You are here

Article Text

Article menu
XML
Immunology (including allergy)
Protocol
Sibship size, birth order and risk of asthma and allergy: protocol for a systematic review and meta-analysis
  1. Daniil Lisik1,
  2. Athina Ioannidou1,
  3. Gregorio Milani2,3,
  4. Sungkutu Nyassi1,
  5. Saliha Selin Özuygur Ermis4,
  6. Giulia C I Spolidoro3,
  7. Emma Goksör1,
  8. Göran Wennergren1,5,
  9. Bright I Nwaru1
  1. 1Krefting Research Centre, University of Gothenburg, Gothenburg, Sweden
  2. 2Pediatric Unit, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy
  3. 3Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy
  4. 4Department of Pulmonology, Dokuz Eylül University, Izmir, Turkey
  5. 5Department of Paediatrics, Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden
  1. Correspondence to Daniil Lisik; daniil.lisik{at}gmail.com

Abstract

Introduction The hygiene hypothesis suggests that reduced exposure to microbes might have contributed to the increase in prevalence and incidence of asthma and allergy observed during the second half of the last century. Following this proposal, several studies have investigated the role of sibship size and birth order in the development of asthma and allergic diseases, but the underlying evidence is conflicting. The objective of the present systematic review will be to identify, critically appraise and synthesise previous primary studies investigating the association of sibship size and birth order with the risk of asthma and allergic diseases.

Methods and analysis The following databases will be searched: AMED, CABI, CINAHL, Embase, Google Scholar, OAIster, Open Access Theses and Dissertations, Open Grey, ProQuest Dissertations & Theses Global, PsycINFO, PubMed, SciELO, Scopus, Web of Science and WHO Global Index Medicus. Studies published up until 31 December 2020 will be eligible. There will be no restrictions by language and geographical location. Risk of bias in the included studies will be assessed using the Effective Public Health Practice Project quality assessment tool. The produced evidence will be synthesised narratively, and studies that present comparable numerical data will be included in meta-analyses using random effects model.

Ethics and dissemination Only data from the published literature will be included in this systematic review. Therefore, no ethical approval is required. The final review paper will be published in a peer-reviewed journal.

PROSPERO registration number CRD42020207905.

  • asthma
  • allergy
  • eczema
  • immunology
  • paediatric infectious disease & immunisation
  • paediatric dermatology
https://creativecommons.org/licenses/by/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.

https://doi.org/10.1136/bmjopen-2020-045795

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • This will be the first systematic review encompassing a comprehensive spectrum of the most common allergic and respiratory outcomes, in relation to sibship size and birth order.

    • Inclusion of the leading databases, including search of the grey literature, enables a comprehensive identification of the relevant studies addressing the research question.

    • The reproducibility of our work is enhanced through a priori outline of the review processes before the actual review starts.

    • Self-reported diagnoses of the study outcomes are expected to make up a significant source of data from included studies, which gives the possibility of assessment bias.

    Introduction

    The incidence and prevalence of asthma, along with allergic diseases such as allergic rhinitis and atopic eczema, were observed to have increased during the second half of the last century, in particular in the developed world.1 2 More recent trends remain unclear, as both increase3 and levelling off1 4 5 have been suggested. Around 300 million people have asthma globally.1 For allergic diseases, evidence indicates that there is still a global increase in prevalence.5 6 Asthma and allergic diseases account for significant morbidity for individuals, as well as a substantial socioeconomic burden on the society.2 Asthma results in roughly 14 million missed school days each year in the USA alone, and the morbidity is even higher for adults.7 Allergic rhinitis is also associated with significant loss in productivity.8 Furthermore, the WHO estimates that roughly 250 000 cases of death annually, worldwide, are due to asthma.7 Identifying risk factors for asthma and allergy is therefore of great interest in order to reduce the burden associated with these diseases.

