Vaccines for preventing influenza in healthy adults

  • Review
  • Intervention

Authors

  • V Demicheli,

  • D Rivetti,

  • JJ Deeks,

  • TO Jefferson


Dr Vittorio Demicheli, Servizo Sovrazonale di Epidemiologia, ASL 20, Via Venezia 6, Alessandria, Piemonte, 15100, ITALY. Vittorio.DeMicheli@regione.piemonte.it.

Abstract

Background

Three different types of influenza vaccines are currently produced worldwide. None is traditionally targeted to healthy adults. Despite the publication of a large number of clinical trials, there is still substantial uncertainty about the clinical effectiveness of influenza vaccines and this has negative impact on the vaccines acceptance and uptake.

Objectives

  • To assess the effects of vaccines on influenza in healthy adults.

  • To assess the effectiveness of vaccines in preventing cases of influenza in healthy adults.

  • To estimate the frequency of adverse effects associated with influenza vaccination in healthy adults.

Search strategy

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, Issue 1, 2004) which contains the Cochrane Acute Respiratory Infections Group trials register; MEDLINE (January 1966 to December 2003); and EMBASE (1990 to December 2003). We wrote to vaccine manufacturers and first or corresponding authors of studies in the review.

Selection criteria

Any randomised or quasi-randomised studies comparing influenza vaccines in humans with placebo, control vaccines or no intervention, or comparing types, doses or schedules of influenza vaccine. Live, attenuated or killed vaccines or fractions thereof administered by any route, irrespective of antigenic configuration were considered. Only studies assessing protection from exposure to naturally occurring influenza in healthy individuals aged 14 to 60 (irrespective of influenza immune status) were considered.

Data collection and analysis

Two reviewers independently assessed trial quality and extracted data.

Main results

Twenty five reports of studies involving 59,566 people were included. The recommended live aerosol vaccines reduced the number of cases of serologically confirmed influenza by 48% (95% confidence interval (CI) 24% to 64%), whilst recommended inactivated parenteral vaccines had a vaccine efficacy of 70% (95% CI 56% to 80%). The yearly recommended vaccines had low effectiveness against clinical influenza cases: 15%(95% CI 8% to 21%) and 25% (95% CI 13% to 35%) respectively. Overall the percentage of participants experiencing clinical influenza decreased by 6%. Use of the vaccine significantly reduced time off work but only by 0.16 days for each influenza episode (95% CI 0.04 to 0.29 days); Analysis of vaccines matching the circulating strain gave higher estimates of efficacy, whilst inclusion of all other vaccines reduced the efficacy.

Authors' conclusions

Influenza vaccines are effective in reducing serologically confirmed cases of influenza. However, they are not as effective in reducing cases of clinical influenza and number of working days lost. Universal immunisation of healthy adults is not supported by the results of this review.

Plain language summary

Plain language summary

There is not enough evidence to recommend universal vaccination against influenza in healthy adults

Influenza is an acute, viral respiratory infection with symptoms of headache, fever, muscle pain, cough and a runny nose. It spreads easily and can cause serious illness. New strains develop regularly. Each year, the World Health Organization recommends which strains to include in vaccinations for the next 'season'. People considered 'at risk' of complications may be offered vaccination. This review found that vaccinating healthy adults not at risk of complications reduced their chances of developing flu-like illness by only a quarter, and the number of working days lost by less than half a day. Vaccination had no reported relevant adverse effects.

Background

Influenza is an acute respiratory infection caused by a virus of the Orthomyxoviridae family. Three serotypes are known (A, B and C). Influenza causes an acute febrile illness with myalgia, headache and cough. Although the median duration of the acute illness is three days, cough and malaise can persist for weeks. Complications of influenza include otitis media, pneumonia, secondary bacterial pneumonia, exacerbations of chronic respiratory disease and bronchiolitis in children. These illnesses may require treatment in a hospital and can be life-threatening especially in 'high-risk' people e.g. the elderly and people suffering from chronic heart disease. Additionally, influenza can cause a range of non-respiratory complications including febrile convulsions, Reye's syndrome and myocarditis (Wiselka 1994).

The influenza virus is composed of a protein envelope around an RNA core. On the envelope are two antigens: neuraminidase (N antigen) and hemagglutinin (H antigen). Hemagglutinin is an enzyme that facilitates the entry of the virus into cells of the respiratory epithelium, while neuraminidase facilitates the release of newly produced viral particles from infected cells. The influenza virus has a marked propensity to mutate its external antigenic composition to escape the hosts' immune defences. Given this extreme mutability, a classification of viral subtype A based on H and N typing has been introduced. Additionally, strains are classified on the basis of antigenic type of the nucleoprotein core (A, B ), geographical location of first isolation, strain serial number and year of isolation. Every item is separated by a slash mark (e.g. A/Wuhan/359/95 (H3N2)). Unless otherwise specified such strains are of human origin.

In the last century there have been three pandemics caused by different Influenza A viral subtypes:

Year Influenza A Subtype
1918-9 H1N1
1957 H2N2
1968 H3N2

Pandemics occur when there is no natural immunity in the entire population or in a significant segment of the population. Pandemics are caused by so-called antigenic shift (a major change in H configuration with or without a concomitant change in N and perhaps viral alteration of tissue tropism) leading to the appearance of a new subtype against which there is little circulating natural immunity. Pandemics are thought to originate in southern China where ducks (the animal reservoir and breeding ground for new strains), pigs (which are thought to be the biological intermediate host or 'mixing vessel') and humans live in very close proximity (Bonn 1997). Pigs are considered plausible intermediate hosts as their respiratory epithelium cells have receptors for both avian (i.e. duck) and human viral haemagglutinins (Bonn 1997). Minor changes in viral antigenic configurations, known as 'drift', cause local or more circumscribed epidemics.

Flu infection may also appear as a zoonotic infection with direct spread of the avian virus to humans. In April 2003, 87 persons in Netherlands were infected with avian virus H7N7. Most of them presented only conjunctivitis, but a fifty-seven- year-old veterinary attending to veterinary public health interventions, died of respiratory distress. Moreover an avian virus (H5N1) was transmitted during February 2003 to a Hong Kong family: the nine-year-old son was hospitalised and recovered however the eight-year-old sister and the father died. Adaptation of avian influenza viruses to humans could be the explanation for the high virulence of the 1918 to 1919 pandemic and implies that in this era of global travel, new influenza viruses could cause epidemics and pandemics with little warning. Influenza pandemics are caused by viruses which have the ability to replicate and spread rapidly in humans and against which there are no antibodies in the human population (Claas 1998).

Pandemics by definition cause a very high morbidity and mortality burden (Jefferson 1998). The 1918 to 1919 pandemic is considered to have caused an estimated up to 20 million deaths worldwide. Efforts to prevent or treat influenza have had their mainstay in two separate approaches: vaccines and antivirals (mainly amantadine and rimantadine: neuraminidase inhibitors). The effects of antivirals are the topic of separate reviews. Current influenza vaccines are of three types:
(1) whole virion vaccines which consist of complete viruses which have been "killed" or inactivated, so that they are not infectious but retain their strain-specific antigenic properties;
(2) subunit virion vaccines which are made of surface antigens (H and N) only;
(3) split virion vaccines in which the viral structure is broken up by a disrupting agent. These vaccines contain both surface and internal antigens.
In addition a variety of non-European manufacturers produce live attenuated vaccines. Traditionally whole virion vaccines are thought to be the less well-tolerated because of the presence of a lipid stratum on the surface of the viral particles (a remnant of the host cell membrane coating the virion, when budding from the host cell). Influenza vaccines are produced worldwide.

Periodic antigenic drifts and shifts pose problems for vaccine production and procurement, as a new vaccine closely matching circulating antigenic configuration must be produced and procured for the beginning of each new influenza 'season'. To achieve this, the WHO has established a worldwide surveillance system allowing identification and isolation of viral strains circulating the different parts of the globe. Sentinel practices recover viral particles from the naso-pharynx of patients with influenza-like symptoms and the samples are swiftly sent to the laboratories of the national influenza centres (110 laboratories in 79 countries). When new strains are detected the samples are sent to one of the four WHO reference centres (London, Atlanta, Tokyo and Melbourne) for antigenic analysis. Information on the circulating strain is then sent to the WHO, who in February of each year recommends, through a committee, the strains to be included in the vaccine for the forthcoming 'season'. Individual governments may or may not follow the WHO recommendations. Australia, New Zealand and more recently South Africa, follow their own recommendations for vaccine content. Surveillance and early identification thus play a central part in the composition of the vaccine.

Traditionally, influenza vaccines have been targeted to the elderly and those at serious risk of complications. Despite clear theoretical advantages in the use of vaccines, their uptake has been patchy. The current low level of influenza vaccine uptake in targeted populations may reflect uncertainty on the part of primary care and public health practitioners and health policy decision-makers regarding vaccine effectiveness. One possible problem may be the diversity of regulations for the financing and reimbursement of the vaccines. Other reasons may include perceived low efficacy, due to the mutable viral configuration, the perceived commonality of the disease, which may breed contempt and, strangely, a misperception of the burden imposed by the disease on society. Nowhere is this more marked than in the case of healthy adults in employment, a population which would benefit from protection against influenza. Epidemics in settings such as schools, barracks, prisons, offices, hospitals and industrial complexes cause great losses, but are seldom prevented by vaccination of staff.

Despite the publication over a period of more than five decades of a large number of reports of controlled clinical trials, there remains substantial uncertainty about the clinical effectiveness of influenza vaccine. This uncertainty is manifested in widely varying estimates of vaccine effectiveness in the current health care literature. For example the journal Morbidity and Mortality Weekly Report (MMWR) states: " the effectiveness of influenza vaccine in preventing or attenuating illness varies, depending primarily on the age and immunocompetence of the vaccine recipient and the degree of similarity between the virus strains included in the vaccine and those that circulate during the influenza season. When a good match exists between vaccine and circulating viruses, influenza vaccine has been shown to prevent illness in approximately 70% to 90% of healthy persons aged less than 65 years" (MMWR 1998). There are no systematic reviews of the effects of vaccination on healthy adults known to us while there are several examples in the elderly such as the ones by Vu, Gross and Strassburg (Gross 1995; Strassburg 1986; Vu 2002).

The deficiencies of current and past reviews of influenza vaccine effectiveness can be summarised as follows:
(1) lack of comprehensiveness in the identification of primary studies;
(2) lack of methodological assessment of primary studies;
(3) failure to satisfactorily account for (or in some cases, to acknowledge) the marked variability in vaccine effectiveness among controlled studies;
(4) failure to provide estimates of vaccine effectiveness under conditions of imperfect antigenic matching between vaccines and prevalent viruses (that is, when vaccines contain either a different strain or a different subtype of influenza virus than the prevalent virus);
(5) lack of credible estimates of vaccine effectiveness in specific populations currently targeted for influenza vaccination (for example, institutionalised elderly, community-dwelling elderly and persons with underlying medical conditions associated with a high risk of complications) (MMWR 1998; NACI 1988).

These deficiencies help to explain discrepancies in reported vaccine effectiveness in the existing literature. Moreover, they can be expected to give rise to uncertainty among clinicians and policy-makers regarding the expected effectiveness of influenza vaccine. In this scenario a systematic review of the effects of vaccines against naturally occurring influenza is necessary to enable decision-makers to devise strategies to deal with influenza based on evidence.

Objectives

To identify, retrieve and assess all studies evaluating the effects of vaccines on influenza in healthy adults.
To assess the effectiveness of vaccines in preventing cases of influenza in healthy adults.
To estimate the frequency of adverse effects associated with influenza vaccination in healthy adults.

The following hypotheses were tested:
In comparisons between groups intended for influenza vaccination compared with control/placebo groups:
There is no difference in the number of cases of influenza and their severity (expressed, for example, as numbers of days of bed rest, inactivity, absence from work or school, deaths and admissions to hospital).
There is no difference in the number and severity of adverse effects (both systemic and localised).

Criteria for considering studies for this review

Types of studies

Any randomised or quasi-randomised studies* comparing influenza vaccines in humans with placebo, control vaccines or no intervention or comparing types, doses or schedules of influenza vaccine. Only studies assessing protection from exposure to naturally occurring influenza will be considered.

*A study is randomised when it appears that the individuals (or other experimental units) followed in the study were definitely or possibly assigned prospectively to one of two (or more) alternative forms of health care using random allocation. A study is quasi-randomised when it appears that the individuals (or other experimental units) followed in the study were definitely or possibly assigned prospectively to one of two (or more) alternative forms of health care using some quasi-random method of allocation (such as alternation, date of birth or case record number).

Types of participants

Healthy individuals aged 14 to 60 irrespective of influenza immune status. Studies considering more than 25 percent of individuals outside this age range will be excluded from the review.

Types of intervention

Live, attenuated or killed vaccines or fractions thereof administered by any route, irrespective of antigenic configuration.

Types of outcome measures

Clinical:
Numbers and seriousness of influenza cases (however defined) occurring in vaccine and placebo groups.

Adverse effects:
Number and seriousness of adverse effects (classified as local and systemic). Systemic adverse effects include cases of malaise, nausea, fever, arthralgias, rash, headache and more generalised and serious signs. Local adverse effects include induration, soreness and redness at the site of inoculation.

Search methods for identification of studies

See: Unavailable methods used in reviews.

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, Issue 1, 2004) which contains the Cochrane Acute Respiratory Infections Group's trials register; MEDLINE (January 1966 to December 2003); and EMBASE (1990 to December 2003). There were no language restrictions.

The following terms were searched for reports of trials:

MEDLINE
#1 ("Influenza Vaccine/administration and dosage"[MeSH] OR "Influenza Vaccine/adverse effects"[MeSH] OR "Influenza Vaccine/contraindications"[MeSH] OR "Influenza Vaccine/immunology"[MeSH] OR "Influenza Vaccine/metabolism"[MeSH] OR "Influenza Vaccine/radiation effects"[MeSH] OR "Influenza Vaccine/therapeutic use"[MeSH] OR "Influenza Vaccine/toxicity"[MeSH]) OR ("Influenza/epidemiology"[MeSH] OR "Influenza/immunology"[MeSH] OR "Influenza/mortality"[MeSH] OR "Influenza/prevention and control"[MeSH] OR "Influenza/transmission"[MeSH])
#2 (influenza vaccin*[Title/Abstract]) OR ((influenza [Title/Abstract] OR flu[Title/Abstract]) AND (vaccin*[Title/Abstract] OR immuni*[Title/Abstract] OR inoculati*[Title/Abstract] OR efficacy[Title/Abstract] OR effectiveness[Title/Abstract])
#3 #1 OR #2
# 4 "Randomized Controlled Trial"[Publication Type] OR "Randomized Controlled Trials"[MeSH] OR "Controlled Clinical Trial"[Publication Type] OR "Controlled Clinical Trials"[MeSH] OR "Random Allocation"[MeSH] OR "Double-Blind Method"[MeSH] OR "Single-Blind Method"[MeSH]
#5 controlled clinical trial*[Title/Abstract] OR randomised controlled trial*[Title/Abstract] OR clinical trial*[Title/Abstract] OR random allocation[Title/Abstract] OR random*[Title/Abstract] OR placebo[Title/Abstract] OR double - blind[Title/Abstract] OR single - blind[Title/Abstract] OR RCT[Title/Abstract] OR CCT[Title/Abstract] OR allocation[Title/Abstract] OR follow - up[Title/Abstract]
#6 #4 OR #5
#7 #3 AND #6

The above search strategy was modified and repeated on CENTRAL and EMBASE databases. There were no language restrictions. In order to identify further trials, we read the bibliography of retrieved articles and handsearched the journal Vaccine from its first issue to the end of 2003. Results of handsearches are included in CENTRAL. In order to locate unpublished trials, for the first edition of this review, we wrote to the following: manufacturers; first or corresponding authors of studies in the review.

Methods of the review

Inclusion procedure
Two authors (VD and TOJ) read all trials retrieved in the search and applied inclusion criteria.

Trial quality assessment
Trials fulfilling the review inclusion criteria were assessed for quality and the results analysed by VD and TOJ. Assessments of trial quality were made according to the following criteria:
(1) generation of allocation schedule (defined as the methods of generation of the sequence which ensures random allocation);
(2) measure(s) taken to conceal treatment allocation (defined as methods to prevent selection bias, i.e. to ensure that all participants have the same chance of being assigned to one of the arms of the trial. It protects the allocation sequence before and during allocation);
(3) number of drop-outs of allocated healthcare worker participants from the analysis of the trial (defined as the exclusion of any participants for whatever reason - deviation from protocol, loss to follow-up, withdrawal, discovery of ineligibility while the unbiased approach analyses all randomised participants in the originally assigned groups regardless of compliance with protocol - known as intention to treat analysis);
(4) measures taken to implement double blinding (a single blind study is one in which observer(s) and/or subjects are kept ignorant of the group to which the subjects are assigned, as in an experiment, or of the population from which the subjects come, as in a non-experimental study. When both the observer and participants are kept ignorant of assignment the trial is called double-blind. Unlike allocation concealment, double blinding seeks to prevent ascertainment bias and protects the sequence after allocation).

For criteria two, three and four there is empirical evidence that low quality in their implementation is associated with exaggerated trial results (Schulz 1995) and it is reasonable to infer a quality link between all four items.

Arbitration procedure
When there was disagreement between TOJ and VD on the quality of a trial, JJD arbitrated.

Data collection
The following data were extracted, checked and recorded -
Characteristics of trials:
- Date of trial
- Location of trial
- Setting
- Case definitions used (clinical, serological, virological)
- Surveillance system used
- Type and length of epidemic (definition used, characteristics of circulating virus reporting the virus nomenclature used at the time of the trial)
Characteristics of participants:
- Number of participants
- Age, gender, ethnic group, risk category
- Occupation
Characteristics of interventions:
- Type of vaccine, type of placebo, dose, treatment or prophylaxis schedule, length of follow-up (in days)
- Route of administration
Characteristics of outcome measures:
- Numbers and seriousness of influenza cases (however defined) occurring in vaccine and placebo groups. Other outcome measures used in the trials to assess effects to be included:
- Adverse effects: presence and type
Sponsor of trial (specified, known or unknown)
Publication status

Individual studies are described as follows:
Type of trial
Setting
Year of trial
Type of vaccine
Comparator
Type of population
Population size
Arm size
Length of follow-up
Length of epidemic
Definition of epidemic
Outcomes used
Case definition used
Recommended vaccine for that year
Circulating virus

Four different definitions of "epidemic period" were found.
- the interval between the first and the last virus isolation in the community
- the interval during which influenza virus was recovered from more than a stated percentage of ill subjects
- the period during which an increase of respiratory illness more than a stated % was recorded
- the winter period taken as a proxy for epidemic period
The data were included regardless of the definition of epidemic period used in the primary study.
When data were presented for the epidemic period and the entire follow up period, those which occurred during the former were considered.

