Haemophilus influenzae, mainly serotype b (Hib), is a major cause of childhood meningitis, pneumonia, epiglottitis, septicaemia, cellulitis, osteomyelitis and septic arthritis worldwide.1 In the mid-1980s, the Hib conjugate vaccine was proven to have high short-term efficacy in preventing childhood invasive disease.2^–5 As a result, the vaccine was incorporated into national infant immunisation programmes with a schedule of two or three doses in the first 6 months of life, followed by a booster dose in late infancy or the second year of life.

In the UK, an accelerated immunisation schedule at 2, 3 and 4 months was introduced in 1990 to provide early protection against pertussis and to improve vaccine coverage.6 Prior to implementation, the Hib conjugate vaccine administered at the UK schedule was shown to be well-tolerated and to produce antibody consistent with protection (⩾0.15 μg/ml) in almost all children when given with the combined diphtheria, tetanus, whole-cell pertussis (DTwP) and oral polio vaccines.7 8

In October 1992, the Hib conjugate vaccine was introduced as part of the UK immunisation programme at 2, 3 and 4 months, with no booster dose. At the same time, a 12-month catch-up campaign took place to vaccinate older children who were also at high risk of invasive Hib disease. This campaign involved offering three doses of vaccine to infants between 4 and 12 months of age and a single dose to children aged 12–48 months.

The introduction of the Hib conjugate vaccine along with the catch-up campaign led to a rapid and sustained decline in invasive Hib disease.9 This decline was initially observed in the age group targeted for vaccination.10 However, as toddlers have the highest rates of asymptomatic pharyngeal carriage of Hib11 12 and the Hib conjugate vaccine has been shown to dramatically reduce Hib carriage,13^–16 this campaign also provided indirect protection (herd immunity) by reducing transmission of Hib across all age groups.17 The success of the vaccination programme was, therefore, attributed both to high vaccine coverage and the concomitant catch-up campaign.18

From 1999, the number of cases of invasive Hib disease started to increase in all age groups, including adults, but was most marked in children under 5 years of age.19 The reasons for this increase included a decline in indirect protection,20 a greater than expected decline in vaccine effectiveness among children vaccinated in infancy21 and a temporary change in the Hib vaccine combination offered to young infants (table 1). As a result of this increase, a Hib vaccination booster campaign was conducted between May and September 2003, offering one dose of vaccine to all children who were aged between 6 months and 4 years on 1 April 2003 (ie, were born between 2 April 1999 and 1 October 2002) and to children who reached 6 months of age during the campaign. This study reports the impact of the booster campaign on the number of invasive Hib infection cases reported in England and Wales up to December 2006.

METHODS

In 2004, Hib vaccination status for children targeted in the campaign was collected through computerised child health systems by primary care trusts (PCTs). The population denominator includes all children in the target cohort for whom the PCT is responsible. For each PCT, information was requested on the number of children eligible for vaccination and the number of those who received a Hib booster dose between May and September 2003 in two cohorts: those aged 6–12 months (born between 1 April 2002 and 30 September 2002) and those aged 1–3 years (born between 1 April 1999 and 31 March 2002). PCTs that did not provide data for the catch-up campaign or reported coverage rates below 50% were contacted to discuss possible reasons for this.

As part of on-going surveillance, the Health Protection Agency (HPA) estimates the incidence of Hib in England and Wales from the number of H influenzae isolates referred to the HPA Haemophilus Reference Unit for serotyping and from hospital laboratory reports of confirmed Hib infections to the HPA Centre for Infections.22 Laboratory reports of confirmed infections from microbiologists in nearly 400 hospital and public health laboratories have been received at the HPA Centre for Infections since 1975.22 A case of Hib was defined as H influenzae isolated from a normally sterile site and confirmed as Hib, through standard slide agglutination and molecular techniques,23 by a hospital laboratory or the HPA Haemophilus Reference Unit.21 Children with a clinical diagnosis of epiglottitis were included if associated with isolation of Hib from a sterile site. Mid-year population estimates for children <5 years of age were obtained from the Office for National Statistics website (http://www.statistics.gov.uk).

