Hepatitis B immunisation in persons not previously exposed
to hepatitis B or with unknown exposure status (Review)
Mathew JL, El Dib R, Mathew PJ, Boxall EH, Brok J
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2008, Issue 3
http://www.thecochranelibrary.com
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fig. 1.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fig. 2.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up, Outcome 1
HBsAg at follow-up (all studies reporting this outcome). . . . . . . . . . . . . . . . . . . .
Analysis 1.2. Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up, Outcome 2
HBsAg at follow-up (sub-group analysis based on age of enrolled participants). . . . . . . . . . . .
Analysis 1.3. Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up, Outcome 3
HBsAg at follow-up (sub-group analysis based on prevalence in participating population).. . . . . . . .
Analysis 1.4. Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up, Outcome 4
HBsAg at follow-up (sub-group analysis based on duration of follow-up). . . . . . . . . . . . . .
Analysis 1.5. Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up, Outcome 5
HBsAg at follow-up (sub-group analysis based on vaccination schedule). . . . . . . . . . . . . .
Analysis 2.1. Comparison 2 HB vaccine versus control: presence of antibodies to HBc antigen in serum, Outcome 1
Anti-HBc antibodies in serum. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.2. Comparison 2 HB vaccine versus control: presence of antibodies to HBc antigen in serum, Outcome 2
Anti-HBc antibodies in serum (random-effects model). . . . . . . . . . . . . . . . . . . .
Analysis 3.1. Comparison 3 HB vaccine versus control: adverse events, Outcome 1 Lack of sero-protection. . . .
Analysis 3.2. Comparison 3 HB vaccine versus control: adverse events, Outcome 2 Lack of sero-protection (fixed-effect
model). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 4.1. Comparison 4 HB vaccine versus control: serious adverse events, Outcome 1 Serious adverse events: allcause mortality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 5.1. Comparison 5 HB vaccine versus control: adverse events, Outcome 1 Fever. . . . . . . . . .
Analysis 5.2. Comparison 5 HB vaccine versus control: adverse events, Outcome 2 Fever (fixed-effect model). . .
Analysis 5.3. Comparison 5 HB vaccine versus control: adverse events, Outcome 3 Local reactions (pain, redness,
soreness). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 5.4. Comparison 5 HB vaccine versus control: adverse events, Outcome 4 Other adverse events. . . . .
Analysis 6.1. Comparison 6 HB vaccine versus control: lack of compliance, Outcome 1 Lack of compliance. . . .
Analysis 6.2. Comparison 6 HB vaccine versus control: lack of compliance, Outcome 2 Lack of compliance (fixed-effect
model). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
1
1
2
2
3
3
5
9
10
11
14
15
15
19
32
35
36
38
39
41
43
44
44
45
45
46
46
47
47
48
48
i
[Intervention review]
Hepatitis B immunisation in persons not previously exposed
to hepatitis B or with unknown exposure status
Joseph L Mathew1 , Regina El Dib2 , Preethy J Mathew3 , Elizabeth H Boxall4 , Jesper Brok5
1
Department of Pediatrics, Advanced Pediatrics Centre, Chandigarh, India. 2 McMaster University, Toronto, Canada. 3 Department
of Anaesthesia, Postgraduate Institute of Medial Education and Research (PGIMER), Chandigarh, India. 4 Public Health Laboratory,
Heartlands Hospital, Birmingham, UK. 5 Cochrane Hepato-Biliary Group, Copenhagen Trial Unit, Centre for Clinical Intervention
Research, Copenhagen, Denmark
Contact address: Joseph L Mathew, Department of Pediatrics, Advanced Pediatrics Centre, Postgraduate Institute of Medial Education
and Research (PGIMER), Chandigarh, 160012, India. [email protected]. (Editorial group: Cochrane Hepato-Biliary Group.)
Cochrane Database of Systematic Reviews, Issue 3, 2008 (Status in this issue: New)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
DOI: 10.1002/14651858.CD006481.pub2
This version first published online: 16 July 2008 in Issue 3, 2008.
Last assessed as up-to-date: 4 May 2008. (Dates and statuses?)
This record should be cited as: Mathew JL, El Dib R, Mathew PJ, Boxall EH, Brok J. Hepatitis B immunisation in persons not
previously exposed to hepatitis B or with unknown exposure status. Cochrane Database of Systematic Reviews 2008, Issue 3. Art. No.:
CD006481. DOI: 10.1002/14651858.CD006481.pub2.
ABSTRACT
Background
The benefits and harms of hepatitis B vaccination in persons not previously exposed to hepatitis B infection or with unknown exposure
status have not been established.
Objectives
To assess the benefits and harms of hepatitis B vaccination in people not previously exposed to hepatitis B infection or with unknown
exposure status.
Search strategy
Trials were identified from The Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials
(CENTRAL), MEDLINE, EMBASE, LILACS,Science Citation Index Expanded (last search, March 2007). Additionally, we contacted
experts and vaccine manufacturers, and read through reference lists for eligible trials.
Selection criteria
Randomised clinical trials comparing hepatitis B vaccine versus placebo, no intervention, or another vaccine in persons not previously
exposed to hepatitis B (HBsAg negative) or with unknown exposure status.
Data collection and analysis
The primary outcome was hepatitis B infection (detecting HBsAg, HBeAg, HBV DNA, or anti-HBc). Secondary outcomes were lack of
sero-protection, antibody titre, clinical complications, adverse events, lack of compliance, and cost-effectiveness. Dichotomous outcomes
were reported as relative risk (RR) with 95% confidence interval (CI), using intention-to-treat analysis assuming an unfavourable event
for missing data. Sensitivity analyses based on methodological quality (risk of bias), available data analysis, intention-to-treat analysis
assuming a favourable event for missing data, best-case scenario, and worst-case scenario were conducted.
Main results
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
1
Twelve trials were eligible. All had high risk of bias and reporting was inconsistent. Hepatitis B vaccine did not show a clear effect on
the risk of developing HBsAg (RR 0.96, 95% CI 0.89 to 1.03, 4 trials, 1230 participants) and anti-HBc (RR 0.81, 95% CI 0.61 to
1.07; 4 trials, 1230 participants, random-effects) when data were analysed using intention-to-treat analysis assuming an unfavourable
event for missing data. Analysis based on data of available participants showed reduced risk of developing HBsAg (RR 0.12, 95% CI
0.03 to 0.44, 4 trials, 576 participants) and anti-HBc (RR 0.36, 95% CI 0.17 to 0.76, 4 trials, 576 participants, random-effects).
Intention-to-treat analysis assuming favourable outcome for missing data showed similar reduction in risk. Hepatitis B vaccination had
an unclear effect on the risk of lacking protective antibody levels (RR 0.57, 95% CI 0.26 to 1.27, 3 trials, 1210 participants, randomeffects). Development of adverse events was sparsely reported.
Authors’ conclusions
In people not previously exposed to hepatitis B, vaccination has unclear effect on the risk of developing infection, as compared to no
vaccination. The risk of lacking protective antibody levels as well as serious and non-serious adverse events appear comparable among
recipients and non-recipients of hepatitis B vaccine.
PLAIN LANGUAGE SUMMARY
In children or grown-ups who have not been previously exposed to hepatitis B infection or whose exposure status is not known,
hepatitis B vaccination as compared with no hepatitis B vaccination has an unclear effect on the risk of developing hepatitis B
infection
Several million people world-wide are infected with hepatitis B virus. The infection may cause serious short-term and long-term effects
including portal hypertension, liver failure, liver cancer, and death. Hepatitis B vaccination is reported to be beneficial in some specific
groups of people such as babies born to women infected with hepatitis B, health-care workers, and people with long-standing kidney
failure. Whether hepatitis B vaccine is beneficial in people who have not been exposed to hepatitis B infection or those whose exposure
status is not known is assessed in the present review.
Twelve randomised clinical trials fulfilled the inclusion criteria of this review. Primary analysis of the data based on criteria described
beforehand (intention-to-treat model assigning unfavourable outcome for missing data) showed that hepatitis B vaccination has an
unclear effect on the risk of developing hepatitis B infection. Analysis of data of available participants in the various trials showed that
as compared to not vaccinating, hepatitis B vaccination reduces the risk of developing hepatitis B infection; by 88% for hepatitis B
surface antigen marker and 62% for anti-core antibody marker. One should note, that these findings are based on only four randomised
clinical trials of poor methodological quality involving 1230 participants. When compared with other vaccines or placebo, hepatitis B
vaccination results in comparable risk of developing adverse events. This includes serious adverse events such as admission to hospital
and convulsions, as well as less serious events such as fever, local redness, and pain. This shows that the risk of developing these adverse
events is not more than with other vaccinations. There was not enough data to draw definite conclusions on the effect of hepatitis B
vaccination on compliance and cost-effectiveness.
BACKGROUND
Hepatitis B virus (HBV) infection is a global problem. Current estimates suggest that over two billion people all over the world have
been infected, of whom approximately 350 million have chronic
infection (Beasley 1984; Maddrey 2000; Walsh 2001; Rizzetto
2002). About one million deaths each year are attributable to hepatocellular carcinoma and cirrhosis, related to chronic hepatitis B
virus infection (WHO 2004).
Clinically, hepatitis B can result in a spectrum of clinical problems
ranging from asymptomatic infection to symptomatic infection
(McMahon 1985; Walsh 2001). The long-term danger is that
some infected persons develop persistent infection, of whom a
fraction develop chronic liver disease and associated complications
such as cirrhosis, portal hypertension, hepatocellular carcinoma,
and hepatocellular failure (Liaw 1998; Rehermann 2003). The risk
of developing persistent infection and its complications is inversely
proportional to the age at infection; the younger the person at the
time of initial infection, the greater is the risk of persistence, and
vice versa (McMahon 1985; Moyer 1994; Hyams 1995; Gall 2001;
Aggarwal 2004). Therefore, children infected at the time of birth
have the highest risk of persistent hepatitis B infection. Persons
with persistent infection, even if asymptomatic themselves, can still
transmit the infection to others. The major routes of transmission
are through sexual contact, sharing contaminated objects such as
needles, razors, etc, and transfusion of infected blood or blood
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
2
products. Another important route is from an infected pregnant
woman to her child before or during delivery, ie, perinatal or
vertical transmission (Gitlin 1997; Walsh 2001).
Vaccination against hepatitis B is expected to protect people from
infection. If we assume that hepatitis B vaccine is safe and efficacious, every individual residing in geographical regions with high
incidence of infection should receive hepatitis B vaccination early
in life. This is often not feasible due to logistic and economic
considerations. Therefore, to obtain maximum cost-benefit two
strategies are possible (CDC 1999; Aggarwal 2004). The ’high
risk strategy’ focuses on people believed to be at highest risk of infection, eg, health-care professionals, recipients of blood products,
intravenous drug abusers, babies born to infected mothers, etc.
This strategy depends on identification of all high-risk people and
vaccination compliance. The high risk strategy is less cost intensive, but likely to work only in areas of low population prevalence
of hepatitis B infection.
This contrasts the strategy of universal immunisation, wherein all
individuals are vaccinated, irrespective of individual risk. Universal immunisation can be used in populations with high prevalence
of hepatitis B infection and where it is difficult to identify individuals at highest risk (Chen 1996). This is economically and logistically more demanding and its success also depends on vaccination
compliance. Individuals at high risk frequently refuse vaccination
or fail to complete the vaccination series (Wong 1994; Nystrom
2000).
Previous systematic reviews have shown the benefit of hepatitis B
immunisation in newborns of mothers with hepatitis B infection
(Lee 2006a; Lee 2006b) and other high risk groups such as healthcare workers (Chen 2005) and patients with chronic renal failure
(Schroth 2004). However, no meta-analyses or systematic reviews
have addressed the issue of immunisation in persons who are not
at high risk of hepatitis B exposure or individuals in the general
population whose hepatitis B exposure status is not known. As
the majority of the world’s population fits into either of these
categories, this review was undertaken to assess whether hepatitis
B vaccination protects against hepatitis B virus infection in such
persons.
Types of studies
We included randomised clinical trials, irrespective of blinding,
publication status, or language.
Types of participants
We included trials that recruited participants of any age who were
either hepatitis B surface antigen (HBsAg) negative or whose status was not known. Absence of HBsAg suggests absence of hepatitis B infection, although this may not always be the case. We
excluded trials on newborns who were born to mothers known to
be infected with hepatitis B, and also trials on participants suspected to be at high risk of hepatitis B infection (health-care workers, intravenous drug abusers, homosexual persons, patients undergoing haemodialysis, those with haemato-oncological malignancy, frequent recipients of blood or blood products, participants
with immune compromised status, and those who were receiving
immunoglobulins for any reason). These groups were excluded
because they constitute “high-risk groups” and there already exist Cochrane reviews on some of these groups of people (Schroth
2004; Chen 2005; Lee 2006a).
Types of interventions
We analysed hepatitis B vaccine versus placebo, or no intervention, or another vaccine. We considered hepatitis B immunisation
with either type of vaccine (plasma derived or recombinant), administered by any route (intramuscularly, subcutaneously, or intradermally), using any schedule, with or without simultaneous
administration of other vaccines, administered singly or as a combination vaccine. Trials that only compared different doses, schedules or types of hepatitis B vaccine, without a comparator group
who received placebo or another vaccine or no vaccination were
excluded.
Types of outcome measures
The following outcome measures were considered.
Primary outcome
1. Hepatitis B occurrence determined by detection of HBsAg,
HBe antigen (HBeAg), HBV DNA, or antibody to HBc antigen
in serum (anti-HBc).
OBJECTIVES
To assess the benefits and harms of hepatitis B vaccination in
persons not previously exposed to hepatitis B infection or with
unknown exposure status.
METHODS
Criteria for considering studies for this review
Secondary outcomes
2. Lack of sero-protection, ie, number of vaccine recipients with
anti-HBs antibody titre less than 10 IU/L, which is generally considered protective (sufficient to prevent hepatitis B virus infection)
(Mahoney 1999; Walsh 2001).
3. Geometric mean titre (GMT) of anti-HBs antibodies.
4. Number of people developing clinical complication related to
chronic hepatitis B infection including ascites, variceal bleeding,
encephalopathy, hepatocellular carcinoma, and liver-related death.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
3
5. Number of people who developed serious adverse events. Serious adverse event was defined as per the International Conference
on Harmonisation - Good Clinical Practice guidelines (ICH-GCP
1996) as any event that would lead to death; was life-threatening;
required inpatient hospitalization; resulted in a persistent or significant disability; or any important medical event, which might
have jeopardised the patient or required intervention to prevent
it.
6. Number of people who developed local or systemic adverse
events.
7. Lack of compliance to vaccination series, ie, number of enrolled
participants who failed to complete the full vaccination series.
8. Cost-effectiveness.
If a trial reported outcomes at multiple time intervals following
vaccination series, the data at longest follow-up were included in
the analysis.
Search methods for identification of studies
We searched the following databases: The Cochrane Hepato-Biliary
Group Controlled Trials Register (searched on 17 January 2007)
(Gluud 2007a), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library Issue 2/2007); MEDLINE
(1966 to 31 March 2007); EMBASE (1980 to 31 March 2007);
LILACS (1982 to 31 March 2007), and Science Citation Index
Expanded (1945 to 31 March 2007) (Royle 2003). The search
terms used, time span of search, and date of search are shown in
Appendix 1.
We also contacted experts in the field, manufacturers of hepatitis
B vaccine, and searched the reference lists of eligible trials for
additional trials.
Data collection and analysis
We followed the instructions given in the Cochrane Handbook for
Systematic Reviews of Intervention (Higgins 2006) and TheCochrane
Hepato-Biliary Group Module (Gluud 2007a).
Searching for trials
RED conducted searches of all electronic databases; this was coordinated with The Cochrane Hepato-Biliary Review Group. RED
conducted additional searches in LILACS and Science Citation
IndexExpanded. JLM contacted the corresponding authors and
first authors of the included trials by e-mail requesting for unpublished data. This was done on 20 July 2007 through the
Cochrane Hepato-Biliary Group. Vaccine manufacturing companies were jointly identified by JLM and The Cochrane HepatoBiliary Group through an Internet search on www.google.com using the search terms ’hepatitis B vaccine’ and ’hepatitis B vaccine
manufacture# ’on 18 July 2007. Ten companies with correspondence addresses were identified in this manner in addition to the
seven identified by The Cochrane Hepato-Biliary Group. Postal
and/or fax communication was sent to these manufacturers on 25
July 2007. RED searched reference lists of eligible trials for additional trials.
