Trends in Meningococcal Disease in the United States Military, 1971–2010

When you consider the risks undertaken by US military personnel, do you include risk for disease? Public health officials do. Military personnel are at risk for infectious disease because of crowding, the rigors of physical training, and sometimes unhygienic field conditions. Meningococcal disease (usually manifested as bacterial meningitis or blood-borne infection) can be rapidly fatal. It has historically affected the military more than the general US population. One hundred years' worth of data support this trend from as long ago as World War I. However, in 1970, a policy requiring vaccination of military recruits started lowering the rate of infection, although the rate remained higher than that for the general population. Since 1982, improvements in vaccines have lowered rates even further. As a result of these vaccination efforts, the meningococcal disease rate among military personnel has reached a historic low, which now matches that of the general population.

C ases of meningococcal disease have long plagued the United States military, with incidences (defi ned as cases per 100,000 person-years) as high as 150 during World War I (1) and 80 during World War II (2). Corresponding incidences among the US general population were 9 and 16 during World War I and World War II, respectively (2). In 1969, the incidence among US Army basic trainees was 81 (2). Such elevated incidences relative to the nonage-matched US population were attributed to crowding and unhygienic conditions in unique environments. In particular, these factors imposed higher incidences of disease at military basic training centers. Contributing factors likely included the convergence of people from a wide geographic area and the extreme physical demands of basic military training.
During the past century, US military incidence of meningococcal disease has markedly decreased to converge with that of the (non-age-adjusted) US general population ( Figure 1). Since the 1970s, the military has maintained a policy of universal meningococcal vaccination for all persons entering all branches of service. After the US Army's early 1970s introduction to all incoming personnel of a vaccine targeting N. meningitidis serogroup C, disease rates dropped by >90% (1,4,5). However, during 1971-1989, the Army's mean annual incidence remained signifi cantly higher than that among the non-age-matched general population (3.6 vs. 1.02; p<0.0001; general population data from [3]).
In 1982, a quadrivalent polysaccharide vaccine (MPSV-4; Menomune, Sanofi Pasteur, Bridgewater, NJ, USA) was introduced; this vaccine targets serogroups A, C, W-135, and Y. No broad-coverage vaccine against serogroup B exists (6). During 1982During -1989, meningococcal disease incidence among members of the military was 2.1 (1); for 1990-2009, rates among both the Army and the US general population dropped signifi cantly, with Army rates not signifi cantly different from those observed in the general population (0.5 vs. 0.7; p = 0.19; general population data from [5]).
Despite declining incidence during the past 4 decades, the elevated susceptibility to meningococcal disease among members of the US military makes this population of interest regarding the performance of current vaccines. We report the epidemiology of 26 cases of meningococcal disease that occurred in members of the US military during 2006-2010. Demographics, geographic location, clinical syndrome, vaccination, and death rates are reviewed. Historical and current trends in the military are evaluated and compared with those of the US general population.

Case Information Collection
The Naval Health Research Center (NHRC) monitors reports from the Armed Forces Health Surveillance Center (AFHSC) Defense Medical Surveillance System database (7) and from the EpiData Center at the Navy and Marine Corps Public Health Center. Possible cases of meningococcal disease are identifi ed from the Defense Medical Surveillance System database by using diagnostic code 036 from the International Classifi cation of Diseases, 9th Revision, or by reports from military and US civilian laboratories that have identifi ed N. meningitidis infection in active-duty military members. Possible cases without a confi rmatory laboratory diagnosis are defi ned as probable by the presence of the appropriate clinical syndrome, with laboratory identifi cation of encapsulated, gram-negative diplococci that are not identifi ed as N. meningitidis. A case is defi ned as confi rmed if the patient has the appropriate clinical syndrome and a laboratory identifi cation of N. meningitidis is made by either PCR or culture. PCR and direct fl uorescent antibody testing were used in our laboratories to identify a specifi c serogroup. The NHRC obtains case information and a variety of patient specimen types from the treating hospital.

Statistical Analyses
Analysis of variance and Student t tests were used to compare incidences for which we did not have numerators or denominators over different periods within the Army and military populations and between the Army and military and the general population. The normal approximation was used for the comparison of incidences of >100 cases/year; for incidences <100 cases/year, Poisson modeling was used to evaluate differences between population strata (8). Demographic categories evaluated were patient age, sex, military rank, military branch, region, and military versus US general population; N. meningitidis serogroups were also compared.

