, AABRs, or both OAE and AABRs), and outcomes, therefore, a meta-analysis is unlikely to be feasible. The primary approach to summarise the collected information will be narrative synthesis. If data allow, meta-analysis will be conducted through random-effects modelling. In the narrative synthesis all studies will be included without taking into consideration their quality. However, if meta-analysis is possible only studies with minor deficiencies will be included.

Subgroup analysis

If feasible, synthesis will be categorised into 3 groups: (i) population (ii) interventions (iii) geographical location of data collection.

For the first grouping category (i) the following subgroups will be compared:

–          Universal community-based NHS

–          Targeted community-based NHS

For targeted screening infants will be classified as being at-risk if there is intrauterine TORCH infection (including toxoplasmosis, cytomegalovirus, syphilis, rubella or herpes), familial hearing loss or cranio-facial anomalies (Karaca et al., 2014; Olusanya, Luxon and Wirz, 2004). Other risk factor will include: low birthweight (<1.5kg), parental consanguinity, bacterial meningitis, ototoxic medications, hyperbilirubinemia or hospitalisation in intensive unit (Karaca et al., 2014; Olusanya, Luxon and Wirz, 2004).

For the second grouping category (ii) the following subgroups will be compared:

–          Community-based NHS consisting only of OAE

–          Community-based NHS consisting only of AABRs

–          Community-based NHS consisting of OAE and AABRs sequentially

–          Community-based NHS consisting of DPOAE

For the third grouping category (iii), regions will be classified following the World Bank classifications (The World Bank, 2019).

–          East Asia and Pacific

–          Europe and Central Asia

–          Latin America & the Caribbean

–          Middle East and North Africa

–          South Asia

–          Sub-Saharan Africa

A common metric will be used to synthesise data. In this case effect direction will be used. Following the WHO guidelines and other studies in this field, the outcome thresholds for efficacy would include: age at first screening as 1 month, follow-up rate as 70% and screening coverage rate of 90% (Ravi et al., 2016; WHO, 2009). For effectiveness, age at diagnosis will be set up at 6 months (Firoozbakht et al., 2014). If the overall age at first screening for an individual study is equal or less than 1 month; the screening coverage is equal or above 90% and the age at diagnosis is equal or less than 6 months the intervention of study will be considered as having a positive effect on the outcome. In the opposite scenario, the intervention of study will be considered as having a negative effect on the outcome. An overall effect (number of positive studies) will be calculated for each of the subgroups and compared to other sets in the same groups. Any later changes since the protocol will be reported and justified.

Meta-bias

If meta-analysis is feasible, publication bias across included studies will be assessed with funnel plots and Begg and Egger tests.

Confidence in cumulative evidence

GRADE guidelines will be used to rate the strength of evidence on the base of risk of bias, publication bias, imprecision, consistency, directness and accuracy (Guyatt, Oxman and Schünemann, 2013)

CONCLUSION

Newborn hearing screening has been proved to be an effective secondary measure to improve the prognosis of children with hearing impairment. In LMIC, most women deliver outside hospital facilities making difficult to establish hospital-based programmes. Although community-based approaches have been suggested as an effective alternative, little is known about its effectiveness. This systematic review will help evaluate the effectiveness of community-based screening to  establish public health recommendations in LMIC. 

PERSONAL REFLECTION

Systematic reviews summarise quality literature around a relevant topic. In public health, these studies have an essential role in synthesising the available evidence on potentially effective interventions to generate guidelines or consensus to protect, promote or improve the health of a population. Generally, systematic reviews in public health are complex to develop, as usually participants, interventions and outcomes are widely heterogenic (Jackson and Waters, 2005). Before, taking this course when thinking about a systematic review or meta-analysis I always thought of it as a statistical synthesis of individual studies. However, the term systematic review englobes more scenarios such as narrative synthesis for heterogenous quantitative data or for qualitative data. In fact, this kind of approaches are widely used as the synthesis methods in public health systematic reviews. 

When thinking of a topic for a systematic literature review it was challenging to find an original research question since all the ideas that came initially to my mind were already addressed in other systematic reviews or meta-analysis. In this systematic review protocol, the definition of the population, intervention and outcomes was remarkably challenging because of the heterogenous data subject to this health issue. Multiple interventions have been previously defined to perform newborn hearing screening, including different participants (universal versus targeted screening) and different diagnosis test (DPOAE, OAE, AABRs, or both) therefore, it was difficult to narrow down which ones would be included in the review. For the evaluation of outcomes multiple measurements were considered and it was challenging to decide which ones would answer more accurately the current research question. Similarly, the lenient exclusion criteria for study designs in this protocol will generate remarkable heterogenicity in data extraction.

