Pneumococcal conjugate vaccines for preventing invasive pneumococcal disease and pneumonia in children aged up to five years

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the effects and safety of pneumococcal conjugate vaccines for preventing invasive pneumococcal disease and pneumonia in children aged up to five years.

Background

Description of the condition

The bacterium Streptococcus pneumoniae is a gram‐positive bacterium presenting in pairs (diplococcus) and is a leading cause of serious illness in children (Thorrington 2018). It causes a range of infectious conditions, some of which are serious, such as bacteraemia, sepsis, meningitis, and pneumonia, and some that are less serious but more common, such as otitis media and sinusitis. Pneumococcus has a capsule that is an essential factor for its ability in causing disease and defines the differentiation of known serotypes. There are currently more than 90 distinct identified serotypes of S pneumoniae, and their distribution and prevalence varies over time and by age, disease severity, disease syndrome, geographical area, and the presence of antimicrobial resistant genes (WHO 2019; Yildirim 2015). Transmission occurs through respiratory droplets. The presence of bacteria on nasopharyngeal mucosa usually occurs during childhood, but rarely results in illness. However, some serotypes may enter the bloodstream and cause an infection of the blood (bacteraemia) with subsequent infections such as meningitis, or invade close tissues to cause otitis media and sinusitis.

Pneumonia is often caused by aspiration of S pneumoniae from the nasopharynx into the lungs. Invasive pneumococcal disease is commonly defined as a systemic morbidity associated with the isolation of pneumococci from a normally sterile site, such as the bloodstream, or conditions spread by the infected blood, such as meningitis or septic arthritis (WHO 2012). Meningitis, bacteraemia, arthritis, peritonitis, and pneumonia associated with bacteraemia are considered to be invasive pneumococcal diseases (WHO 2019). About 75% of invasive pneumococcal diseases and 83% of pneumococcal meningitis occur in children aged under two years (Russell 2011). Studies show that the rate of prevalence for S pneumoniae nasopharyngeal colonisation in children aged up to five years ranges from 27% in high‐income countries to 85% in low‐income countries (WHO 2012). This varies according to age, environment, presence of associated upper airway disease, and populations studied. For example, the rate of colonisation is known to increase in infants aged under two years, day‐care attendants, presence of exposure to indoor smoking, absence of breastfeeding, and during the winter months (Yildirim 2015).

Prevalence and mortality are higher in low‐income countries, and amongst people with chronic conditions and HIV infection (O'Brien 2009).

According to the National Center for Immunization and Respiratory Diseases from the Centers for Disease Control and Prevention (CDC), overall invasive pneumococcal disease incidence in children in the USA before 2000 was 95 cases per 100,000. Invasive pneumococcal disease incidence caused by pneumococcal serotypes 1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, and 23F; the incidence was 88 cases per 100,000 (CDC 2018).

Following the introduction of the pneumococcal conjugate vaccines (PCV) (i.e. PCV7 in 2000 and PCV13 in 2010), the CDC reported a dramatic decline in the overall invasive pneumococcal disease incidence to 9 cases per 100,000 in 2015, and invasive pneumococcal disease caused by PCV13 serotypes declined to 2 cases per 100,000 in 2015 (CDC 2018).

In the UK, the national childhood immunisation programme implemented the PCV7 vaccine for children aged up to two years in 2006. Invasive pneumococcal disease rates dropped by 98% in children aged up to two years, and more than 75% in older age groups, with an overall decline in incidence of 34%. In 2010, PCV13 replaced PCV7, and four years later, invasive pneumococcal disease rates fell by a further 32% compared with the pre‐PCV13 baseline, and by 56% compared with the pre‐PCV7 baseline. Amongst children aged under five years, invasive pneumococcal disease due to extra PCV13 serotypes dropped by 69% in 2013 and 2014 (Makwana 2018).