The following outcomes were included in the comparisons:
- cases of influenza clinically defined
- cases of influenza clinically defined on the basis of a specific list of symptoms and/or signs
- cases of influenza confirmed by laboratory tests
- hospital admissions
- complications
- working days lost (WDL)
- local adverse events
- systemic adverse events

A clinically defined case was assumed as:
- any case definition based on symptoms without further specification

A clinically defined case (specific definition) was assumed as:
- "flu-like illness" according to a predefined lists of symptoms (including the Centres for Disease Control (CDC) case definition for surveillance)
- "upper respiratory illness" according to a predefined lists of symptoms
When more than one definition was given for the same trial, data related to the more specific definition were included.

The laboratory confirmation of cases found were:
- virus isolation from culture
- four-fold antibody increase (hemagglutinin) in acute or convalescent phase sera
- four-fold antibody increase (hemagglutinin) in post-vaccination or post-epidemic phase sera
When more than one definition was given for the same trial, data related to the more sensitive definition (seroconversion) were included.

Hospital admissions rates were calculated as proportion of cases hospitalised for respiratory causes. Complications were considered as proportion of cases complicated by bronchitis, pneumonia or otitis. Working days lost in episodes of sickness absence regardless of cause were also considered. Only five studies used working days lost as an outcome measure and four of them measured the work absence in terms of difference of the average number of days lost in the two arms of the trial. These studies presented a value of standard error measured accordingly. The remainder Nichol (Nichol 1999) expresses the work absence in terms of rate ratio and this does not allow the recalculation of the correct estimate of the standard error. Therefore this study was excluded from the pooled analysis.

Local symptoms are presented separately from systemic symptoms. Individual adverse effects have been considered in the analysis, as well as a combined endpoint (any or highest symptom).

All the data included in the analysis were used as presented by the authors in the primary study regardless of the number of drop-outs. This approach was decided because the majority of the studies did not present any attempt at using an intention to treat analysis, did not mention the reasons for the loss to follow up and did not contain detailed information to allow estimate the real number of participants.

The tables of comparisons were constructed according to the following criteria:
(1) Influenza vaccine versus placebo.
All studies comparing any influenza vaccine against a placebo (inert substances or non-influenza vaccines) were included in this group.
Subgroup analysis were carried out for the following type of vaccine:
- live aerosol vaccine
- inactivated parenteral vaccine
- inactivated aerosol vaccine

The parenteral route comprised both intramuscular and subcutaneous. Different dosages and schedules of the vaccine and the presence of different adjuvants were not compared and data from arms of trials comparing only vaccine composition or dosage were pooled in the analysis.

(2) At least one vaccine recommended for that year versus placebo or other vaccines. All the trials in which the vaccine in study contained at least one of the A strains recommended for that year by WHO or single governments were included in this group of comparison, independently from substances used in the control arm (WHO recommendations on content of vaccines have been published since 1973).
Subgroup analysis was carried out according to control group (three strata: recommended vaccine against placebo, recommended vaccine against influenza B vaccine, recommended vaccine against other non recommended A strains). Vaccines containing only an influenza B recommended strain were excluded from this specific comparison since a number of authors used monovalent B vaccine as placebo in the control arm and some confusion may be generated. The compliance of the study vaccine with the official recommendations was checked by reviewing WHO records when possible. In case of uncertainty due to ambiguity of wording used (in the oldest trials), the opinion stated by authors was taken into account. The compliance of a live attenuated vaccine with the recommendation has been decided according to the antigenic comparability of the wild strains.

(3) Vaccine matching circulating strain versus placebo or other vaccines. All the trials in which the vaccine in the study contained the strain matching the circulating virus (or at least one of several circulating viruses) were included in this group of comparison, independently from substances used in the control arm. Subgroup analysis was carried out according to the control group (three strata: matching vaccine against placebo, matching vaccine against influenza B vaccine, matching vaccine against other non recommended A strains). In case of uncertainty due to incomplete match or ambiguity of wording used, the opinion stated by authors was also taken into account. Minor viral drift clearly stated was assumed as non-matching.

Data synthesis
We performed separate analyses for live aerosol vaccines, inactivated parenteral vaccines and inactivated aerosol vaccines. Clinical influenza outcomes were specified according to whether specific criteria were or were not used, for which estimates were produced separately, and combined together.

As between study heterogeneity is to be expected in influenza vaccine trials as there are unpredictable systematic differences between trials regarding the circulating strains and the levels of immunity presented by different population in different settings, the DerSimonian and Laird method (random effects model) was applied for the all comparisons and outcomes presented. Once the relative risk (RR) had been obtained, vaccine efficacy (VE) was calculated as VE = 1-RR and expressed as a percentage. Similar analyses were also undertaken for other events, such as complications, hospital admissions and adverse effects.

In addition to the traditional estimate of vaccine efficacy, we estimated the effect of vaccination on the number of clinical cases by averaging the risk differences (influenza rate in vaccinated group minus influenza rate in control group). Where the total number of clinical influenza cases depends more on the number of other influenza-like illnesses than true influenza A illnesses, it is more likely that an intervention will appear to reduce the total number of cases by an absolute amount (i.e. a constant risk difference) than by a relative amount (i.e. a constant relative effect).

As the data on average time off work was reported as a continuous measurement, these results were expressed as differences in means, and combined using the weighted mean difference (WMD) method. Caution should be exercised in interpreting these results, as the data are very skewed.

Several trials included more than one active vaccine arm. Where several active arms from the same trial were included in the same analysis, the placebo group was split equally between the different arms, so that the total number of participants in any one analysis did not exceed the actual number in the trials.

Description of studies

The first version of the review in The Cochrane Library Issue 4, 1999, contained 20 studies. This update searched from 1997 to December 2003 and yielded 1046 titles. From review of the titles and abstracts, 29 reports of studies were retrieved of which five met the inclusion criteria. Reasons for exclusion are reported in the table of excluded studies. In total we identified 25 reports of studies. Some of them had more than two arms, comparing different vaccines, routes of administration, schedules or dosages and reported data from different settings and epidemic seasons. We split these studies into sub-studies up to a total of 47 trials. For the remainder of this review the words "report of study" will identify the original study report. The word "trial" will identify the sub-study. Details of the division of the reports of studies into trials are given in the table of included studies. Some reports of studies took into account the history of previous natural or vaccine immunisations.

Included trials assessed three types of vaccine: live attenuated aerosol, inactivated aerosol and inactivated parenteral. Eight trials of live attenuated vaccine were included. Four of them were placebo controlled. They involved 30,676 participants, 17,072 in the treatment arms and 13,604 in the control groups. We included two studies assessing inactivated vaccine aerosol. Both were placebo controlled and involved 1,507 subjects, 1,339 in treatment arm and 168 in control group. Most studies assessed the efficacy of inactivated parenteral vaccines against placebo or other influenza vaccines (some of them used a monovalent influenza B vaccine as placebo). They involved 27,383 subjects, 16,190 in treatment arm and 11,193 in control group.

Surveillance methods were prospective or retrospective, active (by phone interview or questionnaire compilation) or passive (ill participants spontaneously presenting). Mean length of follow up was 87 days (median 79 days, 25 th percentile 61 days, 75 th percentile 119 days).
The duration of the epidemic was specified in 17 trials. Mean length of the epidemic period was 62 days (median 63 days, 25 th percentile 42 days, 75 th percentile 77 days).

Identified trials are listed and described below using the name of the first author and the publication year.

Keitel 1988
Keitel is an RCT conducted in the USA on healthy volunteers, 30 to 60 years old, working in the Texas Medical Center in Houston, Texas, or in surrounding industrial companies. Volunteers were randomly allocated to receive vaccine or placebo using a table of random numbers according to prior vaccination experience. Groups were evaluated in a double-blind fashion. Keitel 1988 compares a trivalent, killed whole, intramuscularly administered vaccine against placebo. Each dose of the commercial vaccine for the fall of 1983 contained 15 mg of hemagglutinin of each of the following influenza strains: A/Philippines/2/82 (H3N2), A/Brazil/11/78 (H1N1) and B/Singapore/222/79. Each dose of the commercial vaccine for the fall of 1984 contained 15 mg of hemagglutinin of each of the following influenza strains: A/Philippines/2/82 (H3N2), A/Chile/1/83 (H1N1) and B/USSR/100/83. Placebo was sterile saline for injection. The protective efficacy of the influenza vaccine was evaluated during the 1983 to 1984 outbreak (Keitel 1988a) and the 1984 to 1985 outbreak (Keitel 1988b). Clinical and laboratory confirmed cases were detected in three groups: first vaccinated, multi vaccinated and placebo. Each arm counted, respectively, 162, 138 and 298 participants in the Keitel 1988a and 171, 285 and 241 in the Keitel 1988b. Adverse effects were not considered. Outcomes were detected during the whole epidemic period, defined as the interval during which community surveillance recovered influenza viruses from 10% or more of persons with febrile respiratory illness per calendar week. The epidemic period was from January 8 to March 17, 1984 and from January 6 to March 9, 1985. Illnesses was classified in "any", "flu-like" (lower respiratory and/or systemic illness) and "febrile" (oral temperature of 37.8 or higher) according to findings at initial examination in 1983 to 1984 and according to a combination of findings at the time of examination and when the convalescent blood specimen was collected in 1984 to 1985. Symptoms and signs of disease of the trachea, bronchi, and/or lung were designated lower respiratory illness. A designation of systemic illness was based on occurrence of myalgia, malaise, fatigue, lethargy, chills, sweats and anorexia.

In our review "any" illness was used as "clinically defined case" and "flu like" was used as "specific clinical definition of case".
Laboratory confirmation was based on culture and/or four-fold or greater rise in antibody titre occurred between postvaccination (pre-epidemic), acute, convalescent and/or spring (post-epidemic) sera. The laboratory confirmed outcome was reported separately for influenza A or influenza B and for culture or serological positivity. The authors do not specify when culture and serological confirmation were obtained on the same participant. In our review antibody rise was used as outcome measure for laboratory confirmed case and A and B cases were added up. Viral strains isolated from cases were A/Victoria/7/83 (H1N1) and B/USSR/100/83 in the winter 1983 to 1984, and A/Philippines/2/82 (H3N2)-like in the winter 1984 to 1985. The authors underlined the imperfect antigenic matches between vaccine and epidemic viruses for the combined influenza A/H1N1-B outbreak in 1983 to 1984.

The WHO recommended formula was:
1983 to 1984: A/Philippines/2/82 (H3N2), A/Brazil/11/78 (H1N1), B/Singapore/222/79.
1984 to 1985: A/Philippines/2/82 (H3N2), A/Chile/1/83 (H1N1) and B/USSR/100/83.
Keitel 1988a was included among studies comparing recommended vaccines against placebo or other vaccines
Keitel 1988b was included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.

Mair 1974
Mair is a controlled clinical trial (CCT) conducted in UK on 507 volunteer patients from general practices. The vaccines were allocated according to date of birth. The identity of each vaccine was concealed by a code number known only to the laboratory. Complete follow up was obtained for 465 volunteers, 186 male and 279 female. Two monovalent, inactivated, intramuscularly administered influenza A vaccines were compared against a monovalent inactivated influenza B vaccine used as placebo. Vaccine strains were respectively: A/Hong Kong/1/68 (H3N2) (recombinant X31, 600 IU), A/England/42/72 (H3N2) (recombinant XPR8, 600 IU), B/Victoria/98926/70 (400 IU). Each arm counted, respectively, 141,169 and 155 participants. Data were extracted and loaded as follow:
Mair 1974a: monovalent influenza A matching circulating strain (n = 85) against monovalent B (n = 155).
Mair 1974b: monovalent influenza A matching circulating strain (n = 84) against monovalent A (n = 141).
Mair 1974c: monovalent recommended influenza A strain (n = 71) against monovalent B (n = 155).
Mair 1974e: monovalent recommended influenza A strain (n = 70) against monovalent A (n = 169).
Outcomes were clinical and laboratory confirmed cases and adverse effects. Illness was defined as "any symptoms of influenza". Cases were laboratory confirmed when an influenza virus was isolated from throat swabs or a four-fold or greater rise in antibody titre occurred between acute and convalescent sera. Local adverse effects were classified as mild (slight swelling, aching) and severe (painful, red and swollen arm). Systemic adverse effects were classified as mild (slight pyrexia, mild aching of the back and limbs) and severe (influenza-like symptoms). Data on adverse effects were not reported separately for each study arm. Mair was conducted during the 1972 to 1973 influenza season. The follow up period lasted 13 weeks. The authors do not specify the length of the epidemic period. The viral strain isolated from cases was A/England .The recommended formula for that year in the USA was A2/Aichi/2/68, B/Massachusetts/1/72.

Hammond 1978
Hammond 1978 is a CCT, double blinded, conducted in Australia on 225 volunteers: medical students or staff members at Monash University. Coded identical-looking vials were sequentially administered to enrolled participants. A trivalent, subunit vaccine containing 200 IU of A/Victoria/3/75, 250 IU of A/Scotland/840/74 and 300 IU of B/Hong Kong/8/73 (as recommended locally), was compared against "placebo" (diphtheria and tetanus toxoids). Vaccines were administered subcutaneously. Arm size was: 116 participants in the vaccine group and 109 participants in the placebo group. Outcomes were clinical and laboratory confirmed cases. Participants were regarded as suffering from influenza when they suffered a respiratory illness which was associated with the isolation of influenza virus, a four-fold or greater rise in antibody titre occurring between postvaccination and post-epidemic sera, or both. Adverse effects are not considered.
Epidemic influenza was defined by virus isolation and serology tests and lasted from middle April to middle August 1976. Outcomes were detected during the whole epidemic period. A/Vic/3/75-like virus was isolated from patients. The recommended formula for that year was: A/Victoria/3/75, A/Scotland/840/74 and of B/Hong Kong/8/73.

Edmonson 1970
Edmonson 1970 is a randomised, double blind, not placebo controlled trial, conducted in Australia on 1,983 healthy volunteers (1,284 males and 699 females) from amongst Canberra public servants. The bottles of vaccine were numbered in a non consecutive random manner by a clerk, who remained the only person knowing the code. The coded vials were sequentially administered to enrolled participants. A bivalent vaccine (influenza A and B), identical with the commercial one used that year, was compared against a monovalent B influenza vaccine. The bivalent subunit vaccine contained a Hong Kong type virus (A2/N.T./60/68, 8000 HA units) and B/Victoria/2/65 (3000 HA units) strain, while the monovalent B vaccine had a B/Victoria/2/65 (3000 HA units) strain. Vaccines were administered by injection in two doses (the second 20 days after the first one).
Outcomes are clinically defined cases, working days lost and adverse effects. Each arm counted 975 and 878 participants for adverse effects and 933 and 841 participants for symptoms defined cases. Sick leave records were used for dividing diagnoses into groups: influenza, other respiratory diseases, injuries and all other illnesses. Local adverse effects were classified as minimal (when not affecting activities), mild (discomfort when the arm was used; redness, swelling but no absence from work) and severe (causing absence from work). Systemic adverse effects were classified as minimal (no interference with normal activities), mild (interference with normal activities but no absence from work) and severe (absence from work). The trial was conducted during the winter of 1969. Epidemic period length is unknown but may be the whole follow-up period. The circulating virus, isolated from participants, was A2/H.K./68. The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68. The vaccine strain was considered to be recommended and matching the circulating strain.

Weingarten 1985
Weingarten 1985 is an RCT conducted in USA on 179 healthy volunteer hospital employees, 21 to 65 year olds. Participants were assigned using a random-number generator to receive either the influenza vaccine (a split trivalent containing 15 mg each of A/Chile/1/83 (H1N1), A/Philippines/2/82 (H3N2), and B/USSR/100/83 hemagglutinin antigens) or placebo (saline). Each arm counted 91 and 88 participants respectively. All injections were given intramuscularly. The injections were given and the data were collected in a double blind manner. Outcomes were clinical cases symptoms defined, working days lost (WDL), and adverse effects. Influenza illness was defined by the CDC case definition: a documented temperature greater than 100°F and at least the symptoms of cough or sore throat. Data regarding WDL and adverse effects were not complete and they were not considered. Trial was conducted during the 1985 to 1986 influenza season. Epidemic influenza was defined according to population surveillance data (without better explanation), begun in December 1985 and concluded in February 1986. The follow up comprised the epidemic period and lasted 67 days. Most of the influenza infections were caused by type B. In the authors' opinion there was poor closeness of fit between the virus in the vaccine and the circulating virus. The recommended formula for that year was A/Chile/1/83 (H1N1), A/Philippines/2/82 (H3N2), and B/USSR/100/83.

Waldman 1972
Waldman is an RCT conducted in a USA university campus, on 864 volunteer students and staff members. Volunteers were randomly allocated to study groups and were immunised in a double blind manner. Two inactivated vaccines (monovalent A and bivalent AB) and two different routes of administration (subcutaneous and aerosol) were compared against placebo (saline). Data were extracted and loaded as follows:
Waldman 1972a: monovalent A (A2/Aichi/1/68) 200 CCA aerosol against placebo. 195 treated, 25 placebo.
Waldman 1972b: monovalent A (A2/Aichi/1/68) 200 CCA subcutaneous against placebo. 190 treated, 25 placebo.
Waldman 1972c: bivalent AB (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA and B/Massachusetts/3/66 200 CCA) aerosol against placebo. 194 treated, 25 placebo.
Waldman 1972d: bivalent AB (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA and B/Massachusetts/3/66 200 CCA) subcutaneous against placebo. 187 treated and 23 placebo.
Outcomes were clinical cases and adverse effects. Clinical cases were defined as febrile respiratory illness with oral temperature higher then 99.5 F. Local adverse effects were defined as pain and/or tenderness and redness and/or swelling. Systemic adverse effects were defined as general (fever, muscle pain, nausea or vomiting, diarrhoea and malaise) or respiratory (runny and/or stuffy nose, sore throat, cough, shortness of breath). The protective efficacy of the influenza vaccines was evaluated during the 1968 to 1969 outbreak. Study follow-up lasted 88 days and is likely to have covered the whole epidemic period. Circulating viral strain was A2/Hong Kong virus and it matched the monovalent but not the polyvalent vaccine. The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68. The recommended polyvalent formula for that year in the USA was: A/PR/8/34, A1/AA/1/57, A2/Japan/170/62, A2/Taiwan/1/64, B/Mass/3/66
Viruses used in the vaccines matched the recommended strains.