    Over the last five decades, numerous hypotheses have been proposed to explain the observed increase in the prevalence of these diseases, a substantial part of the studies focusing on the role of environmental factors. One of the main hypotheses is the hygiene hypothesis, which was first proposed by Strachan9 in 1989, and suggests that reduced microbial exposure during childhood increases the risk of developing asthma and allergy. One of the first proposed underlying biological mechanisms to the hygiene hypothesis was the observed skewing of balance towards T helper 2 cells (which have been associated with allergic sensitisation) in subjects lacking microbial stimuli and conversely a T helper 1 cell domination in subjects exposed to greater quantities of microbes.10 Further research has broadened the explanatory model with additional factors, such as T regulatory cells and T helper 17 cells,11 but the pathophysiology is yet to be fully understood.12 Connected to the hygiene hypothesis is the proposed sibling effect, which suggests that the number of siblings and/or the birth order of a child in a family may play a role in the development of asthma and allergy, as a result of varying degrees of microbial exposure during childhood, depending on the number of siblings in total and/or the number of younger/older siblings.13

    While several studies have investigated the association of sibship (group of individuals sharing the same pair of parents) size and birth order (the sequence in which members of a sibship are born) with risk of asthma and allergic diseases, findings are conflicting.14 Karmaus and Botezan have estimated the proportion of cases attributable to the sibling effect to be 34% for atopic dermatitis, 56% for allergic rhinitis and 28% for asthma. Karmaus and Botezan have also argued that at least 30% of cases of asthma and allergy could be prevented if the causal factors for these conditions were better understood,15 further indicating warranty for elucidating the sibling effect in relation to asthma and allergy. So far, there are no systematic reviews synthesising evidence from previous studies on the topic. A systematic synthesis of previous studies investigating the association of sibship size and birth order with risk of asthma and allergy will provide a clearer appreciation of the strength, magnitude and quality of the underlying evidence.

    Aim

    To identify, critically appraise and synthesise previous primary studies investigating the association of sibship size and birth order with risk of asthma and allergic diseases.

    Methods and analyses

    This protocol is reported according to the recommendations of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P),16 which provides guidelines for a standardised, transparent and reproducible reporting of systematic review protocols. The PRISMA-P checklist is presented in online supplemental appendix 1. Updates to the protocol will be documented, and deviations from the protocol will be described in the final review paper. The protocol for this systematic review has been prospectively registered with the international prospective register of systematic reviews (PROSPERO, https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=207905) with registration number CRD42020207905.

    Study eligibility criteria

    Study types and publication status

    We will include observational epidemiological studies, including prospective and retrospective cohort studies, case–control studies and cross-sectional studies. Randomised controlled studies, quasirandomised controlled studies, controlled before-after studies and controlled clinical trials will not be considered, as interventional studies are not relevant for this research question. Animal studies, reviews, case studies, case series, expert opinions will also be excluded. Studies of any publication status will be eligible, and data used from them if available.

    Participants

    Offspring of any age, gender, ethnic background and medical background, where the study context is that the participants are part of defined sibships. Studies with any amount of participants will be eligible.

    Exposures

    Sibship size, birth order (number of older siblings) and number of younger siblings in the studied sibships.

    Outcome measures

    Self-reported or objectively measured/diagnosed asthma and allergic disease in the sibships. For the purpose of encompassing all relevant literature on the topic, asthma and allergic disease will be defined broadly. Asthma will be defined as any type of asthma, including those based on symptom definition, such as wheezing, and those based on spirometry findings of variable expiratory airflow limitation.17 Allergic disease will encompass any of the following: (A) allergic rhinitis/(rhino)conjunctivitis, food allergy, atopic eczema, urticaria, angio-oedema, anaphylaxis18 and (B) indicators of hypersensitivity (and indirectly of allergic disease), which includes allergen-specific serum IgE test, skin prick test and provocation/challenge test. Conditions with primarily a genetic aetiology, such as hereditary angioedema,19 will not be included in these definitions.