In Europe in 2014 invasive pneumococcal disease was predominantly found in infants and the elderly, with 11.3 confirmed cases per 100,000 people in children aged up to 12 months (ECDPC 2016).

Data from the year 2000 indicated that each year S pneumoniae caused 1.2 million deaths worldwide due to pneumonia, of which about 40% occurred in children under five years of age (Farhat 2002). In 2008, the World Health Organization (WHO) estimated that from 8.8 million overall deaths occurring in children aged up to five years, 476,000 were caused by pneumococcal infection (O'Brien 2009). In 2015, an analysis by Liu and colleagues showed that from 2000 to 2015, amongst the 5.9 million deaths in children aged under five years, the leading causes were preterm birth complications (1.055 million), pneumonia (0.921 million), and intrapartum‐related events (0.691 million). In the two regions with the most deaths, the leading cause was pneumonia in sub‐Saharan Africa and preterm birth complications in southern Asia. Stratified by mortality rates under five years of age, pneumonia was the leading cause in countries with very high mortality rates. Preterm birth complications and pneumonia were both important in high‐, medium‐high‐, and medium‐child mortality countries (Liu 2016).

The cost burden of diseases caused by pneumococcus is significant. One study showed that the direct cost of pneumococcal disease amongst people of all ages in the USA was USD 3.500 million in 2004 (Huang 2011), of which USD 370 million was the cost in children aged under five years. A study performed in Taiwan including almost 30,000 children demonstrated that the total direct cost in 2013 was more than USD 17 million (Ho 2015).

Vaccination has been demonstrated to be cost‐effective. A recent systematic review of the cost‐effectiveness of pneumococcal vaccination in children in low‐ and middle‐income countries, which included evidence from 22 studies, showed that both 10‐valent and 13‐valent PCVs are probably cost‐effective compared with the 7‐valent PCV or no vaccination (Saokaew 2016). Chen and colleagues published a global modelling analysis on effects and cost‐effectiveness of vaccination. The authors estimated that global PCV13 use could prevent 0.399 million child deaths and 54.6 million disease episodes annually. Global vaccine costs (in 2015 international dollars) of ID 15.500 million could be partially offset by healthcare savings of ID 3.190 million and societal cost savings of ID 2.640 million. The expected cost of PCV vaccination globally is around ID 16,000 million per year (Chen 2019).

Description of the intervention

Treatment of invasive pneumococcal disease involves general supportive measures and antibiotics. Nevertheless, antibiotic resistance is a serious problem in many countries (WHO 2019). Pneumococcal vaccines have been developed and added to routine childhood immunisation schedules over the last 15 years (WHO 2012). Currently, there are three types of conjugated vaccines: 10‐valent (PCV10); 13‐valent (PCV13); and 15‐valent (PCV15).

PCV10 includes capsular polysaccharides purified from serotypes 1, 4, 5, 6B, 7F, 9V, 14, 18C, 19F, and 23F. It is conjugated with D protein from Haemophilus strains, and contains aluminium phosphate as an adjuvant. It is used in Europe and other regions around the world.

PCV13 replaced PCV7 in the USA in 2010, and is approved for use in children aged over six weeks. PCV13 has six additional serotypes (compared with PCV7) that are responsible for 63% of invasive pneumococcal disease in children aged up to five years old in the USA. This vaccine includes polysaccharides antigens of pneumococcal capsular serotypes 1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, and 23F. It is conjugated with non‐toxic toxoid diphtheric CRM 197 and contains alluminum phosphate as an adjuvant.

PCV15 is the newest vaccine type and has been tested in children and adults (McFetridge 2015; Sobanjo‐ter Meulen 2015). It added two serotypes to the PCV13 list: 22F and 33F.

PCV10 and PCV13 are licenced for use in children aged six weeks to five years. They are administered by injection into the side of the thigh in infants, and into the upper arm in children from two years of age (WHO 2019).