Selection of trials
The initial results of the search were independently screened by
JLM and PJM, and potentially relevant trials were identified on
the basis of title, abstract, and key words/MeSH headings. Those
that appeared eligible for inclusion were obtained in full. JLM
read through these and assessed the eligibility for inclusion in the
review. RED validated the list of included and excluded trials.
Data extraction
JLM and RED extracted the following data from the selected trials:
characteristics of the trial design, participant characteristics, interventions (dosage, type, schedule), primary and secondary outcome
measures, assessment of methodological quality (risk of bias), and
sub-groups into which the trial could be included. We also marked
the additional information sought from the authors.
Assessment of bias risk
The methodological quality (and hence risk of bias due to the
methodology) of the trials was assessed following the Cochrane
methodology (Higgins 2006) and as per the recommendations of
The Cochrane Hepato-Biliary Group (Gluud 2007a). This was
done by assessment of each trial as given below.
Generation of the allocation sequence
• Adequate, if the allocation sequence was generated by a computer or random number table. Drawing of lots, tossing of
a coin, shuffling of cards, or throwing dice were also considered as adequate if a person who was not otherwise involved
in the recruitment of participants performed the procedure.
• Unclear, if the trial was described as randomised, but the
method used for the allocation sequence generation was not
described.
• Inadequate, if a system involving dates, names, or admittance
numbers were used for the allocation of patients.
Allocation concealment
• Adequate, if the allocation of patients involved a central independent unit, on-site locked computer, identically appearing numbered drug bottles or containers prepared by an independent pharmacist or investigator, or sealed envelopes.
• Unclear, if the trial was described as randomised, but the
method used to conceal the allocation was not described.
• Inadequate, if the allocation sequence was known to the
investigators who assigned participants or if the study was
quasi-randomised.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
4
Blinding
• Adequate, if the trial was described as double blind and both
the investigator assessing outcomes as well as participants
were blinded to the nature of intervention. Blinding was also
considered adequate if trials reported single blinding of the
investigator assessing outcomes.
• Unclear, if the trial was described as double blind, but the
method of blinding was not described.
• Not performed, if the trial was not double blind.
Trials were assumed to have low risk of bias if at least two of the
three quality components were adequate. Otherwise, trials were
classified as having high risk of bias.
Data analysis
Data from trials were combined for meta-analysis using Review
Manager 4.2. (RevMan 2003). We used relative risk (RR) for dichotomous data and weighted mean difference (WMD) for continuous data both with 95% confidence intervals (CI). Data were
analyzed by the fixed-effect model. If there was significant heterogeneity interpreted as defined below, the data were also analysed
using the random-effects model and both analyses are presented.
We performed intention-to-treat analysis for dichotomous data
assuming an unfavourable outcome for participants whose data
were not reported.
Heterogeneity
We qualified inconsistency among the pooled estimates using the
I2 = [(Q - df )/Q] x 100% test, where Q is the chi-squared statistic
and df is its degrees of freedom. This illustrates the percentage
of the variability in effect estimates resulting from heterogeneity
rather than sampling error. We assumed significant heterogeneity
when I2 is more than 50% and analysed such outcomes using both
a fixed-effect model and random-effects model (Higgins 2003;
Higgins 2006).
• Schedule of vaccination, ie, three doses with conventional
0,1,6 months schedule, three doses with any accelerated
schedule, and four doses by any schedule.
• Administration of hepatitis B vaccine alone versus administration as part of a combination or along with other vaccines
in the same syringe.
We used the test for interaction to estimate the difference between
two subgroups (Altman 2003).
Sensitivity analyses
We performed sensitivity analyses whenever appropriate in order
to explore causes of heterogeneity and the robustness of the results
of the primary outcome. We separated studies according to:
1. Risk of bias: including only trials with low risk of bias.
2. Attrition bias: including only trials with low drop-out rate (less
than 10%).
3. Follow-up:
(a) available data analysis, ie, including only data of those participants who were followed up and ignoring those whose data were
not reported. For this analysis, the total number at follow-up was
used as the denominator;
(b) intention-to-treat analysis assuming favourable outcome for
those whose data were not reported, ie, none of the drop-outs in
the experimental (hepatitis B vaccine) and control group had the
primary outcomes;
(c) best-case scenario favouring heaptitis B vaccine, ie, none of the
dropouts in the experimental (hepatitis B vaccine) group had the
primary outcome, but all dropouts from the control group had
the primary outcome;
(d) worst-case scenario favouring control, ie, all the dropouts from
the experimental (hepatitis B vaccine) group had the primary outcome, but none from the control group had the primary outcome.
Bias
Funnel plots were constructed for each primary outcome to test
for possible bias.
Subgroup analyses
We performed the following sub-group analyses on the primary
outcome (markers of hepatitis B occurrence) .
• Age of participants at initiation of vaccination, ie, newborns
(within first 28 days of birth), children (28 days to 10 years),
adolescents (11 to 18 years), and adults (more than 18 years).
• Presence or absence of first dose of vaccine at birth.
• Prevalence of HBsAg among the population from which
the trial participants originated. This was categorized as low
(less than 2%), moderate (2% to 7%), and high (more than
7%) prevalence of hepatitis B (Van Damme 1997; Mahoney
1999).
• Follow-up periods less than 1 year, from 1 to 5 years, and
more than 5 years.
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded
studies.
Search results
Search of the electronic databases identified 2964 citations. Of
these 328 were considered relevant for this review. Of these, 66
publications were considered potential trials for inclusion. Full
text of these was obtained and 14 publications met the inclusion
criteria for this review describing 12 trials (see Included studies
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
5
references) and 52 publications were excluded (see Characteristics
of excluded studies table). One of the included trials (Xueliang
2000) had three publications. Forty-eight experts were contacted
by e-mail on 20 July 2007 and one response was received; however,
the trial details supplied had already been included in the review.
Seventeen vaccine manufacturers were contacted by post or fax in
the last week of July 2007; one was returned owing to incomplete
address. However, no responses were received from the manufacturers. Search of the reference lists of eligible trials identified 10
potential studies; of these 9 had already been included in the review and full text of the tenth showed that it was not eligible for
inclusion.
Included trials
The twelve included trials had different participant characteristics. One included adults; mostly medical students (Ambrosch
1992), three included neonates less than 28 days old (Wang 1985;
Perrin 1986; Aristegui 1995), four recruited infants (Tsega 1990;
Giammanco 1991; Bassily 1995; Greenberg 1996), and four included older children (Coursaget 1984; Sun 1991; Liao 1999;
Xueliang 2000). None of the trials included adolescents.
Four trials were conducted in populations regarded to have low
sero-prevalence of HBsAg (Giammanco 1991; Ambrosch 1992;
Aristegui 1995; Greenberg 1996), one in a population with intermediate prevalence (Bassily 1995), and seven in populations with
high prevalence (Coursaget 1984; Wang 1985; Perrin 1986; Tsega
1990; Sun 1991; Liao 1999; Xueliang 2000).
Design of trials
Nine of the twelve trials were based in a hospital setting and three
were community based (Sun 1991; Liao 1999; Xueliang 2000);
all these three were conducted in China. Only one of the twelve
trials was conducted as a multi-centre trial (Greenberg 1996).
of administration (Perrin 1986; Tsega 1990; Sun 1991; Liao 1999;
Xueliang 2000), five used intramuscular route (Giammanco 1991;
Ambrosch 1992; Aristegui 1995; Bassily 1995; Greenberg 1996),
one used subcutaneous route (Coursaget 1984), and one trial offered vaccine by either of the two routes (Wang 1985). Four trials
reported the site of administration as deltoid muscle or anterolateral thigh (Giammanco 1991; Ambrosch 1992; Aristegui 1995;
Greenberg 1996).
Control interventions
Four trials administered placebo products to participants in the
control group; these included vaccine diluent (Liao 1999), placenta immunoglobulin (Wang 1985), or an unspecified product
(Sun 1991; Xueliang 2000). Eight trials used another vaccine in
the control group; these included hepatitis A vaccine (Ambrosch
1992) and other routine childhood vaccines (Coursaget 1984;
Perrin 1986; Tsega 1990; Giammanco 1991; Aristegui 1995;
Bassily 1995; Greenberg 1996). None of the trials used ’no intervention’ in the control group.
Follow-up
The follow-up period ranged from 1 month to 15 years. Five trials
had follow-up periods less than one year (Wang 1985; Tsega 1990;
Giammanco 1991; Ambrosch 1992; Aristegui 1995), of which
two had follow-up less than three months (Tsega 1990; Ambrosch
1992). Four trials had follow-up for one to five years (Coursaget
1984; Perrin 1986; Bassily 1995; Greenberg 1996), and three trials
followed participants for five or more years (Sun 1991; Liao 1999;
Xueliang 2000).
Risk of bias in included studies
Bias risk
Experimental interventions
Four trials used a plasma derived hepatitis B vaccine (Wang 1985;
Tsega 1990; Liao 1999, Xueliang 2000); six used a recombinant hepatitis B vaccine (Coursaget 1984; Giammanco 1991;
Ambrosch 1992; Aristegui 1995; Bassily 1995; Greenberg 1996),
and two trials did not specify the type of hepatitis B vaccine used
(Perrin 1986; Sun 1991).
The twelve trials incorporated 12 different vaccination schedules;
four trials used the conventional 0,1,6 months schedule (Sun
1991; Ambrosch 1992; Liao 1999; Xueliang 2000), four trials incorporated a four-dose schedule (Coursaget 1984; Wang 1985;
Tsega 1990; Giammanco 1991). Two trials also had arms that
compared different hepatitis B vaccine schedules with a control
group (Giammanco 1991; Aristegui 1995). The dosage of hepatitis B vaccine in the trials varied from 2 µg to 40 µg of hepatitis
B surface antigen. Two trials (Sun 1991; Xueliang 2000) did not
mention the dosage of HBsAg. Five trials did not specify the route
None of the twelve trials met the criteria for high methodological
quality and hence low risk of bias. Almost all of the trials had unclear allocation generation, allocation concealment, and blinding
methodology (see Characteristics of included studies table).
Description of drop-outs and withdrawals
All trials had significant drop-outs; the reasons for this were not
described and intention-to-treat analysis was not used in any of
the trials.
Effects of interventions
Primary outcomes
Hepatitis B occurrence
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
6
Presence of HBsAg
Nine of the twelve trials included HBsAg as an outcome. However,
five trials did not clearly report the exact number of events and/or
participants in each intervention group. Of these, two (Bassily
1995; Giammanco 1991) did not report findings in either arm of
the trial, in two trials the outcome was assessed in a fraction of
subjects, but the relationship of the fraction to the original number enrolled was not clear (Coursaget 1984; Sun 1991). One trial
did not specify the denominator (Perrin 1986). Thus, only four
trials reported data which could be included in the meta-analysis.
Hepatitis B vaccination had an unclear effect on the risk of HBsAg presence (RR 0.96, 95% CI 0.89 to 1.03, 4 trials with 1230
participants, I2 = 0%). Analysis of data of available participants
showed RR 0.12, 95% CI 0.03 to 0.44, 4 trials with 576 participants, I2 = 0%.
HBeAg and HBV DNA
None of the included trials reported these markers of infectivity
as an outcome.
Anti-HBc antibodies
Eight trials included serum anti-HBc antibodies (anti-HBc) as
an outcome, but only four trials reported data that could be included in the meta-analysis (Wang 1985; Tsega 1990; Liao 1999;
Xueliang 2000). These trials also reported presence of HBsAg.
Two trials (Giammanco 1991; Bassily 1995) did not report the
outcome in either arm, one did not specify the denominator in either arm (Coursaget 1984), and one presented this outcome as antibody level rather than presence or absence (Perrin 1986). Three
of the trials whose data could be included in the meta-analysis
were in children (Tsega 1990; Liao 1999; Xueliang 2000) and one
in newborn babies (Wang 1985). Hepatitis B vaccination showed
an unclear effect on the risk of presence of anti-HBc antibodies
(RR 0.81, 95% CI 0.61 to 1.07, 4 trials with 1230 participants;
I2 = 84.4%, random-effects and RR 0.86, 95% CI 0.80 to 0.92,
fixed-effect). When data of available participants were analysed, it
showed RR 0.36, 95% CI 0.17 to 0.76, 4 trials with 576 participants, I2 = 55.9%, random-effects and RR 0.38, 95% CI 0.28 to
0.52, fixed-effect.
Subgroup analyses
Analysis by age of participants showed unclear effect on the risk
of HBsAg presence in children (RR 0.96, 95% CI 0.89 to 1.03,
3 trials with 1210 participants, I2 = 0%). One trial with newborn
babies did not report any event in either arm. Test of interaction
revealed no significant difference (P > 0.05).
Subgroup analysis by presence or absence of birth dose of hepatitis B vaccine could not be done as one trial (Wang 1985) did
not observe any events in either arm and two trials (Perrin 1986;
Aristegui 1995) did not report the primary outcome.
Subgroup analysis comparing different HBsAg prevalence was not
possible as all four trials were conducted in populations with high
prevalence (more than 8%).
Analysis by duration of follow-up showed unclear effect on the
risk of developing HBsAg in trials with less than one year followup (RR 0.76, 95% CI 0.41 to 1.40, 2 trials with 320 participants)
as well as in trials with more than five years follow-up (RR 0.97,
95% CI 0.90 to 1.04; 2 trials with 910 participants, I2 = 0%).
Test of interaction revealed no significant difference (P > 0.05).
Analysis by vaccination schedule showed unclear effect with three
doses of hepatitis B vaccine administered by the conventional 0,1,6
months schedule (RR 0.97, 95% CI 0.90 to 1.04, 2 trials with 910
participants, I2 = 0%), and the four dose schedule (RR 0.76, 95%
CI 0.41 to 1.40, 2 trials with 320 participants). Test of interaction
revealed no significant difference between these subgroups (P >
0.05).
Subgroup analysis comparing administration of hepatitis B vaccine
alone compared to part of a combination could not be done as the
included trials did not report this aspect.
Sensitivity analysis
Methodological quality (risk of bias)
Sensitivity analysis based on methodological quality could not be
performed as all trials were of low methodological quality, ie, had
high risk of bias.
Attrition
Only one trial had less than 10% drop-outs (Wang 1985), but
there were no events in either arm, hence could not be analysed.
Follow-up
Analysis of data of available participants showed that hepatitis B
vaccination reduced risk of developing HBsAg (RR 0.12, 95% CI
0.03 to 0.44, 4 trials with 576 participants, I2 = 0%).
Intention-to-treat analysis assuming that all participants whose
data were not reported had a favourable outcome, ie, none developed HBsAg, showed identical reduction in risk (RR 0.12, 95%
CI 0.03 to 0.44, 4 trials with 1230 participants, I2 = 0%). Bestcase senario analysis showed reduced risk (RR 0.01, 95% CI 0.00
to 0.03, 4 trials with 1230 participants, I2 = 18.3%).
Worst case-scenario showed RR 9.45, 95% CI 2.14 to 41.67, 4
trials with 1230 participants, I2 = 90.7%, random-effects model
and RR 15.59, 95% CI 10.29 to 23.64, fixed-effect model.
Table 1 summarises the relative risk using different analyses models
as described.
Table 1. Table 1. Primary outcomes using different methods of analysis
Comparison
Trials
Participants
RR (95% CI)
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
I2
7
Table 1. Table 1. Primary outcomes using different methods of analysis (Continued )
HBsAg (ITT assuming unfavourable outcome for all drop-outs)
4
1230
0.96 (0.89-1.03)
0%
HBsAg (ITT analysis assuming favourable outcome for all drop-outs)
4
1230
0.12 (0.03-0.44)
0%
HBsAg (analysis of available participants).
4
576
0.12 (0.03-0.44)
0%
HBsAg (best-case scenario).
4
1230
0.01 (0.00-0.03)
18.3%
HBsAg: worst-case scenario (random-effects model).
4
1230
9.45 (2.14-41.67)
90.7%
HBsAg: worst-case scenario (fixed-effect model).
4
1230
15.59 (10.29-23.64)
90.7%
Anti-HBc antibodies (ITT analysis assuming unfavourable outcome for all
drop-outs) Random-effects model
4
1230
.0.81 (0.61-1.07)
84.4%
Anti-HBc antibodies (ITT analysis assuming unfavourable outcome for all
drop-outs) Fixed-effect model.