Edwards 1994
Edwards is an RCT conducted in the USA on healthy children and adults of metropolitan Nashville aged one to 65 years old. Participants were recruited from seven organisations and assigned to one of the study groups using a permuted block randomisation scheme that was stratified by treatment center and age group. The study was double blinded. Participants and clinical staff were kept unaware of the assigned vaccine group through the use of sealed randomisation envelopes that contained vaccine codes. A total of 5210 people participated in the trial. Two types of vaccine (a bivalent, live cold adapted, aerosol administered influenza A vaccine and the commercial inactivated intramuscularly administered influenza vaccine) were compared against placebo. Equal proportions of participants were included in the three vaccine arms. The study lasted five years (from 1985 to 1989). Participants were immunised each fall staying in the same assigned vaccine group. Each year the viral strains used in the vaccines were different. In the four years of the study the commercial inactivated vaccine contained influenza A and B viruses. Since cold adapted influenza B vaccines were not sufficiently characterised to include in the study, in those years authors used monovalent inactivated influenza B vaccine as intramuscular placebo. Otherwise placebo was saline solution used for injections. Allantoic fluid was the placebo for aerosol administration. The influenza A viral strains used in the vaccines were:
- winter 1985 to 1986: cold adapted: Dunedin/6/83 H1N1 and Korea/1/82 H3N2 (107-107,6 plaque forming units per ml (pfu/ml)); inactivated: Chile/1/83 H1N1 and Philippines/2/82 H3N2 (15 mg each strain)
- winter 1986 to 1987: cold adapted: Texas/1/85 H1N1 and Bethesda/1/85 H3N2 (107-107,6 pfu/ml); inactivated: Chile/1/83 H1N1 and Mississippi/1/85 H3N2 (15 mg each strain).
- winter 1987 to 1988: cold adapted: Kawasaki/9/86 H1N1 and Bethesda/1/85 H3N2 (107-107,6 pfu/ml); inactivated: Taiwan/1/86 H1N1 and Leningrad/360/86 H3N2 (15 mg each strain).
- winter 1988 to 1989: cold adapted: Kawasaki/9/86 H1N1 and Los Angeles/2/87 H3N2 (107-107,6 pfu/ml); inactivated: Taiwan/1/86 H1N1 and Sichuan/2/87 H3N2 (15 mg each strain).
- winter 1989 to 1990: cold adapted: Kawasaki/9/86 H1N1 and Los Angeles/2/87 H3N2 (107-107,6 pfu/ml); inactivated: Taiwan/1/86 H1N1 and Shanghai/11/87 H3N2 (15 mg each strain).
Influenza B strain contained in the commercial and monovalent vaccines was not described. Strains used yearly to develop cold adapted and inactivated vaccines were antigenically comparable. Cases were detected each year on the amount of participants enrolled up-to-date, independently of the number of injections received and the duration of participation in the 5-years study. During the 1985 to 1986 season no influenza A illness was detected. An outbreak of influenza B/Ann Arbor/1/86 occurred, but no data relating to this outbreak are included in the report. The epidemic period in any study year started on the day that the first influenza A virus isolate was obtained in Nashville and ended on the day that the last isolate was obtained. Data were extracted and loaded as follows:
Edwards 1994a: 1986 to 1987 outbreak; live cold adapted against placebo (allantoic fluid); 872 treated, 439 placebo; circulating virus Taiwan/1/86 (antigen drift from vaccine strain); epidemic period 56 days.
Edwards 1994b: 1986 to 1987 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 878 treated, 439 placebo; circulating virus Taiwan/1/86 (antigen drift from vaccine strain) epidemic period 56 days.
WHO recommended formula for that year: A/Chile/1/83, A Singapore/6/86, A/Mississippi/1/85, B/Ann Arbor/1/86. These trials were included among studies comparing recommended vaccines against placebo or other vaccines.
Edwards 1994c: 1987 to 1988 outbreak; live cold adapted against placebo (allantoic fluid); 1,029 treated, 532 placebo; circulating viruses Sichuan/2/87 (H3N2) (antigen drift from vaccine strain) and B/Victoria/2/87; epidemic period 101 days.
Edwards 1994d: 1987 to 1988 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 1,060 treated, 532 placebo; circulating viruses Sichuan/2/87 (H3N2) (antigen drift from vaccine strain) and B/Victoria/2/87; epidemic period 101 days.
WHO recommended formula for that year: A Singapore/6/86, A/Leningrad/360/86, B/Ann Arbor/1/86. These trials were included among studies comparing recommended vaccines against placebo or other vaccines.
Edwards 1994e: 1988 to 1989 outbreak; live cold adapted against placebo (allantoic fluid); 1,114 treated, 562 placebo; circulating viruses Taiwan/1/86 (H1N1) and B/Yamata/16/88; epidemic period 78 days.
Edwards 1994f: 1988 to 1989 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 1,126 treated, 562 placebo; circulating viruses Taiwan/1/86 (H1N1) and B/Yamata/16/88; epidemic period 78 days.
WHO recommended formula for that year: A Singapore/6/86, A/Sichuan/2/87, B/Beijing/1/87. These trials were included among studies comparing recommended vaccines against placebo or other vaccines and those comparing vaccines matching the circulating virus against placebo or other vaccines.
Edwards 1994g: 1989 to 1990 outbreak; live cold adapted against placebo (allantoic fluid); 999 treated, 508 placebo; circulating virus Shanghai/11/87 (H3N2); epidemic period 80 days.
Edwards 1994h: 1989 to 1990 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 1016 treated, 508 placebo; circulating virus Shanghai/11/87 (H3N2); epidemic period 80 days.
WHO recommended formula for that year: A Singapore/6/86, A/Shanghai/11/87, B/Yamagata/16/88. These trials were included among studies comparing recommended vaccines against placebo or other vaccines and those comparing vaccines matching the circulating virus against placebo or other vaccines.

Outcomes were clinically specific, with laboratory confirmed cases and adverse effects. Participants had influenza-like illness if they had fever of abrupt onset with at least one of the following: chills, headache, malaise, myalgia, cough, pharyngitis or other respiratory complaints, and presented for throat culture. They had a culture-positive illness if the culture revealed an influenza A virus. They had retrospectively reported illness if they stated at the time of the spring blood sampling that they had an influenza-like illness during the previous winter. Participants seroconverted if they had a fourfold rise in titre from post-vaccination to spring sera for the influenza A virus that circulated that season. Adverse effects were those experienced by all the participants vaccinated in the five-years study. Local adverse effects were redness, induration, tenderness. Systemic adverse effects were fever, sore throat, coryza, lethargy, chills, nausea, headache, muscle ache and cough.

Nichol 1995
Nichol 1995 is an RCT conducted in the USA on healthy full-time employed volunteers, 18 to 64 years old. Randomisation was performed according to a computer-generated randomisation schedule. Double blinding was ensured by preloaded, coded identical looking syringes.
A subvirion, trivalent, parenteral influenza A and B vaccine was compared against placebo (vaccine diluent). Vaccine strains were A/Texas/36/91, A/Shangdong/9/93, B/Panama/45/90, 15 mg each. A total of 849 participants were enrolled. Follow up interviews to ascertain side effects were completed for 841 participants. Complete follow up data were obtained for 416 of 422 placebo recipients and 409 of 419 vaccine recipients. Outcomes were clinical cases symptoms defined, working days lost (WDL), and adverse effects. Patients were defined as cases if they had at least one upper respiratory illness (a sore throat associated with either fever or cough that lasted at least 24 hours).
Local adverse effects were defined as arm soreness. Systemic adverse effects were defined as fever, tiredness, "feeling under the weather", muscle ache, headache. The trial was conducted during the 1994 to 1995 influenza season. The follow up period was defined as December 1, 1994 through to March 31, 1995 (the influenza season). The WHO recommended formula for that year was: A/Texas/36/91, A/Shangdong/9/93, B/Panama/45/90.

Rytel 1977
Rytel 1977 is a single blind RCT, conducted in the USA on 143 young adult female student nurse volunteers with low antibody titre against influenza A or B virus. Efficacy of a single or double doses of a live attenuated, bivalent, intranasal influenza A (containing 107,2 EID50) and B (containing 107,8 EID50 ) vaccines was tested against placebo (5% sucrose). Vaccine virus strains were A/England/42/72 (H3N2) and B/Hong Kong/5/72. Study participants were randomly assigned into three subgroups to receive either two doses of the vaccine (n = 47), one dose of vaccine and one dose of placebo (n = 48) or two doses of placebo (n = 48) at 14 days apart. Outcomes were laboratory confirmed cases and adverse effects. A laboratory confirmed case was defined as the presence of a flu-like illness (three or more symptoms of acute respiratory disease and temperature greater then 37.2) and virus isolation and/or four fold rise in antibody titre in sera obtained at 30 days and six months following immunisation. Local adverse effects were upper respiratory infection (URI) symptoms and cough and were subdivided into moderate and severe. A definition of general adverse effects (again distinguished among moderate and severe) was not given. The trial was conducted during the 1974 to 1975 influenza season in which a two-month epidemic period is described by the authors with no reference to a definition. The follow-up period of the study was generically defined as "winter and spring", lasted six months and comprised the epidemic period.
The circulating strain was A/PortChalmers/1/73 (H3N2) and presented a slight antigenic drift when compared with the vaccine strain.

Powers 1995
Power is an RCT conducted at the university of St. Louis, USA, on 127 healthy volunteers aged 18 to 45 years old. Participants were randomly assigned to receive one of the following five vaccine preparations in a double-blinded manner: 15 mg of rHA0, 15 mg of rHA0 plus alum, 90 mg of rHA0, licensed subvirion trivalent AB vaccine and placebo (saline). The recombinant HA vaccine contained full-length uncleaved haemagglutinin (HA0) glycoprotein from the influenza A/Beijing/32/92 (H3N2) virus. The licensed vaccine contained 15 mg/dose of each the HAs from influenza A/Texas/36/91 (H1N1), A/Beijing/32/92 (H3N2) and B/Panama/45/90 viruses. All the vaccines were administered intramuscularly. Data were extracted and loaded as follows:
Powers 1995a: recombinant HA0 (n = 77) against placebo (n = 12)
Powers 1995b: licensed trivalent (n = 26) against placebo (n = 12)
Outcomes were clinical and laboratory confirmed cases and adverse effects. An "influenza like illness" was defined as the presence of any respiratory symptom(s) for greater than or equal to two days, accompanied by fever or systemic symptoms of myalgias or chills. Laboratory evidence of influenza A (H3N2) virus infection was defined as either or both of the isolation of virus from nasopharyngeal secretion and a greater than or equal to four-fold increase in serum HAI antibody titre between the three week post-vaccination (pre-season) specimen and the corresponding post-season specimen collected in the following spring. Adverse effects were defined as follows: local adverse effects: erythema, pain, tenderness, induration, arm stiffness. Systemic adverse effects: headache, generalized myalgia, diarrhoea, nausea, feverishness, temperature > 37.8°C. The protective efficacy of the influenza vaccines was evaluated during the 1993 to 1994 influenza season. Vaccinations were done during the last November and the first December. The study ended in the following spring, at least two to three weeks after influenza viruses were no longer circulating in the local communities. The circulating strain was an A (H3N2). The USA recommended formula for that year was: A/Texas/36/91, A/Beijing/32/92, B/Panama/45/90. Vaccines matched the recommended strains.

Eddy 1970
Eddy 1970 is a CCT carried out in South Africa among 1,758 healthy male African employees aged 18 to 65 and housed in a large factory in the Western Cape. Of these, 91 failed to appear for vaccination and were thus excluded from the experiment. 75% of the participants (1,254) were given the monovalent inactivated A2/Aichi/2/68 (Hong Kong variant) influenza vaccine (administered subcutaneously) and 25% (413) received a placebo (sterile water). The control participants were selected by drawing a one in four systematic sample from a ranked list of the personnel numbers. The word "double blinding" was not used, but the control group received an injection of "dummy vaccine" consisting in sterile water. Outcomes were clinical cases of "influenza" (no better defined) and inpatient days. The first clinical case of influenza appeared on May, 21 1969, and the last six weeks later. All ill persons were admitted to hospital, no definition of influenza-like illness was given. The outbreak was due to an A2/Hong Kong/68 virus. The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68. The vaccine strain was considered to be recommended and matching the circulating strain.

Hoskins 1973
Hoskins 1973 is a quasi-randomised, double blind, not placebo controlled trial, conducted in a UK boarding school on 800 boys 11 to nine years old. It compares a monovalent inactivated influenza A vaccine containing the A/Hong Kong/68 strain against a monovalent inactivated influenza B vaccine. The boys were allocated to the two arms of the study according to date of birth. The vaccines were labelled according to a code. The study lasted three years, from 1970 to 1972: boys were re-vaccinated staying in the same vaccine group, and new entrants were allocated as before. The viral strain used in the vaccines varied during the period as follow:
- winter 1970 to 1971: A2/Hong Kong/X31/68 (H3N2) 200 IU and B/England/5/66 200 IU
- winter 1971 to 1972: A2/Hong Kong/X31/68 (H3N2) 200 IU and B/England/5/66 200 IU
- winter 1972 to 1973: A2/Hong Kong/1/68X (H3N2) 200 IU and B/Victoria/98926/70 200 IU
Each vaccine was given subcutaneously. Outcomes were clinical and laboratory confirmed cases and adverse effects. Clinical cases were defined as "febrile respiratory infection". Cases were laboratory confirmed when an influenza virus was isolated from throat swabs or a four-fold or greater rise in antibody titre occurred between acute and convalescent sera. Reactions to vaccination had insufficient definition to be extracted. The protective efficacy of the influenza vaccines was evaluated during an outbreak of influenza due to A/England/42/72, a strain which had undergone a recent antigenic shift.
Group receiving influenza A vaccine counted 384 participants.
Group receiving influenza B vaccine counted 340 participants.
The epidemic period lasted 11 days, from December 5 to December 15, 1972. Authors discuss protection conferred by natural antibody and repeated inoculations. The recommended formula for those years in the USA was A2/Aichi/2/68, B/Massachusetts/1/72. The vaccine was considered to match one of the recommended strains.

Mogabgab 1970
Mogabgab is a randomised trial, without mention of blinding, carried out in a USA Air Force Base, on 3000 airmen, 18 to 21 years old, previously unvaccinated. A monovalent inactivated influenza A vaccine and a polyvalent influenza A and B vaccine (the 1967 military formula) were compared against placebo (saline treated with formaldehyde solution). All vaccines were administered subcutaneously. Data were extracted and loaded as follows:
- Mogabgab 1970a: monovalent A2/Aichi 2/68 300 CCA (n = 881) against placebo (n = 521).
- Mogabgab 1970b: polyvalent A/Swine/33 100 CCA, A/PR8/34 100 CCA, A1/AA/1/57 100 CCA, A2/Taiwan 1/64 400 CCA, B/Lee/40 100 CCA, B/Mass 3/66 200 CCA (n = 1030) against placebo (n = 521).
Outcomes were clinical and laboratory confirmed cases, complications and admissions. Adverse effects are not considered. All respiratory illnesses were classified as febrile (38.3°C or greater), afebrile, pharyngitis, bronchitis or pneumonia (complication). Laboratory confirmation was obtained (by culture or four fold antibody titre increase in acute - convalescent sera) on 20 men randomly selected each week among the ill. The protective efficacy of the influenza vaccines was evaluated during the 1968 to 1969 influenza outbreak, lasting from December 9 to February 3. Virus isolated from patients was an A2/Hong Kong and matched the monovalent but not the polyvalent vaccine strain. The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68. The recommended polyvalent formula for that year in the USA was: A/PR/8/34, A1/AA/1/57, A2/Japan/170/62, A2/Taiwan/1/64, B/Mass/3/66. Viruses used in the vaccines matched the recommended strains.

Monto 1982
Monto 1982 is a randomised, single blind, placebo controlled trial conducted in USA on 306 volunteer university students to compare a monovalent, live attenuated, intranasal influenza B vaccine against placebo (vaccine diluent). Each subject was given a serial number that had previously been assigned randomly by a code to either the vaccine or the placebo group. The vaccine virus, cold recombinant, was produced by recombining the attenuated B/Ann Arbor/1/66 with a wild strain B/Hong Kong/8/73. There were 154 participants in the vaccine arm, and 152 in the placebo arm. The numbers of participants who completed the study were 144 and 140 respectively. Outcomes were clinical and laboratory confirmed cases and adverse effects. Patients suffered a respiratory illness if they had at least two respiratory symptoms. Cases were laboratory confirmed if they had an increase in antibody titre against three influenza B virus antigens, i.e. if there was a four-fold increase from an initial sample. Adverse effects were defined as sore throat, coryza, hoarseness, cough, muscle aches, temperature >100 F occurring during the first three days after vaccination. The protective efficacy of the influenza vaccine was evaluated during the 1979 to 1980 outbreak. The epidemic period was defined by first and last isolation (February 11 to march 18). Circulating virus strains were B/Singapore/79-like and B/Buenos Aires/79-like, antigenically different from that in the vaccine. The WHO recommended formula for that year was: A/USSR/90/77 or A/Brazil/11/78, A/Texas/1/77, B/Hong Kong/5/72, and did not match the strains used to produce the recombinant virus.

Maynard 1968
Mainard is a randomised, not placebo controlled trial without mention of blinding, conducted in Alaska among 502 volunteer students 14 to 18 years old, not previously immunised. Four hundred and eighty eight completed the study. It compares a monovalent inactivated influenza A and a monovalent inactivated influenza B vaccine against the standard commercial polyvalent AB formulation. All the vaccines were administered subcutaneously. Study arms were the following:
- monovalent A2/Taiwan/1/64 400 CCA - 171 participants
- monovalent B/Maryland/1/59 400 CCA - 158 participants
- polyvalent A/PR8 100 CCA, A1/Ann Arbor/1/57 100 CCA, A2/Jap/170/62 100 CCA, A2/Taiwan/1/64 100 CCA, B/Maryland/1/59 200 CCA, 159 participants.
Outcomes were clinical and laboratory confirmed cases. Adverse effects are not considered. A case was defined "influenza-like" if he had three or more symptoms of the following: fever, cough, headache, myalgia and sore throat. Blood for serological studies was obtained in a one-third random sample from each of the three vaccine groups. A case was laboratory confirmed if a four-fold rise in antibody titre between post-vaccination and post-epidemic sera was seen. Throat swabs and acute convalescent sera were collected from a small number of ill individuals to confirm influenza virus circulating. Attack rates by vaccine groups were reported only for clinical illness during the whole epidemic period and during selected period of time (three days) representing the two peaks of A and B influenza illness. The protective efficacy of the influenza vaccines was evaluated during an outbreak due simultaneously to influenza A and influenza B virus strains that peaked in February and March 1966 respectively. The epidemic period started on February 17 and ended on March 18: it was defined according to increase in absenteeism. Circulating strains were B/Alaska/1/66 and an A2 virus; both had only minor antigenic variation from B/Maryland and A2/Taiwan contained in the vaccines. The recommended formula for that year in the USA was: A/PR8, A1/Ann Arbor/1/57, A2/Jap/170/62, A2/Taiwan/1/64, B/Maryland/1/59 and completely matched the polyvalent vaccine.