RESULTS

Coverage data were available for 288/303 (95%) PCTs in England. The average coverage rate for the Hib booster programme was 71% (190 457/266 618) for the 6–12-month-old group and 63% (1 020 294/1 617 957) for the 13–48-month-old group. All regions except for London (54% and 45%, respectively) reported a >70% coverage for the 6–12-month-old group and >65% coverage for the 13–48-month-old group. Among the 15 PCTs that did not provide coverage data, the main reason given was problems with computer software. Of the 288 PCTs that provided data, only 19 (7%) reported coverage of <50%. Of these, 15 (79%) were in London, with the remaining four scattered around England. Reasons given by immunisation leads for most PCTs with poor coverage included a very deprived, highly mobile population where it is difficult to maintain accurate data for place of residence and registration with general practitioners. In addition, some PCTs reported an affluent population that tended to access private health care, which does not provide immunisation data. Although the number of cases in each PCT was low, there was no obvious excess of Hib cases in PCTs with low vaccine coverage (data not shown).

Table 2 shows the number of cases of culture-confirmed invasive Hib disease in different age groups. In children under 5 years of age, the number of cases began to rise after 1999 and peaked in 2002. This rise was greatest in children aged 1 and 2 years. The number of cases in the older age group also increased during this period, particularly among those aged 15 years or above, but lagged a year behind the paediatric peak. The change in combination vaccine type and the introduction of the booster campaign in 2003 resulted in a rapid decline in the number of cases within the targeted age group (table 2), particularly among 1-year-old children, where the number of cases fell by 82% between 2002 (56 cases) and 2004 (10 cases). In children <5 years of age, the incidence fell from 4.6 (95% confidence interval (CI) 3.9 to 5.4) per 100 000 in 2002 (139 cases) to 1.1 (95% CI 0.8 to 1.1) per 100 000 in 2004 (34 cases). In 2005, however, the number of cases increased among 2-year-old children and, to a lesser extent, among 3-year-old children, and this upward trend continued in 2006. Of the 18 cases in children aged 1–4 years in 2004, 33% (6/18 cases) were too young to be eligible for vaccination in the booster campaign. This proportion rose to 70% in 2005 (19/27 cases) and 76% in 2006 (26/34 cases).

The booster campaign also led to a significant decline in the number of Hib cases among older children and adults. In the latter group, the number of cases fell by 51% within 3 years of the campaign (table 2). This fall is similar to the 49% reduction observed within 3 years in the number of adult cases following the introduction of the Hib conjugate vaccine and the associated catch-up campaign in 1992 (table 2).

DISCUSSION

The introduction of the Hib conjugate vaccine in 1992 alongside a catch-up campaign for children up to 4 years of age led to a rapid, dramatic and sustained reduction in invasive Hib disease across all age groups. In children under 5 years of age, the number of reported cases in England and Wales fell from almost 500 in the pre-vaccine era to around 20 within 2 years of implementation of the vaccination programme. At that time, follow-up studies showed that anti-Hib antibodies remained above protective levels in almost all children up to 12 months after vaccination.5 24 In addition, although antibody levels were subsequently shown to fall significantly after this time,18 estimated vaccine effectiveness only fell from 99% in children aged 5–11 months to 97% in those aged 12–71 months.18 Therefore, because the evidence suggested that clinical protection was sustained, a booster dose was not considered necessary.

In 1999, the number of reported invasive Hib disease cases began to rise and, in children under 5 years of age, peaked at 120 in 2002. A major reason for this increase was the fact that the impact of the catch-up campaign had expired. In this campaign, most of the children had been vaccinated above the age of 12 months, when even a single dose of vaccine results in high antibody levels leading to high and prolonged direct protection in these cohorts.25 By 1998, all cohorts under 5 years of age were only eligible for the routine infant immunisation schedule. In addition, the catch-up campaign also contributed to the elimination of asymptomatic carriage,15 resulting in high levels of indirect protection (herd immunity), which is estimated to have contributed to at least a third of the >95% reduction in invasive Hib disease since the introduction of the Hib vaccine.21

Another major contributor to the observed increase from 1999 is the greater than expected decline in vaccine effectiveness among children vaccinated in infancy.21 In children under 5 years of age who developed invasive Hib disease between 1993 and 2000, for example, vaccine effectiveness measured by the screening method, unlike previous estimates that only measured direct protection,18 was estimated to be only around 57% (95% CI 43 to 67%). In addition, vaccine effectiveness declined significantly with time since vaccination (p = 0.004).21