    Search methods

    The search queries were developed using the PEO model: population, exposure and outcome (PEO). PEO is a specific implementation of population, intervention, control, outcome (PICO), used as a framework to produce effective search queries from formulated research questions, especially befitting retrieval of interventional and observational studies.20 Since the population (P) will be defined broadly, that is, including both studies in children and adults, the actual search queries will be composed of two blocks: exposure (E) and outcome (O). A scoping search was performed in PubMed to identify previous studies on the topic and map relevant search terms. The search terms identified were: Medical Subject Headings (MeSH) and their corresponding alternatives in other databases, entry terms, free-text words and phrases. Subsequent scoping searches were made in PubMed with Boolean operator ‘NOT’ between various MeSH and free-text terms, alternately, in order to identify more synonyms and related search terms. The developed search terms have been piloted and refined before they will be used to identify relevant studies. The search queries have been modified for each database to be searched in regards to, inter alia, support for controlled vocabulary and syntax. Peer Review of Electronic Search Strategies has been used to identify potential weaknesses in the search strategy. Details of the search strategy are presented in online supplemental appendix 2.

    Studies will be retrieved from the following databases: AMED (via Ovid), CABI, CINAHL (via EBSCO), Embase (via Ovid), Google Scholar, PsycINFO (via ProQuest), PubMed, SciELO, Scopus, Web of Science and WHO Global Index Medicus. In addition, unpublished articles and grey literature will be retrieved through searches of OAIster, Open Access Theses and Dissertations, Open Grey and ProQuest Dissertations & Theses Global. Finally, studies will also be included from reference lists of the studies included in the review, as well as through contact with experts who have published in the field. All databases will be searched for articles published from inception of respective database up until 3 December 2020; an updated search will be performed at the completion of the review to ensure inclusion of studies published after the first search. There will be no language restrictions, and articles will be translated into English where possible. Articles that could not be translated will be reported in the final review paper. In Google Scholar, due to the fact that the amount of results is sometimes overwhelming, results will be retrieved from the first 300 hits.21 Furthermore, the search query for Google Scholar has been significantly simplified, including only the most important terms in each block, due to an upper limit of 256 characters for search strings.22

    Data management

    EndNote will be used for deduplication, full-text retrieval, secondary screening and for general management of retrieved studies. For primary screening, the articles will be imported to Rayyan QCRI.

    Screening/selection process

    The first stage of screening will be based on the title and/or abstract of the articles. Articles that are clearly not relevant to the research question or clearly meet any of the exclusion criteria will be excluded. Articles, where there is doubt about relevancy, will be included to the next step. In the second stage of screening, the full text of the articles will be retrieved and assessed for eligibility. The reason for each article not being included will be documented and presented in a Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram in the final review paper.23 The screening/selection will be independently performed by two reviewers. A third reviewer will arbitrate any disagreement.

    Data extraction

    A data extraction form (online supplemental appendix 3) has been developed to extract data from included studies in a standardised and reproducible fashion. The form will be piloted and revised before being used in the review. If a study does not present needed data, authors of the study will be contacted. Extracted data will be presented in table form. The extraction will be independently conducted by two reviewers. A third reviewer will arbitrate any disagreement.

    Data items

    The following data items will be summarised from each study: author of publication; country of origin of study; publication year; type of study design; sample size of study; source from where study participants were recruited; definition and assessment of sibship and birth order; duration of follow-up; confounding factors adjusted included in studies; study outcomes and their assessment; analysis methods; and main results.

    Quality assessment

    Quality and risk of bias in the individual, included studies will be assessed using the Effective Public Health Practice Project Quality Assessment Tool (EPHPP).24 The EPHPP contains six domains of assessment for each study, including study design, selection bias, confounding, blinding, study collection, withdrawals and dropouts. Based on the grading of each of the six domains, a global quality grading will be derived for each study. Detailed results will be presented in a separate table in the final review paper. Appraisal of quality and risk of bias will be independently performed by two reviewers. A third reviewer will arbitrate any disagreement.

    Data synthesis

    Descriptive tables will be generated to present the key characteristics of the included studies. The produced evidence will be synthesised narratively. In addition, studies that present numerically comparable and reasonably homogenous (in terms of clinical and epidemiological settings of study participants) data will be synthesised quantitatively with meta-analyses in RevMan V.5 to produce pooled effect size estimates. Random effects model will be applied in the meta-analyses, because the included studies, solely based on published literature, are anticipated to not be similar in every aspect and thus do not estimate the same effect. This model is more conservative and provides a realistic scenario in the context of studies gathered solely from published literature.25 Separate meta-analyses will be undertaken for each of the factors investigated (sibship size and birth order) in relation to each asthma and allergy outcome. The results of the meta-analyses will be presented in forest plots.