4
1230
0.86 (0.80-0.92)
84.4%
Anti-HBc antibodies (ITT analysis assuming favourable outcome for all
drop-outs).
4
1230
0.45 (0.25-0.50)
42%
Anti-HBc antibodies (analysis based on available participants). Fixed-effect
model.
4
576
0.38 (0.28-0.52)
55.9%
Anti-HBc antibodies (analysis based on available participants). Randomeffects model.
4
576
0.36 (0.17-0.76)
55.9%
Anti-HBc antibodies: best-case scenario (random-effects model).
4
1230
0.08 (0.01-1.03)
96.7%
Anti-HBc antibodies: best-case scenario (fixed-effect model).
4
1230
0.08 (0.06-0.12)
96.7%
Anti-HBc antibodies: worst-case scenario (random-effects model).
4
1230
3.45 (0.38-31.57)
99%
Anti-HBc antibodies: worst-case scenario (fixed-effects model).
4
1230
3.66 (3.03-4.42)
99%
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
8
Sensitivity analysis on anti-HBc antibodies showed reduction in
risk based on analysis of data of available participants (RR 0.36,
95% CI 0.17 to 0.76, 4 trials with 576 participants, I2 = 55.9%,
random-effects model and RR 0.38, 95% CI 0.28 to 0.52, fixedeffect model).
Intention-to-treat analysis assuming that all participants whose
data were not reported had a favourable outcome, ie, none developed anti-HBc at follow-up showed similar result (RR 0.35, 95%
CI 0.25 to 0.50, 4 trials with 1230 participants, I2 = 42%). Bestcase scenario analysis showed RR 0.08, 95% CI 0.01 to 1.03, 4
trials with 1230 participants, I2 = 96.7%, random-effects model
and RR 0.08, 95% CI 0.06 to 0.12, fixed-effect model.
Worst-case scenario analysis showed RR 3.45, 95% CI 0.38 to
31.57, 4 trials with 1230 participants, I2 = 99.0%, random-effects
model and RR 3.66, 95% CI 3.03 to 4.42, fixed-effect model.
Bias
The funnel plots appeared symmetric and did not suggest significant bias. Figure 1 shows the funnel plot for HBsAg at follow-up
for all trials reporting the outcome, and Figure 2 shows the funnel
plot for anti-HBc in serum.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
9
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
10
Secondary outcomes
Lack of sero-protection
Nine trials included lack of sero-protection as an outcome. Of
these, six did not measure antibody level in the control group
(Giammanco 1991; Sun 1991; Ambrosch 1992; Aristegui 1995;
Bassily 1995; Greenberg 1996). Three trials reported sero-protection; meta-analysis showed an unclear effect of vaccination (RR
0.57, 95% CI 0.26 to 1.27, 3 trials with 1210 participants, I2 =
99.1%, random-effects model); however, the fixed-effect model
showed RR 0.73, 95% CI 0.69 to 0.78.
Geometric mean titre (GMT) of anti-HBs antibodies
Six trials measured the GMT of anti-HBs antibodies, but only
one reported data in both arms of the trial (Xueliang 2000) and
showed a mean GMT of 9.71 among vaccinees and 1.23 among
controls. However, standard deviation was neither reported nor
could be calculated. The other five trials did not report the GMT
in participants of the control group (Tsega 1990; Giammanco
1991; Ambrosch 1992; Aristegui 1995; Bassily 1995).
Clinical complications attributable to hepatitis B infection
None of the trials reported an outcome that could be included as
a clinical complication of hepatitis B infection.
Serious adverse events
One trial (Greenberg 1996) reported three serious adverse events;
mortality (all-cause), seizures, and hospitalisation. Hepatitis B vaccination did not significantly affect all-cause mortality (RR 1.16,
95% CI 0.47 to 2.84, 10317 participants). Data for seizures and
hospitalisation were not reported in the control group. Another
trial (Sun 1991) mentioned serious adverse events among hepatitis B vaccine recipients, but did not report the number in either
group.
Systemic, local, and other adverse events
The reported systemic adverse events were fever (Aristegui
1995; Greenberg 1996), headache (Ambrosch 1992), diarrhoea
(Aristegui 1995), etc. The trials recorded adverse events by providing diary cards to care-givers of children or participants themselves
(adults). Six trials recorded local adverse events such as redness
at injection site (Aristegui 1995), local pain (Giammanco 1991),
soreness, swelling (Aristegui 1995), etc. One trial reported clinical
hepatitis as an outcome of interest (Xueliang 2000) and another
(Ambrosch 1992) reported levels of hepatic enzymes as an outcome.
For the meta-analysis, adverse events were analysed in three groups:
systemic, local, and other. The denominator for these was calculated based on the number of potential events (calculated as number of participants multiplied by the number of doses). The risk of
developing fever with hepatitis B vaccine was not significantly affected (RR 1.42, 95% CI 0.60 to 3.38, 2 trials with 14169 potential events, I2 = 92.3%, random-effects model). The fixed-effect
model showed RR 1.03, 95% CI 0.90 to 1.18. The two trials compared hepatitis B vaccine co-administered with other childhood
vaccines versus routine childhood vaccines alone (Bassily 1995;
Greenberg 1996).
Local reactions included pain, redness, and soreness, singly or in
combination. The risk of developing local reactions was significantly increased among hepatitis B vaccine recipients (RR 2.93,
95% CI 1.72 to 4.99, 2 trials with 1904 potential events, I2 =
36.9%). The two trials compared hepatitis B vaccine either alone
or co-administered with other childhood vaccine versus routine
childhood vaccine (Bassily 1995; Giammanco 1991).
One trial reported headache (Ambrosch 1992), and the risk of
developing this event was unclear (RR 2.33, 95% CI 0.64 to 8.56,
1 trial with 110 participants). The comparator group received
hepatitis A vaccine.
Lack of compliance
Six trials reported lack of compliance among enrolled participants.
However, three did not report data for controls (Giammanco 1991;
Greenberg 1996; Liao 1999) and one (Xueliang 2000) mentioned
lack of compliance in 4 of 265 participants without specifying the
group(s) to which the four participants belonged. The two trials
with data that could be combined in meta-analysis had controls
that received routine childhood vaccinations. Although the risk
of failure to comply was remarkably different in the two trials,
the pooled effect showed no significant difference in the risk of
failing to comply with hepatitis B vaccination series as compared
to controls (RR 0.91, 95% CI 0.03 to 24.39, 2 trials with 1976
participants, I2 = 98.0%, random-effects model). The fixed-effect
model showed RR 0.26, 95% CI 0.21 to 0.32.
Cost-effectiveness
There were no randomised clinical trials comparing cost effectiveness.
DISCUSSION
Justification for this systematic review
Hepatitis B vaccination has become increasingly popular, especially because it is recommended both for universal vaccination
and protection of high-risk groups by the World Health Organization and many professional bodies. Therefore, the very idea of
performing a systematic review to assess the benefits and harms
of this vaccine may appear strange to some people. Some critics
could suggest that this review was undertaken for the purpose of
’proving a foregone conclusion in favour of the vaccine’, making it an unnecessary exercise. Skeptics of the concept and process of evidence-based medicine may also argue that irrespective
of the conclusions of the review, one cannot but continue with
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
11
hepatitis B vaccination; hence such a review is nothing more that
a frivolous scientific exercise. The authors of this review may even
be (mis)labeled as evidence-based medicine fanatics demanding
proof for ’the obvious’.
We must emphasize that despite being aware of such opinions,
we have undertaken this review for three important reasons. The
first aim was to test whether the popular belief in the beneficial
effects of hepatitis B vaccine can be proven (or otherwise) in a scientifically rigorous manner. Second and more important, we have
tried to assess whether there is evidence to answer a central question that is missed in most trials on hepatitis B vaccine, namely,
whether hepatitis B vaccine prevents the occurrence of hepatitis
B, in other words the protective efficacy of the vaccine. Numerous
trials have implied this conclusion through the use of surrogate
outcome measures, especially sero-protection (development of a
’protective’ titre of antibodies) within a limited time frame (usually
four weeks after the last vaccine dose), but this review shows that
only a limited number of trials have evaluated protective efficacy
measuring relevant outcomes. We believe that this is an important
issue to resolve before recommending whether or not hepatitis B
vaccine should be used in people not previously exposed to the
infection. Thirdly, we have tried to assess the benefits and harms of
hepatitis B vaccine with reference to several other outcome measures such as adverse events, cost effectiveness, and compliance
pattern among recipients. These issues also are important when
hepatitis B vaccination is being considered in a country or region.
Explanation of the methodology used in this
systematic review
This review is primarily focused on assessing the benefits and
harms of hepatitis B vaccination in people who are either hepatitis
B surface antigen negative, or have unknown exposure status; because these groups comprise the vast majority of the world’s population. We have deliberately excluded trials in ’high-risk’ groups
since their response to the vaccine could be different from the socalled ’low-risk’ patient group. Therefore this review is likely to be
helpful for countries planning to initiate hepatitis B vaccination
for ’low risk people’.
Many vaccination trials recruit and randomise a reasonably large
number of participants, but fail to follow all of them. Some trials
report results in only a small fraction of those randomised, failing
to account for the missing participants. Since vaccination cannot
be assumed to have only beneficial effects, ignoring the participants whose data is not reported can affect the results of the systematic review. There are a number of ways of dealing with this
problem, the least appropriate being to ignore it and use only the
data of available participants during data analysis. Other analysis
models are intention-to-treat analysis assuming unfavourable or
favourable outcomes for all participants whose data are missing,
in both arms of trials. Another possible analysis could be the bestcase scenario and worst-case scenario models. Yet other analysis
could be done, by assuming that a proportion of dropouts (in
each arm of the trials) had favourable outcomes, while the rest
had unfavourable outcomes. However it is not clear how the correct proportion can be calculated, hence this was not a realistic
option. We also felt that intention-to-treat analysis merely carrying forward last-reported data is inaccurate because the primary
outcomes considered in this review can change with time. It is
difficult to judge which of these analysis models is likely to give
the most correct result that represents the ’true’ effect. Therefore,
we decided a priori to use an intention-to-treat analysis assuming
unfavourable outcomes for all missing data and undertake two
sensitivity analyses, ie, analysis of data of available participants at
follow-up and intention-to-treat analysis assuming that all participants in either arm whose data were not reported had a favourable
outcome.
At the time of initiating this review, we did not realize that most
of the trials eligible for inclusion had a large number of participants whose data was neither reported nor its absence explained.
Since most trials reported outcomes in only a small fraction of randomised participants, the Contact Editors of this review opined
that our original analysis using the intention-to-treat model (assuming unfavourable outcome for all participants whose data were
not reported) and the two sensitivity analyses (available data analysis and intention-to-treat analysis assuming favourable outcomes
for all missing data) would not be appropriate because the pooled
effect could vary depending on the number of participants whose
data are missing, in eiether arm of the trials. They therefore suggested that the data of available participants be presented as the
primary analysis and the analysis described in the published protocol of this review be reported as a sensitivity analysis. However,
this suggestion has not been accepted by us for two reasons. First,
the published protocol specified the analysis performed by us a
priori and second, we believe that it is inappropriate to change
the reporting format after the results are evident. However, on the
suggestion of the Contact Editors, two additional sensitivity analyses have been conducted and reported, ie, the best-case scenario
and worst-case scenario. Although these have not been mentioned
in the protocol, the Handbook describes both these analyses as
methods to deal with data that is not reported. The four sensitivity
analyses are presented only for the primary outcome, ie, ocurrence
of hepatitis B infection.
Since it is not clear which of the analysis models is closest to
the ’truth’, we recommend that all systematic reviews on vaccine
efficacy report results using all these analysis models.
Conclusions that can be drawn from this review
This systematic review demonstrates that hepatitis B vaccination
has an unclear effect on the risk of developing hepatitis B virus
markers of infection (HBsAg and anti-HBc antibodies) in participants without previous knowledge about exposure to hepatitis B
infection. However, alternate methods of analysis using the data of
available participants shows a clear and dramatic reduction in the
risk of developing hepatitis B infection. A similar result is observed
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
12
with an analysis assuming favourable outcomes for all those whose
data are missing and the best-case scenario. Although it is clear
that none of these analyses represents the perfect method of analysis, it is difficult to specify which of these gives the most accurate
result. Whichever method of analysis is chosen to represent the
truth, one must realize that the findings are based on a small number of methodologically poor, randomised clinical trials. Therefore, based on the findings in this review, countries planning investments in a vaccination programme should exercise individual
judgement on the expected magnitude of benefit vis a vis the cost
of a vaccination programme. Although this is unintentional, this
review cannot provide any further guidance on this subject.
It should further be noted that the methodological quality of the
included trials was generally low and hence the risk of bias is high.
Furthermore, a number of trials did not report relevant outcomes
or reported outcomes unclearly which raises the risk of outcome
measure reporting bias (Chan 2004a; Chan 2004b; Chan 2005;
Furukawa 2007) although, these observations usually increase the
risk of positive bias. Yet another point to be considered is that
the trials in this review recruited participants who were HBsAg
negative and not necessarily those with unknown status. Since a
vaccination programme will also include participants in the latter
category (who may be HBsAg positive unless proven through a
screening programme and excluded from vaccination), the overall
benefit of a vaccination programme may not be similar to that
expected from the findings of this review.
It is worth remembering that clinical trials are not the only way
to determine vaccine efficacy. For many vaccines, data from clinical trials may need to be considered along with other important
information over a period of time, such as the change in morbidity and mortality statistics with the inception of a vaccination
programme, observation(s) of disease epidemiology when vaccination programmes are suspended or discontinued for any reason,
the relationship between attack rate of disease and the proportion
of vaccinated individuals in a community, etc. These important
assessments were outside the scope of this review.
This review has also shown that there is an unclear effect on the
presence of protective level of anti-HBs antibodies following vaccination. One way of interpreting this finding could be to assert
that it supports the theory that protection depends on immunological memory and not on the presence of protective antibody
levels (Chan 1991; West 1993). The counter argument is that
immunological memory itself depends on the development of a
protective level of antibodies after vaccination (sero-protection),
that may later decline or become undetectable. Therefore, some
experts may consider the lack of sero-protection as unacceptable.
Since this review is not designed to demonstrate whether sero-protective antibody levels decline with time (or otherwise), we suggest caution in interpreting results of anti-HBs antibodies. This
is another reason why trials should report occurrence of infection
based on clinical or laboratory measurements rather than surrogate outcomes such as antibody levels (Gluud 2007b).
There is lack of robust data to demonstrate whether or not hepatitis B vaccination is associated with higher or lower risk of serious adverse events as well as non-serious adverse events. Caution
must be exercised because trials reporting adverse events administered other childhood vaccines in addition to hepatitis B. However, one trial (Bassily 1995) that used concomitant childhood
vaccines showed a lower risk of fever among hepatitis B vaccinees,
indirectly suggesting a protective effect of the vaccine. However,
such an interpretation needs verification.
The impact of hepatitis B vaccine on vaccination compliance is
difficult to judge since only two trials reported this and showed
very different results.
It is disappointing that there are no randomised trials to determine
development of clinical complications of hepatitis B as well as cost
effectiveness; both these are important issues in deciding vaccination guidelines at the local, national, and regional levels. However,
such information may not be available through randomised trials
and could be evaluated through other study designs that are outside the scope of the present review.
Strengths and limitations of this review
Some of the strengths of this systematic review are that we assessed
the effect of hepatitis B vaccine on multiple outcome measures, the
most important being protective efficacy. We did this by considering whether vaccination can prevent infection, rather than merely
evaluating antibody levels or sero-conversion following vaccination. For this, we planned to analyse all serological markers of hepatitis B infection. Protective efficacy is the real proof of the beneficial effect of any vaccine; all other potential surrogate measurements are only indirect proof (Gluud 2007b). Another strength
of this review is that we evaluated the effect of vaccination at the
longest follow-up reported. This is important because hepatitis B
can result in persistent infection with risk of chronic liver disease
and its complications. Therefore, the real test of the vaccine is
whether or not it can prevent infection in the long term and not
only the short term (in contrast to some other vaccines where the
infection predominantly affects children). We also determined the
safety profile in terms of serious and non-serious adverse events.