Hobson 1970
Hobson 1970 is an RCT carried out in the UK on 1601 volunteers, industrial workers aged 16 to 64. The trial compared a monovalent inactivated vaccine of the Hong Kong strain of influenza A virus against a polyvalent vaccine containing only pre-1968 Asian viruses.
Volunteers were allocated to one of the three groups by random numbers. A double-blind procedure, ensured by coded preparation, was maintained throughout the trial. In two groups the vaccines were administered intramuscularly and a placebo (saline) was administered in nose. The third group received both the vaccines: the monovalent in nose and the polyvalent intramuscularly. Study arms were the following:
- polyvalent (A2/Eng/12/64 3000 HA units, A2/Eng/76/66 6000 HA units, B/Eng/5/66 3000 HA units, B/Swiss/265/67 3000 HA units) I.M. and saline in nose - 545 participants;
- polyvalent I.M. and monovalent in nose - 491 participants;
- monovalent (A2/Eng/344/68 7000 HA units) I.M. and saline in nose - 508 participants.
Outcomes were clinical cases and WDL. Clinical cases were defined as "respiratory illness" or "clinical influenza" according to the clinical syndromes characterised by Stuart-Harris (1965).
Vaccine efficacy was evaluated during the whole 1968 to 1969 influenza season (23 weeks) and in a smaller fraction of the epidemic period representing the peak of incidence of respiratory illness. The epidemic period lasted 10 weeks. The viral strain isolated from participants was antigenically identical to A2/Eng/344/68. The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68 (a Hong Kong virus). The monovalent A vaccine was considered to be recommended and matching the circulating strain.

Waldman 1969
Waldman is an RCT conducted in USA on 2,100 school teachers. The volunteers were randomly divided into nine groups. Vaccine and placebo were administered in a double blind manner on two occasions three weeks apart. The study was completed by 99% of the volunteers. The trial compared two inactivated vaccines (monovalent A and the commercial polyvalent AB), administered according to different schedule (one or two doses) and route of administration (intramuscular and aerosol), against placebo (saline). Data were extracted and loaded as follows:
Waldman 1969a: monovalent (A/Hong Kong/68) intramuscularly administered (n = 465) against placebo (n = 59);
Waldman 1969b: polyvalent (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA, B/Massachusetts/3/66 300 CCA) intramuscularly administered (n = 471) against placebo (n = 59);
Waldman 1969c: monovalent A/Hong Kong/68 aerosol administered (n = 479) against placebo (n = 59);
Waldman 1969d: polyvalent (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA, B/Massachusetts/3/66 300 CCA) aerosol administered (n = 471) against placebo (n = 59);
Outcomes were clinical cases and side effects. Clinical case definition was based on the presence of a "temperature > 100°F or a feverish feeling plus any two of the following symptoms: sore throat, muscle or joint pain, cough, stuffy or runny nose". Data concerning adverse effects were only partially reported by graph. The protective efficacy of the influenza vaccines was evaluated during the 1968 to 1969 influenza season. The follow up period lasted 10 weeks and comprised the epidemic period. The outbreak was caused by an A2/Hong Kong/68 influenza virus, that matched the monovalent but not the polyvalent vaccine. The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68. The recommended polyvalent formula for that year in the USA was: A/PR/8/34, A1/AA/1/57, A2/Japan/170/62, A2/Taiwan/1/64, B/Mass/3/66. Viruses used in the vaccines matched the recommended strains.

Williams 1973
Williams reports the results of three double-blind randomised field trials carried out in different Canadian military bases, on healthy volunteer military recruits. Volunteers' details were entered, in order of attendance, on vaccination registers that indicated by code letter the vaccine to be given. The allocation of vaccine code was at random. The vials were coded in such a manner that the vaccinator did not know which type of vaccine was being used. All the vaccines were inoculated subcutaneously. The studies were all conducted during the 1968 to 1969 influenza season. Data were extracted as follows:
Williams 1973b: a monovalent aqueous killed influenza A vaccine (A/Hong Kong/68 400 CCA) was compared against a monovalent aqueous killed influenza B vaccine (B/Mass766 400 CCA); participants in the two arms were 2,214 and 2,231 respectively. In bases that took part in the trial no influenza outbreak occurred, but influenza-like illnesses had been noted during a five weeks period (January 1969). The follow up period lasted 18 weeks. Occurrence of respiratory disease was reported for both the follow up and the epidemic period. The study was excluded from the review because it was not placebo controlled, both the vaccines contained a recommended strain and circulating virus was not identified.
Williams 1973: the trial compared three "standard" Institute of Microbiology and Hygiene (IMH) aqueous killed vaccines: a monovalent influenza A vaccine (A2/HK/68), a monovalent influenza B vaccine (B/Mass/66) and a bivalent one (A2/Mtl/68, B/Mass/66). Participants in the three arms were 1947, 1955 and 1961 respectively. In one of the bases that took part in the trial an influenza outbreak occurred during the last week of December 1968 and January 1969. Three viral strains, identified as A2/Hong Kong, were isolated from ill men and were considered to match the monovalent A vaccine.
The follow up period lasted 16 weeks. Occurrence of respiratory disease was reported for both the follow up and the epidemic period. Only the cases occurring during the epidemic period were considered. Only comparison between monovalent A and B vaccines was taken into account, owing the uncertain matching between bivalent vaccine and circulating viruses.
Williams 1973c: the trial compared three "standard" IMH aqueous killed vaccines: a monovalent influenza A vaccine (A2/HK/68), a monovalent influenza B vaccine (B/Mass/66) and a bivalent one (A2/Mtl/68, B/Mass/66). Participants in the three arms were 987, 988 and 996 respectively. In bases that took part in the trial no influenza outbreak occurred. Occurrence of respiratory disease was reported for the follow up period only (13 weeks). The study was excluded from the review because was not placebo controlled, all the vaccines contained a recommended strain and circulating virus was not identified.
Outcomes were clinical cases and side effects. Clinical cases were classified as "respiratory illness" or "non-respiratory illness", without further specification. Data relating to adverse effects had unclear definitions and was not included.

Tannock 1984
Tannock 1984 is a CCT double blind, placebo controlled trial, carried out in Australia in 88 volunteer staff, aged 16 to 64, from Newcastle Hospital and the Commonwealth Steel Corporation. Volunteers were allocated by alternation to receive one or two doses of a trivalent subunit vaccine (A/Brazil/11/78, A/Bangkok/1/79, B/Singapore/222/79 7 mg each) or placebo (sterile saline). Vaccine or placebo was administered by the subcutaneous route. Initially 46 volunteers were assigned to each treatment, but only individuals who received the expected doses of vaccine were considered as part of the study (29, 27 and 32 for each group described above). Complete follow up data were obtained for 18, 19, and 20 participants respectively. Outcomes laboratory confirmed cases and adverse effects. A case was laboratory confirmed if a respiratory illness, retrospectively reported, was associated with a four fold antibody titre increase between postvaccination and post-epidemic sera. Reactions to vaccination were classified as follows: local reaction: redness, swelling, warmth or irritation; local pain: pain on contact, pain with pressure, continuous pain, or restriction of arm movement; systemic reaction: fever, chills, sweating, drowsiness or insomnia. The follow up period lasted 20 weeks, during 1981 winter, and comprised the epidemic period. Viral strain isolated from patients was A/Bangkok/1/79. The recommended formula for that year was: A/Brazil/11/78, A/Bangkok/1/79, B/Singapore/222/79.

Sumarokow 1971
Sumarokow 1971 is a field trial conducted in Russia among 76,150 persons.
A first study group counted 36,665 children three to 12 years old. A second study group counted 3,945 participants aged 13 to 25. Only data regarding the latter were extracted and loaded. The trial compared two live vaccines against placebo. Vaccinations were carried out using coded preparation. Strains used in the vaccines were not reported. Study arms were as follows:
- allantoic intranasal vaccine; n = 9,945
- placebo; n = 9,942
- tissue vaccine for oral administration; n = 9,817
- placebo; n = 9,781
Outcomes were clinical and laboratory confirmed cases, deaths, severity of illness. Clinical outcomes were all the acute respiratory infections. Laboratory confirmation was obtained on a sample of ill participants by virus isolation or demonstration of seroconversion.
Bronchitis, otitis and pneumonia were considered as complications. Outcomes were detected during the entire period of influenza outbreak.
The epidemic period was defined as the period of highest influenza morbidity and lasted 11 weeks from the last ten days of January to the first ten days of April. Virus circulating was A2/Hong Kong/68, not matching the vaccine strains.

Nichol 1999
Nichol 1999 is an RCT conducted in the USA on healthy working adults aged 18 to 64. Participants were randomized 2:1 according to a predetermined computer-generated randomisation schedule. Double blinding was ensured by using pre-labelled, coded identical looking vials. The placebo was indistinguishable in appearance and smell. A trivalent, live attenuated influenza A and B vaccine was compared against a placebo (e.g. allantoic fluid). Vaccine strains were A/Shenzhen/227/95, A/Wuhan/395/95, B/Harbin/7/94-like. A total of 4,561 participants were enrolled: 3,041 vaccine recipients and 1,520 placebo recipients. Follow up interviews to ascertain side effects were available for 98.2% of vaccine recipients and 98.0% of placebo recipients. Complete follow up data were obtained for 2874 participants in the treatment arm and for 1,433 participants in the placebo arm. Outcomes were clinical cases symptoms defined, working days lost (WDL) and adverse effects. Case definition had three specification: febrile illness (fever for at least one day and two or more symptoms for at least two days: fever, chills, headache, cough, runny nose, sore throat, muscle aches, tiredness); severe febrile illness (three days of symptoms and one day of fever); febrile upper respiratory tract illness URI (three days of upper respiratory tract symptoms and one day of fever). For analysis we chose the URI definition. The outcome working days lost is presented as rate ratio, even if data are presented in a way that allows calculations on difference in mean days lost but not calculations on the standard error. Local adverse effects were defined as runny nose, sore throat and cough. Systemic adverse effects were defined as headache, muscle aches, chills, tiredness and fever. The trial was conducted during the 1997 to 1998 influenza season. The follow up period was defined as the whole epidemic period (December 14, 1997 through to March 21, 1998).
The viral strains were antigenically equivalent to those recommended for 1997 to 1998 season, but did not match the circulating strain.

Bridges 2000
Bridges is an RCT conducted in USA on healthy factory employees aged 18 to 64. Volunteers were randomly allocated to receive vaccine or placebo using a table of random numbers. Groups were evaluated in a double-blind fashion. The study compares a trivalent, inactivated, intramuscularly administered vaccine against placebo. The protective efficacy of the influenza vaccine was evaluated during the 1997 to 1998 outbreak (Bridges 2000a) and the 1998 to 1999 outbreak (Bridges 2000b). For the fall of 1997 vaccine influenza strains were: A/Johannesburg/82/96, A/Nanchang/933/95 and B/Harbin/7/94. For the fall of 1998 vaccine influenza strains were: A/Beijing/262/95, A/Sydney/5/97 and B/Harbin/7/94. Placebo was sterile saline for injection. Outcome were: clinical and laboratory confirmed cases, working days lost (WDL), admissions and adverse effects. Treatment and placebo arm counted, respectively, 595 and 589 participants in the Bridges 2000a and 587 and 604 in Bridges 2000b. Outcomes were detected during the whole follow up period, from November to March, in each study year. Illnesses was defined as influenza-like illness (fever = 37,7 °C with cough or sore throat) or upper respiratory illness (cough with sore throat or fever = 37,7 °C). For analysis we chose ILI definition. Laboratory confirmation was based on four fold or greater rise in antibody titre occurred between the three week post-injection and end-of-season serum sample. Local adverse effects were arm soreness and redness. Systemic adverse effects were fever, sore throat, coryza, myalgia, headache and fatigue, but authors reported no data. Viral strains isolated from cases were A/Sidney/5/97-like in 1997 to 1998 winter (antigenically distinct from vaccine components), and A/Sidney/5/97-like and B/Beijing/184/93-like in 1998 to 1999 winter (both similar to vaccine components).
Bridges 2000a was included among studies comparing recommended vaccines against placebo or other vaccines
Bridges 2000b was included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.

Mesa Duque 2001
Mesa Duque 2001 is an RCT conducted in Columbia on healthy volunteers, bank employees aged 18 to 60. The study compared a sub-unit inactivated, intramuscularly administered vaccine against placebo (vitamin C). Volunteers were randomly allocated to receive vaccine or placebo using a table of random numbers. Groups were evaluated in a double-blind fashion, ensured by pre-labelled, coded identical looking vials. The protective efficacy of the influenza vaccine was evaluated during the 1997 season. Vaccine influenza strains were: A/Wahan/359/95, A/Texas/36/91 and B/Beijing/184/93, the same strains recommended by WHO for the 1997 to 1998 season. Outcomes were: clinical cases, working days lost (WDL), and adverse effects. Treatment arm counted 247 participants and placebo arm 246 participants. An intention to treat analysis was performed. Illness was defined as upper respiratory illness (fever, sore throat and cough lasting more than 24 hours) according to ICD IX codes 381, 382, 460, 466, 480 and from 487 to 490. Clinical cases were detected from March, 15 to August, 31; WDL were detected all the year round. Local adverse effects were oedema, erythema, pain, swelling. Systemic adverse effects were fever, headache and indisposition. Circulating strains were not isolated from local cases but by WHO and Columbia surveillance system, and matched vaccine components. Mesa Duque (Mesa Duque 2001) was included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.

Mixéu 2002
Mixéu 2002 is a double-blind, randomised, placebo controlled trial conducted in Brazil on flight crews of an airline company. Volunteers were aged 18 to 64. The study compared a split trivalent, intramuscularly administered vaccine against placebo (containing the vaccine diluent only). Authors did not describe the methods used to ensure randomisation and blinding. The protective efficacy of the influenza vaccine was evaluated during the 1997 to 1998 influenza season. Vaccine influenza strains were: A/Nanchang/933/95, A/Texas/36/91 and B/Harbin/7/94, according to WHO recommendations for the 1997-98 season. Outcomes were: clinical cases, working days lost (WDL), and adverse effects.
Treatment arm counted 405 participants and placebo arm 408 participants. 294 and 299 participants respectively were considered protocol compliant and were included in the outcome analysis. Illness was defined as flu like illness (fever > 37.6°C and cough, headache, myalgia, rhinorrhoea and sore throat lasting at least 24 hours). Authors counted the episodes of illness regardless of the number of episodes in each subject. Local adverse effects considered were pain, induration, redness and swelling. Systemic adverse effects were fever, headache, insomnia and myalgia. Local and systemic effects were reported together and therefore not included in the review. Follow up was monthly and participants who returned three or more follow up diaries were included in the analysis. Circulating strains matched vaccine components.
Mixéu 2002 was included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.

Wilde 1999
Wilde is a RCT conducted in USA on health care professionals without chronic medical problems and younger than 50 years. Volunteers were randomly allocated to receive a trivalent, inactivated, intramuscularly administered vaccine or a control (meningococcal vaccine, pneumococcal vaccine or placebo) using a four-units block randomisation. Groups were evaluated in a double-blind fashion. The protective efficacy of the influenza vaccine was evaluated during the 1992 to 1993 outbreak (Wilde 1999a), the 1993 to 1994 outbreak (Wilde 1999b) and the 1994 to 1995 outbreak (Wilde 1999c). For the fall of 1992 vaccine influenza strains were: A/Texas/36/91, A/Beijing/353/89 and B/Panama/45/90 and control was meningococcal vaccine; for the fall of 1993 vaccine influenza strains were: A/Texas/36/91, A/Beijing/32/92 and B/Panama/45/90 and control was pneumococcal vaccine; for the fall of 1994 vaccine influenza strains were: A/Texas/36/91, A/Shangdong/9/32 and B/Panama/45/90 and control was sterile saline for injection. Outcomes were: laboratory confirmed cases and working days lost (WDL). Treatment and placebo arm counted, respectively, 52 and 50 participants in Wilde 1999a, 51 and 52 participants in Wilde 1999b, 77 and 77 participants in Wilde 1999c. Outcomes were detected during the whole follow up period, from October to November to March to April, in each study year. Laboratory confirmation was based on four-fold or greater rise in antibody titre occurred between the four week post-injection and end-of-season serum sample. Viral strains isolated from cases were A(H3N2) and B strains: a good match was achieved between the epidemic influenza subtypes and the vaccine components during 1993 to 1994 influenza season; there was a partial match during 1992 to 1993 and 1994 to 1995 influenza season. Wilde 1999c was included among studies comparing influenza vaccine against placebo. All the three trials were included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.

Methodological quality

Two reviewers (VD and TOJ) independently assessed allocation method, allocation concealment, blinding and completeness of follow-up.
There were 25 reports of studies in all, that were further split into 47 trials, 34 of these were placebo controlled. Nine of them used an influenza B vaccine in the control arm, considering it as a placebo. Four studies compared two or more influenza vaccines and did not consider a control arm. Sixteen studies reported data on adverse effects, but only ten were included in analysis: one did not have sufficient reporting and five studies did not have a placebo arm.
The overall quality of the studies was good. Among the 25 papers assessed, allocation concealment was adequate in 16, inadequate in seven and unclear in two. Twenty studies were properly randomised, four stated that the allocation method was quasi random and one did not report information about randomisation. Assessment was double blinded in 19 studies. Two studies were single blinded and four did not mention blinding. Two studies were field trials.

Results

Effect of vaccination on clinical cases of influenza
Trial data for the two definitions of influenza (no case definition and specific case definition) are presented separately for each of the three types of vaccine: live aerosol (including cold adapted), inactivated parenteral and inactivated aerosol. Significant heterogeneity was detected between trial results for most comparisons, and the figures quoted are average treatment effects estimated from random effects models.

Live aerosol vaccines were not effective in the studies reporting no cases definition and showed a poor level of efficacy (15% reduction, 95% CI 5 to 24%) in the studies using specific case definition. A combined analysis of data from the three studies estimated the total vaccine efficacy to be 5% (95% CI -1% to 10%).

Inactivated vaccines did offer significant protection. Taking the data from the eight trials presenting no case definition, the parenteral vaccine reduced the number of cases by 31% (95% CI: 5% to 51%). Considering the ten trials presenting specific case definition, the parenteral vaccine reduced the number of cases by 22% (95% CI: 9% to 33%). The efficacy of the inactivated aerosol vaccine was higher for the unspecified case definition (vaccine efficacy (VE) = 65%, 95% CI 32% to 82%) but not the specific influenza case definition (VE = 27%, 95% CI 0% to 47%).

