Elucidating the impact of the pneumococcal conjugate vaccine programme on pneumonia, sepsis and otitis media hospital admissions in England using a composite control

Abstract

BackgroundThe seven-valent pneumococcal conjugate vaccine (PCV) was introduced in England in September 2006, changing to the 13-valent vaccine in April 2010. PCV impact on invasive pneumococcal disease (IPD) has been extensively reported, but less described is its impact on the burden of pneumonia, sepsis and otitis media in the hospital.MethodsUsing details on all admissions to hospitals in England, we compared the incidence of pneumococcal-specific and syndromic disease endpoints in a 24-month pre-PCV period beginning April 2004 to the 24-month period ending March 2015 to derive incidence rate ratios (IRRs). To adjust for possible secular trends in admission practice, IRRs were compared to the IRRs for five control conditions over the same period and the relative change assessed using the geometric mean of the five control IRRs as a composite, and individually for each control condition to give the min-max range. Relative changes were also compared with IRRs for IPD from the national laboratory database. The effect of stratifying cases into those with and without clinical risk factors for pneumococcal infection was explored.ResultsRelative reductions in pneumococcal pneumonia were seen in all age groups and in those with and without risk factors; in children under 15 years old reductions were similar in magnitude to reductions in IPD. For pneumonia of unspecified cause, relative reductions were seen in those under 15 years old (maximum reduction in children under 2 years of 34%, min-max: 11–49%) with a relative increase in 65+ year olds most marked in those with underlying risk conditions (41%, min-max: 0–82%). Reductions in pneumococcal sepsis were seen in all age groups, with the largest reduction in children younger than 2 years (67%, min-max 56–75%). Reductions in empyema and lung abscess were also seen in under 15 year olds. Results for other disease endpoints were varied. For disease endpoints showing an increase in raw IRR, the increase was generally reduced when expressed as a relative change.ConclusionsUse of a composite control and stratification by risk group status can help elucidate the impact of PCV on non-IPD disease endpoints and in vulnerable population groups. We estimate a substantial reduction in the hospitalised burden of pneumococcal pneumonia in all age groups and pneumonia of unspecified cause, empyema and lung abscess in children under 15 years of age since PCV introduction. The increase in unspecified pneumonia in high-risk 65+ year olds may in part reflect their greater susceptibility to develop pneumonia from less pathogenic serotypes that are replacing vaccine types in the nasopharynx.Electronic supplementary materialThe online version of this article (doi:10.1186/s12916-018-1004-z) contains supplementary material, which is available to authorized users.

The changing epidemiology of community‐acquired pneumonia: nationwide register‐based study in Sweden

“…Reductions in AC‐CAP hospitalizations of 36%, 20% and 16% in children <2, 2–4 and 5–17 years of age, respectively, and even more pronounced decreases in PL‐CAP, are consistent with reports from most other countries . These findings are concordant with previous publications from Sweden that reported up to 37% reduction of inpatient all‐cause pneumonia in children < 2 years in 2012 .…”

“…However, in a recent study that adjusted for secular trends using the synthetic control method, there was no observed reduction of hospitalization for all‐cause pneumonia after PCV‐CIP introduction in any of five countries in the Americas (the United States, Mexico, Ecuador, Brazil and Chile) . Neither was there a reduction in pneumonia among the elderly in Australia, and studies from the UK and Denmark have observed similar increases of AC‐CAP in the elderly as in our study . Our results are in accordance with the moderate impact of the PCV‐CIP on the IPD incidence in the elderly observed both in Sweden, and in a multinational European study .…”

“…The point estimates suggested a reduction of PL‐CAP among adults without comorbidities for most age groups, and an increase among 40–84‐year olds with underlying diseases, although the results should be interpreted with caution due to the statistical uncertainty. Analogous findings of an attenuated reduction in AC‐CAP hospitalizations among adults with underlying diseases was recently observed from two healthcare claims repositories in the United States, and the prevalence of comorbidities has increased over time among patients with CAP in England, providing support for a differential impact of PCV‐CIP on CAP incidence in adults with and without underlying diseases . In addition, we now demonstrate that there has been an increase of the proportion of individuals with comorbidities in the whole population over time, which may be one reason for the unchanged or even increased incidence of AC‐CAP in adults.…”

“…Furthermore, we did not control for underlying trends in the analyses of patients with and without comorbidities since we did not have access to complete data on control conditions stratified according to underlying diseases. Indeed, it has been shown in a previous study that adjustment for admittance practices has a greater effect on estimates in patients with than without comorbidities . Second, we did not perform interrupted time‐series analyses to assess trends.…”

“…Yet, the conjugate vaccines only cover 7-13 pneumococcal serotypes and due to expansion of nonvaccine serotypes, the overall impact on IPD in adults has been small in many settings, especially in older persons [5,8]. The impact of PCV childhood immunization programmes (PCV-CIP) on CAP among children show a reduced burden of disease for both all-cause pneumonia and pneumonia with more pneumococcal-specific diagnoses, but there are discrepant results concerning the impact among the elderly [9][10][11][12][13][14][15][16][17].…”