Besides these, we made efforts to determine cost effectiveness of
vaccinating versus not vaccinating.
We performed a comprehensive search that included contacting
experts and vaccine manufacturers, and unlike most Cochrane reviews, we have reported the process and outcome of these additional searches. At least two authors worked independently at all
stages of the review.
We also performed a rigorous intention-to-treat analysis assuming unfavourable outcome for participants whose data were not
reported in the various trials. This was done in order to avoid the
bias that could be created by ignoring the number of drop-outs
from the trials. The importance of this has been described above.
Besides this, we also performed four sensitivity analyses besides
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
13
analysis of subgroups of importance. Our methodology has been
explicit and transparent.
Nevertheless we acknowledge that this review has some important
limitations, particularly that all included trials had low methodological quality, ie, had high risk of bias. Furthermore, the risk of
bias is further increased by the fact that many of the included trials
did not report on outcomes that seemed highly relevant to report
on. This could lead to outcome reporting bias (Chan 2004a; Chan
2004b; Chan 2005; Furukawa 2007), which may also lead to significant overestimation of intervention effects.
We did not include publications that reported non-randomised
trial design, some of which may be relevant, despite the high risk
of bias.
Paucity of high quality data
This review highlights the paucity of methodologically sound trials
(trials with low risk of bias) to address the central issue of whether
or not hepatitis B vaccination protects against hepatitis B. As is
the case with many vaccine trials, most hepatitis B vaccine trials
also report limited outcomes such as antibody levels one month
after vaccination along with recording a few adverse events. It has
already been emphasised that post-vaccination antibodies are only
a surrogate outcome that is believed or expected to correlate with
protective efficacy.
Our review highlights several gaps in existing knowledge about
hepatitis B vaccine. Some of these include information about serious adverse events, the concentration of antibodies generated by
vaccination, duration of protective effect, the risk of developing
infection if exposed during the vaccination series and the nature
of long-term protection afforded by vaccination. A major missing
piece is the cost-effectiveness as compared to not vaccinating a
cohort of individuals.
Design of future trials
This review gives information that can help to design future trials
in order to enhance knowledge about hepatitis B vaccine and its
benefits or harms. Ideally, future trials should be conducted on
low risk people by including those with unknown exposure status
who form the vast majority of the population as well as those already known to be hepatitis B surface antigen negative, without
resorting to a screening programme. This would make the participants in the trial similar to the population where a vaccination programme is proposed (should beneficial effect be demonstrated).
Furthermore, the participants should be recruited consecutively
without resorting to very strict exclusion criteria as is often the
case with randomised trials. Such a trial (assuming the usual alpha of 0.05 and power of 0.80) in a population with low prevalence (2%) would require 861 participants in each arm, to demonstrate a reduction in HBsAg by 75%, ie, to 0.5%. Demonstration
of a similar reduction in a population of intermediate prevalence
(4%) would require 424 participants in each arm; 279 participants
would be required in a population with 6% prevalence. For a high
prevalence population of 8%, 206 subjects in each arm will be sufficient to demonstrate 75% reduction in risk. Trials should address
protective efficacy by measuring clinical and serological outcomes
that suggest hepatitis B infection indisputably (HBsAg, anti-HBc,
HBV DNA), rather than limiting to surrogate measures of efficacy (sero-protection, GMT of antibodies, etc). These measurements should be performed serially, for the longest duration possible making effort to restrict drop-outs and carefully accounting
for any that do occur. However, it is difficult to spell out for how
long such trials must be conducted, since it depends on resources
available. Intention-to-treat analysis and calculation of numbers
needed-to-treat and numbers needed-to-harm will help to get a
clearer picture of the effects of hepatitis B vaccination. It need
not be emphasised that future trials ought to be conducted and
reported according to the recommendations of the CONSORT
Statement (www.consort-statement.org). However, it should be
recognised that the practicalities of actually undertaking and completing such a trial, especially in the low-risk population, are substantial.
Issues that need to be resolved
Certain issues were outside the scope of this review. These include
comparison of various doses of hepatitis B vaccine, ideal number
of doses, best vaccination schedule, comparison of different routes
of vaccination, comparison between vaccinating separately versus
combining with other vaccines, relevance of a birth dose in infancy,
duration of protection, impact on event-free survival, etc. These
could become the topic of future trials and reviews.
Updating this review
We believe (and hope) that this review will generate considerable
interest and correspondence from various quarters, particularly industry. We also recognise that it might bring to light some relevant
trials that have not been disclosed so far. Recognising that additional data will be very useful to get a clearer understanding, we
propose to submit the first update of this review, one year following its initial publication, instead of the more common practice
of waiting for two to three years.
AUTHORS’ CONCLUSIONS
Implications for practice
Current evidence suggests that hepatitis B vaccination has an unclear effect on the risk of developing hepatitis B infection in persons not previously exposed to hepatitis B or with unknown exposure status.
Implications for research
There are several gaps in the knowledge about hepatitis B vaccination including serious adverse events, systemic adverse events, best
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
14
dose, best vaccination schedule, and long-term protective effect.
The low methodological quality of available data argues strongly
in favour of designing new randomised trials that can cover the
lacunae in existing data.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the constant encouragement
and assistance offered by Dimitrinka Nikolova. We are also grateful to Sarah Louise Klingenberg for assistance in developing and
conducting the search strategies.
Peer Reviewers: G Bjelakovic, Serbia; G Aloj, Italy.
Contact Editors: RL Koretz, USA, and C Gluud, Denmark.
REFERENCES
References to studies included in this review
Ambrosch 1992 {published data only}
Ambrosch F, Andre FE, Delem A, D’Hondt E, Jonas S, Kunz C, et
al.Simultaneous vaccination against hepatitis A and B: results of a
controlled study. Vaccine 1992;10(Suppl 1):S142–5.
Aristegui 1995 {published data only}
Aristegui J, Muñiz J, Pérez Legorburu A, Imaz M, Arrate JP, Suárez
MD, et al.Newborn universal immunisation against hepatitis B: immunogenicity and reactogenicity of simultaneous administration of
diphtheria/tetanus/pertussis (DTP) and oral polio vaccines with hepatitis B vaccine at 0, 2 and 6 months of age. Vaccine 1995;13(11):
973–7.
Bassily 1995 {published data only}
Bassily S, Kotkat A, Gray G, Hyams KC, Brown FM, Imam IZ,
et al.Comparative study of the immunogenicity and safety of two
dosing schedules of hepatitis B vaccine in neonates. American Journal
of Tropical Medicine and Hygiene 1995;53(4):419–22.
Coursaget 1984 {published data only}
Coursaget P, Deciron F, Tortey E, Barin F, Chiron JP, Yvonnet B, et
al.Immune response to hepatitis B vaccine in infants and newborns:
control trial in an endemic area (Senegal). IARC Scientific Publications
1984;63:319–35.
Giammanco 1991 {published data only}
Giammanco G, Li Volti S, Mauro L, Bilancia GG, Salemi I, Barone
P, et al.Immune response to simultaneous administration of a recombinant DNA hepatitis B vaccine and multiple compulsory vaccines
in infancy. Vaccine 1991;9(10):747–50.
Greenberg 1996 {published data only}
Greenberg DP, Vadheim CM, Marcy SM, Partridge S, Jing J, Chiu
CY, et al.Safety and immunogenicity of a recombinant hepatitis B
vaccine administered to infants at 2, 4 and 6 months of age. The
Kaiser-UCLA Vaccine Study Group. Vaccine 1996;14(8):811–6.
Liao 1999 {published data only}
Liao SS, Li RC, Li H, Yang JY, Zeng XJ, Gong J, et al.Long-term
efficacy of plasma-derived hepatitis B vaccine: a 15-year follow-up
study among Chinese children. Vaccine 1999;17(20-21):2661–6.
Perrin 1986 {published data only}
Perrin J, Coursaget P, Ntareme F, Chiron JP. Hepatitis B immunization of newborns according to a two dose protocol. Vaccine 1986;4
(4):241–4.
Sun 1991 {published data only}
Sun Z, Zhu Y, Stjernsward J, Hilleman M, Collins R, Zhen Y, et
al.Design and compliance of HBV vaccination trial on newborns to
prevent hepatocellular carcinoma and 5-year results of its pilot study.
Cancer Detection and Prevention 1991;15(4):313–8.
Tsega 1990 {published data only}
Tsega E, Tafesse B, Nordenfelt E, Wolde-Hawariat G, Hansson BG,
Lindberg J. Immunogenicity of hepatitis B vaccine simultaneously
administered with the expanded programme on immunisation (EPI).
Journal of Medical Virology 1990;32:232–5.
Wang 1985 {published data only}
Wang YJ, Kang Y, Gong ZL, Mei YX, Xu JY, Meng LX, et al.Hepatitis
B vaccine in neonates Induction of anti-HBs immune response and
interruption of maternal HBsAg transmission. Chinese Medical Journal 1985;98(4):265–70.
Xueliang 2000 {published data only}
∗
Xueliang W, Huiwen X, Guihua Z. Long-term efficacy and immunological memory of plasma-derived hepatitis B vaccine 11 years
after initial inoculation. Journal of Xi’An Medical University, English
Edition 2000;12(2):122–5.
Xu H, Zhuang G, Wang X, Chen Q, Chen Z. Efficacy and immune
memory of plasma-derived hepatitis B vaccine 11 years after primary
immunization. Zhonghua yu fang yi xue za zhi (Chinese Journal of
Preventive Medicine) 2000;34(2):113–5.
Xu H, Zuo H, Zhuang G. Evaluation of the effectiveness nine years
after primary immunization with local produced plasma-derived hep-
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
15
atitis B vaccine. Zhonghua yu fang yi xue za zhi [Chinese journal of
preventive medicine] 1998;32(4):205–7.
References to studies excluded from this review
Aach 1982 {published data only}
Aach RD. Hepatitis prophylaxis. The Johns Hopkins Medical Journal
1982;151(6):318–25.
Abbas 1999 {published data only}
Abbas R, Ahenad V. Evaluation of the immunogenicity of hepatitis
B vaccine in students of the Isfahan university of medical sciences.
Asian Pacific Journal of Allergy Immunology 1999;17 Suppl 1:25.
Abdurrahman 1984 {published data only}
Abdurrahman MB. Hepatitis B infection and immunization: A review. Saudi Medical Journal 1984;5(4):369–76.
Akbar 1997 {published data only}
Akbar SM, Kajino K, Tanimoto K, Kurose K, Masumoto T, Michitaka K, et al.Placebo-controlled trial of vaccination with hepatitis B
virus surface antigen in hepatitis B virus transgenic mice. Journal of
Hepatology 1997;26(1):131–7.
Akram 2005 {published data only}
Akram DS, Maqbool S, Khan DS, Jafri R, Randhawa S, ValenzuelaSilva C, et al.Immunogenicity of a recombinant, yeast-derived, antihepatitis B vaccine after alternative dosage and schedule vaccination
in Pakistani children. Vaccine 2005;23(50):5792–7.
vaccines administered to students in three different doses containing
half the antigen amount routinely used for adult vaccination. Revista
do Instituto de Medicina Tropical de Sao Paulo 2004;46(2):103–7.
Chunsuttiwat 1997 {published data only}
Chunsuttiwat S, Biggs BA, Maynard J, Thamapalo S, Laoboripat
S, Bovornsin S, et al.Integration of hepatitis B vaccination into the
expanded programme on immunization in Chonburi and Chiangmai
provinces, Thailand. Vaccine 1997;15(6-7):769–74.
Coutinho 1983 {published data only}
Coutinho RA, Lelie N, Albrecht-Van Lent P, Reerink-Brongers EE,
Stoutjesdijk L, Dees P, et al.Efficacy of a heat inactivated hepatitis B
vaccine in male homosexuals: outcome of a placebo controlled double
blind trial. BMJ (Clinical Research Ed.) 1983;286(6374):1305–8.
Coutinho RA, Lelie PN, Albrecht-Van Lent P, Stoutjesdijk L, Huisman J, Kuipers H, et al.Efficacy trial of heat-inactivated hepatitis
B vaccine (CLB) in male homosexuals in the Netherlands. Developments in Biological Standardization 1983;54:287–92.
∗
Duclos 2003 {published data only}
Duclos P. Safety of immunization and adverse events following vaccination against hepatitis B. Journal of Hepatology 2003;39:S83–8.
Fortuin 1993 {published data only}
Fortuin M, Chotard J, Jack AD, Maine NP, Mendy M, Hall AJ,
et al.Efficacy of hepatitis B vaccine in the Gambian expanded programme on immunization. Lancet 1993;341(8853):1129–31.
Alikasifoglu 2001 {published data only}
Alikasifoglu M, Cullu F, Kutlu T, Arvas A, Tastan Y, Erginöz E,
et al.Comparison study of the immunogenicity of different types
and dosages of recombinant hepatitis B vaccine in healthy neonates.
Journal of Tropical Pediatrics 2001;47:60–2.
Gomez 2002 {published data only}
Gomez J, Vesga BE, Bucaramanga I. No English title [Eficacia de la
vacunacin contra hepatitis B en trabajadores de la salud en un hospital
colornbiano]. Acta Medica Colombiana 2002;27:103–7.
Ambrosch 1994 {published data only}
Ambrosch F, Wiedermann G, André FE, Delem A, Gregor H, Hofmann H, et al.Clinical and immunological investigation of a new
combined hepatitis A and hepatitis B vaccine. Journal of Medical Virology 1994;44(4):452–6.
Guihua 1997 {published data only}
Guihua Z, Huiwen X, Xueliang W, Hong Z, Pengbo L, Lingbin K, et
al.Efficacy of plasma-derived hepatitis B vaccine produced by China 9
years after initial injection. Journal of Xi’an Medical University, English
Edition 1997;9(2):144–50.
Artan 2004 {published data only}
Artan R, Erol M, Velipasaoglu S, Yegin O. The effect of concurrent
use of hepatitis B and Bacille Calmette-Guerin vaccination on antihepatitis B response. Saudi Medical Journal 2004;25(12):1939–42.
Guihua 1998 {published data only}
Guihua Z, Huiwen X, Hong Z, Xueliang W, Pengbo L, Lingbin K.
Six-year efficacy of hepatitis B revaccination: a double-blind, placebocontrolled and randomized field trial. Journal of Xi’an Medical University, English Edition 1998;10(1):44–8.
Assateerawatt 1993 {published data only}
Assateerawatt A, Tanphaichitr VS, Suvatte V, Yodthong S. Immunogenicity and efficacy of a recombinant DNA hepatitis B vaccine, GenHevac B Pasteur in high risk neonates, school children and healthy
adults. Asian Pacific Journal of Allergy Immunology 1993;11(1):85–
91.
Ayoola 1984 {published data only}
Ayoola EA. Hepatitis B vaccine in developing countries: problems
and prospects. IARC Scientific Publications 1984;63:297–305.
Ayoola 1986 {published data only}
Ayoola EA, Atoba MA, Johnson AO. Intradermal vaccination against
hepatitis B virus infection in an endemic area (Nigeria), two year
results. Archives of Virology 1986;91(3-4):291–6.
Baldy 2004 {published data only}
Baldy JL, de Lima GZ, Morimoto HK, Reiche EMV, Matsuo T, de
Mattos ED, et al.Immunogenicity of three recombinant hepatitis B
Hall 1993 {published data only}
Hall AJ. Hepatitis B vaccination: protection for how long and against
what?. BMJ (Clinical Research Ed) 1993;307(6899):276–7.
Jones-Lecointe 2001 {published data only}
Jones-Lecointe A, Swanston WM, Charles WP, Simeon D. Immunization status of thirty patients with sickle cell disease five years
post hepatitis B vaccination. West Indian Medical Journal 2001;50
(4):317–8.
Koff 1994 {published data only}
Koff RS, Needles CF, Germanaud J, Barthez JP. Immunogenicity of
hepatitis B vaccines. JAMA 1994;272(3):201–2.
Liao 1998 {published data only}
Liao S, Li R, Li H, Yang J, Zeng X. Long-term efficacy of plasmaderived hepatitis B vaccine among Chinese children: a 12-year followup. Zhongguo Yi Xue Ke Xue Yuan Xue Bao 1998;20(2):89–94.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
16
Lok 2004 {published data only}
Lok AS. Prevention of hepatitis B virus-related hepatocellular carcinoma. Gastroenterology 2004;127 Suppl 1:S303–9.