Effect of vaccination on serologically confirmed cases of influenza
Data from two trials showed that aerosol live vaccines reduced the number of serologically confirmed cases of influenza by 79% (95% CI 44% to 92%). Nine trials provided data for inactivated parenteral vaccines, showing a similar efficacy of 67% (95% CI 51% to 78%). No trials of inactivated aerosol vaccine reported numbers of serological confirmed cases.

Effect of vaccination on other outcomes
Six trials evaluated time off work, estimating that vaccination saved on average around 0.12 working days. This result was not statistically significant. All these trials were of parenteral inactivated vaccine. Hospital admissions (evaluated in four trials) were also lower in vaccinated arms, but the difference was not statistically significant. There was little difference in complication rates between vaccinated and unvaccinated groups. Only one trial, assessing a live aerosol vaccine, considered death as an outcome and did not register any event.

Adverse effects - aerosol live vaccines
Significantly more recipients experienced sore throats and coryza after vaccine administration than placebo administration (sore throats relative rate = 1.84, 95% CI 1.27 to 2.67; coryza relative rate = 1.56; 95% CI 1.26 to 1.94). There was no significant increase in systemic side effects, although rates of fever and myalgia were higher in vaccine than placebo groups. Overall 43% of vaccine recipients reported our combined endpoint for local reactions, whilst only 8% reported the combined endpoint for systemic effects.

Adverse effects - inactivated vaccines
Local tenderness and soreness was more than twice as common among parenteral vaccine recipients than those in the placebo group (relative rate = 2.04; 95% CI 1.4 to 2.99). There were also increases in erythema (RR = 2.34; 95% CI 1.83 to 3.00), but not induration or arm stiffness. The combined local effects endpoint was significantly higher for those receiving the vaccine (RR = 3.56; 95% CI 3.14 to 4.03), with 59% reporting some effect.

None of the systemic effects were individually more common in parenteral vaccine recipients than in placebo recipients. However the combined endpoint was increased with 28% vaccine recipients reporting some side effects (RR = 1.18; 95% CI 0.97 to 1.43).

None of the trials on inactivated aerosol vaccines reported side effects that could be included in the analysis. The two trials which evaluated these vaccines included parenteral components using an influenza B control group so that the side-effects of the oral vaccine could not be estimated separately.

Recommended vaccines
Thirty eight trials evaluated the effect of the vaccines recommended (by WHO or single governments) on clinical cases of influenza. Twenty seven of these trials were placebo controlled, nine made comparisons with a group who received influenza B vaccine and two compared recommended influenza A vaccines with non-recommended influenza A and B vaccines.

Live aerosol, inactivated parenteral and inactivated aerosol all had similar clinical efficacies, although the estimate for inactivated aerosol vaccines was only based on the results of one study. When compared to placebo, live aerosols had a vaccine efficacy of 15% (95% CI 8% to 21%), inactivated parenteral had a vaccine efficacy of 25% (95% CI 13% to 35%), and inactivated aerosols had a vaccine efficacy of 40% (95% CI 13% to 59%). Combining the data from all three vaccine types from the placebo controlled trials, the overall estimate of vaccine efficacy was 22% (95% CI 14% to 30%). The estimate decreased to 21% (95% CI 15 to 27%) when the non-placebo controlled trials were included.

The individual trial results were more consistent when expressed as risk differences than relative effects. Overall the percentage of participants experiencing clinical influenza decreased by 6% (95% CI 4% to 8%) using data from the placebo controlled trials. The reductions were 3%, 7% and 9% for the live aerosol, inactivated parenteral and inactivated aerosol vaccines respectively, the first figure not being statistically significant.

There were significant reductions in serologically confirmed cases of influenza for live aerosol and inactivated parenteral preparations. There were no data for inactivated aerosol vaccines. Vaccine efficacy, in the placebo controlled comparisons, was estimated as 48% (95% CI 24% to 64%) for live aerosol vaccines, and 70% (95% CI 56% to 80%) for inactivated parenteral vaccines.

The number of working days lost was lower for participants treated with the recommended parenteral inactivated vaccines compared to placebo or non-influenza vaccines. The overall reduction was significant but of only 0.16 days (95% CI 0.04 to 0.29). In the placebo controlled trials the reduction is 0.12 days (95% CI 0.00 to 0.24).

Vaccine matching the circulating strain
The highest estimates of vaccine efficacy come from the analyses of vaccines that were shown to match the circulating vaccine strain. Twenty eight trials were included in these analyses, and 16 were placebo controlled. Since several studies had more than two arms, the efficacy of the vaccines containing the matching strain was compared against non-matching A or B influenza vaccines.

Estimate of the efficacy of parenteral inactivated vaccines in reducing cases of clinical influenza was 39% (95% CI 21% to 53%). Overall vaccine efficacy based on results of the placebo controlled trials was 33% (95% CI 20% to 44%). The estimate declined to 30% (95% CI 21% to 38%) when the non-placebo controlled trials were included. Expressing the efficacy as a risk difference, on average 9% (95% CI 5% to 12%) fewer participants who received matched vaccine suffered influenza like illnesses compared to placebo recipients.

The effect of the matched vaccine on serologically confirmed cases was also larger than in any other analysis. Overall the results of thirteen trials reporting serologically confirmed cases estimated the vaccine efficacy to be 75% (95% CI 62% to 84%).

Discussion

Although this review presents a large number of comparisons and outcomes based on a number of different groupings of studies and trials, the mainstream of the discussion was based on the results of the analysis of a WHO recommended vaccine against placebo. Parenterally administered influenza vaccines appear significantly better than their comparators and can reduce the incidence of serologically confirmed influenza by around 70%, if the WHO recommendations are adhered to. However, whilst the vaccines do prevent serologically confirmed cases, this is only one part of the spectrum of "clinical effectiveness" as they reduce total clinical influenza rates only by 25%.

It is not possible to give a definite indication on the practical use of live aerosol vaccines, because the assessment of their effectiveness is based on a limited number of studies presenting conflicting results. The effectiveness, according to WHO criteria, appears relatively low. Results regarding inactivated aerosol vaccine are based on the analysis of a few trials reporting only clinical outcomes not directly comparable, owing to non-homogeneous definitions. It does not seem wise to draw conclusions from these data.

Rates of complications caused by influenza in these trials were very low and analysis of the few trials which contained this outcome, did not reveal a significant reduction with the influenza vaccine. This result appears to contrast with the conclusions reported in the literature regarding elderly people, probably due to the general rarity of complications caused by respiratory infection in healthy adults. Hospitalisation was assessed in four trials and did not show a significant benefit from vaccination.

Working days lost in placebo recipient and vaccine recipient were significantly reduced in the vaccinated group, but by less than half a day on average.

Safety does not appear to be a particular problem with any of the vaccines, either from an incidence or severity point of view, although, given the low effectiveness of the aerosol vaccines, the effects classified as adverse may be caused by influenza. Either way, the public health safety profile of the vaccines is acceptable. The public health use of the vaccines assessed in this review needs to be carefully considered.

While the parenteral vaccine efficacy (its ability to avoid serologically confirmed cases) is around 70% for the WHO recommended strain, its impact on the global incidence of clinical cases of influenza is limited (around 25% and no more then 40% in best case scenario). The universal immunisation of healthy adults should achieve a number of specific goals: reducing the spread of the disease, reducing the economic loss due to working days lost and reducing morbidity and hospitalisation. None of the studies included in the review presented results evaluating the ability of this vaccination to interrupt the spread of the disease. Some studies presented data on reduction of working days lost and showed a very limited effect. Similarly a very limited effect was found on morbidity and no effect was found on hospitalisation. Given the limited availability of resources for mass immunisation, the use of influenza vaccines should be primarily directed to the prevention of complications in high-risk groups where it is likely to produce a more effective impact on public health. Overall the results of this review seem to discourage the utilisation of universal vaccination against influenza in healthy adults.

A number of problems should be taken into consideration when interpreting the results of this review:
- none of the live aerosol vaccines included in the review were registered
- methods of vaccine standardisation have changed significantly
- recent vaccines present significant differences in purity when compared with older ones
- different doses and schedules were pooled in the analysis

Authors' conclusions

Implications for practice

The results of this review seem to discourage the utilisation of vaccination against influenza in healthy adults as a public health measure.
As healthy adults have a low risk of complications due to respiratory disease, the use of the vaccine may be only advised as an individual protection measure in very specific cases.

Implications for research

The major differences in effect size between outcomes highlight the need for careful consideration of the best study design to assess the effects of public health measures such as vaccines. Additionally more work needs to be carried out on the combined use of vaccines and antivirals.

Potential conflict of interest

None

Acknowledgements

The authors gratefully acknowledge the help received from Drs Brian Hutchison, Alan Hampson, James Irlam, Andy Oxman and Barbara Treacy. The authors would like to thanks also the help received in updating the review by Carlo Di Pietrantonj, Alessandro Rivetti and Gabriella Morandi. While the original review was funded by the UK Ministry of Defence, the 2004 update was supported by the two Italian Local Health Authorities in which two of the reviewers are employed.