The rise in number of Hib cases after 1998 was amplified in 2000 and 2001 when some children received diphtheria-tetanus-pertussis-Hib (DTP-Hib) combination vaccines containing acellular pertussis (aP) because of supply problems with combined vaccines containing whole cell pertussis (wP). Several DTaP-Hib vaccines have been shown to have reduced Hib immunogenicity,26 27 especially if given in an accelerated immunisation schedule such as that used in the UK.26 The reduced immunogenicity of these vaccines, however, was not considered to be clinically relevant27 28 until the rise in Hib cases was observed in the UK. A case–control study at that time showed that a higher proportion of children with Hib vaccine failure had received three doses of DTaP-Hib vaccine for their primary immunisation schedule compared with vaccinated controls who had not developed invasive Hib disease.29 Furthermore, there was a trend for an increased risk of vaccine failure with each additional dose of the DTaP-Hib vaccine, an observation confirmed by a second study.30 In a separate study, analysis of vaccine effectiveness using the screening method also showed reduced vaccine efficacy in children immunised between 2000 and 2002, when the DTaP-Hib vaccine was in circulation.21

In addition to the increase in invasive Hib disease in children, the number of cases in adults also rose between 1998 and 2003 to levels that were higher than the annual average of around 75 cases in the pre-vaccine era. This was attributed to a reduction in opportunities for natural boosting of immunity, resulting in a significant decline in serum antibody levels to Hib among adults after the vaccine was introduced into the infant immunisation programme.20 The resurgence of invasive Hib disease in younger children after 1999 then increased the risk of exposure to the organism, and therefore risk of infection, among susceptible adults.20

The 2003 booster campaign had an immediate impact in rapidly reducing the total number of cases across all age groups. Vaccine coverage for the Hib booster campaign was lower than for the infant immunisation programme, which has been consistently above 90% throughout most of the UK.31 This is not surprising as systems for inviting infants for their routine infant immunisations at the appropriate time have been in place for many years. In contrast, the Hib booster campaign was a one-off programme, over only a short period of a few months, that invited children at an age when parents would not be expecting their child to be vaccinated.

Vaccine coverage for the Hib booster campaign was also lower than for similar age groups targeted in the meningococcal C catch-up campaign where coverage was 80% among infants aged 5–11 months (this age group received two doses of vaccine), 84% among 1-year-old children and 74% among 2–4-year-old children.32 Reasons for the lower coverage include the much higher profile of the meningococcal C campaign, accompanied by extensive publicity, and the longer duration of the campaign (approximately 12 months compared with 5 months for the Hib booster programme). Despite this, the Hib booster campaign has had a major impact on disease control by targeting the age group with the largest increase in cases, many of whom will have received the less effective DTaP-Hib in their primary infant immunisation schedule. In addition, this age group may have become colonised with Hib16 and therefore could have transmitted the organism to others. The decline in Hib cases following the booster campaign was initially observed in the age groups targeted, but was soon followed by a reduction in older children and adults. In the latter group, the reduction in the number of cases was similar to that in the initial vaccination campaign in 1992, suggesting that the booster campaign has, once again, contributed significantly to indirect protection among adults.

In 2003, the implicated DTaP preparation was withdrawn and the use of DTwP preparations resumed. In September 2004, the recommended vaccine for infants was changed to a vaccine containing DTaP, inactivated polio and Hib (eg, Pediacel). This combination vaccine has a different acellular pertussis component to that implicated in the increase in invasive Hib disease in children and has been shown to have a satisfactory immune response to the Hib component.33 This vaccine is preferred to the DTwP combination because it is less reactogenic than whole cell pertussis34^–36 and removes the risk of vaccine-associated paralytic poliomyelitis with oral polio vaccine.37

In September 2006, the Department of Health introduced a booster dose of the Hib conjugate vaccine at 12 months of age into the national immunisation schedule. It is hoped that this booster dose will reduce the incidence of invasive Hib disease in toddlers and, by reducing asymptomatic carriage, will also help prevent disease in older children and adults. However, one particular birth cohort born between March 2003 and September 2005 (approximately 1.5 million children) was not eligible for the booster campaign or for the scheduled booster dose at 12 months of age. As protection wanes, these children may be at higher risk of infection compared to those who have received a booster dose. This observation is supported by our surveillance data, which show an increase in the number of cases in the respective age groups in 2005 and 2006. This group of children has been targeted in the catch-up programme announced by the Chief Medical Officer on 23 July 2007 (www.dh.gov.uk/en/Publicationsandstatistics/Lettersandcirculars/Professionalletters/Chiefmedicalofficerletters/DH_076964). A Hib booster is being offered to children born between 13 March 2003 and 3 September 2005 at the pre-school stage and should further reduce the number of invasive Hib cases in children and adults over the next few years.