    Risk ratio (RR) will be used as the outcome measure in the meta-analyses, because of its intuitive interpretative feature.26 Data from studies presenting effect measures as OR, incidence rate ratio (IRR) or HR will be converted to estimates of RR before combining with other studies, using the following formulas:

    1. RR ≈ IRR.

    2. RR ≈ HR or OR (if outcome is <15% by the end of follow-up).

    3. RR ≈ Embedded Image or Embedded Image (if outcome is ≥15% by the end of follow-up).27

    Calculation of I2 will quantify heterogeneity between the included studies.28 Consideration will be taken, regarding that this statistic can be biased in meta-analyses with few studies.29 Subgroup analysis will be performed to explore potential reasons for heterogeneity between studies with the following subgroup variables: (A) study design; (B) quality appraisal of studies; (C) classification of the study country into ‘high-income’, ‘upper-middle-income’, ‘lower-middle-income’ and ‘low-income’ economy, as defined by the World Bank30 ; (D) time during which the study was conducted, grouped into <1990, 1990–1999, 2000–2009 and 2010–2020; (E) participant age, grouped arbitrarily into <1 year, 1–6 years, 7–14 years and ≥15 years; and (F) gender, grouped into ‘male’ and ‘female’. Subgroup analysis will be performed if there will be at least four (arbitrarily chosen cut-off31) studies in at least two subgroups. In addition, if more than 10 included studies present comparable numerical data,32 meta-regression will be performed to explore the impact of explanatory variables (covariates) on the observed heterogeneity in estimates across studies.

    To investigate whether the conclusions of the review are independent of arbitrary decisions, sensitivity analysis will be performed by only including studies that: (A) reach either ‘strong’ or ‘moderate’ global rating of quality in accordance to EPHPP and (B) have objectively verified diagnosis of asthma or allergic disease as outcome, with either ICD codes or verified medical examination as the basis for diagnosis. The sensitivity analysis will be reported in a summary table.

    Publication bias

    Publication bias will be assessed with Funnel plot, as well as Begg’s rank test and Egger’s regression test,33 34 with p<0.05 being defined as statistically significant. In case of (significant) publication bias, the trim-and-fill method will be implemented to analyse its influence on the review results.35

    Patient and public involvement

    No patients or participants were involved in the development of this protocol or the design of this study.

    Discussion

    The conclusions that will potentially be drawn from this systematic review will be limited by the quality of the included studies. For this research question, the fact that all included studies will be observational limits the establishment of causality between sibship size, birth order and risk of asthma and allergic diseases.36

    A strength of this study is the comprehensiveness of the search strategy, including 15 of the leading databases of formally published literature, as well as grey literature. There will be no restrictions in terms of language or geographical location. All these enable comprehensive identification of relevant studies for this research question. Furthermore, this systematic review will encompass a comprehensive spectrum of the most common asthma and allergic outcomes in relation to sibship size and birth order, thereby contributing to a broad overview of the existing evidence on the topic.

    Asthma and allergic diseases pose a significant burden on both individuals and society. While the role of sibship size and birth order in the development of these diseases have been investigated in several studies, although with conflicting evidence, a systematic review of existing studies is essential in providing a clearer appreciation of the underlying evidence. This protocol presents the methodology to perform a comprehensive systematic review and meta-analysis of existing literature on the topic.