Characteristics of included studies

StudyBridges 2000a
MethodsRCT conducted in USA. Volunteers were randomly allocated to receive vaccine or placebo using a table of random numbers. Groups were evaluated in a double-blind fashion. A trivalent, inactivated, intramuscularly administered vaccine was compared against placebo. Placebo was sterile saline for injection.
ParticipantsHealthy factory employees 18-64 years old. Treatment and placebo arm counted, respectively, 595 and 589 participants. Laboratory confirmed cases were evaluated in 138 and 137 participants respectively.
InterventionsThe protective efficacy of the influenza vaccine was evaluated during the 1997-98 outbreak. Vaccine influenza strains were: A/Johannesburg/82/96, A/Nanchang/933/95 and B/Harbin/7/94.
OutcomesOutcome were: clinical and laboratory confirmed cases, working days lost (wdl), admissions and adverse effects. Outcomes were detected during the whole follow up period, from November to March, in each study year. Illness was defined as influenza-like illness (fever = 37,7 °C with cough or sore throat) or upper respiratory illness (cough with sore throat or fever = 37,7 °C). For analysis we chose ILI definition.
Laboratory confirmation was based on four-fold or greater rise in antibody titre occurred between the 3-week-post injection and end-of-season serum sample.
Local adverse effects were arm soreness and redness. Systemic adverse effects were fever, sore throat, coryza, myalgia, headache and fatigue, but authors didn't report data.
NotesViral strains isolated from cases were A/Sidney/5/97-like (antigenically distinct from vaccine component).
The study was included among studies comparing recommended vaccines against placebo or other vaccines.
Allocation concealmentD – Not used
StudyBridges 2000b
MethodsSee Buxton Bridges 1
ParticipantsHealthy factory employees 18-64 years old. Treatment and placebo arm counted, respectively, 587 and 604 participants. Laboratory confirmed cases were evaluated in 141 and 137 participants respectively.
InterventionsThe protective efficacy of the influenza vaccine was evaluated during the 1998-99 outbreak. Vaccine influenza strains were: A/Beijing/262/95, A/Sidney/5/97 and B/Harbin/7/94.
OutcomesSee Buxton Bridges 1
NotesViral strains isolated from cases were A/Sidney/5/97-like and B/Beijing/184/93-like (both similar to vaccine component).
The study was included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.
Allocation concealmentD – Not used
StudyEddy 1970
MethodsCCT carried out in South Africa
75% of the participants (1254) were given the monovalent inactivated and 25% (413) received a placebo. The control participants were selected by drawing a 1-in-4 systematic sample from a ranked list of the personnel numbers. The word "double blinding" was not used, but the control group received an injection of "dummy vaccine".
Participants1758 healthy male Bantu employees aged 18-65 and housed in a large factory in the Western Cape. Of these 91 failed to appear for vaccination and were thus excluded from the experiment.
InterventionsMonovalent inactivated: A2/Aichi/2/68 (Hong Kong variant) administered subcutaneously.
Placebo: sterile water.
OutcomesOutcomes were clinical cases of "influenza" (no better defined) and inpatient days.
The first clinical case of influenza appeared on May, 21 1969, and the last 6 weeks later. All ill persons were admitted to hospital, no definition of influenza-like illness was given.
NotesOutbreak was due to an A2/Hong Kong/68 virus.
The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68.
The vaccine strain was considered to be recommended and matching the circulating strain.
Allocation concealmentC – Inadequate
StudyEdmonson 1970
MethodsRandomised, double blind, not placebo controlled trial, conducted in Australia
The bottles of vaccine were numbered in a non consecutive random manner by a clerk, who remained the only person knowing the code. The coded vials were sequentially administered to enrolled participants.
Participants1983 healthy volunteers (1284 males and 699 females) from amongst Canberra public servants.
Each arm counted 975 and 878 participants for adverse effects and 933 and 841 participants for symptoms defined cases.
InterventionsBivalent vaccine (influenza A and B) against a monovalent B influenza vaccine.
The bivalent subunit vaccine contained A2/N.T./60/68 (8000 HA units) and B/Victoria/2/65 (3000 HA units) strains, while the monovalent B vaccine had a B/Victoria/2/65 (3000 HA units) strain.
Vaccines were administered by injection in two doses (the second 20 days after the first one).
OutcomesOutcomes are clinically defined cases, working days lost and adverse effects.
Sick leave records were considered in order to classify diagnoses into groups: influenza, other respiratory diseases, injuries and all other illnesses.
Local adverse effects were classified as minimal (when not affecting activities), mild (discomfort when the arm was used; redness, swelling but no absence from work) and severe (causing absence from work).
Systemic adverse effects were classified as minimal (no interference with normal activities),mild (interference with normal activities but no absence from work) and severe (absence from work).
NotesTrial was conducted during the winter of 1969.
Epidemic period length is unknown but probably the whole follow-up period comprises it. The circulating virus, isolated from patients, was A2/H.K./68. The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68.
The vaccine strain was considered to be recommended and matching the circulating strain.
Allocation concealmentA – Adequate
StudyEdwards 1994a
MethodsRCT conducted in USA. Participants were recruited from seven organisation and assigned to one of the study groups using a permuted block randomisation scheme that was stratified by treatment center and age group. The study was double blinded.
Participants and clinical staff were kept unaware of the assigned vaccine group trough the use of sealed randomisation envelopes that contained vaccine codes.
Two kinds of vaccine (a bivalent, live cold adapted, aerosol administered influenza A vaccine and the commercial inactivated intramuscularly administered influenza vaccine) were compared against placebo. Equal proportion of participants were included in the three vaccine arms.
The study lasted 5 years (from 1985 to 1989). Participants were immunised each fall staying in the same assigned vaccine group. Each year the viral strains used in the vaccines were different, but the strains used to develop cold adapted and inactivated vaccines were antigenically comparable. Cases were detected each year on the number of participants rolled up-to-date, independently of the number of injections received and the duration of participation in the 5-years study.
The epidemic period in any study year started on the day that the first influenza A virus isolate was obtained in Nashville and ended on the day that the last isolate was obtained.
ParticipantsA total of 5210 healthy children and adults of metropolitan Nashville aged 1-65 participated in the trial.
InterventionsData were extracted and loaded as follows:
Edwards 1: 1986-87 outbreak; live cold adapted against placebo; 872 treated, 439 placebo.
Cold adapted: Texas/1/85 H1N1 and Bethesda/1/85 H3N2 (107-107,6 pfu/ml); placebo: allantoic fluid.
OutcomesOutcomes were clinical and laboratory confirmed cases and adverse effects.
Patients had influenza like illness if they had fever or abrupt onset with at least one of the following: chills, headache, malaise, myalgia, cough, pharyngitis or other respiratory complaints, and presented for throat culture. They had a culture-positive illness if the culture revealed an influenza A virus. They had retrospectively reported illness if they stated at the time of the spring blood sampling that they had an influenza-like illness during the previous winter. Patients seroconverted if they had a fourfold rise in titre from post-vaccination to spring sera for the influenza A virus that circulated that season.
Side effects were those experienced by all the participants in the five-years study. Local adverse effects were redness, induration, tenderness. Systemic adverse effects were fever, sore throat, coryza, lethargy, chills, nausea, headache, muscle ache, cough.
NotesCirculating virus: Taiwan/1/86 (antigenic drift from vaccine strain). Epidemic period 56 days.
The WHO recommended vaccine for that year contained A/Chile/1/83, A/Singapore/6/86, A/Mississippi/1/85 and B/Ann Arbor/1/86.
The A strains administered in the study matched the recommended ones.
Allocation concealmentA – Adequate
StudyEdwards 1994b
MethodsSee Edwards 1
ParticipantsSee Edwards 1
InterventionsData were extracted and loaded as follows:
Edwards 2: 1986-87 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 878 treated, 439 placebo.
Inactivated vaccine formula: Chile/1/83 H1N1 and Mississippi/1/85 H3N2 (15 mg each strain). B strain contained in the vaccines was not described.
OutcomesSee Edwards 1
NotesSee Edwards 1
Allocation concealmentA – Adequate
StudyEdwards 1994c
MethodsSee Edwards 1
ParticipantsSee Edwards 1
InterventionsData were extracted and loaded as follows:
Edwards 3: 1987-88 outbreak; live cold adapted against placebo; 1029 treated, 532 placebo.
Cold adapted: Kawasaki/9/86 H1N1 and Bethesda/1/85 H3N2 (107-107,6 pfu/ml); placebo: allantoic fluid.
OutcomesSee Edwards 1
NotesCirculating viruses Sichuan/2/87 (H3N2) (antigen drift from vaccine strain) and B/Victoria/2/87; epidemic period 101 days.
The WHO recommended vaccine for that year contained A/Singapore/6/86, A/Leningrad/360/86 and B/Ann Arbor/1/86.
Only one of the A strains contained in the study vaccine was recommended.
Allocation concealmentA – Adequate
StudyEdwards 1994d
MethodsSee Edwards 1
ParticipantsSee Edwards 1
InterventionsData were extracted and loaded as follows:
Edwards 4: 1987-88 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 1060 treated, 532 placebo.
Inactivated vaccine formula: Taiwan/1/86 H1N1 and Leningrad/360/86 H3N2 (15 mg each strain). B strain contained in the vaccines was not described.
OutcomesSee Edwards 1
NotesSee Edwards 3
Allocation concealmentA – Adequate
StudyEdwards 1994e
MethodsSee Edwards 1
ParticipantsSee Edwards 1
InterventionsData were extracted and loaded as follows:
Edwards 5: 1988-89 outbreak; live cold adapted against placebo; 1114 treated, 562 placebo.
Cold adapted: Kawasaki/9/86 H1N1 and Los Angeles/2/87 H3N2 (107-107,6 pfu/ml); placebo: allantoic fluid.
OutcomesSee Edwards 1
NotesCirculating viruses Taiwan/1/86 (H1N1) and B/Yamagata/16/88; epidemic period 78 days.
The WHO recommended vaccine for that year contained A/Singapore/6/86, A/Sichuan/2/87 and B/Beijing/1/87.
Only one of the A strains contained in the study vaccine was recommended.
The vaccine was considered to be recommended and matching the circulating strain.
Allocation concealmentA – Adequate
StudyEdwards 1994f
MethodsSee Edwards 1
ParticipantsSee Edwards 1
InterventionsData were extracted and loaded as follows:
Edwards 6: 1988-89 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 1126 treated, 562 placebo;
Inactivated vaccine formula: Taiwan/1/86 H1N1 and Sichuan/2/87 H3N2 (15 mg each strain). B strain contained in the vaccines was not described.
OutcomesSee Edwards 1
NotesSee Edwards 5
Allocation concealmentA – Adequate
StudyEdwards 1994g
MethodsSee Edwards 1
ParticipantsSee Edwards 1
InterventionsData were extracted and loaded as follows:
Edwards 7: 1989-90 outbreak; live cold adapted against placebo; 999 treated, 508 placebo.
Cold adapted: Kawasaki/9/86 H1N1 and Los Angeles/2/87 H3N2 (107-107,6 pfu/ml); placebo: allantoic fluid.
OutcomesSee Edwards 1
NotesCirculating virus Shanghai/11/87 (H3N2); epidemic period 80 days.
The WHO recommended vaccine for that year contained A/Singapore/6/86, A/Shanghai/11/87 and B/Yamagata/16/88.
Only one of the A strains contained in the study vaccine was recommended.
The vaccine was considered to be recommended and matching the circulating strain.
Allocation concealmentA – Adequate
StudyEdwards 1994h
MethodsSee Edwards 1
ParticipantsSee Edwards 1
InterventionsData were extracted and loaded as follows:
Edwards 8: 1989-90 outbreak; trivalent inactivated commercial vaccine against inactivated influenza B vaccine; 1016 treated, 508 placebo
Inactivated vaccine formula: Taiwan/1/86 H1N1 and Shanghai/11/87 H3N2 (15 mg each strain). B strain contained in the vaccines was not described.
OutcomesSee Edwards 1
NotesSee Edwards 7
Allocation concealmentA – Adequate
StudyHammond 1978
MethodsCCT, double blinding, conducted in Australia.
Coded identical looking vials were sequentially administered to enrolled participants.
Participants225 volunteers, medical students or staff members of the Monash University
Arm size was: 116 subjects in the vaccine group and 109 subjects in the placebo group.
InterventionsA trivalent, subunit vaccine containing 200 IU of A/Victoria/3/75, 250 IU of A/Scotland/840/74 and 300 IU of B/Hong Kong/8/73 (as issued for general use), was compared against placebo (diphtheria and tetanus toxoids). Vaccines were administered subcutaneously.
OutcomesOutcomes were clinical and laboratory confirmed cases.
Participants were regarded as suffering from influenza when they suffered a respiratory illness which was associated with the isolation of influenza virus, a four-fold or greater rise in antibody titre occurring between post-vaccination and post-epidemic sera, or both.
Adverse effects are not considered.
NotesEpidemic influenza was defined by virus isolation and serology tests and lasted from mid April to mid August 1976. Outcomes were detected during the whole epidemic period. A/Vic/3/75 was isolated from patients.
The recommended formula for that year was: A/Victoria/3/75, A/Scotland/840/74 and of B/Hong Kong/8/73.
Allocation concealmentA – Adequate
StudyHobson 1970
MethodsRCT carried out in the UK.
The trial compared a monovalent inactivated vaccine of the Hong Kong strain of influenza A virus against a polyvalent vaccine containing only pre-1968 Asian viruses. Volunteers were allocated to one of the three groups by random numbers. A double-blind procedure, ensured by coded preparation, was maintained throughout the trial.
In two groups the vaccines were administered intramuscularly and a placebo (saline) was administered in nose. The third group received both the vaccines: the monovalent in nose and the polyvalent intramuscularly.
Participants1601 volunteers, industrial workers aged 16-64.
InterventionsData were extracted and loaded as follows:
Polyvalent intramuscular (n= 545) against monovalent intramuscular influenza A (n= 508)
polyvalent formula was: A2/Eng/12/64 3000 HA units, A2/Eng/76/66 6000 HA units, B/Eng/5/66 3000 HA units, B/Swiss/265/67 3000 HA units
monovalent was a Hong Kong virus: A2/Eng/344/68 7000 HA units
OutcomesOutcomes were clinical cases and working days lost.
Clinical cases were defined as "respiratory illness" or "clinical influenza" according to the clinical syndromes characterised by Stuart-Harris (1965).
NotesThe epidemic period lasted 10 weeks.
Vaccine efficacy was evaluated during the whole 1968-69 influenza season (23 weeks).
The viral strain isolated from participants was antigenically identical to A2/Eng/344/68.
The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68 (a Hong Kong virus).
The monovalent A vaccine was considered to be recommended and matching the circulating strain.
Allocation concealmentA – Adequate
StudyHoskins 1973
MethodsQuasi-randomised, double blind, not placebo controlled trial, conducted in a UK boarding school.
The boys were allocated to the two arms of the study according to date of birth. The vaccines were labelled according to a code.
Participants800 boys 11-19 year-old.
Group receiving influenza A vaccine counted 384 participants.
Group receiving influenza B vaccine counted 340 participants.
InterventionsIt compares a monovalent inactivated influenza A vaccine containing the A/Hong Kong/68 strain against a monovalent inactivated influenza B vaccine.
The study lasted three years, from 1970 to 1972: boys were re-vaccinated staying in the same vaccine group, and new entrants were allocated as before. The viral strain used in the vaccines varied during the period as follow:
- winter 1970-71: A2/Hong Kong/X31/68 (H3N2) 200 IU and B/England/5/66 200 IU
- winter 1971-72: A2/Hong Kong/X31/68 (H3N2) 200 IU and B/England/5/66 200 IU
- winter 1972-73: A2/Hong Kong/1/68X (H3N2) 200 IU and B/Victoria/98926/70 200 IU
Each vaccine was given subcutaneously.
OutcomesOutcomes were clinical and laboratory confirmed cases and adverse effects.
Clinical cases were defined as "febrile respiratory infection".
Cases were laboratory confirmed when an influenza virus was isolated from throat swabs or a four-fold or greater rise in antibody titre occurred between acute and convalescent sera.
Reactions to vaccination had insufficient definition to be extracted.
NotesThe protective efficacy of the influenza vaccines was evaluated during an outbreak of influenza due to A/England/42/72, a strain which had undergone a recent antigenic shift.
The epidemic period lasted 11 days, from December 5 to December 15, 1972.
The recommended formula for those years in the USA was A2/Aichi/2/68, B/Massachusetts/1/72.
The vaccine was considered to match one of the recommended strains. It was not totally clear wether less then 25% of the population was outside the age range for inclusion
Allocation concealmentA – Adequate
StudyKeitel 1988a
MethodsRCT
Volunteers were randomly allocated to receive vaccine or placebo using a table of random numbers according to prior vaccination experience. Three groups were formed: first vaccinated, multi vaccinated and placebo. Each arm counted, respectively, 162, 138 and 298 participants.
Groups were evaluated in a double-blind fashion.
Participants598 healthy volunteers, 30-60 years old, working in the Texas Medical Center in Houston, Texas, or in surrounding industrial companies.
InterventionsData were extracted and loaded as follows:
Keitel 1: 1983-84 outbreak; trivalent inactivated, killed whole, intramuscularly administered, commercial vaccine against placebo; 300 treated, 298 placebo
Each dose of the commercial vaccine contained 15 mg of hemagglutinin of each of the following influenza strains: A/Philippines/2/82 (H3N2), A/Brazil/11/78 (H1N1) and B/Singapore/222/79. Placebo was sterile saline for injection.
OutcomesClinical and laboratory confirmed cases.
Illnesses were classified as "any", "flu-like" (lower respiratory and/or systemic illness) and "febrile" (oral temperature of 37.8 or higher) according to findings at initial examination. Symptoms and signs of disease of the trachea, bronchi, and/or lung were designated lower respiratory illness. A designation of systemic illness was based on occurrence of myalgia, malaise, fatigue, lethargy, chills, sweats and anorexia.
Laboratory confirmation was based on culture and/or four-fold or greater rise in antibody titre occurring between post-vaccination (pre-epidemic), acute, convalescent and/or spring (post-epidemic) sera. The laboratory confirmed outcome is reported separately for influenza A or influenza B and for culture or serological positivity. Authors do not specify when culture and serological confirmation are obtained on the same subject.
Adverse effects are not considered.
Outcomes were detected during the whole epidemic period, defined as the interval during which community surveillance recovered influenza viruses from 10% or more of person with febrile respiratory illness per calendar week.
NotesThe epidemic period was from January 8 to March 17, 1984.
Viral strains isolated from cases were A/Victoria/7/83 (H1N1) and B/USSR/100/83.
The WHO recommended formula for that year was: A/Brazil/11/78, A/Philippines/2/82 and of B/Singapore/222/79.
Allocation concealmentB – Unclear
StudyKeitel 1988b
MethodsSee Keitel 1
ParticipantsSee Keitel 1
InterventionsData were extracted and loaded as follows:
Keitel 2: 1984-85 outbreak; trivalent inactivated, killed whole, intramuscularly administered, commercial vaccine against placebo; 456 treated, 241 placebo
Each dose of the commercial vaccine for the fall of 1984 contained 15 mg of hemagglutinin of each of the following influenza strains: A/Philippines/2/82 (H3N2), A/Chile/1/83 (H1N1) and B/USSR/100/83. Placebo was sterile saline for injection.
OutcomesSee Keitel 1
NotesThe epidemic period was from January 6 to March 9, 1985.
The viral strain isolated from cases was an A/Philippines/2/82 (H3N2)-like
The WHO recommended formula for that year was: A/Philippines/2/82 (H3N2), A/Chile/1/83 (H1N1) and B/USSR/100/83.
The vaccine was considered to be recommended and matching the circulating strain.
Allocation concealmentB – Unclear
StudyMair 1974a
MethodsCCT conducted in UK
Allocation was made by day of birth. Double blind.
It compares two monovalent, inactivated, intramuscularly administered influenza A vaccines against a monovalent inactivated influenza B vaccine used as placebo.
Participants507 volunteers patients from general practices. Complete follow up was obtained for 465 of them.
Three arms which counted, respectively, 141,169 and 155 participants.
InterventionsData were extracted and loaded as follows:
Mair 1: monovalent influenza A matching circulating strain (n = 85) against monovalent B (n = 155).
Vaccine strains were respectively: A/England/42/72 (H3N2) (recombinant XPR8, 600 IU), B/Victoria/98926/70 (400 IU).
OutcomesOutcomes are clinical and laboratory confirmed cases and adverse effects. Illness was defined as "any symptoms of influenza". Cases were laboratory confirmed when an influenza virus was isolated from throat swabs or a four-fold or greater rise in antibody titre occurred between acute and convalescent sera.
Local adverse effects were classified as mild (slight swelling, aching) and severe (painful, red and swollen arm).
Systemic adverse effects were classified as mild (slight pyrexia, mild aching of the back and limbs) and severe (influenza-like symptoms).
NotesTrial was conducted during the 1972-73 influenza season. The follow up period lasted 13 weeks. Authors do not specify the length of the epidemic period. Viral strain isolated from cases was A/England.
The recommended formula for that year in the USA was A2/Aichi/2/68, B/Massachusetts/1/72.
Allocation concealmentA – Adequate
StudyMair 1974b
MethodsSee Mair 1
ParticipantsSee Mair 1
InterventionsData were extracted and loaded as follows:
Mair 2: monovalent influenza A matching circulating strain (n=84) against monovalent A (n=141).
Vaccine strains were respectively: A/England/42/72 (H3N2) (recombinant XPR8, 600 IU), A/Hong Kong/1/68 (H3N2) (recombinant X31, 600 IU)
OutcomesSee Mair 1
NotesSee Mair 1
Allocation concealmentA – Adequate
StudyMair 1974c
MethodsSee Mair 1
ParticipantsSee Mair 1
InterventionsData were extracted and loaded as follows:
Mair 3: monovalent recommended influenza A strain (n=71) against monovalent B (n=155).
Vaccine strains were respectively: A/Hong Kong/1/68 (H3N2) (recombinant X31, 600 IU), B/Victoria/98926/70 (400 IU).
OutcomesSee Mair 1
NotesSee Mair 1
Allocation concealmentA – Adequate
StudyMair 1974e
MethodsSee Mair 1
ParticipantsSee Mair 1
InterventionsData were extracted and loaded as follows:
Mair 4: monovalent recommended influenza A strain (n=70) against monovalent A (n=169).
Vaccine strains were respectively: A/Hong Kong/1/68 (H3N2) (recombinant X31, 600 IU), A/England/42/72 (H3N2) (recombinant XPR8, 600 IU).
OutcomesSee Mair 1
NotesSee Mair 1
Allocation concealmentA – Adequate
StudyMaynard 1968
MethodsMainard is a randomised, not placebo controlled trial without mention of blinding, conducted in Alaska.
It compares a monovalent inactivated influenza A and a monovalent inactivated influenza B vaccine against the standard commercial polyvalent AB formulation. All the vaccines were administered subcutaneously.
Participants502 volunteer students 14-18 year-old, not previously immunised. 488 completed the study.
InterventionsData were extracted and loaded as follows:
Standard commercial polyvalent AB formulation (n=159) against monovalent B (n= 158)
polyvalent formula: A/PR8 100 CCA, A1/Ann Arbor/1/57 100 CCA, A2/Jap/170/62 100 CCA, A2/Taiwan/1/64 100 CCA, B/Maryland/1/59 200 CCA.
monovalent B: B/Maryland/1/59 400 CCA
OutcomesOutcomes were clinical and laboratory confirmed cases. Adverse effects are not considered.
A case was defined "influenza-like" if there were three or more symptoms of the following: fever, cough, headache, myalgia, sore throat.
Blood for serological studies was obtained in a one-third random sample from each of the three vaccine groups. A case was laboratory confirmed if a four-fold rise in antibody titre between post-vaccination and post-epidemic sera was seen.
Throat swabs and acute convalescent sera were collected from a small number of ill individuals to confirm influenza virus circulating.
Attack rates by vaccine groups were reported for clinical illness during the whole epidemic period.
NotesThe protective efficacy of the influenza vaccines was evaluated during an outbreak due simultaneously to influenza A and influenza B virus strains that peaked in February and March 1966 respectively. The epidemic period started on February 17 and ended on March 18: it was defined according to increase in absenteeism. Circulating strains were B/Alaska/1/66 and an A2 virus; both had only minor antigenic variation from B/Maryland and A2/Taiwan contained in the vaccines.
The recommended formula for that year in the USA was: A/PR8, A1/Ann Arbor/1/57, A2/Jap/170/62, A2/Taiwan/1/64, B/Maryland/1/59 and completely matched the polyvalent vaccine.
Allocation concealmentB – Unclear
StudyMesa Duque 2001
MethodsRCT conducted in Columbia. The study compared a sub-unit inactivated, intramuscularly administered vaccine against placebo (C vitamin). Volunteers were randomly allocated to receive vaccine or placebo using a table of random numbers. Groups were evaluated in a double-blind fashion, ensured by pre labeled, coded identical looking vials.
Participants593 healthy volunteers, bank employees 18-60 years old, 247 in vaccine arm and 246 in placebo arm. Drops out were respectively 15 in vaccine arm and 19 in placebo arm.
InterventionsThe protective efficacy of the influenza vaccine was evaluated during the 1997 season. Vaccine influenza strains were: A/Wahan/359/95, A/Texas/36/91 and B/Beijing/184/93, the same strains recommended by WHO for the 1997-98 season.
OutcomesOutcome were: clinical cases, working days lost (wdl), and adverse effects. Treatment arm counted 247 subjects and placebo arm 246 subjects. An intention to treat analysis was performed.
Illnesses was defined as upper respiratory illness (fever, sore throat and cough lasting more than 24 hours) according to 381, 382, 460, 466, 480 and from 487 to 490 ICD IX codes. Clinical cases were detected from March, 15 to August, 31; Wdl were detected all over the year.
Local adverse effects were edema, erythema, pain, swelling. Systemic adverse effects were fever, headache and indisposition.
NotesCirculating strains were not isolated from local cases but by WHO and Columbia surveillance system, and matched vaccine components.
The study was included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.
Allocation concealmentD – Not used
StudyMixéu 2002