There are several immunisation schedules, but manufacturers recommend three primary doses with an interval of at least four weeks between doses, plus a booster (3p + 1). The first should be given as early as six weeks of age. The most common schedule recommends three doses in the first year of life, at two, four, and six months, with a booster 12 months later to enhance protection. If the first dose is given after the infant is six months of age, two doses with an interval of eight weeks are recommended in the first year of life, with a booster 12 months later. Finally, if the first dose is given to children aged over two years, a single dose of PCV13 is recommended, with the exception of children who are immunocompromised. For non‐immunocompromised children, two doses are recommended with an interval of eight weeks (WHO 2012). In a recent report, the WHO recommended a three‐dose schedule, administered either as (2p + 1) or (3p + 0), leaving the decision about which schedule to follow up to each country (WHO 2019).

The licencing studies of VCP10 and VCP13 conjugate vaccines revealed an excellent safety profile, with mild and transient adverse events. Most common local reactions are pain, swelling, and redness. Systemic symptoms such as irritability, crying, and fever may occur, but they are usually short term.

How the intervention might work

Although there are more than 90 S pneumoniae serotypes, only a few lead to invasive pneumococcal disease (Yildirim 2015). There are two types of vaccines available for disease prevention: conjugated and polysaccharide (Johnson 2010).

Polysaccharide vaccine is associated with a poor immunogenic response in children under two years of age due to its poor production of specific antibodies, absence of memory response, and rapid decrease of serum concentrations. Unfortunately, this age group has the highest incidence of invasive disease (Bocchini 2010).

The development of conjugated vaccines, through a conjugation between an immunogenic non‐pneumococcal protein to individual pneumococcal components, induces a memory response and antibody production, thereby improving the effectiveness of vaccines in this group of people.

To assess vaccine efficacy, the concentration of antibodies related to protection against pneumococcal disease is proposed to be equal to or more than 0.35 mcg/mL one month after the first immunisation, as recommended by the WHO (WHO 2019).
 
 Following the licencing and use of the 7‐valent pneumococcal conjugate vaccine, a significant reduction in disease burden was demonstrated not only in the vaccinated population but also in different age groups not covered by the vaccine, revealing the great potential of the vaccine to induce collective immunity through the reduction of the nasopharyngeal carrier status of the S pneumoniae serotypes included in the vaccine (Makwana 2018).
 
 Pneumococcal immunity is specific for each serotype, therefore the inclusion of different serotypes in polyvalent vaccines has the objective of amplifying their coverage in different regions. The most prevalent serotypes may have different antimicrobial resistance patterns. The inclusion of new serotypes aims to reduce morbidity and mortality by pneumococcal disease.
 
 The risk of developing invasive pneumococcal disease is higher in children and adults with an impaired immune system, whether due to congenital or acquired immunodeficiency, immunosuppressive, functional (e.g. asplenia of sickle cell anaemia), chronic renal failure, or nephrotic syndrome (Madhi 2005).
 
 With respect to children with HIV in particular, it is known that the immune response to PCVs should be evaluated not only by the blood antibody concentrations achieved, but also by the memory response provoked by the antibodies (Madhi 2005; Nunes 2012).

Why it is important to do this review

PCV13 and PCV10 are currently the most used vaccines worldwide (WHO 2019). At the time of their licencing, these vaccines were compared with PCV7 in observational and non‐inferiority trials (WHO 2012). Most systematic reviews that assessed the effectiveness of these vaccines were based on observational studies (Conklin 2014; Fleming‐Dutra 2014; Loo 2014; Scott 2011).