    Ethics and dissemination

    Ethical approval will not be required due to the study being a systematic review of already published primary studies available in the public domain. Furthermore, patient consent will not be needed since data will stay aggregated.37 38

    Ethics statements

    Patient consent for publication

    References

    1. ↵
      1. Dharmage SC,
      2. Perret JL,
      3. Custovic A
      . Epidemiology of asthma in children and adults. Front Pediatr 2019;7:246.doi:10.3389/fped.2019.00246pmid:http://www.ncbi.nlm.nih.gov/pubmed/31275909
      OpenUrlPubMed
    2. ↵
      1. Nunes C,
      2. Pereira AM,
      3. Morais-Almeida M
      . Asthma costs and social impact. Asthma Res Pract 2017;3:1.doi:10.1186/s40733-016-0029-3pmid:http://www.ncbi.nlm.nih.gov/pubmed/28078100
      OpenUrlCrossRefPubMed
    3. ↵
      1. Lundbäck B,
      2. Backman H,
      3. Lötvall J, et al
      . Is asthma prevalence still increasing? Expert Rev Respir Med 2016;10:39–51.doi:10.1586/17476348.2016.1114417pmid:http://www.ncbi.nlm.nih.gov/pubmed/26610152
      OpenUrlPubMed
    4. ↵
      1. Martinez FD
      . Trends in asthma prevalence, admission rates, and asthma deaths. Respir Care 2008;53:561–5. discussion 565-567.pmid:http://www.ncbi.nlm.nih.gov/pubmed/18426610
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Hicke-Roberts A,
      2. Åberg N,
      3. Wennergren G, et al
      . Allergic rhinoconjunctivitis continued to increase in Swedish children up to 2007, but asthma and eczema levelled off from 1991. Acta Paediatr 2017;106:75–80.doi:10.1111/apa.13433pmid:http://www.ncbi.nlm.nih.gov/pubmed/27102081
      OpenUrlPubMed
    6. ↵
      1. Doll RJ,
      2. Joseph NI,
      3. McGarry D
      . Epidemiology of allergic diseases. allergy and asthma: the basics to best practices. Cham: Springer International Publishing, 2019: 31–51.
    7. ↵
      1. Pawankar R
      . Allergic diseases and asthma: a global public health concern and a call to action. World Allergy Organ J 2014;7:12.doi:10.1186/1939-4551-7-12pmid:http://www.ncbi.nlm.nih.gov/pubmed/24940476
      OpenUrlPubMed
    8. ↵
      1. Bousquet J,
      2. Khaltaev N,
      3. Cruz AA, et al
      . Allergic Rhinitis and its Impact on Asthma (ARIA) 2008 update (in collaboration with the World Health Organization, GA(2)LEN and AllerGen). Allergy 2008;63 Suppl 86:8–160.doi:10.1111/j.1398-9995.2007.01620.xpmid:http://www.ncbi.nlm.nih.gov/pubmed/18331513
      OpenUrlPubMed
    9. ↵
      1. Strachan DP
      . Hay fever, hygiene, and household size. BMJ 1989;299:1259–60.doi:10.1136/bmj.299.6710.1259pmid:http://www.ncbi.nlm.nih.gov/pubmed/2513902
      OpenUrlFREE Full Text
    10. ↵
      1. Pfefferle PI,
      2. Teich R,
      3. Renz H
      . The immunological basis of the hygiene hypothesis. Springer Japan, 2009: 325–48.
    11. ↵
      1. Stiemsma LT,
      2. Reynolds LA,
      3. Turvey SE, et al
      . The hygiene hypothesis: current perspectives and future therapies. Immunotargets Ther 2015;4:143–57.doi:10.2147/ITT.S61528pmid:http://www.ncbi.nlm.nih.gov/pubmed/27471720
      OpenUrlCrossRefPubMed
    12. ↵
      1. Lambrecht BN,
      2. Hammad H
      . The immunology of the allergy epidemic and the hygiene hypothesis. Nat Immunol 2017;18:1076–83.doi:10.1038/ni.3829pmid:http://www.ncbi.nlm.nih.gov/pubmed/28926539
      OpenUrlCrossRefPubMed
    13. ↵
      1. Bloomfield SF,
      2. Rook GA,
      3. Scott EA, et al
      . Time to abandon the hygiene hypothesis: new perspectives on allergic disease, the human microbiome, infectious disease prevention and the role of targeted hygiene. Perspect Public Health 2016;136:213–24.doi:10.1177/1757913916650225pmid:http://www.ncbi.nlm.nih.gov/pubmed/27354505