MethodsMixéu is a double blind, randomised, placebo controlled trial conducted in Brazil on flight crews of an airline company. Volunteers were 18-64 years old. The study compared a split trivalent, intramuscularly administered vaccine against placebo (vaccine diluent). Authors did not describe the methods used to ensure randomisation and blindness.
ParticipantsTreatment arm counted 405 participants and placebo arm 408 participants. 294 and 299 participants respectively were considered protocol compliant and were included in the outcome analysis.
InterventionsThe protective efficacy of the influenza vaccine was evaluated during the 1997-98 influenza season. Vaccine influenza strains were: A/Nanchang/933/95, A/Texas/36/91 and B/Harbin/7/94, according to WHO recommendations for the 1997-98 season.
OutcomesOutcome were: clinical cases, working days lost (wdl), and adverse effects. Illness was defined as flu like illness (fever 37.°C and cough, headache, myalgia, rhinorrhoea, sore throat lasting at least 24 hours). Authors counted the episodes of illness, without respect for the number of episodes in each participant
Local adverse effects were pain, induration, redness and swelling. Systemic adverse effects were fever, headache, insomnia and myalgia. Adverse effects reports were cumulative for local and systemic effects and were not considered for the review.
NotesFollow up was monthly and participants who returned 3 or more follow up diary were included in the analysis.
Circulating strains matched vaccine components.
Mixéu 2002 was included among studies comparing recommended vaccines against placebo or other vaccines and among studies comparing vaccines matching the circulating virus against placebo or other vaccines.
Allocation concealmentB – Unclear
StudyMogabgab 1970a
MethodsRandomised trial, without mention of blinding, carried out in a USA Air Force Base
ParticipantsAirmen, 18-21 year-old, previously unvaccinated.
InterventionsData were extracted and loaded as follows:
Mogabgab 1: monovalent inactivated influenza A vaccine (A2/Aichi 2/68 300 CCA) against placebo (saline treated with formaldehyde solution). Arm sizes: monovalent A (n = 881); placebo (n = 521)
The vaccines were administered subcutaneously.
OutcomesOutcomes were clinical and laboratory confirmed illnesses, complications and admissions. Numbers of illnesses were counted, and not numbers of participants. Adverse effects are not considered.
All respiratory illnesses were classified as febrile (38.3°C or greater), afebrile, pharyngitis, bronchitis or pneumonia (complication).
Laboratory confirmation was obtained (by culture or 4-fold antibody titre increase in acute - convalescent sera) on 20 men randomly selected each week among the ill.
NotesThe protective efficacy of the influenza vaccines was evaluated during the 1968-69 influenza outbreak, lasting from December 9 to February 3.
Virus isolated from patients was an A2/Hong Kong and matched the vaccine strain.
The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68.
The vaccine strain was considered to be recommended and matching the circulating strain.
Allocation concealmentB – Unclear
StudyMogabgab 1970b
MethodsSee Mogabgab 1
ParticipantsSee Mogabgab 1
InterventionsData were extracted and loaded as follows:
Mogabgab 2: Polyvalent influenza A and B vaccine (the 1967 military formula A/Swine/33 100 CCA, A/PR8/34 100 CCA, A1/AA 1/57 100 CCA, A2/Taiwan 1/64 400 CCA, B/Lee/40 100 CCA, B/Mass 3/66 200 CCA) against placebo (saline treated with formaldehyde solution). Arm sizes: polyvalent (n=1030); placebo (n=521)
All the vaccines were administered subcutaneously.
OutcomesSee Mogabgab 1
NotesThe protective efficacy of the influenza vaccines was evaluated during the 1968-69 influenza outbreak, lasting from December 9 to February 3.
Virus isolated from patients was an A2/Hong Kong and did not match the vaccine strains.
The recommended polyvalent formula for that year in the USA was: A/PR/8/34, A1/AA/1/57, A2/Japan/170/62, A2/Taiwan/1/64, B/Mass/3/66
The vaccine strain was considered to be recommended.
Allocation concealmentB – Unclear
StudyMonto 1982
MethodsRandomised, single blind, placebo controlled trial conducted in USA.Each participant was given a serial number that had previously been assigned randomly by a code to either the vaccine or the placebo group.
Participants306 volunteer university students.
Vaccine arm counted 154 participants, placebo arm 152. Subjects completing the study were 144 and 140 respectively.
InterventionsMonovalent, live attenuated, intranasal influenza B vaccine against placebo (vaccine diluent).
The vaccine virus, cold recombinant, was produced by recombining the attenuated B/Ann Arbor/1/66 with a wild strain B/Hong Kong/8/73.
OutcomesOutcomes were clinical and laboratory confirmed cases and adverse effects.
Patients suffered a respiratory illness if they had at least 2 respiratory symptoms.
Cases were laboratory confirmed if they had an increase in antibody titre against 3 influenza B virus antigens, i.e. if there was a four-fold increase from an initial sample.
Adverse effects were defined as sore throat, coryza, hoarseness, cough, muscle aches, temperature >100 F occurred during the first three days after vaccination.
NotesThe protective efficacy of the influenza vaccine was evaluated during the 1979-80 outbreak.
The epidemic period was defined by first and last isolation (February 11 to March 18). Circulating virus strains were B/Singapore/79-like and B/Buenos Aires/79-like, antigenically different from that in the vaccine.
The WHO recommended formula for that year was: A/USSR/90/77 or A/Brazil/11/78, A/Texas/1/77, B/Hong Kong/5/72, and didn't match the strains used to produce the recombinant virus.
Allocation concealmentA – Adequate
StudyNichol 1995
MethodsRCT conducted in USA
Randomisation was performed according to a computer-generated randomisation schedule.
Double blinding was ensured by preloaded, coded identical looking syringes.
ParticipantsA total of 849 healthy full-time employed volunteers, 18-64 year-old were enrolled.
Follow up interviews to ascertain side effects were completed for 841 subjects. Complete follow up data were obtained for 416 of 422 placebo recipients and 409 of 419 vaccine recipients.
InterventionsA subvirion, trivalent, parenteral influenza A and B vaccine was compared against placebo (vaccine diluent). Vaccine strains were A/Texas/36/91, A/Shangdong/9/93, B/Panama/45/90.
OutcomesOutcomes were clinical cases symptoms defined, WDL, and adverse effects. Patients were defined as cases if they had at least one upper respiratory illness (a sore throat associated with either fever or cough that lasted at least 24 hours).
Local adverse effects were defined as arm soreness. Systemic adverse effects were defined as fever, tiredness, feeling under the weather, muscle ache, headache.
NotesTrial was conducted during the 1994-95 influenza season. The follow up period was defined as December 1, 1994 through March 31, 1995 (the influenza season).
The USA recommended formula for that year was: A/Texas/36/91, A/Shangdong/9/93, B/Panama/45/90.
Allocation concealmentA – Adequate
StudyNichol 1999
MethodsRCT conducted in USA. Participants were randomized 2:1 according to a predetermined computer-generated randomisation schedule. Double blinding was ensured by pre labeled, coded identical looking vials. The placebo was indistinguishable in appearance and smelling.
Participants4561 healthy working adults, 18-64 year-old.
Vaccine arm counted 3041 participants, placebo arm counted 1520 participants.
InterventionsA trivalent, live attenuated influenza A and B vaccine was compared against placebo (egg allantoic fluid). Vaccine strains were A/Shenzhen/227/95, A/Wuhan/395/95, B/Harbin/7/94-like.
Complete follow up data were obtained for 2874 participant in the treatment arm and for 1433 subject in the placebo arm.
OutcomesOutcomes were clinical cases symptoms defined, working days lost (wdl), and adverse effects. Case definition had three specification: febrile illness (fever for at least 1 day and two or more symptoms for at least 2 days: fever, chills, headache, cough, runny nose, sore throat, muscle aches, tiredness); severe febrile illness (3 days of symptoms and 1 day of fever); febrile upper respiratory tract illness (3 days of upper respiratory tract symptoms and 1 day of fever). For analysis we chose upper respiratory tract infection URI definition.
Local adverse effects were defined as runny nose, sore throat and cough. Systemic adverse effects were defined as headache, muscle aches, chills, tiredness and fever. The trial was conducted during the 1997-98 influenza season. The follow up period was defined as the whole epidemic period (December 14, 1997 through to March 21, 1998).
NotesThe viral strain were antigenetically equivalent to those recommended for 1997-98 season, but did not match the circulating strain.
Allocation concealmentD – Not used
StudyPowers 1995a
MethodsRCT conducted in USA.
127 participants were randomly assigned to receive one of five vaccine preparations in a double-blinded manner.
ParticipantsHealthy volunteers in university of St. Louis. 18-45 year-old.
InterventionsData were extracted and loaded as follows:
Powers 1: recombinant HA0 (n = 77) against placebo (n = 12)
The recombinant HA vaccine contained full-length uncleaved haemagglutinin (HA0) glycoprotein from the influenza A/Beijing/32/92 (H3N2) virus. Three vaccine preparations (15 mg of rHA0, 15 mg of rHA0 plus alume, 90 mg of rHA0).
Placebo was saline for injection.
All the vaccines were administered intramuscularly.
OutcomesOutcomes were clinical and laboratory confirmed cases and adverse effects.
An "influenza like illness" was defined as the presence of any respiratory symptom(s) for >=2 days, accompanied by fever or systemic symptoms of myalgias or chills.
Laboratory evidence of influenza A (H3N2) virus infection was defined as either or both of the isolation of virus from nasopharyngeal secretion and a >=four-fold increase in serum HAI antibody titre between the 3-week post-vaccination (preseason) specimen and the corresponding post-season specimen collected in the following spring.
Adverse effects were defined as follows:
local adverse effects: erythema, pain, tenderness, induration, arm stiffness;
systemic adverse effects: headache, generalized myalgia, diarrhoea, nausea, feverishness, temperature > 37.8°C.
NotesThe protective efficacy of the influenza vaccines was evaluated during the 1993-94 influenza season. Vaccinations were done during the last day of November and the first of December. The study ended in the following spring, at least 2-3 weeks after influenza viruses were no longer circulating in the local communities.
The circulating strain was an A (H3N2).
The USA recommended formula for that year was: A/Texas/36/91, A/Beijing/32/92, B/Panama/45/90.
Vaccine matched one of the recommended strains.
Allocation concealmentB – Unclear
StudyPowers 1995b
MethodsSee Powers 1
ParticipantsSee Powers 1
InterventionsData were extracted and loaded as follows:
Licensed trivalent (n=26) against placebo (n=12)
Licensed subvirion trivalent AB vaccine contained 15 mg/dose of each the HAs from influenza A/Texas/36/91 (H1N1), A/Beijing/32/92 (H3N2) and B/Panama/45/90 viruses
Placebo was saline for injection.
All the vaccines were administered intramuscularly.
OutcomesSee Powers 1
NotesSee Powers 1
Vaccine matched the recommended formula.
Allocation concealmentB – Unclear
StudyRytel 1977
MethodsRCT, single blind, conducted in USA
Study participants were randomly assigned to three subgroups.
Participants143 young adult female student nurse volunteers with low antibody titre against influenza A or influenza B virus.
Three subgroups received either two doses of the vaccine (n = 47), one dose of vaccine and one dose of placebo (n = 48) or two doses of placebo (n = 48) at 14 days apart.
InterventionsEfficacy of a single or double dose of a live attenuated, bivalent, intranasal influenza A (containing 10 exp7,2 EID50) and B (containing 10 exp7,8 EID50 ) vaccine was tested against placebo (5% sucrose). Vaccine virus strains were A/England/42/72 (H3N2) and B/Hong Kong/5/72.
OutcomesOutcomes were laboratory confirmed cases and adverse effects.
Laboratory confirmed case was defined as the presence of a flu-like illness (three or more symptoms of acute respiratory disease and temperature greater then 37.2 degrees) and virus isolation and/or four fold rise in antibody titre in sera obtained at 30 days and 6 months following immunisation.
Local adverse effects were URI symptoms and cough and were subdivided into moderate and severe. A definition of general adverse effects (again distinguished among moderate and severe) was not given.
NotesTrial was conducted during the 1974-75 influenza season in which a two months epidemic period is described by the authors with no reference to a definition.
The follow-up period of the study was generically defined as "winter and spring", lasted six months and comprised the epidemic period.
The circulating strain was A/PortChalmers/1/73 (H3N2) and presented a slight antigenic drift when compared with vaccine strain.
Allocation concealmentB – Unclear
StudySumarokow 1971
MethodsField trial conducted in Russia. Vaccinations were carried out using coded preparation.
Participants76150 persons.
A first study group counted 36665 children 3-12 years old. A second study group counted 39485 participants aged 13-25. Only data regarding the latter were extracted and loaded.
InterventionsThe trial compared two live vaccines against placebo. Strains used in the vaccines were not reported.
Study arms were the following:
- allantoic intranasal vaccine; n = 9945
- placebo; n = 9942
- tissue vaccine for oral administration; n = 9817
- placebo; n = 9781
Only data from the first two groups were included in the meta-analysis.
OutcomesOutcomes were clinical and laboratory confirmed cases, deaths, severity.
Clinical outcomes were all the acute respiratory infections.
Laboratory confirmation was obtained on a sample of ill subjects by virus isolation or demonstration of seroconversion.
Bronchitis, otitis and pneumonia were considered as complications.
NotesOutcomes were detected during the entire period of influenza outbreak.
The epidemic period was defined as the period of highest influenza morbidity and lasted 11 weeks from the last ten days of January to the first ten days of April.
Virus circulating was A2/Hong Kong/68, and did not match the vaccine strains.
Allocation concealmentA – Adequate
StudyTannock 1984
MethodsCCT double blind, placebo controlled, carried out in Australia.
Volunteers were allocated by alternation to receive one or two doses of a trivalent subunit vaccine or placebo.
Participants88 volunteers, aged 16-64, staff from a Newcastle Hospital and the Commonwealth Steel Corporation.
Initially 46 volunteers were assigned to each treatment, but only individuals who received the expected doses of vaccine were considered as part of the study (29, 27 and 32 for each group described above). Complete follow up data were obtained for 18, 19, and 20 subjects respectively.
InterventionsTrivalent vaccine (A/Brazil/11/78, A/Bangkok/1/79, B/Singapore/222/79 7 mg each) against placebo (sterile saline).
Vaccine or placebo was administered by the subcutaneous route.
OutcomesOutcomes were laboratory confirmed cases and adverse effects.
A case was laboratory confirmed if a respiratory illness, retrospectively reported, was associated with a 4-fold antibody titre increase between post-vaccination and post-epidemic sera.
Reactions to vaccination were classified as follows:
local reaction: redness, swelling, warmth or irritation;
local pain: pain on contact, pain with pressure, continuous pain, or restriction of arm movement;
systemic reaction: fever, chills, sweating, drowsiness or insomnia.
NotesThe follow up period lasted 20 weeks and comprised the epidemic period (year 1981)
Viral strain isolated from patients was A/Bangkok/1/79.
The recommended formula for that year was: A/Brazil/11/78, A/Bangkok/1/79, B/Singapore/222/79
Allocation concealmentC – Inadequate
StudyWaldman 1969a
MethodsRCT conducted in USA.
The trial compared two inactivated vaccines (monovalent A and the commercial polyvalent AB), administered according to different schedules (1 or 2 doses) and route of administration (intramuscular and aerosol), against placebo (saline).
The volunteers were randomly divided into 9 groups. Vaccine and placebo were administered in a double blind manner on two occasion 3 weeks apart. The study was completed by 99% of the volunteers.
Participants2100 school teachers.
InterventionsData were extracted and loaded as follows:
Waldman 5: monovalent (A/Hong Kong/68) intramuscularly administered (n = 465) against placebo (n = 59)
OutcomesOutcomes were clinical cases and side effects.
Clinical case definition was "temperature > 100°F or feverish feeling plus any 2 of the following symptoms: sore throat, muscle or joint pain, cough, stuffy or runny nose".
Data concerning adverse effects were only partially reported by graph.
NotesThe protective efficacy of the influenza vaccines was evaluated during the 1968-69 influenza season. The follow up period lasted 10 weeks and comprised the epidemic period. Outbreak was caused by an A2/Hong Kong/68 influenza virus.
The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68.
Vaccine was considered to be recommended and matching the circulating strain.
Allocation concealmentB – Unclear
StudyWaldman 1969b
MethodsSee Waldman 5
ParticipantsSee Waldman 5
InterventionsData were extracted and loaded as follows:
Waldman 6: polyvalent (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA, B/Massachusetts/3/66 300 CCA) intramuscularly administered (n = 471) against placebo (n = 59);
OutcomesSee Waldman 5
NotesThe protective efficacy of the influenza vaccines was evaluated during the 1967-68 influenza season. The follow up period lasted 10 weeks and comprised the epidemic period. Outbreak was caused by an A2/Hong Kong/68 influenza virus.
The recommended polyvalent formula for that year in the USA was: A/PR/8/34, A1/AA/1/57, A2/Japan/170/62, A2/Taiwan/1/64, B/Mass/3/66
Viruses used in the vaccines matched the recommended strains.
Allocation concealmentB – Unclear
StudyWaldman 1969c
MethodsSee Waldman 5
ParticipantsSee Waldman 5
InterventionsData were extracted and loaded as follows:
Waldman 7: monovalent A/Hong Kong/68 aerosol administered (n = 479) against placebo (n = 59);
OutcomesSee Waldman 5
NotesSee Waldman 5
Allocation concealmentB – Unclear
StudyWaldman 1969d
MethodsSee Waldman 5
ParticipantsSee Waldman 5
InterventionsData were extracted and loaded as follows:
Waldman 8: polyvalent (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA, B/Massachusetts/3/66 300 CCA) aerosol administered (n = 471) against placebo (n = 59);
OutcomesSee Waldman 5
NotesSee Waldman 6
Allocation concealmentB – Unclear
StudyWaldman 1972a
MethodsRCT conducted in a USA university campus.
Volunteers were randomly allocated to study groups and were immunised in a double blind manner.
Two inactivated vaccines (monovalent A and bivalent AB) and two different routes of administration (subcutaneous and aerosol) were compared against placebo (saline).
Participants864 volunteer students and staff members.
InterventionsData were extracted and loaded as follows:
Monovalent A (A2/Aichi/1/68) 200 CCA aerosol against placebo. 195 treated, 98 placebo.
OutcomesOutcomes were clinical cases and adverse effects.
Clinical cases were defined as febrile respiratory illness with oral temperature higher then 99.5 degrees F.
Systemic adverse effects were defined as respiratory (runny and/or stuffy nose, sore throat, cough, shortness of breath).
NotesThe protective efficacy of the influenza vaccines was evaluated during the 1968-69 outbreak.
Study follow-up lasted 88 days and it is likely to have covered the whole epidemic period. Circulating viral strain was A2/Hong Kong virus and it matched the vaccine strain.
The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68.
The vaccine strain was considered to be recommended and matching the circulating strain.
Allocation concealmentA – Adequate
StudyWaldman 1972b
MethodsSee Waldman 1
ParticipantsSee Waldman 1
InterventionsData were extracted and loaded as follows:
Monovalent A (A2/Aichi/1/68) 200 CCA subcutaneous against placebo. 190 treated, 25 placebo.
OutcomesSee Waldman 1
NotesSee Waldman 1
Allocation concealmentA – Adequate
StudyWaldman 1972c
MethodsSee Waldman 1
ParticipantsSee Waldman 1
InterventionsData were extracted and loaded as follows:
Bivalent AB (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA and B/Massachusetts/3/66 200 CCA) aerosol against placebo. 194 treated, 25 placebo.
OutcomesSee Waldman 1
NotesThe protective efficacy of the influenza vaccines was evaluated during the 1968-69 outbreak.
Study follow-up lasted 88 days and it is likely to have covered the whole epidemic period. Circulating viral strain was A2/Hong Kong virus.
The recommended polyvalent formula for that year in the USA was: A/PR/8/34, A1/AA/1/57, A2/Japan/170/62, A2/Taiwan/1/64, B/Mass/3/66
The vaccine strain was considered to be recommended.
Allocation concealmentA – Adequate
StudyWaldman 1972d
MethodsSee Waldman 1
ParticipantsSee Waldman 1
InterventionsData were extracted and loaded as follows:
Bivalent AB (A2/Japan/170/62 150 CCA, A2/Taiwan/1/64 150 CCA and B/Massachusetts/3/66 200 CCA) subcutaneous against placebo. 187 treated and 23 placebo.
OutcomesSee Waldman 1
NotesSee Waldman 3
Allocation concealmentA – Adequate
StudyWeingarten 1985
MethodsRCT conducted in USA
Participants were assigned using a random-number generator.
The injections were given and the data were collected in a double blind manner.
Participants179 healthy volunteer hospital employees, 21-65 year-old.
Each arm counted 91 and 88 participants respectively.
InterventionsSplit trivalent containing 15 mg each of A/Chile/1/83 (H1N1), A/Philippines/2/82 (H3N2), and B/USSR/100/83 hemagglutinin antigens)
Saline placebo.
All injections were given intramuscularly.
OutcomesOutcomes were clinical cases symptoms defined, WDL, and adverse effects.
An influenza illness was defined by the CDC case definition: a documented temperature greater than 100°F and at least the symptoms of cough or sore throat. Data regarding adverse effects were not complete and therefore not considered.
NotesTrial was conducted during the 1985-86 influenza season.
Epidemic influenza was defined according to population surveillance data (without better explanation), began in December 1985 and concluded in February 1986. The follow up comprised the epidemic period and lasted 67 days.
Most of the influenza infections were caused by type B. In authors' opinion there was poor closeness of fit between the virus in the vaccine and the circulating virus.
The recommended formula for that year was A/Chile/1/83 (H1N1), A/Philippines/2/82 (H3N2), and B/USSR/100/83 .
Allocation concealmentA – Adequate
StudyWilde 1999a
MethodsWilde is a RCT conducted in USA. Volunteers were randomly allocated to receive a trivalent, inactivated, intramuscularly administered vaccine or a control (meningococcal vaccine, pneumococcal vaccine or placebo) using a four-units block randomisation. Groups were evaluated in a double-blind fashion.
ParticipantsHealth care professionals without chronic medical problems and younger than 50 years. Treatment and placebo arm counted respectively 52 and 50 participants in 1992/93
InterventionsThe protective efficacy of the influenza vaccine was evaluated during the 1992-93 outbreak. Vaccine influenza strains were: A/Texas/36/91, A/Beijing/353/89 and B/Panama/45/90 and control was meningococcal vaccine.
OutcomesOutcomes were: laboratory confirmed cases and working days lost (wdl). Treatment and placebo arm counted, respectively, 52 and 50 subjects in 1992/93, 51 and 52 subjects in 1993/94, 77 and 77 subjects in 1994/95.
Outcomes were detected during the whole follow up period, from October-November to March-April, in each study year. Laboratory confirmation was based on four-fold or greater rise in antibody titre occurred between the 4-week-post injection and end-of-season serum sample.
NotesViral strains isolated from cases were A(H3N2) and B strains: a good match was achieved between the epidemic influenza subtypes and the vaccine components during 1993/94 influenza season; there was a partial match during 1992/93 and 1994/95 influenza season.
As every year the vaccine strains matched at least one of several circulating viruses the study was included among those comparing recommended vaccines against placebo or other vaccines and among those comparing vaccines matching the circulating virus against placebo or other vaccines.
Allocation concealmentA – Adequate
StudyWilde 1999b
MethodsSee Wilde 1
ParticipantsHealth care professionals without chronic medical problems and younger than 50 years. Treatment and placebo arm counted respectively 51 and 52 subjects in 1993/94,
InterventionsThe protective efficacy of the influenza vaccine was evaluated during the 1993-94 outbreak. Vaccine influenza strains were: A/Texas/36/91, A/Beijing/32/92 and B/Panama/45/90 and control was pneumococcal vaccine.
OutcomesSee Wilde 1
NotesSee Wilde 1
Allocation concealmentA – Adequate
StudyWilde 1999c
MethodsSee Wilde 1
ParticipantsHealth care professionals without chronic medical problems and younger than 50 years. Treatment and placebo arm counted respectively 77 and 77 participants in 1994/95.
InterventionsThe protective efficacy of the influenza vaccine was evaluated during the 1994-95 outbreak. Vaccine influenza strains were: A/Texas/36/91, A/Shangdong/9/32 and B/Panama/45/90 and control was sterile saline for injection.
OutcomesSee Wilde 1
NotesSee Wilde 1
Allocation concealmentA – Adequate
StudyWilliams 1973
MethodsWilliams reports the results of a double-blind randomised field trial carried out in Canadian military bases
Volunteers particulars were entered, in order of attendance, on vaccination registers that indicate by code letter the vaccine to be given. The allocation of vaccine code was at random. The vials were coded in such a manner that the vaccinator didn't know which type of vaccine was being used.
ParticipantsHealthy volunteer military recruits.
InterventionsData were extracted and loaded as follow: a monovalent "standard" IMH (Institute of Microbiology and Hygiene) aqueous killed influenza A vaccine (A2/HK/68) against a monovalent influenza B vaccine (B/Mass/66).
Participants in the arms were: 1947 (monovalent A), 1955 (monovalent B)
All the vaccines were inoculated subcutaneously.
OutcomesOutcomes were clinical cases and side effects.
Clinical cases were classified as "respiratory illness" without further specification. Adverse effects had unsatisfactory definition to be extracted and loaded.
Only cases occurred during the epidemic period were evaluated.
NotesStudy was conducted during the 1968-69 influenza season.
In one of the bases an influenza outbreak occurred during the last week of December 1968 and January 1969.
3 strains, identified as A2/Hong Kong, were isolated from ill men and were considered matching the monovalent A vaccine. The follow up period lasted 16 weeks.
The monovalent vaccine recommended for that year in the USA was A2/Aichi/2/68.
The vaccine strain was considered to be recommended and matching the circulating strain.
Allocation concealmentA – Adequate

