Review
Strengthening the Reporting of Observational Studies in Epidemiology for Newborn Infection (STROBE-NI): an extension of the STROBE statement for neonatal infection research

https://doi.org/10.1016/S1473-3099(16)30082-2Get rights and content

Summary

Neonatal infections are estimated to account for a quarter of the 2·8 million annual neonatal deaths, as well as approximately 3% of all disability-adjusted life-years. Despite this burden, few data are available on incidence, aetiology, and outcomes, particularly regarding impairment. We aimed to develop guidelines for improved scientific reporting of observational neonatal infection studies, to increase comparability and to strengthen research in this area. This checklist, Strengthening the Reporting of Observational Studies in Epidemiology for Newborn Infection (STROBE- NI), is an extension of the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement. STROBE-NI was developed following systematic reviews of published literature (1996–2015), compilation of more than 130 potential reporting recommendations, and circulation of a survey to relevant professionals worldwide, eliciting responses from 147 professionals from 37 countries. An international consensus meeting of 18 participants (with expertise in infectious diseases, neonatology, microbiology, epidemiology, and statistics) identified priority recommendations for reporting, additional to the STROBE statement. Implementation of these STROBE-NI recommendations, and linked checklist, aims to improve scientific reporting of neonatal infection studies, increasing data utility and allowing meta-analyses and pathogen-specific burden estimates to inform global policy and new interventions, including maternal vaccines.

Introduction

Progress in improving child survival has been one of the greatest successes in international development.1 However, there is an unfinished agenda,2 since the mortality reduction has been slowest for neonates. Almost half (44%) of all child deaths now occur in the neonatal period (0–27 days),3 with a substantial burden of mortality in the first few days after birth.4 The Every Newborn Action Plan sets out a United Nations-led platform, endorsed by all countries, to end preventable neonatal deaths, but requires data to implement and inform innovation.2, 5

Estimates by WHO for 195 countries suggest that infection accounts for around 680 000 deaths—a quarter of all neonatal deaths yearly;6 and half of all neonatal deaths in settings with high neonatal mortality.2 The closely linked 2·6 million annual stillbirths have an as yet poorly quantified infection burden.7 Significant neurodevelopmental impairment affects approximately a quarter of neonates following meningitis, but few data exist regarding impairment worldwide, particularly for common infection syndromes such as sepsis and pneumonia.8, 9

An estimated 6·9 million neonates have possible serious bacterial infection annually in sub-Saharan Africa, south Asia, and Latin America.8 Approximately 84% of neonatal deaths attributed to infections could be averted by increasing coverage of prevention and access to treatment, yet currently the gap is high, especially in the poorest countries.10 Recent large clinical trials have assessed the safety and efficacy of improving access to treatment through outpatient care, in cases for which referral is not possible.11, 12, 13

Aetiology-specific data for neonatal infections are scarce, and challenging to combine. Hospital-based studies suggest that Staphylococcus aureus, Escherichia coli, Klebsiella spp, and group B streptococci might be the most common pathogens globally.14 As yet, there are no community-based aetiological studies from Africa, and few from south Asia. There is an urgent need to improve data on aetiology (bacterial, viral, and fungal), incidence (especially in the first days following birth), antimicrobial sensitivity, and outcomes. These data are essential to understand the burden and risk factors, refine treatment algorithms, support potential interventions (eg, maternal vaccines for respiratory syncytial virus and group B streptococcus),15, 16, 17 and mitigate antimicrobial resistance, which threatens current treatment strategies.18, 19, 20

Recording, reporting, and interpreting neonatal infection data poses specific challenges. More than 95% of neonatal deaths occur in countries without adequate birth and death certification to capture cause-specific mortality,2, 6 let alone pathogen-specific surveillance. Systematic clinical assessment, with investigations providing microbiological data, is also uncommon.8 Most available neonatal infection data are from tertiary referral hospitals, with recruitment bias, by missing those not accessing higher levels of care, or any care.21 In population-based studies, which are extremely few in high-burden settings,22, 23, 24 even if women are recruited in pregnancy, the challenge remains that many neonates die within hours of birth before being assessed, meaning counting, investigations, and treatment are missed.25 In a population-based Bangladeshi cohort, 62% of neonates who died were never clinically assessed, with 59% of deaths occurring within 48 h of birth.22 Even when cases are captured in the numerator and denominator, case definitions are often inconsistent. Diagnosis is usually based on clinical expertise, or in settings with fewer health workers, on simplified clinical algorithms designed to be highly sensitive. For example, the most commonly used WHO algorithm to classify young infants with possible serious bacterial infection is very sensitive (85%) and fairly specific (75%).26, 27, 28 Additionally, unlike childhood infections, gestational age has a major effect on incidence, aetiology, and outcomes of neonatal infections. Neonates of 25 and 35 weeks' gestation are both preterm, yet differentiation between the two is often missing in reported data, which is crucial for interpretation.