      OpenUrlPubMed
    14. ↵
      1. Goldberg S,
      2. Israeli E,
      3. Schwartz S, et al
      . Asthma prevalence, family size, and birth order. Chest 2007;131:1747–52.doi:10.1378/chest.06-2818pmid:http://www.ncbi.nlm.nih.gov/pubmed/17413052
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Karmaus W,
      2. Botezan C
      . Does a higher number of siblings protect against the development of allergy and asthma? A review. J Epidemiol Community Health 2002;56:209–17.doi:10.1136/jech.56.3.209pmid:http://www.ncbi.nlm.nih.gov/pubmed/11854343
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Moher D,
      2. Shamseer L,
      3. Clarke M, et al
      . Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015;4:1. doi:10.1186/2046-4053-4-1pmid:http://www.ncbi.nlm.nih.gov/pubmed/25554246
      OpenUrlCrossRefPubMed
    17. ↵
      1. Zeller TA
      . Respiratory conditions update: asthma. FP Essent 2016;448:11–19.pmid:http://www.ncbi.nlm.nih.gov/pubmed/27576231
      OpenUrlPubMed
    18. ↵
      1. Pinart M,
      2. Albang R,
      3. Maier D, et al
      . Systematic review on the definition of allergic diseases in children: the MeDALL study. Int Arch Allergy Immunol 2015;168:110–21.doi:10.1159/000442414pmid:http://www.ncbi.nlm.nih.gov/pubmed/26657241
      OpenUrlPubMed
    19. ↵
      1. Banerji A
      . Hereditary angioedema: classification, pathogenesis, and diagnosis. Allergy Asthma Proc 2011;32:403–7.doi:10.2500/aap.2011.32.3492pmid:http://www.ncbi.nlm.nih.gov/pubmed/22221432
      OpenUrlPubMed
    20. ↵
      1. Khan KS,
      2. Kunz R,
      3. Kleijnen J, et al
      . Five steps to conducting a systematic review. J R Soc Med 2003;96:118–21.doi:10.1177/014107680309600304pmid:http://www.ncbi.nlm.nih.gov/pubmed/12612111
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Haddaway NR,
      2. Collins AM,
      3. Coughlin D, et al
      . The role of Google Scholar in evidence reviews and its applicability to grey literature searching. PLoS One 2015;10:e0138237. doi:10.1371/journal.pone.0138237pmid:http://www.ncbi.nlm.nih.gov/pubmed/26379270
      OpenUrlCrossRefPubMed
    22. ↵
      1. Gusenbauer M,
      2. Haddaway NR
      . Which academic search systems are suitable for systematic reviews or meta-analyses? evaluating retrieval qualities of Google Scholar, PubMed, and 26 other resources. Res Synth Methods 2020;11:181–217.doi:10.1002/jrsm.1378pmid:http://www.ncbi.nlm.nih.gov/pubmed/31614060
      OpenUrlPubMed
    23. ↵
      1. Liberati A,
      2. Altman DG,
      3. Tetzlaff J, et al
      . The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009;339:b2700. doi:10.1136/bmj.b2700pmid:http://www.ncbi.nlm.nih.gov/pubmed/19622552
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Armijo-Olivo S,
      2. Stiles CR,
      3. Hagen NA, et al
      . Assessment of study quality for systematic reviews: a comparison of the Cochrane collaboration risk of bias tool and the effective public health practice project quality assessment tool: methodological research. J Eval Clin Pract 2012;18:12–18.doi:10.1111/j.1365-2753.2010.01516.xpmid:http://www.ncbi.nlm.nih.gov/pubmed/20698919
      OpenUrlCrossRefPubMed
    25. ↵
      1. Borenstein M,
      2. Hedges LV,
      3. Higgins JPT, et al
      . A basic introduction to fixed-effect and random-effects models for meta-analysis. Res Synth Methods 2010;1:97–111.doi:10.1002/jrsm.12pmid:http://www.ncbi.nlm.nih.gov/pubmed/26061376
      OpenUrlPubMed
    26. ↵
      1. Davies HT,
      2. Crombie IK,
      3. Tavakoli M
      . When can odds ratios mislead? BMJ 1998;316:989–91.doi:10.1136/bmj.316.7136.989pmid:http://www.ncbi.nlm.nih.gov/pubmed/9550961