A previous Cochrane Review evaluated randomised trials that tested PCV7, PCV9, and PCV11 with placebo, which were the available vaccines at the time (Lucero 2009). Lucero 2009 included six randomised controlled trials in which the pooled vaccine efficacy (VE) for vaccine‐serotype invasive pneumococcal disease was 80% (95% confidence interval (CI) 58% to 90%, P < 0.001); all‐serotypes invasive pneumococcal disease, 58% (95% CI 29% to 75%, P = 0.001); WHO x‐ray‐defined pneumonia, 27% (95% CI 15% to 36%, P < 0.001); clinical pneumonia, 6% (95% CI 2% to 9%, P = 0.001); and all‐cause mortality, 11% (95% CI −1% to 21%, P = 0.08). Analysis involving HIV‐1 positive children had similar findings. However, our review will consider some different clinical and methodological aspects compared with Lucero 2009. First, we will include children aged up to five years, instead of limiting the population to children aged up to two years. We will include different vaccine valences from those included in Lucero 2009, covering additional serotypes, which could affect the outcomes of our review.

Finally, there is discussion concerning the optimal schedule policy with the new vaccines (WHO 2019).

In the United Nations' Sustainable Development Goals (SDG) campaign, countries are advised to prioritise child survival policy and programmes based on their child cause‐of‐death composition (Liu 2016). Efforts to scale up proven lifesaving interventions are needed to achieve the SDG child survival target. Vaccination is one intervention that could improve survival rates, hence a systematic review assessing the current evidence on the efficacy and safety of pneumococcal vaccines should have an impact on immunisation policies around the world.

Objectives

To assess the effects and safety of pneumococcal conjugate vaccines for preventing invasive pneumococcal disease and pneumonia in children aged up to five years.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised controlled trials (RCTs) and cluster‐RCTs. We will include studies reported as full text, those published as abstract only, and unpublished data.

Types of participants

We will include children aged up to five years participating in trials comparing pneumococcal conjugated vaccines (PCV) to placebo, no intervention, or different valences of vaccines. If we identify studies including other age ranges, we will use the data for children up to five years of age, provided this age group is more likely to develop the disease of interest. If these data are not available, we will contact the trial authors for this information.

Types of interventions

We will include trials of PCV, irrespective of valency (PCV10, PCV13, and PCV15). We will include trials in which participants have received:

1. two primary doses and one booster; or

2. three primary doses with or without a booster dose.

We will include trials that compared:

1. vaccine versus placebo;

2. vaccine versus nothing; or

3. different valences of PCV vaccines (e.g. PCV10 versus PCV13, etc.).

All vaccines are delivered intramuscularly, at regular doses, as recommended by the WHO (WHO 2019).

Types of outcome measures

We will analyse the effect of the intervention on diseases caused by:

1. vaccine serotype (VS), defined as serotypes contained in the study vaccine;

2. vaccine‐related serotypes (VRS), defined as isolates of the same serogroup but not of the same serotype as those in the study vaccine; or

3. non‐vaccine serotypes (NVS), defined as all other serotypes not contained in the study vaccine.

Primary outcomes

1. Incidence of VS invasive pneumococcal disease, in intervention and control groups.

2. Incidence of x‐ray defined pneumonia according to WHO standardised guidelines (WHO x‐ray pneumonia) and clinical pneumonia (non‐x‐ray‐diagnosed pneumonia) amongst intervention and control groups.

3. Mortality related to invasive pneumococcal disease or pneumonia, in intervention and control groups.

4. Serious adverse events of the intervention (according to US Food and Drug Administration definitions, i.e. any untoward medical occurrence that at any dose results in death, is life‐threatening and requires inpatient hospitalisation or causes prolongation of existing hospitalisation, results in persistent or significant disability/incapacity, may have caused a congenital anomaly/birth defect, or requires intervention to prevent permanent impairment or damage).

Secondary outcomes

1. Incidence of VRS and NVS invasive pneumococcal disease, in intervention and control groups.

2. All‐cause mortality.

3. Minor adverse events.

Reporting one or more of the outcomes listed here is not an inclusion criterion for the review. We plan to contact study authors if a study does not report these outcomes.