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)29 and Consolidated Standards of Reporting Trials (CONSORT)30 statements were developed to improve scientific reporting. Several extensions of these statements have been published with additional recommendations for specialised fields of research—for example, the Strengthening the Reporting of Molecular Epidemiology for Infectious Diseases (STROME-ID)31 and the Outbreak Reports and Intervention Studies of Nosocomial Infection (ORION)32 statement. These extensions build on the principles of STROBE and CONSORT but explicitly address additional, problematic methods or settings. There are reporting guidelines under development that are specific to child health trials (SPIRIT-C; CONSORT-C),33 and for systematic reviews and meta-analyses (PRISMA-C; PRISMA-PC).34 We aim to address the specific challenges in reporting neonatal infections, using the STROBE29 model. If these recommendations are applied by upcoming epidemiological and interventional studies on neonatal infections, the value of new data will increase, avoiding research waste.35

Section snippets

Aims of STROBE-NI

The purpose of these guidelines is to promote transparency, clarity, and comparability of scientific reporting, specifically for neonatal infection research. We focus on observational studies (although many elements will be true for other study designs), and include detailed consideration of aetiological (bacterial, viral, and fungal) data. Through improved reporting, we aim to facilitate reliable comparison of emerging newborn infection data across settings worldwide, and the synthesis of

Development of the STROBE-NI checklist

The STROBE-NI checklist was developed following recommended methods.36 The participants, processes, and outputs are shown in figure 1. We searched the scientific literature to identify highly cited publications on neonatal infection from different regions worldwide (1996–2015), and more recent (2011–15) articles from high impact journals (see appendix for literature search criteria). Additional searches were done for reporting guidelines relevant to neonatal infections.

Through these reviews we

STROBE-NI standards

The final STROBE-NI checklist is an extension of the 22-item STROBE checklist, with 28 additional elements relating to neonatal infection. The STROBE-NI checklist includes a suggested flow diagram for both the recruitment and follow-up of mothers and newborn babies, for which a template is provided in figure 2. Here, we describe the additional recommendations for STROBE-NI that are not already outlined in detail in STROBE or other extensions.

Implications of STROBE-NI

The STROBE-NI checklist provides a tool for researchers, funders, reviewers, and publishers to improve neonatal infection data, which have specific, previously unaddressed, requirements for scientific reporting. Building on the STROBE29 statement and its related extensions, the checklist mainly targets observational studies.29 However, STROBE-NI checklist items should also be considered for randomised controlled trials, alongside other guideline extensions.33, 34 To our knowledge, there are no

Search strategy and selection criteria

We searched Scopus to identify the three leading research articles published (most citations per year) on neonatal infection, from seven regions worldwide (1996–2015; last search Feb 27, 2015), and PubMed for more recent research articles (2011–15; last search March 15, 2015) in paediatric and/or infectious disease journals with the highest impact factors (Thomson Reuters) in 2015. We used the search terms “neonat*” OR “newborn*” OR “newborn infant*” OR “young infant*” AND “infect*” OR “sepsis”

References (59)

  • K Modjarrad et al.

    WHO consultation on respiratory syncytial virus vaccine development report from a World Health Organization meeting held on 23–24 March 2015

    Vaccine

    (2016)
  • SJ Schrag

    Group B streptococcal vaccine for resource-poor countries

    Lancet

    (2011)
  • CJ Murray et al.

    Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010

    Lancet

    (2012)
  • N Field et al.

    Strengthening the Reporting of Molecular Epidemiology for Infectious Diseases (STROME-ID): an extension of the STROBE statement

    Lancet Infect Dis

    (2014)
  • SP Stone et al.

    The ORION statement: guidelines for transparent reporting of outbreak reports and intervention studies of nosocomial infection

    Lancet Infect Dis

    (2007)
  • A Saint-Raymond et al.

    CONSORT 2010

    Lancet

    (2010)
  • I Chalmers et al.

    Avoidable waste in the production and reporting of evidence

    Lancet

    (2009)
  • R Emmadi et al.

    Molecular methods and platforms for infectious diseases testing: a review of FDA-approved and cleared assays

    J Mol Diagn

    (2011)
  • AC Seale et al.

    Maternal and early onset neonatal bacterial sepsis: burden and strategies for prevention in sub-Saharan Africa

    Lancet Infect Dis

    (2009)
  • PT Heath et al.

    Group B streptococcal disease in UK and Irish infants younger than 90 days

    Lancet

    (2004)
  • H Blencowe et al.

    National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications

    Lancet

    (2012)
  • D You et al.

    Levels and trends in child mortality. Report 2014. Estimates developed by the UN Inter-agency Group for Child Mortality Estimation

  • Every newborn—an action plan to end preventable deaths

  • AC Seale et al.

    Neonatal severe bacterial infection impairment estimates in south Asia, sub-Saharan Africa, and Latin America for 2010

    Pediatr Res

    (2013)
  • GL Darmstadt et al.

    Neonatal infections: a global perspective

  • AS Meulen et al.

    Path to impact: a report from the Bill and Melinda Gates Foundation convening on maternal immunization in resource-limited settings; Berlin—January 29–30, 2015

    Vaccine

    (2015)
  • AC Seale et al.

    Rational development of guidelines for management of neonatal sepsis in developing countries

    Curr Opin Infect Dis

    (2014)
  • Antimicrobial resistance: global report on surveillance

  • HK Herbert et al.

    Care seeking for neonatal illness in low- and middle-income countries: a systematic review

    PLoS Med

    (2012)
  • Cited by (115)

    View all citing articles on Scopus

    Members listed at end of paper

    View full text