Margolis 1995 {published data only}
Margolis HS, Coleman PJ, Brown RE, Mast EE, Sheingold SH,
Arevalo JA. Prevention of hepatitis B virus transmission by immunization - an economic-analysis of current recommendations. JAMA
1995;274(15):1201–8.
Martins 2004 {published data only}
Martins RM, Bensabath G, Arraes LC, Oliveira ML, Miguel JC,
Barbosa GG, et al.Multicenter study on the immunogenicity and
safety of two recombinant vaccines against hepatitis B. Memórias do
Instituto Oswaldo Cruz 2004;99(8):865–71.
Maupas 1976 {published data only}
∗
Maupas P, Chiron JP, Barin F, Coursaget P, Goudeau A, Perrin J, et
al.Efficacy of hepatitis B vaccine in prevention of early HBsAg carrier
state in children. Controlled trial in an endemic area (Senegal). Lancet
1981;1(8215):289–92.
Maupas P, Goudeau A, Coursaget P, Drucker J, Bagros P. Immunisation against hepatitis B in man. Lancet 1976;1(7974):1367–70.
Maupas 1981 {published data only}
Maupas P, Chiron JP, Barin F, Coursaget P, Goudeau A, Perrin J, et
al.Efficacy of hepatitis B vaccine in prevention of early HBsAg carrier
state in children. Controlled trial in an endemic area (Senegal). Lancet
1981;1(8215):289–92.
Montesano 2002 {published data only}
Montesano R. Hepatitis B immunization and hepatocellular carcinoma: the Gambia hepatitis intervention study. Journal of Medical
Virology 2002;67(3):444–6.
Perrilo 1987 {published data only}
Perrilo R, Bodicky C, Campbell C, Davis E, Veit H. A randomized,
controlled-study of the recombinant yeast hepatitis-B vaccine in the
mentally-retarded. Journal of Medical Virology 1987;21(4):A102–A3.
Perroni 1989 {published data only}
Perroni L, Mattioli R, Ensoli G, Albertoni F, Duchini B, Ungaro
P, et al.Vaccination against hepatitis B: 3 years’ experience at the
Latium Operative Unit [Vaccinazione contro l’epatite B: tre anni
di esperienza presso l’Unità Operativa di Latina]. Annali di igiene :
medicina preventiva e di comunità 1989;1(6):1307–15.
Petersen 1993 {published data only}
Petersen KM, Bulkow LR, Mcmahon BJ, Zanis C, Getty M, Peters H,
et al.Duration of hepatitis B immunity in low risk children receiving
hepatitis B vaccinations from birth. Pediatric Infectious Disease Journal
2004;23(7):650–5.
Phanuphak 1989 {published data only}
Phanuphak P, Phanpanich T, Wongurai S, Sirivichayakul S, Sriwanthana B, Panmuong W, et al.Comparative immunogenicity study of
four plasma-derived hepatitis B vaccines in Thai young adults. Vaccine 1989;7(3):253–6.
Pongpipat 1987 {published data only}
Pongpipat D, Suvatte V, Assateerawatts A, Bhethraratt S. Active preexposure immunisation against hepatitis B virus: immunogenicity of
hepatitis B vaccine in healthy Thai adults and children. Asian Pacific
Journal of Allergy and Immunology 1987;5(1):63–5.
Ricciardi 1990 {published data only}
Ricciardi G, Graziano G. Safety and immunogenicity of an antihepatitis-B vaccine obtained using the recombinant DNA technique:
results of a longitudinal study in hospital personnel [Sicurezza ed
immunogenicita di un vaccino anti–epatite B ottenuto con la tecnica del DNA–ricombinante: risultati di un’indagine longitudinale
su personale ospedaliero]. Bollettino dell’Istituto Sieroterapico Milanese
1990;69(2):385–90.
Rodrigo 1992 {published data only}
Escudero A, Gilabert MS, García F, Gonzalez R, del Olmo JA, Wassel
AH, et al.Immune response to hepatitis B vaccine in parenteral drug
abusers. Vaccine 1992;10(11):798–801.
Stevens 1985 {published data only}
Stevens CE, Taylor PE, Rubinstein P. Safety of the hepatitis B vaccine.
New England Journal of Medicine 1985;312(6):375–6.
Szmuness 1980 {published data only}
Szmuness W, Stevens CE, Harley EJ, Zang EA, Oleszko WR, William
DC, et al.Hepatitis B vaccine: demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States.
New England Journal of Medicine 1980;303(15):833–41.
Szmuness 1981 {published data only}
Szmuness W, Stevens CE, Harley EJ, Zang EA, Oleszko WR, William
DC, et al.Hepatitis B vaccine: demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States
[Hepatitis B vaccine: darstellung der Wirksamkeit in einem kontrollierten lkinischen Versuch an einer Persobegruppe mit hohen Risikofaktoren in den Vereinigten Staaten]. Innere Medizin 1981;8:79–80.
Thanavala 2005 {published data only}
Thanavala Y, Mahoney M, Pal S, Scott A, Richter L, Natarajan N,
et al.Immunogenicity in humans of an edible vaccine for hepatitis B.
Proceedings of the National Academy of Sciences of the United States of
America 2005;102(9):3378–82.
Thies 2002 {published data only}
Thies N. Universal neonatal immunization for Hepatitis B. Journal
of Paediatric Child Health 2002;38(2):211–2.
Tong 2005 {published data only}
Tong NK, Beran J, Kee SA, Miguel JL, Sanchez C, Bayas JM, et
al.Immunogenicity and safety of an adjuvanted hepatitis B vaccine
in pre-hemodialysis and hemodialysis patients. Kidney International
2005;68(5):2298–303.
Turchi 1997 {published data only}
Turchi MD, Martelli CM, Ferraz ML, Silva AE, Cardoso D das
D, Martelli P, et al.Immunogenicity of low-dose intramuscular and
intradermal vaccination with recombinant hepatitis B vaccine. Revista
do Instituto de Medicina Tropical de Sao Paulo 1997;39(1):15–9.
Xu 1985a {published data only}
Xu ZY, Francis DP, Liu CB, Purcell RH, Duan SC, Chen RJ, et
al.Prevention of hepatitis B virus carriage of infants using HBV vaccine in Shanghai. Preliminary report of a randomized double-blind
placebo-controlled trial. Chinese Medical Journal 1985;98(9):623–6.
Xu 1985b {published data only}
Xu ZY, Liu CB, Francis DP, Purcell RH, Gun ZL, Duan SC,
et al.Prevention of perinatal acquisition of hepatitis B virus carriage using vaccine: preliminary report of a randomized, double-
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
17
blind placebo-controlled and comparative trial. Pediatrics 1985;76
(5):713–8.
Health Research. Canadian Medical Association Journal 2004;171:
735–40.
Xu 1992 {published data only}
Xu H. Evaluation of long-term efficacy on plasma-derived hepatitis
B vaccine. Zhonghua Liu Xing Bing Xue Za Zhi 1992;13(6):362–5.
Chan 2005
Chan AW, Altman DG. Identifying outcome reporting bias in randomised trials on PubMed: review of publications and survey of authors. BMJ (Clinical Research Ed.) 2005;330:753.
Xu 1993 {published data only}
Xu HW, Li DS, Xu JW. Evaluation of long-term efficacy of inoculation of hepatitis B vaccine on subjects with isolated low level of antiHBs. Journal of Xian Medical University 1993;14(4):335–9.
Xu 1995 {published data only}
Xu H, Sui X, Men B. The efficacy of plasma-derived hepatitis B
vaccine in the sixth year after initial injection. Journal of Xi’an Medical
University 1995;16(4):368–71.
Yuen 2004 {published data only}
Yuen MF, Lim WL, Chan AO, Wong DK, Sum SS, Lai CL. 18-year
follow-up study of a prospective randomized trial of hepatitis B vaccinations without booster doses in children. Clinical Gastroenterology
and Hepatology 2004;2(10):941–5.
Yvonnet 1987 {published data only}
Yvonnet B, Coursaget P, Leboulleux D, Barres JL, Fritzell B, Sarr G,
et al.Low-dose hepatitis B vaccine immunisation in children. Lancet
1987; Vol. 1, issue 8525:169.
Zhuang 1998 {published data only}
Zhuang G, Xu H, Zuo H, Wang X, Liu P, Kong L. Six-year efficacy
of hepatitis B revaccination: A double-blind, placebo-controlled and
randomized field trial. Journal of Xi’an Medical University 1998;10
(1):44–8.
Additional references
Aggarwal 2004
Aggarwal R, Ranjan P. Preventing and treating hepatitis B infection.
BMJ (Clinical Research Ed.) 2004;329:1080–6.
Altman 2003
Altman DG, Bland JM. Interaction revisited: the difference between
two estimates. BMJ (Clinical Research Ed.) 2003;326:219.
Chen 1996
Chen HL, Chang MH, Ni YH, Hsu HY, Lee PI, Lee CY, et
al.Seroepidemiology of hepatitis B virus infection in children: ten
years of mass vaccination in Taiwan. JAMA 1996;276:906–8.
Chen 2005
Chen W, Gluud C. Vaccines for preventing hepatitis B in healthcare workers. Cochrane Database of Systematic Reviews 2005, Issue 4.
[DOI: 10.1002/14651858.CD000100.pub3]
Furukawa 2007
Furukawa TA, Watanabe N, Omori IM, Montori VM, Guyatt GH.
Association between unreported outcomes and effect size estimates
in Cochrane meta-analyses. JAMA 2007;297:468–70.
Gall 2001
Gall SA. Hepatitis B virus immunization today. Infectious diseases in
obstetrics and gynecology 2001;9:63–4.
Gitlin 1997
Gitlin 1997. Hepatitis B: diagnosis, prevention, and treatment. Clinical chemistry 1997;43:1500–6.
Gluud 2007a
Gluud C, Nikolova D, Klingenberg SL, Als-Nielsen B, D’Amico
G, Davidson B, et al.Cochrane Hepato-Biliary Group. About The
Cochrane Collaboration (Cochrane Review Groups (CRGs)) 2007,
Issue 2. Art. No.: LIVER.
Gluud 2007b
Gluud C, Brok J, Gong Y, Koretz RL. Hepatology may have problems with putative surrogate outcome measures. Journal of Hepatology 2007;46(4):734–42.
Beasley 1984
Beasley RP, Hwang LY. Hepatocellular carcinoma and hepatitis B
virus. Seminars in Liver Disease 1984;4:113–21.
Higgins 2003
Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analysis. BMJ (Clinical Research Ed.) 2003;327
(7414):557–60.
CDC 1999
Centers for Disease Control and Prevention. Recommendations
to prevent hepatitis B virus transmission-United States, updated.
MMWR. Morbidity and Mortality Weekly Report 1999;48:33–4.
Higgins 2006
Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews
of Interventions 4.2.6 [updated September 2006]. The Cochrane Library, Issue 4, 2006. The Cochrane Collaboration, 2006.
Chan 1991
Chan CY, Lee SD, Tsai YT, Lo KJ. Booster response to recombinant yeast-derived hepatitis B vaccine in vaccinees whose anti-HBs
responses were elicited by a plasma-derived vaccine. Vaccine 1991;9
Suppl A:765–7.
Hyams 1995
Hyams K C. Risk of chronicity following acute hepatitis B virus
infection. Clinical Infectious Diseases 1995;20:992–1000.
Chan 2004a
Chan AW, Hróbjartsson A, Haahr MT, Gøtzsche PC, Altman DG.
Empirical evidence for selective reporting of outcomes in randomized
trials: comparison of protocols to published articles. JAMA 2004;
291:2457–65.
Chan 2004b
Chan AW, Krleza-Jeric’ K, Schmid I, Altman DG. Outcome reporting bias in randomized trials funded by the Canadian Institutes of
ICH-GCP 1996
International conference on harmonisation of technical requirements
for registration of pharmaceuticals for human use. Guideline for
Good Clinical Practice. E6 (R1). ICH Harmonised Tripartite Guideline 1996.
Lee 2006a
Lee C, Gong Y, Brok J, Boxall EH, Gluud C. Hepatitis B immunisation for newborn infants of hepatitis B surface antigen-positive mothers. Cochrane Database of Systematic Reviews 2006, Issue 2. [DOI:
10.1002/14651858.CD004790.pub2]
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
18
Lee 2006b
Lee C, Gong Y, Brok J, Boxall EH, Gluud C. Effect of hepatitis B
immunisation in newborn infants of mothers positive for hepatitis B
surface antigen: systematic review and meta-analysis. BMJ (Clinical
Research Ed.) 2006;332(7537):328–36.
Liaw 1998
Liaw YF, Tai DI, Chu CM, Chen TJ. The development of cirrhosis in
patients with chronic type B hepatitis: a prospective study. Hepatology
1988;8:493–6.
Maddrey 2000
Maddrey WC. Hepatitis B: an important public health issue. Journal
of Medical Virology 2000;61:362–6.
Mahoney 1999
Mahoney FJ. Update on Diagnosis, Management, and Prevention of
Hepatitis B Virus Infection. Clinical Microbiology Review 1999;12
(2):351–66.
McMahon 1985
McMahon BJ, Alward WL, Hall DB, Heyward WL, Bender TR,
Francis DP, et al.Acute hepatitis B virus infection: relation of age
to the clinical expression of disease and subsequent development of
carrier state. The Journal of Infectious Diseases 1985;151:599–603.
Moyer 1994
Moyer LA, Mast EE. Hepatitis B: virology, epidemiology, disease,
and prevention, and an overview of viral hepatitis. American journal
of preventive medicine 1994;10 Suppl:45–55.
Nystrom 2000
Nystrom RJ, Timmons AJ. Completion of hepatitis B vaccination
series in school-based health centers. The Journal of Adolescent Health
2000;26:320–1.
Rehermann 2003
Rehermann B. Immune responses in hepatitis virus infection. Seminars in Liver Disease 2003;23:21–38.
RevMan 2003
Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). 4.2 for Windows. Copenhagen:
The Nordic Cochrane Centre, The Cochrane Collaboration, 2003.
Rizzetto 2002
Rizzetto M, Zanetti AR. Progress in the prevention and control of
viral hepatitis B. Closing remarks. Journal of Medical Virology 2002;
67:463–6.
Royle 2003
Royle P, Milne R. Literature searching for randomized controlled
trials used in Cochrane reviews: rapid versus exhaustive searches.
International Journal of Technology Assessment in Health Care 2003;
19(4):591–603.
Schroth 2004
Schroth RJ, Hitchon CA, Uhanova J, Noreddin A, Taback SP, Moffatt
MEK, et al.Hepatitis B vaccination for patients with chronic renal
failure. Cochrane Database of Systematic Reviews 2004, Issue 3. [DOI:
10.1002/14651858.CD003775.pub2]
Van Damme 1997
Van Damme P, Kane M, Meheus A on behalf of the Viral Hepatitis
Prevention Board. Integration of hepatitis B vaccination into national
immunisation programmes. BMJ (Clinical Research Ed.) 1997;314:
1033.
Walsh 2001
Walsh K, Alexander GJM. Update on chronic viral hepatitis. Postgraduate Medical Journal 2001;77:498–505.
West 1993
West D J, Calandra G B, Hesley T M, Idoli V, Miller W J. Control
of hepatitis B through routine immunization of infants: the need
for flexible schedules, and new combination vaccine formulations.
Vaccine 1993;11:S21–7.
WHO 2004
World
Health
Organisation.
Hepatitis B. http://www.who.int/mediacentre/factsheets/fs204/en/
(Accessed February 2007) 2004.
Wong 1994
Wong VK, Woodruff C, Shapiro R. Compliance of hepatitis B vaccination in patients presenting to a teenage clinic. The Pediatric Infectious Disease Journal 1994;13:936.
∗
Indicates the major publication for the study
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Ambrosch 1992
Methods
Randomised clinical trial
Generation of allocation sequence: unclear. Text mentions that ’participants were allocated at random’ to
one of the three intervention groups.
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: 1 month after last dose, corresponding to 7 months after first dose vaccine.
Withdrawals and drop-outs: not described.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
19
Ambrosch 1992
(Continued )
Intention-to-treat: no.
Participants
Healthy adults, including volunteers and medical personnel found to be HBsAg negative on screening.
Mean age of participants: 18.4 years (Group I 22.3 years; group II 22.7 years group III 22.9 years).