Search methods for identification of studies

Electronic searches

We will search the following databases from inception to present:

1. the Cochrane Central Register of Controlled Trials (CENTRAL), which contains the Cochrane Acute Respiratory Infections Group Specialised Register;

2. MEDLINE (PubMed);

3. Embase;

4. CINAHL (Cumulative Index to Nursing and Allied Health Literature);

5. LILACS (Latin American and Caribbean Health Science Information database).

We will also search Web of Science if relevant.

We will use the search strategy described in Appendix 1 to search MEDLINE. We will combine the MEDLINE search with the Cochrane Highly Sensitive Search Strategy for randomised trials: sensitivity and precision‐maximising version (2008 revision) (Lefebvre 2011). We will impose no language or publication date restrictions.

We will also search grey literature databases and clinical trials registers.

Grey literature databases

1. Health Management Information Consortium database (www.ovid.com/site/catalog/DataBase/99.jsp)

2. National Technical Information Service database (ntis.gov/products/ntisdb.aspx)

3. OpenGrey (opengrey.eu)

Clinical trials registers

1. US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov)

2. World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (apps.who.int/trialsearch/)

Searching other resources

We will check reference lists of all primary studies and review articles for additional references. We will contact authors of identified studies and ask for information about other published and unpublished studies. We will also contact manufacturers and experts in the field.

Data collection and analysis

Selection of studies

Two review authors (CCM, FM) will independently screen the titles and abstracts of articles identified by the search to determine those that are potentially relevant. We will use the Covidence platform to code records as 'include', 'maybe', or 'exclude' (Covidence). We will retrieve the full‐text study reports or publications to code records as 'include', 'maybe', or 'exclude' , and two review authors (CCM, FM) will independently screen the results. We will identify studies for inclusion, and record reasons for exclusion of the ineligible studies. Any disagreements will be resolved through discussion or by consulting a third review author (RR) if required. We will identify and exclude duplicates and collate multiple reports of the same study so that each study, rather than each report, is the unit of interest in the review. We will record the selection process in sufficient detail to complete a PRISMA flow diagram and 'Characteristics of excluded studies' table (Moher 2009).

Data extraction and management

We will use an electronic Cochrane standard data collection form through Covidence for study characteristics and outcome data, which we will pilot on at least one study in the review. Two review authors (CCM, LESF) will identify and independently extract the following study characteristics from the included studies.

1. General information of the study: report title, year of publication, author contacts, and publication type (abstract or full report).

2. Methods: aim of study, study design, unit of allocation, start date, end date, duration of participation, and ethical approval.

3. Participants: population description, setting, inclusion criteria, exclusion criteria, age, method of recruitment, informed consent obtained, total number randomised, baseline imbalances, withdrawals and exclusions, gender, race/ethnicity, comorbidities, subgroups measured, subgroups reported, and other relevant sociodemographics.

4. Interventions: number randomised in each group, type of conjugate vaccine (valency, serotypes contained, carrier protein), dose, duration of treatment period, timing, delivery, providers, co‐interventions, economic information, resource requirements, integrity of delivery, and compliance.

5. Outcomes: primary and secondary outcomes specified and collected, time points measured and reported, outcome definition, person measuring/reporting, unit of measurement, scales, imputation of missing data, assumed risk estimates, and power.

6. Notes: funding for study and notable conflicts of interest of study authors.

Any disagreements will be resolved by consensus or by involving a third review author (RR). One review author (CCM) will enter data into the Review Manager 5 file (Review Manager 2014). We plan to double‐check that the data have been entered correctly by comparing the study reports with how the data will be presented in the review.

Assessment of risk of bias in included studies

Two review authors (LESF, CCM) will independently assess the risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019). Any disagreements will be resolved by discussion or by involving a third review author (RR). We will assess risk of bias according to the following domains.

