Background The incidence of meningococcal disease (MD) in the United Kingdom increased dramatically from 2.4 per 100,000 in the early 1990's to 4.4 per 100,000 by 1996. The age distribution also shifted towards older teenagers in whom the risk profile for disease is poorly understood and the mortality rate higher. Adolescence is a time of biological, social and psychological change and these changes may contribute towards the risk. This study is a prospective population-based case-control study to identify risk and protective factors for MD in adolescents between the ages of 15 and 19 years. Methods Adolescent subjects with MD were recruited at hospital admission in six regions of England from January 1999 to June 2000. One age- and sex-matched control from the same geographical location was recruited per case. Blood samples plus pernasal and throat swabs were taken from cases at hospitalisation and from cases and controls at interview. Data on potential risk and protective factors were gathered at interview. Results 144 case control pairs were recruited (51% male median age 17.6) of which 114 cases (79%) were confirmed microbiologically. Significant independent risk factors for MD were history of preceding illness (matched OR 2.9 95% CI: 1.4-5.9), intimate kissing with multiple partners (OR 3.7, 95% CI: 1.7-8.1), being a university student (OR 3.4, 95% CI: 1.2-10) and preterm birth (OR 3.7, 95% CI: 1.0-13.5). Religious observance (OR 0.09, 95% CI: 0.02-0.6) and meningococcal vaccination (OR 0.12, 95% CI: 0.04-0.4) were protective. Conclusions This is the first large study examining risk factors for MD within an adolescent population. Activities and events increasing risk for MD in adolescence are different from childhood. Students are at higher risk for disease than their counterparts. While altering personal behaviours may moderate risk of MD, it is unlikely that behaviour- based health promotion can significantly reduce disease burden. The development of effective meningococcal vaccines remains a key public health priority.

Neisseria meningitidis is an important cause of meningitis and bacteraemia worldwide and is associated with high case-fatality rates. Meningococcal disease continues to remain a public health issue in Scotland and the rest of Europe. Typing methods are used for epidemiological purposes to investigate outbreaks and the spread of meningococci and to examine the population structure of the organism in order to better understand its variation and evolution. Reference institutes have employed such methods for a number of decades for the diagnosis and detection of meningococci. However, phenotypic methods for serogrouping, serotyping and serosubtyping meningococci, although providing good strain information, can lead to endemic strains appearing identical using these methods when they are in fact quite different. More recently methods have been developed to further characterise bacteria. These methods have included PCR for the detection of meningococcal disease within blood, serogrouping and sequencing of housekeeping genes (MLST) and antigen genes such as PorA. These molecular epidemiological methods were used for the retrospective typing of invasive meningococci in Scotland, 1972-1998, using a fully automated procedure. The results of these were then analysed using statistical packages to examine the population structure of the organism. In total there were 2517 invasive isolates, received by the Scottish Meningococcus and Pneumococcus Reference Laboratory (SMPRL) from the start of 1972 to the end of 1998. Serogroup distribution changed from year to year during the time period 1972-1998 but serogroups B and C were dominant throughout this period. Serogroup B was the dominant serogroup throughout the seventies and early eighties until serogroup C became dominant during the mid 1980s. This increase in dominance of serogroup C has been found in this study not to be associated with one particular sequence type (ST) but is associated with a number of STs, which include ST-8, ST-11, ST-206 and ST-334. This is in contrast to the increase in serogroup C disease in the 1990s that was due to the ST-11 clonal complex. While there was much diversity in the STs (309 different STs among the 2517 isolates), only ten accounted for 1562 isolates (59.9%). These were ST-11, ST-8, ST-41, ST-153, ST-1, ST-32, ST-33, ST-269, ST-334 and ST-60. There were 177 new STs found during the time period. The STs were further differentiated into 31 clonal complexes, with 57 singleton types. As with the STs, although there was much diversity in the clonal complexes, only seven accounted for 1993 isolates. It was found that with PorA variable region (VR) types there were certain combinations significantly more common than others. There was a strong link with PorA type and ST and more so with clonal complex. This link was evident with the PorA type 5, 2-1, 36-2, which occured in 70 isolates representing the ST-11 complex and in all but two isolates representing ST-11. Similarly PorA type 18-3, 1, 35-1 was associated with 15 isolates belonging to the ST41-44 complex. However, this was not the case with all PorA combinations as the PorA type 19, 15, 36 was associated with 10 different complexes. There was some association between serogroup and PorA VR types. There was strong evidence of certain VR1, 2 and 3 regions being associated with certain serogroups, although this was not definitive. For example, of 192 isolates with PorA type 19, 15, 36, 85.4% were associated with serogroup B. Genosubtyping of the porA gene has been shown to increase the power of differentiation within clonal meningococcal populations. For, example, seven isolates that had the same serogroup, ST, VR1 and VR2 could be differentiated by their VR3 type. Using cluster detection software SaTScan to analyse all isolates, it was found there were 29 clusters in Scotland, from 1972-1998. These clusters included 63 cases, which accounted for 2.5% of all cases. A range of different strains caused the clusters that were identified in this study, some caused by hypervirulent strains. These strain types were responsible for a number of cases throughout the world as well as in Scotland during the period of this study. However it was also shown that there were clusters identified in this study caused by lesser-known strain types that were not responsible for many cases and that appear to be unique to Scotland or the UK. This study is the first to look at the detection of clusters over a time period of 26 years and to identify clusters that would have previously been unidentified due to lack of suitable characterisation techniques. The results in this study indicate that the multivalent preparation produced by the Netherlands Vaccine Institute (Nonavalent vaccine) had the potential, based on the PorA types that it contains, to prevent the majority of serogroup B infection that had occurred in Scotland, from 1972-1998. It also had the potential, although not to the same extent as serogroup B, to protect against other serogroups. For the age groups that would potentially have been the first to be immunised with any vaccine as part of the childhood vaccination programme, the 0-4 years old group, the potential coverage was over 92% which is comparable with the coverage seen with the serogroup C meningococcal conjugate (MCC) vaccine, of approximately 90%.