Gender ratio (male to female): Group I 29/26; Group II 30/25; Group III 31/24.
Country: Austria
Inclusion criteria:
Those negative for HBsAg and anti HBc and anti HBs with normal SGOT (AST) and SGPT (ALT) as
defined in the trial.
Exclusion criteria: none specified.
Interventions
Three groups were compared: group I received HAV vaccine, group II: received HBV vaccine, and group
III: received HAV and HBV vaccine in different arms.
Type of HBV vaccine: recombinant vaccine.
Type of product in control group: HAV vaccine (formalin inactivated preparation containing 720 ELISA
units per ml).
Schedule: 0,1,6 months.
Dosage: 20 mcg.
Route: intramuscular.
Site: deltoid muscle.
Outcomes
Lack of sero-protection (Ab < 10 mIU/ ml).
GMT of anti-HBs antibodies.
Adverse events: reported as mean total local symptom score (TLSS) comprising fever, nausea and malaise,
without break-up of individual symptoms.
Headache
Mean SGOT (IU/ml).
Mean SGPT (IU/ml).
Notes
Centre: single centre.
Setting: hospital based.
The authors described this as a randomised prospective study designed to compare single and simultaneous
administration of two vaccines, ie, hepatitis A and hepatitis B. The interventions in each group are
described under Interventions.
The trial does not report two outcomes (lack of sero-protection and geometric mean titre of antibodies)
for participants in the control group.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Aristegui 1995
Methods
Randomised clinical trial
Generation of allocation sequence: unclear.
Allocation concealment: unclear.
Blinding: unclear.
Follow-up: 40 to 60 days after last dose.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
20
Aristegui 1995
(Continued )
Withdrawals and drop-outs: described. 590 of 655 and 329 of 731 babies were available for the follow-up
assessment.
Intention to treat: no.
Participants
Newborns.
Country: Spain
Inclusion criteria: healthy newborns.
Exclusion criteria: none specified.
Mean weight (range): Group A: 3260g (2060 to 4630) and Group B: 3300 g (2000 to 6600).
Interventions
Babies receiving simultaneous vaccination with HB vaccine and DPT vaccine plus OPV were compared
with babies receiving only DPT plus OPV.
Group A: HB plus DPT vaccines.
Group B: DPT vaccine alone.
Type of vaccine/ placebo: recombinant HB vaccine/DPT vaccine.
Schedule: 0,2,6 months for HB vaccine and 2,4,6 months for DPT vaccine for babies in both groups.
Dosage: 10 mcg in 0.5 ml.
Route: intramuscular in antero-lateral thigh.
Outcomes
Lack of sero-protection (anti HBs Ab <10 mIU/ ml).
Adverse events.
Lack of compliance.
Notes
Prospective randomised clinical trial comparing babies receiving simultaneous vaccination with HB
vaccine and DPT vaccine plus OPV against those receiving only DPT plus OPV.
590 of 655 and 329 of 731 babies were available for the follow-up assessment
Single centre.
Setting: hospital based.
Both groups received diphtheria/pertussis/ tetanus and oral polio vaccines.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Bassily 1995
Methods
Randomised clinical trial
Generation of allocation sequence: adequate (random number table).
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: 1 year.
Withdrawals and drop-outs: described.
Intention to treat: no.
Participants
Infants including a group at birth itself.
Country: Egypt
Inclusion criteria: Babies born to mothers screened for hepatitis B markers and found to be negative.
Exclusion criteria: renal, haematologic, liver, chest and cardiac disease;
babies on corticosteroids/ immunosuppressive drugs; birth weight less than 2000 g.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
21
Interventions
Group A: HB vaccine plus routine childhood vaccines (HB given with 0,2,6 months schedule).
Group B: HB plus routine childhood vaccines (HB vaccine given with 2,4,9 months schedule), and
Group C: routine childhood vaccine only (BCG, polio, DPT, and measles vaccines).
Type of vaccine/ placebo: recombinant vaccine/ another vaccine as above.
Schedule:
Group A = 0,2,6 months;
Group B = 2,4,9 months.
Dosage: 2.5 mcg.
Route: intramuscular.
Outcomes
HBsAg.
HBe antigen
HBc antibodies in serum
Lack of sero-protection (Ab < 10 mIU/ ml)
GMT of anti-HBs antibodies.
Adverse events.
Notes
Adverse events were expressed in percentage (Absolute numbers were not mentioned).
This trial randomised 198 (0,2,6 months schedule) and 196 (2,4,9 months schedule) participants to
receive the hepatitis B vaccine, but reported data in 90 and 107 respectively.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Coursaget 1984
Methods
Randomised clinical trial
Generation of allocation sequence: adequate (random number table).
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: maximum follow-up to 3 years after first dose.
Withdrawals and drop-outs: described, but reasons not mentioned.
Intention to treat: no.
Participants
Children aged 3 to 24 months.
Country: Senegal
Inclusion criteria: none specified.
Exclusion criteria: none specified.
Interventions
Group A: HB vaccine.
Group B: DT-polio vaccine.
Type of vaccine/placebo: recombinant/another vaccine.
Schedule: 3 injections one month apart followed by booster after 1 year.
Dosage: 5 mcg
Route: subcutaneous.
Outcomes
HBsAg.
anti-HBc antibodies in serum.
Notes
Four part study comprising:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
22
Coursaget 1984
(Continued )
I: HB vaccine versus control (DT and polio vaccines)
II: HB vaccine versus control versus both vaccines in diffreent arms
III: 2 HB doses followed by booster (not randomised clinical trial)
The total number randomised is not mentioned.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Giammanco 1991
Methods
Prospective trial with four arms; three of which included hepatitis B vaccine.
Generation of allocation sequence: unclear.
Allocation concealment: unclear.
Blinding: yes.
Follow-up: 1 month after final dose.
Withdrawals and drop-outs: yes.
Intention to treat: no.
Participants
Infants sero-negative for antibodies to HBsAg.
Country: Italy
Inclusion criteria: not described.
Exclusion criteria: not described.
Interventions
Groups A,B,C: HB vaccine.
Group D: DT vaccine plus OPV.
Type of HB vaccine/placebo: recombinant vaccine/ another vaccine.
Group A: (0,3,4,10 mo schedule).
Group B: (0,1,3,10 mo schedule).
Group C: (3,4,10 mo schedule).
Dosage: 10 mcg in 0.5 ml.
Route: intramuscular.
Site: deltoid.
Outcomes
HBsAg.
HBc antibodies in serum.
Lack of sero-protection (Ab < 10 mIU/ ml
GMT of anti HBs antibodies.
Adverse events.
Notes
Authors do not specify how the arms in the trial were formed. They report that 111 infants received
hepatitis B vaccine according to one of three different schedules, in addition to routine childhood vaccines
(DT and oral polio vaccine) and ’control group’ of 21 infants received only DT and oral polio vaccine.
Data were extracted for the adverse events.
Single centre.
Risk of bias
Item
Authors’ judgement
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Description
23
Giammanco 1991
(Continued )
Allocation concealment??
Unclear
B - Unclear
Greenberg 1996
Methods
Randomised clinical trial
Generation of allocation sequence: adequate.
Allocation concealment: unclear.
Blinding: unclear.
Follow-up:
Withdrawals and drop-outs: no
Intention to treat: no.
Participants
6 to 15 week old infants.
Country: United States of America
Inclusion criteria: not described.
Exclusion criteria: neurologic disease; seizures; altered immune function including maternal HIV
infection; those who had received immunoglobulin or blood products in the preceding 3 months; acute
fever.
Interventions
Group A: HB plus DT vaccine.
Group B: Hib plus DT vaccine.
Type of HB vaccine and control product: recombinant HB vaccine/PRP-T Hib vaccine.
Schedule: 2,4,6 months.
Dosage: 10 mcg.
Route: Intramuscular.
Site of administration: anterolateral thigh.
Outcomes
Lack of sero-protection (Ab < 10 mIU/ ml).
Serious adverse events.
Mortality (all cause).
Seizures
Hospitalizations
Adverse events.
Lack of compliance.
Notes
Multi-centre
Setting: 13 clinics in Southern California
The trial randomised 5056 participants to receive hepatitis B vaccine, but reported lack of sero-protection
in only 135 of them. This data was not presented for participants in the control arm.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
24
Liao 1999
Methods
Randomised clinical trial
Generation of allocation sequence: unclear.
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: 15 years.
Withdrawals and drop-outs: no.
Intention to treat: no.
Participants
3 to 36 month-old children recruited from vaccination clinics.
Country: China
Inclusion criteria: not described.
Exclusion criteria: not described.
Interventions
Group A: HB vaccine
Group B: vaccine diluent.
Type of HB vaccine and control product: plasma derived/placebo.
Schedule: 0,1,6 months.
Dosage: 17.5 mcg.
Outcomes
HBsAg (at 15 years).
Lack of sero-protection (Ab < 10 mIU/ ml).
Lack of compliance.
Notes
Multi centre: seven vaccination clinics in one county of a province in China.
This trial randomised 308 participants to the intervention arm and 341 to the control arm, but reported
data at fifteen years for only 52 and 54 respectively.
The authors also reported the cumulative number of participants who became HBsAg positive during the
fifteen year follow-up, despite the high attrition rate.
Sero-protection in this trial was defined as greater than 10 ’Sample to Negative control ratio’ based on a
radio-immunoassay method to detect anti-HBs antibodies.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Perrin 1986
Methods
Randomised clinical trial
Generation of allocation sequence: unclear.
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: 12 months after the third dose. Although 480 babies were recruited, the number in each
intervention arm has not been described.
Withdrawals and drop-outs: no.
Intention to treat: no.
Participants
Newborn babies.
Country: Burundi
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
25
Perrin 1986
(Continued )
Inclusion criteria: not specified.
Exclusion criteria: not specified.
Interventions
Group A; HB vaccine.
Group B: other vaccines in routine immunization programme (not mentioned).
Type of HB vaccine and control product: Not specified/another vaccine.
Schedule: 0,2,12 months.
Dosage: 5 mcg.
Route: not specified.
Outcomes
HBsAg at 24 months after first dose.
HBc antibodies in serum.
Notes
Single centre.
Setting: hospital.
This trial does not mention the number of participants randomised.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Sun 1991
Methods
Randomised clinical trial
Generation of allocation sequence: unclear.
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: 5 years. 261 out of 265 completed vaccination; but break up in each arm is not mentioned
Withdrawals and drop-outs: no.
Intention to treat: no.
Participants
Children between 5 to 9 years recruited from primary schools.
Country: China
Inclusion criteria: not described.
Exclusion criteria: not described.
Interventions
Group A: HB vaccine
Group B: Not mentioned.
Type of HB vaccine and control product: HB vaccine/placebo but nature not specified.
Schedule: 0,1,6 months.
Dosage: not mentioned.
Route: not specified.
Outcomes
HBsAg (at 11 years)
HBc antibodies in serum
GMT of anti HBs antibodies
Clinical hepatitis
Lack of compliance
Notes
Setting: community.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
26
Sun 1991
(Continued )
Single centre.
This trial randomised almost 38000 participants in each arm, but reported HBsAg in 682 and 312
participants in the intervention and control arms respectively.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Tsega 1990
Methods
Randomised clinical trial
Generation of allocation sequence: inadequate.
Allocation concealment: unclear.
Blinding: no.
Follow-up: 1 month after third dose.
Withdrawals and drop-outs: yes.
Intention to treat: no.
Participants
Infants between 6 weeks and 6 months of age.
Country: Ethiopia
Inclusion criteria: described as healthy infants.
Exclusion criteria: not specified.
Interventions
Group A: HBV vaccine plus routine EPI vaccines.
Group B: EPI vaccines.
Type of HB vaccine and control product: plasma derived vaccine/another vaccine.
Schedule: 0,1,2,6 months.
Dosage: 2 mcg.
Route: not mentioned.
Outcomes
HBsAg.
HBc antibodies in serum.
Lack of sero-protection (Ab < 10 mIU/ ml).
GMT of anti HBs antibodies.
Notes
Prospective hospital based.
Single centre.
This trial randomised 150 participants each to the intervention and control arms, but reported outcomes
in 134 and 132 respectively.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
27
Wang 1985
Methods
Randomised clinical trial
Generation of allocation sequence: unclear.
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: 6 months after the last (fourth) dose.
Withdrawals and drop-outs: no.
Intention to treat: no.
Participants
Newborn babies
Country: China
Inclusion criteria: not mentioned except that babies were born to mothers who were HBsAg negative (for
the arms being included in the systematic review).
Exclusion criteria: not described.
Interventions
Group A: HB vaccine.
Group B: placebo.
Type of HB vaccine and control product: plasma derived/placenta immunoglobulin.
Schedule: 0,1,3,6 months.
Dosage: 30 mcg.
Route: intramuscular or subcutaneous.
Outcomes
HBsAg
HBc antibodies in serum
Notes
Prospective trial comparing four arms of which two were babies born to HBsAg negative mothers who
received HB vaccine or placebo.
Single centre.
This trial reports that at least 10 babies were present in each group, although it does not specify the exact
number. The data in the trial are reported for 10 babies in each group.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Xueliang 2000
Methods
Randomised clinical trial
Generation of allocation sequence: unclear.
Allocation concealment: unclear.
Blinding: not performed.
Follow-up: 11 years after vaccination.
Withdrawals and drop-outs: no.
Intention to treat: no.
Participants
Children between 5 to 9 years recruited from primary schools.
Country: China
Inclusion criteria: not described.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
28
Xueliang 2000
(Continued )
Exclusion criteria: not described.
Interventions
Group A: HB vaccine.
Group B: placebo.
Type of HB vaccine and control product: plasma derived/placebo, but nature not specified.
Schedule: 0,1,6 months.
Dosage: not mentioned.
Route: not specified.
Outcomes
HBsAg (at 11 years)
HBc antibodies in serum.
Lack of sero-protection (Ab < 10 mIU/ ml).
Clinical hepatitis.
Notes
Authors report this as a randomised, double-blind, placebo-controlled trial. However, there is no
description of the ’placebo’ used. The authors report that a “blind code” was used in the laboratory assay.
Single centre.
This trial randomised 126 and 135 participants to the intervention and control arms, but reported
outcomes in 84 and 100 respectively.
Setting: community.
Risk of bias
Item
Authors’ judgement
Description
Allocation concealment??
Unclear
B - Unclear
Ab = antibodies
anti-HBc = antibodies to hepatitis B core antigen
DPT = diphtheria, pertussis, tetanus vaccine
DT = diphtheria, tetanus vaccine
HAV = hepatitis A vaccine
HB = hepatitis B
HBsAg = hepatitis B surface antigen
Hib = Haemophilus influenzae type b
mcg = micrograms
mIU = milli international units
OPV = oral polio vaccine
RCT = randomised clinical trial.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
29
Characteristics of excluded studies [ordered by study ID]
Study
Reason for exclusion
Aach 1982
This is not a randomised clinical trial comparing hepatitis B vaccine versus control.
Abbas 1999
Trial is not designed as a randomised clinical trial to compare HB vaccine versus placebo or a control.
Abdurrahman 1984
This is not a randomised clinical trial comparing hepatitis B vaccine versus control.
Akbar 1997
Trial was conducted in mice.
Akram 2005
Trial is not designed as a randomised clinical trial to compare HB vaccine versus placebo or a control.
Alikasifoglu 2001
Trial is a comparison of different brands, doses, and schedules, and not a randomised clinical trial of HB
vaccine versus versus placebo or a control.
Ambrosch 1994
Trial is a comparison of a combined HB and hepatitis A vaccine administered either singly, separately, or in
combination and is not a randomised clinical trial of HB vaccine versus placebo or a control.
Artan 2004
Trial is a comparison of different vaccination schedules and is not a randomised clinical trial of HB vaccine
versus placebo or a control.
Assateerawatt 1993
Trial is not a randomised clinical trial of HB vaccine versus control/ placebo but reports observations in high
risk neonates and normal children and adults, with no control group for comparison.
Ayoola 1984
Trial includes comparison of different routes and doses in healthy people, adults who are HBsAg positive and
non-responders.
Ayoola 1986
Trial is not a randomised clinical trial of HB vaccine versus control/ placebo, but compares two routes of
administering HB vaccine.
Baldy 2004
Trial is a comparison of three preparations of HB vaccine and not a randomised clinical trial of HB vaccine
versus placebo or a control.