Critically reflecting my protocol, l can anticipate that the inclusion criteria for geographical location of data collection will be challenging. Generally, fewer researches are conducted in limited-resources settings and this will limit the literature search for this review. Additionally, a systematic review published in the Bulletin of the World Health Organization evidenced that research quality in developing countries is suboptimal in a relevant proportion of studies (Ryman and Dietz, 2008). Therefore, the inclusion of studies with data exclusively collected in LMIC might be subject to a high risk of individual bias. Another limitation of this protocol is that narrative synthesis can be sometimes a limited approach when addressing effectiveness of an intervention. However, although a statistical synthesis might not be feasible with this type of data synthesis, the nature and direction of the intervention effect can be identified. Moreover, narrative synthesis allows to identify patterns in the effects of an intervention.

While preparing this protocol, I realised the importance of this first step in the development of a systematic review. Initially, I had an established idea about what my research question would be. However, as I started writing this protocol I encountered several limitations which made me change my mind and narrow down the research questions for this protocol. Facing these barriers in early stages of the research avoids their later occurrence, when the consequences would be more complex to solve. This fact highlighted me the essential role of a protocol in the design strategy of a systematic review to justify the purpose of the review, clarify the methods and resources that will be applied and overall, minimise bias.

REFERENCES

• Colgan, S., Gold, L., Wirth, K., Ching, T., Poulakis, Z., Rickards, F. and Wake, M. (2012). The Cost-Effectiveness of Universal Newborn Screening for Bilateral Permanent Congenital Hearing Impairment: Systematic Review. Academic Pediatrics, 12(3), pp.171-180.
• Firoozbakht, M., Mahmoudian, S., Alaeddini, F., Esmaeilzadeh, M., Rafiei, M., Firouzbakht, A., Rahimi, F. and Farhadi, M. (2014). Community-based newborn hearing screening programme for early detection of permanent hearing loss in Iran: An eight-year cross-sectional study from 2005 to 2012. Journal of Medical Screening, 21(1), pp.10-17.
• Friderichs, N., Swanepoel, D. and Hall, J. (2012). Efficacy of a community-based infant hearing screening program utilizing existing clinic personnel in Western Cape, South Africa. International Journal of Pediatric Otorhinolaryngology, 76(4), pp.552-559.
• Grill, E., Hessel, F., Siebert, U., Schnell-Inderst, P., Kunze, S., Nickisch, A. and Wasem, J. (2005). Comparing the clinical effectiveness of different new-born hearing screening               strategies. A decision analysis. BMC Public Health, 5(1).
• Guyatt, G., Oxman, A. and Schünemann, H. (2013). GRADE guidelines—an introduction to the 10th–13th articles in the series. Journal of Clinical Epidemiology, 66(2), pp.121-123.
• Huang, L., Zhang, L., Tobe, R., Qi, F., Sun, L., Teng, Y., Ke, Q., Mai, F., Zhang, X., Zhang, M., Yang, R., Tu, L., Li, H., Gu, Y., Xu, S., Yue, X., Li, X., Qi, B., Cheng, X., Tang, W., Xu, L.               and Han, D. (2012). Cost-effectiveness analysis of neonatal hearing screening program in china: should universal screening be prioritized? BMC Health Serv Res, 12(1):97.
• Jackson, N. and Waters, E. (2005). Criteria for the systematic review of health promotion and public health interventions. Health Promotion International, 20(4), pp.367-374.
• Karaca, Ç., Oysu, Ç., Toros, S., Naiboǧlu, B. and Verim, A. (2014). Is Hearing Loss in Infants Associated With Risk Factors? Evaluation of the Frequency of Risk Factors. Clinical and Experimental Otorhinolaryngology, 7(4), p.260.
• Kennedy, C., McCann, D., Campbell, M., Law, C., Mullee, M., Petrou, S., Watkin, P., Worsfold, S., Yuen, H. and Stevenson, J. (2006). Language Ability after Early Detection of Permanent Childhood Hearing Impairment. New England Journal of Medicine, 354(20), pp.2131-2141.
• McPherson, B. (2012). Newborn hearing screening in developing countries: Needs & new directions. Indian J Med Res, 135(2), pp.152–153.
• Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P. and Stewart, L. (2015). Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Reviews, 4(1).
• Montagu, D., Yamey, G., Visconti, A., Harding, A. and Yoong, J. (2011). Where Do Poor Women in Developing Countries Give Birth? A Multi-Country Analysis of Demographic and Health Survey Data. PLoS ONE, 6(2), p.e17155.