Characteristics of excluded studies

StudyReason for exclusion
Ambrosch 1976Data tables and figure missing.
Aoki 1986RCT, single blind. Outcomes were clinical cases and adverse effects.
Follow up data were not reported by arms.
Belshe 2001Non original data
Boyce 2001Outcome measures outside inclusion criteria
Chlibek 2002The study is not RCT
Clover 1991RCT. More than 75% of the study population is out of the range of age stated in the protocol.
Das 2002The study does not contain effectiveness estimates
Davies 1973Data tables and figure missing.
Finklea 1969RCT, double blind. Two bivalent inactivated influenza vaccines, with the same viral composition, differing in purification procedures, were compared.
Outcomes were clinical cases and adverse effects.
Raw data about clinical cases were not reported by arm.
Circulating virus showed significant antigenic differences from the A2 vaccine strain.
Gerstoft 2001The study is not RCT
Gross 1999Outcome measures outside inclusion criteria
Grotto 1998The study is not RCT
Gruber 1994Gruber is a RCT conducted in USA on 41 cystic fibrosis (CF) patients and 89 family members, recruited through a clinic. Participants were randomly assigned in a double-blinded fashion by family to receive either intranasal live cold-adapted influenza A vaccine or the recommended intramuscular trivalent inactivated influenza vaccine.
The study lasted 3 years (from 1989 to 1991). Participants were immunised each fall staying in the same assigned vaccine group. The live vaccine arm counted 20 CF and 33 family members; the trivalent vaccine arm 21 and 56 respectively.
69 of them (17 CF patients and 52 family members) dropped out. The reasons were stated in the article.
The live vaccine was the same all over the period: A/Kawasaki/9/86 (H1N1) 107,3 pfu, A/Los Angeles/2/87 107,3 pfu.
The viral strains used in the inactivated vaccines were:
- 1989-1990: A/Taiwan/1/86 (H1N1), A/Shanghai/11/87 (H3N2), B/Yagamata/16/88,15 mg/dose of each
- 1990-1991: A/Taiwan/1/86 (H1N1), A/Shanghai/16/89 (H3N2), B/Yagamata/16/88,15 mg/dose of each
- 1991-1992: A/Taiwan/1/86 (H1N1), A/Beijing/353/89 (H3N2), B/Panama/45/90, 15 mg/dose of each
Live vaccine recipient also received monovalent inactivated influenza B vaccine (identical to that contained in the trivalent vaccine) as intramuscular placebo. Allantoic fluid was the placebo for aerosol administration.
Data were extracted and loaded for family members only.
Outcomes were clinical and laboratory confirmed cases, working days lost (WDL), admissions, deaths and adverse effects.
Clinical cases were classified as "respiratory illness" or "febrile respiratory illness". Laboratory confirmed cases were defined by an influenza virus isolation from a throat swab.
Adverse effects were defined as temperature > 38°C, rhinorrhoea, sore throat, cough, increasing sputum, redness, swelling, chills. Results are expressed as % of subject-days with symptoms.
Participants were followed throughout the period. Owing to the drop outs, vaccinated were counted as subject-years: 54 in the live vaccine arm; 56 in the trivalent vaccine arm.
The influenza illness surveillance period for study participants was defined as the interval from the date of the first influenza isolate from population under routine surveillance to 2 weeks after the last isolate for each year.
Viral strains circulating during the outbreaks were:
- 1989-1990: A/Shanghai/11/87 (H3N2)
- 1990-1991: A/Beijing/353/89 (H3N2), B/Panama/45/90-like
- 1991-1992: A/Beijing/353/89 (H3N2).
This trial was excluded since it was not placebo controlled and authors did not specify if the strains used to develop cold adapted and inactivated vaccines were antigenetically comparable or not.
Haigh 1973The study is not randomised: all the volunteers were immunised on a single day and the intention to allocate patients randomly was not strictly adhered to.
Halperin 2002Outcome measures outside inclusion criteria
Hobson 1971RCT. Clinical outcomes were side effects only.
Hoskins 1976aThis trial is described in the "included studies" section. Since A/Hong Kong vaccine was shown to protect during the influenza A/England outbreak in 1972, all the 1972 and 1973 entrants were given influenza A/England/42/72 vaccine (200 IU) in October 1973. The 1973 entrants were randomly allocated to receive vaccine A/England alone or combined with B/Victoria (50 IU) and B/Hong Kong (50 IU). Blinding was ensured by coded labels.
Boys in the boarding school had been observed during an outbreak due simultaneously to influenza A/Port Chalmers, B/Hong Kong and B intermediate strains in March 1974.
Efficacy of the vaccines in preventing influenza A or influenza B outbreak was separately assessed in sub-groups of vaccinated boys, as follows.
Hoskins 2: laboratory confirmed influenza A cases were detected in the 1970-71 entry (244 influenza A vaccinates; 226 influenza B vaccinates).
The trial was excluded since it was not placebo controlled, and most of the participants were vaccinated with a formula that was not recommended for that year and did not match the circulating strain.
Laboratory confirmation was based on culture and/or four-fold or greater rise in antibody titre occurring between acute and convalescent sera. Cases without virus isolation but with serological evidence of infection with both influenza A and B viruses were excluded.
The first boy reported sick on Feb. 28, 1974 and the last on March 30.
Hoskins 1976bThe trial was excluded: it was not placebo controlled, most of the participants were vaccinated with a formula that was not recommended for that year and the contemporary presence of influenza A and B viruses in the outbreak make it unwise to consider other comparisons.
Hoskins 1979Hoskins reported the prosecution of a study started in 1970, and described in Flu 65 and Flu 160, during the 1976 outbreak. At the date all the participants were vaccinated against the circulating viral strains, so that a control group was missing.
Howell 1967The study is not prospective. It appears as an historical cohort.
Jianping 1999The study is not RCT
Keitel 2001Outcome measures outside inclusion criteria
Kiderman 2001Tables and text show inconsistencies that do not allow data extraction
Liem 1973Liem reported the results of 9 placebo controlled clinical trials and two field studies, involving a total of about 10000 participants, carried out in several countries to assess the efficacy of killed influenza spray vaccines. Studies were conducted during the years 1969-71.
Allocation of the participants to the arms of the trials was done according to a predetermined randomisation scheme. 8 of them were double-blind. The field studies were not randomised.
The attack rate for influenza among the population study was very low, and in two of the trials vaccination procedure started too late, when outbreak was ongoing.
The attack rates, exclusively based on the serologically confirmed cases, are only reported by a graph and it is impossible to guess the crude data.
Mendelman 2001The study does not report original results
Monto 2000The study is not RCT
Mostow 1977CCT. Clinical outcomes were side effects only.
Muennig 2001The study is not RCT
Nichol 1996RCT. Adverse effects only.
Nichol 1999aThe study is a review
Nichol 2001The study is not RCT
Nichol 2003The study contain data from previous studies
Pyhala 2001The study is not RCT
Pyrhonen 1981RCT. Clinical outcomes were adverse effects and infections (with or without illness) serologically confirmed. It was impossible to work out the number of clinical cases.
Rimmelzwaan 2000Outcome measures outside inclusion criteria
Ruben 1973Ruben is a randomised, not blind, not placebo controlled trial, conducted in a USA university campus on 3103 healthy volunteer students, 17-22 year-old. Authors don't specify the method for randomisation.
Authors compared a split bivalent vaccine intramuscularly administered against a whole inactivated subcutaneous bivalent vaccine. All vaccines contained 400 CCA units of Influenza A/Aichi/68/H3N2 and 300 CCA units of B/Mass/66. The "split vaccine" arm counted 1791 participants, the "whole vaccine" arm 1312.
Outcomes were clinical cases, defined as "febrile upper respiratory illness".
Cases were detected during the whole epidemic period, lasting 3 weeks, from January 15 to early February 1972.
Circulating viral strain was an A/Hong Kong virus and it probably matched the vaccine strain.
The incidence of clinical influenza in the study groups was also compared against the outbreak occurring in the campus among non vaccinated students: this observational comparison was not considered.
The study was excluded since it was not placebo controlled and both the arms contained the same vaccine strains.
Saxen 1999Outcome measures outside inclusion criteria
Smith 1977The article reports a small part of the Hoskins' trial.
It compared illness occurring among a group of vaccinated boys against non vaccinated controls that had no part in the trial.
Spencer 1975Authors did not report crude data concerning the clinical outcomes.
Spencer 1979Reporting does not allow one to understand the methods used to allocate participants and to conceal allocation.
As for the clinical outcomes, authors didn't report the crude data.
Taylor 1966CCT. Side effects only.
Treanor 2001Outcome measures outside inclusion criteria
Treanor 2002Outcome measures outside inclusion criteria
Warshauer 1976The study was not randomised.
Data reporting was not complete.
Williams 1973bA monovalent aqueous killed influenza A vaccine (A/Hong Kong/68 400 CCA) was compared against a monovalent aqueous killed influenza B vaccine (B/Mass766 400 CCA); participants in the two arms were 2214 and 2231 respectively. In the bases that took part in the trial no influenza outbreak occurred, but influenza-like illnesses had been noted during a 5 weeks period (January 1969). The follow up period lasted 18 weeks. Occurrence of respiratory disease was reported for both the follow up and the epidemic period.
Williams 1973cThe trial compared three "standard" IMH aqueous killed vaccines: a monovalent influenza A vaccine (A2/HK/68), a monovalent influenza B vaccine (B/Mass/66) and a bivalent one (A2/Mtl/68, B/Mass/66). Participants in the three arms were 987, 988 and 996 respectively. In bases that took part in the trial no influenza outbreak occurred. Occurrence of respiratory disease was reported for the follow up period only (13 weeks).
Outcomes were clinical cases and side effects.
Clinical cases were classified as "respiratory illness" or "non-respiratory illness", without further specification. Adverse effects had unsatisfactory definition to be extracted and loaded.
Wood 1999The study is not RCT
Wood 2000The study is not RCT

Analyses

Comparison 01. Vaccine versus placebo
Outcome titleNo. of studiesNo. of participantsStatistical methodEffect size
01 Influenza cases (clinically defined)2538130Relative Risk (Random) 95% CI0.77 [0.69, 0.86]
02 Influenza cases (serologically confirmed)112838Relative Risk (Random) 95% CI0.31 [0.21, 0.45]
03 Working days lost65572Weighted Mean Difference (Random) 95% CI-0.12 [-0.24, -0.00]
04 Complications322840Relative Risk (Random) 95% CI0.77 [0.24, 2.47]
05 Hospital admissions45261Relative Risk (Fixed) 95% CI0.65 [0.34, 1.22]
06 Adverse effects for live aerosol vaccines  Relative Risk (Random) 95% CISubtotals only
07 Adverse effects for inactivated parenteral vaccines  Relative Risk (Random) 95% CISubtotals only
Analysis 01.01.

Comparison 01 Vaccine versus placebo, Outcome 01 Influenza cases (clinically defined)

Analysis 01.02.

Comparison 01 Vaccine versus placebo, Outcome 02 Influenza cases (serologically confirmed)

Analysis 01.03.

Comparison 01 Vaccine versus placebo, Outcome 03 Working days lost

Analysis 01.04.

Comparison 01 Vaccine versus placebo, Outcome 04 Complications

Analysis 01.05.

Comparison 01 Vaccine versus placebo, Outcome 05 Hospital admissions

Analysis 01.06.

Comparison 01 Vaccine versus placebo, Outcome 06 Adverse effects for live aerosol vaccines

Analysis 01.07.

Comparison 01 Vaccine versus placebo, Outcome 07 Adverse effects for inactivated parenteral vaccines

Comparison 02. At least one vaccine recommended for that year versus placebo or other vaccine
Outcome titleNo. of studiesNo. of participantsStatistical methodEffect size
01 Influenza cases (clinically defined)3838371Relative Risk (Random) 95% CI0.79 [0.73, 0.85]
02 Influenza cases (serologically confirmed)2114665Relative Risk (Random) 95% CI0.32 [0.24, 0.43]
03 Working days lost77346Weighted Mean Difference (Random) 95% CI-0.16 [-0.29, -0.04]
Analysis 02.01.

Comparison 02 At least one vaccine recommended for that year versus placebo or other vaccine, Outcome 01 Influenza cases (clinically defined)

Analysis 02.02.

Comparison 02 At least one vaccine recommended for that year versus placebo or other vaccine, Outcome 02 Influenza cases (serologically confirmed)

Analysis 02.03.

Comparison 02 At least one vaccine recommended for that year versus placebo or other vaccine, Outcome 03 Working days lost

Comparison 03. Vaccine matching circulating strain versus placebo or other vaccine
Outcome titleNo. of studiesNo. of participantsStatistical methodEffect size
01 Influenza cases (clinically defined)2524809Relative Risk (Random) 95% CI0.70 [0.62, 0.79]
02 Influenza cases (serologically confirmed)139547Relative Risk (Random) 95% CI0.25 [0.16, 0.38]
03 Working days Lost21771Weighted Mean Difference (Random) 95% CI-0.04 [-0.07, -0.02]
Analysis 03.01.

Comparison 03 Vaccine matching circulating strain versus placebo or other vaccine, Outcome 01 Influenza cases (clinically defined)

Analysis 03.02.

Comparison 03 Vaccine matching circulating strain versus placebo or other vaccine, Outcome 02 Influenza cases (serologically confirmed)

Analysis 03.03.

Comparison 03 Vaccine matching circulating strain versus placebo or other vaccine, Outcome 03 Working days Lost

Sources of support

External sources of support

  • Ministry of Defence UK

Internal sources of support

  • Azienda Sanitaria Locale 20, Alessandria ITALY

  • Azienda Sanitaria Locale 19, Asti ITALY

Ancillary