Chunsuttiwat 1997
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Coutinho 1983
Trial was conducted in adult homosexual men.
Duclos 2003
Trial is not a randomised trial comparing HB vaccine versus placebo or a control.
Fortuin 1993
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Gomez 2002
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Guihua 1997
There is later data reported for this trial.
Guihua 1998
Trial is a randomised clinical trial of booster dose versus placebo in a cohort of responders to vaccination and
not a randomised clinical trial of HB vaccine versus placebo or a control.
Hall 1993
This is not a randomised clinical trial comparing hepatitis B vaccine with a control group.
Jones-Lecointe 2001
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Koff 1994
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Liao 1998
This study is succeeded by Liao 1999 that reports results of the same trial at a later follow-up date.
Lok 2004
Trial is not a randomised clinical trial of HB vaccine versus control/placebo, but a review of previously
published data.
Margolis 1995
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
30
Martins 2004
Trial is a comparison of different brands of HB vaccine and not a randomised clinical trial of HB vaccine
versus placebo or a control.
Maupas 1976
Trial does not have a group for comparison and hence is not a randomised clinical trial of HB vaccine versus
placebo or a control.
Maupas 1981
This was not a randomised trial. In addition, the number of participants enrolled in each arm was not specified.
Montesano 2002
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Perrilo 1987
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Perroni 1989
Trial includes babies born to HBsAg positive mothers and other high risk groups that are not part of this
review.
Petersen 1993
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Phanuphak 1989
Trial is a comparison of four brands of the same vaccine (hepatitis B) and has no control group that did not
receive hepatitis B vaccine.
Pongpipat 1987
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Ricciardi 1990
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Rodrigo 1992
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Stevens 1985
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Szmuness 1980
Trial was conducted in adult homosexual men.
Szmuness 1981
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Thanavala 2005
The antigen used in the trial is not a vaccine as yet.
Thies 2002
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Tong 2005
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
Turchi 1997
Trial is a comparison of different routes and not a randomised clinical trial of HB vaccine versus placebo or a
control.
Xu 1985a
Trial is a randomised clinical trial in babies born to HBsAg positive mothers.
Xu 1985b
Trial was conducted in babies born to HBsAg positive mothers.
Xu 1992
There is a later analysis of data of this trial.
Xu 1993
There is nothing mentioned about a control group.
Xu 1995
There is a later analysis of data of this trial.
Yuen 2004
There is no control group in the trial hence not a randomised clinical trial of HB vaccine versus placebo or a
control.
Yvonnet 1987
Trial is not a randomised clinical trial of HB vaccine versus control/placebo, but a comparison of different
doses of HB vaccine without a control group for comparison.
Zhuang 1998
Trial is not a randomised clinical trial of HB vaccine versus placebo or a control.
HB = hepatitis B
HBsAg = hepatitis B surface antigen.
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
31
DATA AND ANALYSES
Table 1. Comparison 1. Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome or subgroup title
1 HBsAg at follow-up (all studies
reporting this outcome)
1.1 Presence of HBsAg at
follow-up (all trials)
2 HBsAg at follow-up (sub-group
analysis based on age of
enrolled participants)
2.3 Children (28 days to 10
years)
2.4 Newborn babies (less than
28 days)
3 HBsAg at follow-up (sub-group
analysis based on prevalence in
participating population).
3.3 High prevalence (> 8%)
4 HBsAg at follow-up (sub-group
analysis based on duration of
follow-up)
4.1 Follow-up less than 1 year
4.3 Follow-up more than 5
years
5 HBsAg at follow-up (sub-group
analysis based on vaccination
schedule)
5.1 Three doses administered
with 0,1,6 schedule
5.3 Four doses administered
by any schedule
No. of
studies
No. of
participants
4
1230
Risk Ratio (M-H, Fixed, 95% CI)
0.96 [0.89, 1.03]
4
1230
Risk Ratio (M-H, Fixed, 95% CI)
0.96 [0.89, 1.03]
4
1230
Risk Ratio (M-H, Fixed, 95% CI)
0.96 [0.89, 1.03]
3
1210
Risk Ratio (M-H, Fixed, 95% CI)
0.96 [0.89, 1.03]
1
20
Risk Ratio (M-H, Fixed, 95% CI)
Not estimable
4
1230
Risk Ratio (M-H, Fixed, 95% CI)
0.96 [0.89, 1.03]
4
4
1230
1230
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
0.96 [0.89, 1.03]
0.96 [0.89, 1.03]
2
2
320
910
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
0.76 [0.41, 1.40]
0.97 [0.90, 1.04]
4
1230
Risk Ratio (M-H, Fixed, 95% CI)
0.96 [0.89, 1.03]
2
910
Risk Ratio (M-H, Fixed, 95% CI)
0.97 [0.90, 1.04]
2
320
Risk Ratio (M-H, Fixed, 95% CI)
0.76 [0.41, 1.40]
Statistical method
Effect size
Table 1. Comparison 2. HB vaccine versus control: presence of antibodies to HBc antigen in serum
Outcome or subgroup title
1 Anti-HBc antibodies in serum
2 Anti-HBc antibodies in serum
(random-effects model)
No. of
studies
No. of
participants
4
4
1230
1230
Statistical method
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Random, 95% CI)
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Effect size
0.86 [0.80, 0.92]
0.81 [0.61, 1.07]
32
Table 1. Comparison 3. HB vaccine versus control: adverse events
Outcome or subgroup title
1 Lack of sero-protection
2 Lack of sero-protection
(fixed-effect model)
No. of
studies
No. of
participants
3
3
1210
1210
Statistical method
Risk Ratio (M-H, Random, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
Effect size
0.57 [0.26, 1.27]
0.73 [0.69, 0.78]
Table 1. Comparison 4. HB vaccine versus control: serious adverse events
Outcome or subgroup title
1 Serious adverse events: all-cause
mortality
No. of
studies
No. of
participants
1
10317
Statistical method
Risk Ratio (M-H, Fixed, 95% CI)
Effect size
1.16 [0.47, 2.84]
Table 1. Comparison 5. HB vaccine versus control: adverse events
Outcome or subgroup title
1 Fever
2 Fever (fixed-effect model)
3 Local reactions (pain, redness,
soreness)
4 Other adverse events
No. of
studies
No. of
participants
2
2
2
14169
14169
1904
1
110
Statistical method
Effect size
Risk Ratio (M-H, Random, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
1.42 [0.60, 3.38]
1.03 [0.90, 1.18]
2.93 [1.72, 4.99]
Risk Ratio (M-H, Fixed, 95% CI)
2.33 [0.64, 8.56]
Table 1. Comparison 6. HB vaccine versus control: lack of compliance
Outcome or subgroup title
1 Lack of compliance
2 Lack of compliance (fixed-effect
model)
No. of
studies
No. of
participants
2
2
1976
1976
Statistical method
Risk Ratio (M-H, Random, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Effect size
0.91 [0.03, 24.39]
0.26 [0.21, 0.32]
33
Table 1. Comparison 7. Sensitivity analysis
Outcome or subgroup title
1 HBsAg based on percentage of
drop-outs in the trial (ITT
analysis)
1.1 Less than 10% drop-outs
2 Anti HBc antibodies based on
percentage of drop-outs (ITT
analysis)
2.1 Less than 10% drop-outs
3 HBsAg: intention to treat
analysis assuming favourable
outcome for all drop-outs
4 HBsAg: available participant
analysis
5 HBsAg: best-case scenario
6 HBsAg: worst-case scenario
7 HBsAg: worst-case scenario
8 Anti HBc antibodies:
intention-to-treat analysis
assuming favourable outcome
for all drop-outs
9 Anti-HBc antibodies in serum
based on analysis of available
participants (fixed-effect
model)
10 Anti-HBc antibodies in serum
based on analysis of available
participants (random-effects
model)
11 Anti-HBc antibodies in serum:
best-case scenario
12 Anti HBc antibodies in serum:
best-case scenario
13 Anti HBc antibodies in serum:
worst-case scenario
14 Anti HBc antibodies in serum:
worst-case scenario
No. of
studies
No. of
participants
1
20
Risk Ratio (M-H, Fixed, 95% CI)
Not estimable
1
1
20
20
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
Not estimable
Not estimable
1
4
20
1230
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
Not estimable
0.12 [0.03, 0.44]
4
576
Risk Ratio (M-H, Fixed, 95% CI)
0.12 [0.03, 0.44]
4
4
4
4
1230
1230
1230
1230
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Random, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
Risk Ratio (M-H, Fixed, 95% CI)
0.01 [0.00, 0.03]
9.45 [2.14, 41.67]
15.59 [10.29, 23.64]
0.35 [0.25, 0.50]
4
576
Risk Ratio (M-H, Fixed, 95% CI)
0.38 [0.28, 0.52]
4
576
Risk Ratio (M-H, Random, 95% CI)
0.36 [0.17, 0.76]
4
1230
Risk Ratio (M-H, Fixed, 95% CI)
0.08 [0.06, 0.12]
4
1230
Risk Ratio (M-H, Random, 95% CI)
0.08 [0.01, 1.03]
4
1230
Risk Ratio (M-H, Random, 95% CI)
3.45 [0.38, 31.57]
4
1230
Risk Ratio (M-H, Fixed, 95% CI)
3.66 [3.03, 4.42]
Statistical method
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Effect size
34
Analysis 1.1. nbsp; Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at
followTeXhyphen;up, Outcome 1 HBsAg at followTeXhyphen;up (all studies reporting this outcome).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 1 HBsAg at follow-up (all studies reporting this outcome)
Study or subgroup
Hepatitis B vaccine
Control
n/N
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
1 Presence of HBsAg at follow-up (all trials)
Liao 1999
257/308
296/341
81.3 %
0.96 [ 0.90, 1.03 ]
Tsega 1990
16/150
21/150
6.1 %
0.76 [ 0.41, 1.40 ]
Wang 1985
0/10
0/10
0.0 %
Not estimable
43/126
45/135
12.6 %
1.02 [ 0.73, 1.44 ]
594
636
100.0 %
0.96 [ 0.89, 1.03 ]
Xueliang 2000
Total (95% CI)
Total events: 316 (Hepatitis B vaccine), 362 (
Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
0.2
0.5
Favours HB vaccine
Review:
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 1 HBsAg at follow-up (all studies reporting this outcome)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
1 Presence of HBsAg at follow-up (all trials)
Liao 1999
257/308
296/341
0.96 [ 0.90, 1.03 ]
Tsega 1990
16/150
21/150
0.76 [ 0.41, 1.40 ]
Wang 1985
0/10
0/10
Not estimable
43/126
45/135
1.02 [ 0.73, 1.44 ]
Xueliang 2000
0.2
0.5
Favours HB vaccine
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
35
Analysis 1.2. nbsp; Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at
followTeXhyphen;up, Outcome 2 HBsAg at followTeXhyphen;up (subTeXhyphen;group analysis based on age
of enrolled participants).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 2 HBsAg at follow-up (sub-group analysis based on age of enrolled participants)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
1 Adults
Subtotal (95% CI)
0
0
0.0 %
Not estimable
0
0
0.0 %
Not estimable
257/308
296/341
81.3 %
0.96 [ 0.90, 1.03 ]
Tsega 1990
16/150
21/150
6.1 %
0.76 [ 0.41, 1.40 ]
Xueliang 2000
43/126
45/135
12.6 %
1.02 [ 0.73, 1.44 ]
584
626
100.0 %
0.96 [ 0.89, 1.03 ]
0/10
0/10
0.0 %
Not estimable
10
10
0.0 %
Not estimable
Total (95% CI)
594
636
100.0 %
0.96 [ 0.89, 1.03 ]
Total events: 316 (Hepatitis B vaccine), 362 (
Control)
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
Test for overall effect: not applicable
2 Adolescents (11 to 18 years)
Subtotal (95% CI)
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
Test for overall effect: not applicable
3 Children (28 days to 10 years)
Liao 1999
Subtotal (95% CI)
Total events: 316 (Hepatitis B vaccine), 362 ( Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
4 Newborn babies (less than 28 days)
Wang 1985
Subtotal (95% CI)
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
Test for overall effect: not applicable
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
0.5
Favours HB vaccine
1
2
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
36
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 2 HBsAg at follow-up (sub-group analysis based on age of enrolled participants)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
3 Children (28 days to 10 years)
Liao 1999
257/308
296/341
0.96 [ 0.90, 1.03 ]
Tsega 1990
16/150
21/150
0.76 [ 0.41, 1.40 ]
Xueliang 2000
43/126
45/135
1.02 [ 0.73, 1.44 ]
584
626
0.96 [ 0.89, 1.03 ]
Subtotal (95% CI)
Total events: 316 (Hepatitis B vaccine), 362 ( Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
0.5
1
Favours HB vaccine
Review:
2
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 2 HBsAg at follow-up (sub-group analysis based on age of enrolled participants)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
4 Newborn babies (less than 28 days)
Wang 1985
Subtotal (95% CI)
0/10
0/10
Not estimable
10
10
Not estimable
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
Test for overall effect: not applicable
0.5
Favours HB vaccine
1
2
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
37
Analysis 1.3. nbsp; Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at
followTeXhyphen;up, Outcome 3 HBsAg at followTeXhyphen;up (subTeXhyphen;group analysis based on
prevalence in participating population)..
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 3 HBsAg at follow-up (sub-group analysis based on prevalence in participating population).
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
1 Low prevalence (< 2%)
Subtotal (95% CI)
0
0
0.0 %
Not estimable
0
0
0.0 %
Not estimable
257/308
296/341
81.3 %
0.96 [ 0.90, 1.03 ]
Tsega 1990
16/150
21/150
6.1 %
0.76 [ 0.41, 1.40 ]
Wang 1985
0/10
0/10
0.0 %
Not estimable
43/126
45/135
12.6 %
1.02 [ 0.73, 1.44 ]
594
636
100.0 %
0.96 [ 0.89, 1.03 ]
636
100.0 %
0.96 [ 0.89, 1.03 ]
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
Test for overall effect: not applicable
2 Intermediate prevalence (2% to 8%)
Subtotal (95% CI)
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
Test for overall effect: not applicable
3 High prevalence (> 8%)
Liao 1999
Xueliang 2000
Subtotal (95% CI)
Total events: 316 (Hepatitis B vaccine), 362 ( Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
Total (95% CI)
594
Total events: 316 (Hepatitis B vaccine), 362 (
Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
0.5
Favours HB vaccine
1
2
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
38
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 3 HBsAg at follow-up (sub-group analysis based on prevalence in participating population).
Study or subgroup
Hepatitis B vaccine
Control
n/N
Risk Ratio
n/N
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
3 High prevalence (> 8%)
Liao 1999
257/308
296/341
0.96 [ 0.90, 1.03 ]
Tsega 1990
16/150
21/150
0.76 [ 0.41, 1.40 ]
Wang 1985
0/10
0/10
Not estimable
43/126
45/135
1.02 [ 0.73, 1.44 ]
594
636
0.96 [ 0.89, 1.03 ]
Xueliang 2000
Subtotal (95% CI)
Total events: 316 (Hepatitis B vaccine), 362 ( Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
0.5
1
Favours HB vaccine
2
Favours control
Analysis 1.4. nbsp; Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at
followTeXhyphen;up, Outcome 4 HBsAg at followTeXhyphen;up (subTeXhyphen;group analysis based on
duration of followTeXhyphen;up).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 4 HBsAg at follow-up (sub-group analysis based on duration of follow-up)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
1 Follow-up less than 1 year
Tsega 1990
16/150
21/150
6.1 %
0.76 [ 0.41, 1.40 ]
Wang 1985
0/10
0/10
0.0 %
Not estimable
160
160
6.1 %
0.76 [ 0.41, 1.40 ]
0
0.0 %
Not estimable
Subtotal (95% CI)
Total events: 16 (Hepatitis B vaccine), 21 ( Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.87 (P = 0.38)
2 Follow-up between 1 and 5 years
Subtotal (95% CI)
0
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
0.2
0.5
Favours HB vaccine
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued . . . )
39
(. . .