1. Random sequence generation.

2. Allocation concealment.

3. Blinding of participants and personnel.

4. Blinding of outcome assessment.

5. Incomplete outcome data.

6. Selective outcome reporting.

7. Other bias.

We will grade each potential source of bias as high, low, or unclear and provide a quote from the study report together with a justification for our judgement in the 'Risk of bias' table. We will summarise the 'Risk of bias' judgements across different studies for each of the domains listed. We will consider blinding and incomplete outcome data separately for different key outcomes where necessary. Where information on risk of bias relates to unpublished data or correspondence with a trialist, this will be noted in the 'Risk of bias' table.

When considering treatment effects, we will take into account the risk of bias for the studies that contribute to that outcome.

Assessment of bias in conducting the systematic review

We will conduct the review according to this protocol and report any deviations from it in the 'Differences between protocol and review' section of the systematic review.

Measures of treatment effect

We will analyse dichotomous data as risk ratios (RR) providing 95% confidence intervals (CI) for the results.

We will undertake meta‐analysis only where this is meaningful, that is if the treatments, participants, and the underlying clinical question are similar enough for pooling to make sense.

Where multiple study arms are reported in a single study, we will include only the relevant arms. If two comparisons (e.g. vaccine A versus placebo and vaccine B versus placebo) must be entered into the same meta‐analysis, we will halve the control group to avoid double‐counting.

If we find no invasive pneumococcal disease events in any included trial, we will conduct analysis using a continuity correction constant. In constant continuity correction, a factor that is often equal to 0.5 will be added to both the treatment and control group counts.

For studies that employed a cluster‐randomised design and did not make an allowance for the design effect, we will calculate the design effect based on a large assumed intracluster correlation coefficient (ICC) of 0.10. We will follow the methods described in Section 16.3 of the Cochrane Handbook for Systematic Reviews of Interventions for the calculations (Higgins 2019).

Unit of analysis issues

The unit of analysis will be the individual, with a single measurement of each outcome for each participant being collected and analysed. For studies that employed cluster‐randomised design and did not make an allowance for the design effect, we intend to calculate the design effect based on a large assumed ICC of 0.10. We will follow the methods described in Section 16.3 of the Cochrane Handbook for Systematic Reviews of Interventions for the calculations.

Dealing with missing data

We will contact the trial authors to verify key study characteristics and to obtain missing numerical outcome data where possible (e.g. when a study is identified in abstract form only). If we are unable to obtain the numerical outcome data, we will make a judgement on the sort of missing data, and classifying it as 'missing at random' or 'not missing at random'. For data classified as 'missing at random', we will only analyse the available data (i.e. ignoring the missing data). Conversely, for data classified as 'not missing at random', we will input the missing data with replacement values, and treat these as if they were observed (e.g. last observation carried forward, imputing an assumed outcome such as assuming all were poor outcomes). Finally, we plan to perform sensitivity analyses to assess how sensitive results are to changes in the assumptions made, and address the potential impact of missing data on the findings of the review in the Discussion section.

If numerical outcome data are missing, such as standard deviations or correlation coefficients, and they cannot be obtained from the trial authors, we will calculate them from other available statistics such as P values according to the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019).

Assessment of heterogeneity

We plan to assess clinical, methodological, and statistical heterogeneity. We will use the Chi² statistic to detect and the I² statistic to measure heterogeneity amongst the studies in each analysis. If we identify substantial heterogeneity, we will explore it by prespecified subgroup analysis. We will investigate statistical diversity by estimates of treatment effect through forest plots using Review Manager 5. We will consider an I² value greater than 50% as substantial heterogeneity, although there is uncertainty in the reliability of the I² statistic when there are few studies in a meta‐analysis. Given that we expect the presence of clinical and/or methodological heterogeneity, we will use a random‐effects model by default, rather than a fixed‐effect model.

Assessment of reporting biases

We intend to contact study authors and ask them to provide missing outcome data. Where it is not possible to obtain these data, and the missing data are thought to introduce serious bias, we will explore the impact of including such studies in the overall assessment of results by a sensitivity analysis.