• Nelson, H., Bougatsos, C. and Nygren, P. (2008). Universal Newborn Hearing Screening: Systematic Review to Update the 2001 US Preventive Services Task Force               Recommendation. PEDIATRICS, 122(1), pp.e266-e276.
• NHS. (2019). Newborn hearing screening. [online] Available at: https://www.nhs.uk/conditions/pregnancy-and-baby/newborn-hearing-test/ [Accessed 29 Apr. 2019].
• NHS Centre for Reviews and Dissemination. (2001) Undertaking systematic reviews of research on effectiveness: CRD’s guidance for those carrying out or commissioning reviews. CRD Report 4. York: University of York. Available at: http://www.crd.york.ac.uk/CRDWeb/ShowRecord.asp?AccessionNumber=32001000984 [Accessed 1 May 2019]
• Olusanya, B. and Akinyemi, O. (2009). Community-based infant hearing screening in a developing country: parental uptake of follow-up services. BMC Public Health, 9(1).
• Olusanya, B. and Newton, V. (2007). Global burden of childhood hearing impairment and disease control priorities for developing countries. The Lancet, 369(9569), pp.1314-              1317.
• Olusanya, B. and Okolo, A. (2006). Early hearing detection at immunization clinics in developing countries. International Journal of Pediatric Otorhinolaryngology, 70(8), pp.1495-1498.
• Olusanya, B., Luxon, L. and Wirz, S. (2004). Benefits and challenges of newborn hearing screening               for developing countries. International Journal of Pediatric Otorhinolaryngology, 68(3), pp.287-305.
• Olusanya, B., Wirz, S. and Luxon, L. (2008). Hospital-based universal newborn hearing screening for early detection of permanent congenital hearing loss in Lagos, Nigeria. International Journal of Pediatric Otorhinolaryngology, 72(7), pp.991-1001.
• Ravi, R., Gunjawate, D., Yerraguntla, K., Lewis, L., Driscoll, C. and Rajashekhar, B. (2016). Follow-up in newborn hearing screening – A systematic review. International Journal of Pediatric Otorhinolaryngology, 90, pp.29-36.
• Ryman, T. and Dietz, V. (2008). A need to improve quality, rigour and dissemination of operations research. Bulletin of the World Health Organization, 86(11), pp.907-907.
• Swanepoel, D., Hugo, R. and Louw, B. (2006). Infant hearing screening at immunization clinics in South Africa. International Journal of Pediatric Otorhinolaryngology, 70(7), pp.1241-1249.
• The World Bank. (2019). World Bank Country and Lending Groups. [online] Available at: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-              country-and-lending-groups [Accessed 29 Apr. 2019].
• Thompson, D., McPhillips, H., Davis, R., Lieu, T., Homer, C. and Helfand, M. (2001). Universal Newborn Hearing Screening. JAMA, 286(16), p.2000.
• Tobe, R., Mori, R., Huang, L., Xu, L., Han, D. and Shibuya, K. (2013). Cost-Effectiveness Analysis of a National Neonatal Hearing Screening Program in China: Conditions for the Scale-Up. PLoS ONE, 8(1), p.e51990.
• Uus, K. and Bamford, J. (2006). Effectiveness of Population-Based Newborn Hearing Screening in               England: Ages of Interventions and Profile of Cases. PEDIATRICS, 117(5), pp.e887-e893.
• Whiting, P., Rutjes, A., Reitsma, J., Bossuyt, P. and Kleijnen, J. (2003). The development of QUADAS:               a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Medical Research Methodology, 3(1).
• Wilson JMG, and Jungner G. (1968). Principles and Practice of Screening for Disease. World Health Organization Public Health Papers, 34. [Online]. [Accessed 27 Apr. 2019]. Available from: http://whqlibdoc.who.int/php/WHO_PHP_34.pdf. Accessed November 28, 2008.
• Wolff, R., Hommerich, J., Riemsma, R., Antes, G., Lange, S. and Kleijnen, J. (2009). Hearing screening in newborns: systematic review of accuracy, effectiveness, and effects of interventions after screening. Archives of Disease in Childhood, 95(2), pp.130-135.
• World Health Organisation (WHO). 2004. Guidelines for hearing aids and services for developing countries [2^nd edition]. [Online]. [Accessed 27 Apr. 2019]. Available from: https://apps.who.int/iris/bitstream/handle/10665/43066/9241592435_eng.pdf?sequ              ence=1
• World Health Organisation (WHO). 2009. Newborn and infant hearing screening. [Online]. [Accessed 1 May. 2019]. Available from: https://www.who.int/blindness/publications/Newborn_and_Infant_Hearing_Screening_Report.pdf?ua=1
• Wroblewska-Seniuk, K., Dabrowski, P., Szyfter, W. and Mazela, J. (2016). Universal newborn hearing screening: methods and results, obstacles and benefits. Pediatr Res, 81(3),               pp.415-422.

APPENDIX