Investigating the pertussis resurgence in England and Wales, and options for future control

Abstract

Background: In 2012 England and Wales experienced a resurgence of pertussis and an increase in infant deaths. This occurred 8 years after acellular pertussis (aP) vaccine replaced whole cell (wP) primary vaccine despite continued high coverage for the primary series and pre-school aP booster. We developed a mathematical model to describe pertussis transmission dynamics in England and Wales since the 1950s and used it to investigate the cause of the resurgence and the potential impact of additional vaccination strategies.

Methods: An age-structured, compartmental, deterministic model of the pertussis transmission dynamics was fitted to 60 continuous years of age-stratified pertussis notification data in England and Wales. The model incorporated vaccine-induced and natural immunity and differentiated between vaccine-induced protection against clinical disease and infection.

Results: The degree of protection of wP vaccine against infection was estimated to be higher than that of aP vaccine. Furthermore, the duration of protection for natural and wP-induced immunity was likely to be at least 15 years, but for aP vaccine it could be as low as 5 years. Model results indicated that the likely cause of the resurgence was the replacement of wP by less efficacious aP vaccine and that an elevated level of pertussis would continue. The collapse in wP vaccine coverage in the 1970s and resultant outbreaks in the late 1970s and early 1980s could not explain the resurgence. Addition of an adolescent or toddler booster was predicted to have little impact on the disease in infants.

Conclusions: Our findings support the recent recommendation by the World Health Organisation that countries currently using wP vaccine for primary immunisation should not change to aP vaccine unless additional strategies to control infant disease such as maternal immunisation can be assured. Improved pertussis vaccines that provide better protection against infection are needed.

Keywords: Intervention; Mathematical model; Pertussis; Resurgence; Transmission; Vaccine.

Fig. 1

Number of annual notifications (a) between 1954 and 2013 and (b) between 1986 and 2013 in England and Wales in seven age groups, and the historical vaccine coverage (course completed) for the primary and pre-school booster programmes (blue line: wP primary vaccine coverage, red line: aP primary vaccine coverage, black line: aP pre-school booster vaccine coverage). Coverage data for 1957 to 1966 not collected but assumed to be the same as in 1967 as it was added into an existing DT vaccine (available to download in Additional file 1)

Fig. 2

Flowchart of the pertussis transmission dynamic model with acellular and whole cell vaccine-protected groups. (The difference equations describing the flows between compartments are given in Appendix 1 in Additional file 2)

Fig. 3

Mean, maximum and minimum values of the force of infection by nine age groups derived from the 1956 pre-vaccination equilibrium in England and Wales (among 45 combined scenarios with 9 different durations of natural immunity and 5 different values of α2/α1)

Fig. 4

Box plots of parameter estimates from the 2000 best fitting aP scenarios to historical notification data between 1956 and 2013 in England and Wales. a Duration of natural immunity; b α2/α1; c duration of wP vaccine protection; d degree of whole cell vaccine protection against infection; e duration of aP vaccine protection; f degree of acellular vaccine protection against infection. Red line for median, blue box for lower and upper quartiles and black bars for minimum and maximum

Fig. 5

Notification data (red line) between 1956 and 2013 in England and Wales and corresponding model outputs from best fitted scenarios (blue box plots minimum, lower quartile, median, upper quartile and maximum) by seven age groups (a–g) and the overall population (h)

Fig. 6

Model outputs of the annual pertussis notification among infants less than a year old between 1956 and 2030 in England and Wales with the ten different vaccination programme options as shown in Table 2 (excludes 6B, as it shows similar results to those of 6A) and blue box plots showing minimum, lower quartile, median, upper quartile and maximum of the annual pertussis notifications from the best fitting scenarios)