      OpenUrlFREE Full Text
    27. ↵
      1. VanderWeele TJ,
      2. Ding P
      . Sensitivity analysis in observational research: introducing the E-Value. Ann Intern Med 2017;167:268–74.doi:10.7326/M16-2607pmid:http://www.ncbi.nlm.nih.gov/pubmed/28693043
      OpenUrlCrossRefPubMed
    28. ↵
      1. Higgins JPT,
      2. Thompson SG,
      3. Deeks JJ, et al
      . Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60.doi:10.1136/bmj.327.7414.557pmid:http://www.ncbi.nlm.nih.gov/pubmed/12958120
      OpenUrlFREE Full Text
    29. ↵
      1. von Hippel PT
      . The heterogeneity statistic I(2) can be biased in small meta-analyses. BMC Med Res Methodol 2015;15:35. doi:10.1186/s12874-015-0024-zpmid:http://www.ncbi.nlm.nih.gov/pubmed/25880989
      OpenUrlPubMed
    30. ↵
      1. World Bank
      . World bank country and lending groups. Washington (DC): World Bank, 2020. https://datahelpdesk.worldbank.org/knowledgebase/articles/906519
    31. ↵
      1. Fu R,
      2. Gartlehner G,
      3. Grant M, et al
      . AHRQ methods for effective health care conducting quantitative synthesis when comparing medical interventions: AHRQ and the effective health care program. methods guide for effectiveness and comparative effectiveness reviews. Rockville (MD: Agency for Healthcare Research and Quality (US), 2008.
    32. ↵
      1. Schmid CH,
      2. Stark PC,
      3. Berlin JA, et al
      . Meta-Regression detected associations between heterogeneous treatment effects and study-level, but not patient-level, factors. J Clin Epidemiol 2004;57:683–97.doi:10.1016/j.jclinepi.2003.12.001pmid:http://www.ncbi.nlm.nih.gov/pubmed/15358396
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Egger M,
      2. Davey Smith G,
      3. Schneider M, et al
      . Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34.doi:10.1136/bmj.315.7109.629pmid:http://www.ncbi.nlm.nih.gov/pubmed/9310563
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Lin L,
      2. Chu H
      . Quantifying publication bias in meta-analysis. Biometrics 2018;74:785–94.doi:10.1111/biom.12817pmid:http://www.ncbi.nlm.nih.gov/pubmed/29141096
      OpenUrlCrossRefPubMed
    35. ↵
      1. Shi L,
      2. Lin L
      . The trim-and-fill method for publication bias: practical guidelines and recommendations based on a large database of meta-analyses. Medicine 2019;98:e15987. doi:10.1097/MD.0000000000015987pmid:http://www.ncbi.nlm.nih.gov/pubmed/31169736
      OpenUrlCrossRefPubMed
    36. ↵
      1. Petrisor B,
      2. Bhandari M
      . The hierarchy of evidence: levels and grades of recommendation. Indian J Orthop 2007;41:11–15.doi:10.4103/0019-5413.30519pmid:http://www.ncbi.nlm.nih.gov/pubmed/21124676
      OpenUrlCrossRefPubMed
    37. ↵
      1. Wormald R,
      2. Evans J
      . What makes systematic reviews systematic and why are they the highest level of evidence? Ophthalmic Epidemiol 2018;25:27–30.doi:10.1080/09286586.2017.1337913pmid:http://www.ncbi.nlm.nih.gov/pubmed/28891724
      OpenUrlPubMed
    38. ↵
      1. Suri H
      . Ethical Considerations of Conducting Systematic Reviews in Educational Research. In: Zawacki-Richter O, Kerres M, Bedenlier S, et al., eds. Systematic reviews in educational research: methodology, perspectives and application. Wiesbaden: Springer Fachmedien Wiesbaden, 2020. : 41–54p..

    Supplementary materials

    Footnotes

    • Contributors BIN conceived the study idea. DL wrote the protocol, with review and suggestions for improvement from BIN, EG, GW, and SSÖE. All authors commented and approved the last version being submitted.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.