Study or subgroup
Hepatitis B vaccine
Control
n/N
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Continued)
Risk Ratio
M-H,Fixed,95% CI
Test for overall effect: not applicable
3 Follow-up more than 5 years
Liao 1999
257/308
296/341
81.3 %
0.96 [ 0.90, 1.03 ]
43/126
45/135
12.6 %
1.02 [ 0.73, 1.44 ]
434
476
93.9 %
0.97 [ 0.90, 1.04 ]
636
100.0 %
0.96 [ 0.89, 1.03 ]
Xueliang 2000
Subtotal (95% CI)
Total events: 300 (Hepatitis B vaccine), 341 ( Control)
Heterogeneity: Chi2 = 0.17, df = 1 (P = 0.68); I2 =0.0%
Test for overall effect: Z = 0.83 (P = 0.41)
Total (95% CI)
594
Total events: 316 (Hepatitis B vaccine), 362 (
Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
0.2
0.5
1
Favours HB vaccine
Review:
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 4 HBsAg at follow-up (sub-group analysis based on duration of follow-up)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
1 Follow-up less than 1 year
Tsega 1990
16/150
21/150
0.76 [ 0.41, 1.40 ]
Wang 1985
0/10
0/10
Not estimable
160
160
0.76 [ 0.41, 1.40 ]
Subtotal (95% CI)
Total events: 16 (Hepatitis B vaccine), 21 ( Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.87 (P = 0.38)
0.2
0.5
Favours HB vaccine
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
40
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 4 HBsAg at follow-up (sub-group analysis based on duration of follow-up)
Study or subgroup
Hepatitis B vaccine
Control
n/N
Risk Ratio
n/N
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
3 Follow-up more than 5 years
Liao 1999
Xueliang 2000
257/308
296/341
0.96 [ 0.90, 1.03 ]
43/126
45/135
1.02 [ 0.73, 1.44 ]
434
476
0.97 [ 0.90, 1.04 ]
Subtotal (95% CI)
Total events: 300 (Hepatitis B vaccine), 341 ( Control)
Heterogeneity: Chi2 = 0.17, df = 1 (P = 0.68); I2 =0.0%
Test for overall effect: Z = 0.83 (P = 0.41)
0.2
0.5
1
Favours HB vaccine
2
5
Favours control
Analysis 1.5. nbsp; Comparison 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at
followTeXhyphen;up, Outcome 5 HBsAg at followTeXhyphen;up (subTeXhyphen;group analysis based on
vaccination schedule).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 5 HBsAg at follow-up (sub-group analysis based on vaccination schedule)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
1 Three doses administered with 0,1,6 schedule
Liao 1999
Xueliang 2000
Subtotal (95% CI)
257/308
296/341
81.3 %
0.96 [ 0.90, 1.03 ]
43/126
45/135
12.6 %
1.02 [ 0.73, 1.44 ]
434
476
93.9 %
0.97 [ 0.90, 1.04 ]
0
0
0.0 %
Not estimable
16/150
21/150
6.1 %
0.76 [ 0.41, 1.40 ]
Total events: 300 (Hepatitis B vaccine), 341 ( Control)
Heterogeneity: Chi2 = 0.17, df = 1 (P = 0.68); I2 =0.0%
Test for overall effect: Z = 0.83 (P = 0.41)
2 Three doses administered with an accelerated schedule
Subtotal (95% CI)
Total events: 0 (Hepatitis B vaccine), 0 ( Control)
Heterogeneity: not applicable
Test for overall effect: not applicable
3 Four doses administered by any schedule
Tsega 1990
0.2
0.5
Favours HB vaccine
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued . . . )
41
(. . .
Study or subgroup
Hepatitis B vaccine
Control
n/N
Risk Ratio
n/N
Wang 1985
Subtotal (95% CI)
Weight
M-H,Fixed,95% CI
Continued)
Risk Ratio
M-H,Fixed,95% CI
0/10
0/10
0.0 %
Not estimable
160
160
6.1 %
0.76 [ 0.41, 1.40 ]
636
100.0 %
0.96 [ 0.89, 1.03 ]
Total events: 16 (Hepatitis B vaccine), 21 ( Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.87 (P = 0.38)
Total (95% CI)
594
Total events: 316 (Hepatitis B vaccine), 362 (
Control)
Heterogeneity: Chi2 = 0.71, df = 2 (P = 0.70); I2 =0.0%
Test for overall effect: Z = 1.13 (P = 0.26)
0.2
0.5
Favours HB vaccine
Review:
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 5 HBsAg at follow-up (sub-group analysis based on vaccination schedule)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
1 Three doses administered with 0,1,6 schedule
Liao 1999
Xueliang 2000
Subtotal (95% CI)
257/308
296/341
0.96 [ 0.90, 1.03 ]
43/126
45/135
1.02 [ 0.73, 1.44 ]
434
476
0.97 [ 0.90, 1.04 ]
Total events: 300 (Hepatitis B vaccine), 341 ( Control)
Heterogeneity: Chi2 = 0.17, df = 1 (P = 0.68); I2 =0.0%
Test for overall effect: Z = 0.83 (P = 0.41)
0.2
0.5
Favours HB vaccine
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
42
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 1 Hepatitis B vaccine versus control/placebo: presence of HBsAg at follow-up
Outcome: 5 HBsAg at follow-up (sub-group analysis based on vaccination schedule)
Study or subgroup
Hepatitis B vaccine
Control
n/N
Risk Ratio
n/N
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
3 Four doses administered by any schedule
Tsega 1990
16/150
21/150
0.76 [ 0.41, 1.40 ]
Wang 1985
0/10
0/10
Not estimable
160
160
0.76 [ 0.41, 1.40 ]
Subtotal (95% CI)
Total events: 16 (Hepatitis B vaccine), 21 ( Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.87 (P = 0.38)
0.2
0.5
Favours HB vaccine
1
2
5
Favours control
Analysis 2.1. nbsp; Comparison 2 HB vaccine versus control: presence of antibodies to HBc antigen in
serum, Outcome 1 AntiTeXhyphen;HBc antibodies in serum.
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 2 HB vaccine versus control: presence of antibodies to HBc antigen in serum
Outcome: 1 Anti-HBc antibodies in serum
Study or subgroup
Hepatitis B vaccine
n/N
Liao 1999
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
259/308
310/341
69.5 %
0.93 [ 0.87, 0.98 ]
Tsega 1990
20/150
24/150
5.7 %
0.83 [ 0.48, 1.44 ]
Wang 1985
0/10
0/10
0.0 %
Not estimable
69/126
109/135
24.9 %
0.68 [ 0.57, 0.81 ]
594
636
100.0 %
0.86 [ 0.80, 0.92 ]
Xueliang 2000
Total (95% CI)
Total events: 348 (Hepatitis B vaccine), 443 (
Control)
Heterogeneity: Chi2 = 12.82, df = 2 (P = 0.002); I2 =84%
Test for overall effect: Z = 4.55 (P < 0.00001)
0.2
0.5
Favours treatment
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
43
Analysis 2.2. nbsp; Comparison 2 HB vaccine versus control: presence of antibodies to HBc antigen in
serum, Outcome 2 AntiTeXhyphen;HBc antibodies in serum (randomTeXhyphen;effects model).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 2 HB vaccine versus control: presence of antibodies to HBc antigen in serum
Outcome: 2 Anti-HBc antibodies in serum (random-effects model)
Study or subgroup
Hepatitis B vaccine
n/N
Liao 1999
Control
Risk Ratio
n/N
Weight
M-H,Random,95% CI
Risk Ratio
M-H,Random,95% CI
259/308
310/341
44.8 %
0.93 [ 0.87, 0.98 ]
Tsega 1990
20/150
24/150
16.7 %
0.83 [ 0.48, 1.44 ]
Wang 1985
0/10
0/10
0.0 %
Not estimable
69/126
109/135
38.5 %
0.68 [ 0.57, 0.81 ]
594
636
100.0 %
0.81 [ 0.61, 1.07 ]
Xueliang 2000
Total (95% CI)
Total events: 348 (Hepatitis B vaccine), 443 (
Control)
Heterogeneity: Tau2 = 0.05; Chi2 = 12.82, df = 2 (P = 0.002); I2 =84%
Test for overall effect: Z = 1.50 (P = 0.13)
0.2
0.5
1
Favours treatment
2
5
Favours control
Analysis 3.1. nbsp; Comparison 3 HB vaccine versus control: adverse events, Outcome 1 Lack of
seroTeXhyphen;protection.
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 3 HB vaccine versus control: adverse events
Outcome: 1 Lack of sero-protection
Study or subgroup
Hepatitis B vaccine
n/N
Liao 1999
Control
Risk Ratio
n/N
Weight
M-H,Random,95% CI
Risk Ratio
M-H,Random,95% CI
282/308
323/341
33.9 %
0.97 [ 0.93, 1.01 ]
Tsega 1990
38/150
150/150
32.6 %
0.25 [ 0.19, 0.33 ]
Xueliang 2000
82/126
120/135
33.5 %
0.73 [ 0.64, 0.84 ]
584
626
100.0 %
0.57 [ 0.25, 1.27 ]
Total (95% CI)
Total events: 402 (Hepatitis B vaccine), 593 (
Control)
Heterogeneity: Tau2 = 0.49; Chi2 = 223.95, df = 2 (P<0.00001); I2 =99%
Test for overall effect: Z = 1.37 (P = 0.17)
0.2
0.5
Favours treatment
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
44
Analysis 3.2. nbsp; Comparison 3 HB vaccine versus control: adverse events, Outcome 2 Lack of
seroTeXhyphen;protection (fixedTeXhyphen;effect model).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 3 HB vaccine versus control: adverse events
Outcome: 2 Lack of sero-protection (fixed-effect model)
Study or subgroup
Hepatitis B vaccine
n/N
Liao 1999
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
282/308
323/341
53.6 %
0.97 [ 0.93, 1.01 ]
Tsega 1990
38/150
150/150
26.2 %
0.25 [ 0.19, 0.33 ]
Xueliang 2000
82/126
120/135
20.2 %
0.73 [ 0.64, 0.84 ]
584
626
100.0 %
0.73 [ 0.69, 0.78 ]
Total (95% CI)
Total events: 402 (Hepatitis B vaccine), 593 (
Control)
Heterogeneity: Chi2 = 223.95, df = 2 (P<0.00001); I2 =99%
Test for overall effect: Z = 10.66 (P < 0.00001)
0.2
0.5
1
Favours treatment
2
5
Favours control
Analysis 4.1. nbsp; Comparison 4 HB vaccine versus control: serious adverse events, Outcome 1 Serious
adverse events: allTeXhyphen;cause mortality.
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 4 HB vaccine versus control: serious adverse events
Outcome: 1 Serious adverse events: all-cause mortality
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
Greenberg 1996
10/5056
9/5261
100.0 %
1.16 [ 0.47, 2.84 ]
Total (95% CI)
5056
5261
100.0 %
1.16 [ 0.47, 2.84 ]
Total events: 10 (Hepatitis B vaccine), 9 ( Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.32 (P = 0.75)
0.2
0.5
Favours HB vaccine
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
45
Analysis 5.1. nbsp; Comparison 5 HB vaccine versus control: adverse events, Outcome 1 Fever.
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 5 HB vaccine versus control: adverse events
Outcome: 1 Fever
Study or subgroup
Hepatitis B vaccine
Control
n/N
Bassily 1995
Risk Ratio
n/N
Weight
M-H,Random,95% CI
Risk Ratio
M-H,Random,95% CI
96/1184
21/588
46.8 %
2.27 [ 1.43, 3.60 ]
Greenberg 1996
358/6213
380/6184
53.2 %
0.94 [ 0.82, 1.08 ]
Total (95% CI)
7397
6772
100.0 %
1.42 [ 0.60, 3.38 ]
Total events: 454 (Hepatitis B vaccine), 401 (
Control)
Heterogeneity: Tau2 = 0.36; Chi2 = 12.98, df = 1 (P = 0.00032); I2 =92%
Test for overall effect: Z = 0.79 (P = 0.43)
0.2
0.5
1
Favours treatment
2
5
Favours control
Analysis 5.2. nbsp; Comparison 5 HB vaccine versus control: adverse events, Outcome 2 Fever
(fixedTeXhyphen;effect model).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 5 HB vaccine versus control: adverse events
Outcome: 2 Fever (fixed-effect model)
Study or subgroup
Hepatitis B vaccine
Control
n/N
Bassily 1995
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
96/1184
21/588
6.9 %
2.27 [ 1.43, 3.60 ]
Greenberg 1996
358/6213
380/6184
93.1 %
0.94 [ 0.82, 1.08 ]
Total (95% CI)
7397
6772
100.0 %
1.03 [ 0.90, 1.18 ]
Total events: 454 (Hepatitis B vaccine), 401 (
Control)
Heterogeneity: Chi2 = 12.98, df = 1 (P = 0.00032); I2 =92%
Test for overall effect: Z = 0.42 (P = 0.67)
0.2
0.5
Favours treatment
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
46
Analysis 5.3. nbsp; Comparison 5 HB vaccine versus control: adverse events, Outcome 3 Local reactions
(pain, redness, soreness).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 5 HB vaccine versus control: adverse events
Outcome: 3 Local reactions (pain, redness, soreness)
Study or subgroup
Hepatitis B vaccine
n/N
Bassily 1995
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
88/1184
14/588
91.8 %
3.12 [ 1.79, 5.44 ]
Giammanco 1991
4/111
1/21
8.2 %
0.76 [ 0.09, 6.44 ]
Total (95% CI)
1295
609
100.0 %
2.93 [ 1.72, 4.99 ]
Total events: 92 (Hepatitis B vaccine), 15 (
Control)
Heterogeneity: Chi2 = 1.58, df = 1 (P = 0.21); I2 =37%
Test for overall effect: Z = 3.94 (P = 0.000080)
0.1
1
Favours treatment
10
Favours control
Analysis 5.4. nbsp; Comparison 5 HB vaccine versus control: adverse events, Outcome 4 Other adverse
events.
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 5 HB vaccine versus control: adverse events
Outcome: 4 Other adverse events
Study or subgroup
Hepatitis B vaccine
n/N
Ambrosch 1992
Total (95% CI)
Total events: 7 (Hepatitis B vaccine), 3 (
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
7/55
3/55
100.0 %
2.33 [ 0.64, 8.56 ]
55
55
100.0 %
2.33 [ 0.64, 8.56 ]
Control)
Heterogeneity: not applicable
Test for overall effect: Z = 1.28 (P = 0.20)
0.2
0.5
Favours treatment
1
2
5
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
47
Analysis 6.1. nbsp; Comparison 6 HB vaccine versus control: lack of compliance, Outcome 1 Lack of
compliance.
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 6 HB vaccine versus control: lack of compliance
Outcome: 1 Lack of compliance
Study or subgroup
Hepatitis B vaccine
Control
n/N
Risk Ratio
n/N
Weight
M-H,Random,95% CI
Risk Ratio
M-H,Random,95% CI
Aristegui 1995
65/655
402/731
50.9 %
0.18 [ 0.14, 0.23 ]
Bassily 1995
49/394
5/196
49.1 %
4.88 [ 1.97, 12.04 ]
1049
927
100.0 %
0.91 [ 0.03, 24.39 ]
Total (95% CI)
Total events: 114 (Hepatitis B vaccine), 407 (
Control)
Heterogeneity: Tau2 = 5.51; Chi2 = 49.42, df = 1 (P<0.00001); I2 =98%
Test for overall effect: Z = 0.06 (P = 0.96)
0.1
1
Favours HB vaccine
10
Favours control
Analysis 6.2. nbsp; Comparison 6 HB vaccine versus control: lack of compliance, Outcome 2 Lack of
compliance (fixedTeXhyphen;effect model).
Review:
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status
Comparison: 6 HB vaccine versus control: lack of compliance
Outcome: 2 Lack of compliance (fixed-effect model)
Study or subgroup
Hepatitis B vaccine
n/N
Control
Risk Ratio
n/N
Weight
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
Aristegui 1995
65/655
402/731
98.3 %
0.18 [ 0.14, 0.23 ]
Bassily 1995
49/394
5/196
1.7 %
4.88 [ 1.97, 12.04 ]
1049
927
100.0 %
0.26 [ 0.21, 0.32 ]
Total (95% CI)
Total events: 114 (Hepatitis B vaccine), 407 (
Control)
Heterogeneity: Chi2 = 49.42, df = 1 (P<0.00001); I2 =98%
Test for overall effect: Z = 12.69 (P < 0.00001)
0.1
Favours HB vaccine
1
10
Favours control
Hepatitis B immunisation in persons not previously exposed to hepatitis B or with unknown exposure status (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
48