If we are able to pool more than 10 trials, we will create and examine a funnel plot to explore possible small‐study and publication biases.

Data synthesis

We will combine the results across studies, undertaking a random‐effects model meta‐analysis for dichotomous outcomes if participants, interventions, comparisons, and outcomes are sufficiently similar to make clinical sense. The explanation for the choice of a random‐effects model is provided in the Assessment of heterogeneity section.

GRADE and 'Summary of findings' table

We will create a 'Summary of findings' table comparing PCV vaccine versus placebo, PCV versus no treatment, and different valences of PCV. We will assess all predefined outcomes. We will use the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the quality of a body of evidence as it relates to the studies that contribute data to the meta‐analysis for the prespecified outcomes (Atkins 2004). We will use the methods and recommendations described in Section 8.5 and Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019), employing GRADEpro GDT software (GRADEpro GDT 2015). We intend to justify all decisions to downgrade or upgrade the quality of studies using footnotes, and make comments to aid readers' understanding of the review where necessary.

Subgroup analysis and investigation of heterogeneity

We plan to conduct a subgroup analysis of the primary outcomes for the following:

1. HIV status (positive/negative), as we expect different immune response in HIV‐positive people.

We will use the Chi² statistic to test for subgroup interactions in Review Manager 5 (Review Manager 2014).

Sensitivity analysis

We will perform the following sensitivity analyses to assess the robustness of our conclusions:

1. excluding studies with a high risk of bias (those classified as high risk in at least one of the following criteria: randomisation, allocation concealment, and blinding);

2. excluding studies with missing data considered 'not missing at random' to assess how sensitive results are to changes in the assumptions that are made; and

3. excluding studies that used cluster‐randomisation, given that different types of trials can lead to important differences in the effects being evaluated (Higgins 2019).

Appendices

Appendix 1. MEDLINE search strategy

1 Child/
 2 Infant/
 3 (children OR infant* OR pediatric OR paediatric OR child OR baby OR babies OR newborn* OR pediatrics OR paediatrics).tw
 4 OR/1‐3
 5 Streptococcus pneumoniae/
 6 streptococcus pneumoniae.tw.
 7 "s. pneumoniae".tw.
 8 exp Pneumococcal Infections/
 9 (pneumococcal adj2 (infection* OR disease*)).tw.
 10 (pneumococc* adj5 (pneumon* OR sepsis OR sinusit* OR meningit* OR otitis media)).tw.
 11 bacteraemic pneumon*.tw.
 12 (invasive pneumococcal disease OR ipd).tw.
 13 OR/5‐12
 14 exp Vaccines/
 15 exp Vaccination/
 16 Immunization/
 17 immunoprophylaxis.tw.
 18 (immuni* OR inocul* OR vaccin*).tw.
 19 Pneumococcal Vaccines/
 20 pneumococcal polysaccharide vaccin*.tw,nm.
 21 ppv*.tw,nm.
 22 pneumovax*.tw.nm

23 prevnar*.tw.nm

24 OR/14‐23

25 ((randomized controlled trial OR controlled clinical trial).pt. OR randomized.ab. OR randomised.ab. OR placebo.ab. OR drug therapy.fs. OR randomly.ab. OR trial.ab. OR groups.ab.) not
 (exp animals/ not humans.sh.)
 26 4 AND 13 AND 24 AND 25

Contributions of authors

CCM: conceiving, designing, writing, and co‐ordinating the protocol.
 LESF: designing search strategy, methodological advice on the protocol.
 CM: statistical advice on the statistical issues.
 FM: technical advice on the Background section.
 RR: general and methodological advice on the protocol.

Sources of support

Internal sources

• Christieny Chaipp Mochdece, Brazil.

  CCM is a professor at Faculdade de Medicina de Petrópolis and receives a salary

External sources

• No sources of support supplied

Declarations of interest

CCM: None known.
 LESF: None known.
 CM: None known.
 FM: None known.
 RR: None known.

New