Search

Background: Several aggregate data meta-analyses have provided estimates of the effectiveness of influenza vaccination in community-dwelling elderly people. However, these studies ignored the effects of patient-level confounders such as sex, age, and chronic diseases that could bias effectiveness estimates. We aimed to assess the confounder-adjusted effectiveness of influenza vaccines on laboratory-confirmed influenza among elderly people by conducting a global individual participant data meta-analysis., Methods: In this individual participant data meta-analysis, we considered studies included in a previously conducted aggregate data meta-analysis that included test-negative design case-control studies published up to July 13, 2014. We contacted all authors of the included studies on Dec 1, 2014, to request individual participant data. Patients were excluded if their unique identifier was missing, their vaccination status was unknown, their outcome status was unknown, or they had had suspected influenza infection more than once in the same influenza season. Cases were patients with influenza-like illness symptoms who tested positive for at least one of A H1N1, A H1N1 pdm09, A H3N2, or B viruses; controls were patients with influenza-like illness symptoms who tested negative for these virus types or subtypes. Influenza vaccine effectiveness against overall and subtype-specific laboratory-confirmed influenza were the primary and secondary outcomes. We used a generalised linear mixed model to calculate adjusted vaccine effectiveness according to vaccine match to the circulating strains of influenza virus and intensity of the virus activity (epidemic or non-epidemic). Vaccine effectiveness was defined as the relative reduction in risk of laboratory-confirmed influenza in vaccinated patients compared with unvaccinated patients. We did subgroup analyses to estimate vaccine effectiveness according to hemisphere, age category, and health status., Findings: We received 23 of the 53 datasets included in the aggregate data meta-analysis. Furthermore, six additional datasets were provided by data collaborators, which resulted in individual participant data for a total of 5210 participants. A total of 4975 patients had the required data for analysis. Of these, 3146 (63%) were controls and 1829 (37%) were cases. Influenza vaccination was significantly effective during epidemic seasons irrespective of vaccine match status (matched adjusted vaccine effectiveness 44·38%, 95% CI 22·63-60·01; mismatched adjusted vaccine effectiveness 20·00%, 95% CI 3·46-33·68; analyses in the imputed dataset). Seasonal influenza vaccination did not show significant effectiveness during non-epidemic seasons. We found substantial variation in vaccine effectiveness across virus types and subtypes, with the highest estimate for A H1N1 pdm09 (53·19%, 10·25-75·58) and the lowest estimate for B virus types (-1·52%, -39·58 to 26·16). Although we observed no significant differences between subgroups in each category (hemisphere, age, and health status), influenza vaccination showed a protective effect among elderly people with cardiovascular disease, lung disease, or aged 75 years and younger., Interpretation: Influenza vaccination is moderately effective against laboratory-confirmed influenza in elderly people during epidemic seasons. More research is needed to investigate factors affecting vaccine protection (eg, brand-specific or type-specific vaccine effectiveness and repeated annual vaccination) in elderly people., Funding: University Medical Center Groningen., (Copyright © 2017 Elsevier Ltd. All rights reserved.)

Background: We aimed to estimate the protection afforded by inactivated influenza vaccine, in both community and hospital settings, in a well characterised urban population in Auckland during 2014., Methods: We used two different comparison groups, all patients who tested negative for influenza and only those patients who tested negative for influenza and had a non-influenza respiratory virus detected, to calculate the vaccine effectiveness in a test negative study design. Estimates were made separately for general practice outpatient consultations and hospitalised patients, stratified by age group and by influenza type and subtype. Vaccine status was confirmed by electronic record for general practice patients and all respiratory viruses were detected by real time polymerase chain reaction., Results: 1039 hospitalised and 1154 general practice outpatient consultations met all the study inclusion criteria and had a respiratory sample tested for influenza and other respiratory viruses. Compared to general practice patients, hospitalised patients were more likely to be very young or very old, to be Māori or Pacific Islander, to have a low income and to suffer from chronic disease. Vaccine effectiveness (VE) adjusted for age and other participant characteristics using all influenza negative controls was 42% (95% CI: 16 to 60%) for hospitalised and 56% (95% CI: 35 to 70%) for general practice patients. The vaccine appeared to be most effective against the influenza A(H1N1)pdm09 strain with an adjusted VE of 62% (95% CI:38 to 77%) for hospitalised and 59% (95% CI:36 to 74%) for general practice patients, using influenza virus negative controls. Similar results found when patients testing positive for a non-influenza respiratory virus were used as the control group., Conclusion: This study contributes to validation of the test negative design and confirms that inactivated influenza vaccines continue to provide modest but significant protection against laboratory-confirmed influenza., (Copyright © 2015 Elsevier Ltd. All rights reserved.)

Aims: This paper describes the recent trends of pertussis and vaccine uptake in New Zealand based on notifications and immunisation registration information since 2011. It highlights the current risk for the infant in the first months after birth and the crucial role a pertussis booster in pregnancy could play. It also aims to show that protection of infants by the acellular pertussis vaccine can be improved by timely immunisation even in a situation of improving overall uptake rates that are nearing the national target of 95%., Methods: We analysed New Zealand notification data for pertussis, extracted from EpiSurv between August 2011 and December 2013, which included the period of the last epidemic. Pertussis immunisation coverage data were extracted from the National Immunisation Register (NIR). Population estimates were based on 2006 census data. Deprivation was analysed using the New Zealand Deprivation Index 2006., Results: Despite immunisation coverage at 12 months having exceeded 90% New Zealand experienced a large epidemic from 2011 to 2014, with several hundred infant hospitalisations and three deaths. Notification data indicated an average annual rate of pertussis in the New Zealand population of 102 per 100,000 with the highest rates in the youngest age groups. While an overall increase in immunisation coverage in New Zealand was evident and the timeliness showed improvement across ethnic groups and deprivation deciles, there was a marked geographical variation within DHBs and between ethnic groups., Conclusions: Given the recent published evidence, pertussis vaccination should be offered to all mothers between weeks 28 and 38 of pregnancy. Further improvements are still possible in coverage at 6 months, particularly in Māori and but also in Pacific populations, as well as in more deprived populations. DHBs work towards achieving the 95% target can contribute to the improvement in the timeliness of immunisation.

Highlights: [•] This study adds to the limited research on VE to influenza from the southern hemisphere. [•] TIV vaccines show low to moderate protection against influenza positive hospitalisation in 2012. [•] VE varied by influenza type and subtype. [•] Vaccination appeared to be less effective in patients aged 65 years and older. [•] Older people did not appear to be significantly protected against infection with influenza A (H3N2). [ABSTRACT FROM AUTHOR]