Cases
Of the 84 possible cases of meningococcal disease within US military that were reviewed during 2006-2010, the NHRC confi rmed 23 cases by culture, PCR, or both; 1 case was determined to be probable. The NHRC also confi rmed 2 cases that had not previously been identifi ed by the AFHSC. The result of the surveillance was 25 confi rmed cases and 1 probable case, of which 5 were fatal (Table 1); N. meningitidis serogroup breakdown was 6 B, 7 C, and 10 Y. The large number of possible cases reported among the military that were not confi rmed were likely a result of miscoding of an initial or subsequent diagnosis; these included miscoding of a patient with viral or aseptic meningitis (n = 34), a fi nal negative laboratory result after an initial diagnosis of meningococcal disease (n = 19), and a positive result on a nonsterile site only (n = 4).

Age
During 2006-2010, incidence of meningococcal disease in the military among those >17 years of age was not signifi cantly different from that among the agematched general population (0.38 vs. 0.26; p>0.05; generalpopulation data from Centers for Disease Control and Prevention [CDC], unpub. data). No signifi cant difference was found between the military and general population in the 17-to 29-year-old age group (0.55 vs. 0.40, respectively; p>0.05). The peak disease incidence in military personnel occurred among the 17-to 19-year-old age group and was signifi cantly higher than that in their general-population counterparts (2.34 vs. 0.62; p<0.05; general population data from CDC, unpub. data). This difference was not present in older age categories; incidences among both military personnel and civilians decreased with age.
In succeeding age ranges (20-24, 25-29, and 35-39 years and ages 17-39 combined), the incidence of meningococcal disease among the military was comparable to that in the general population (general population data from CDC, unpub. data). No cases occurred in the military in the 34-39 and >40 age ranges.

Military Branch and Rank
The mean incidence of meningococcal disease for US military branches ranged from 0.12 for the Air Force to 1.35 for the Marine Corps (Table 2). Of the 11 cases occurring in the 17-to 19-year-old age group, 8 were Marine Corps basic trainees, who are usually 18-21 years of age. Incidence for the Marine Corps was signifi cantly higher than that of the other military branches (p<0.05).
Basic trainees as a group had an incidence of 1.18, signifi cantly higher than enlisted and offi cer personnel (p<0.05; Table 2). Service branch and rank, however, are confounded. Marine Corps basic trainees represented 35% (9/26) of the total military cases but only 2% of the military population.

Geotemporal Clustering
While ≈1 million active-duty military personnel (including all recruits) serve in the United States and an additional 400,000 serve around the world, no geographic clustering of cases was observed. In addition, no signifi cant temporal clustering or seasonality of cases was observed.

Serogroup and Vaccination Status
The historical and recent proportions of N. meningitidis serogroups per year are shown in Figure 2. Among the 26 cases that occurred during 2006-2010, N. meningitidis serogroup was determined for 22; 6 (27%) were B, 7 (32%) C, and 9 (41%) Y (Figure 2). In contrast, in the general population during 2006-2009, 24% were B, 45% C, and Fifteen of the meningococcal disease cases (6 serogroup B) that occurred during 2006-2010 were in persons who had been vaccinated. The mean time from vaccination to illness was 77 months (SD 97) for serogroup B infection. When the polysaccharide vaccine was used, the mean time to illness was 12 months (SD 14) for serogroups C and Y combined; when the conjugate vaccine was used, it was 4 months (SD 2). The apparent difference in mean time to illness for the 2 vaccine groups may be misleading because among the group of 8 persons who received the polysaccharide vaccine was 1 whose time to illness was 43 months.
Seven infections were serogroup C and 8 were serogroup Y; all of these cases occurred in persons vaccinated >21 days before onset of disease. Of those 15 cases, 13 occurred <1 year after vaccination, one <2 years after vaccination, and one 3-4 years after vaccination. Of the 13 cases in which the time between patient vaccination and illness was <1 year, 6 patients had been vaccinated with MPSV-4 and 7 with MCV-4.
Of the 8 cases during 2006-2007 for which the N. meningitidis serogroup was determined, 1 was caused by serogroup Y. During 2008-2010, this serogroup was associated with 8 of the 14 cases with known serogroup, 4 of which occurred in persons who had been previously vaccinated with MPSV-4. Nevertheless, this did not represent a signifi cant increase in the prevalence of serogroup Y infection (p = 0.052).

Fatal Cases
The overall case-fatality rate (CFR) among military personnel with meningococcal disease during 2006-2010 was 19% (5/26 cases), which was not signifi cantly different from the CFR of 13% among the age-matched general population (exact p = 0.24; general population data from CDC, unpub. data). Of the 5 fatal cases, 4 were in persons who had been vaccinated with MPSV-4 (4 deaths among 12 known MPSV-4-vaccinated persons) and 1 in a person vaccinated with MCV-4 (1 death among 12 MCV-4vaccinated persons).
The Army, Air Force, Navy, and Marine Corps had 3, 1, 1, and 0 fatal meningcococcal disease cases, respectively. Along with a review of data made available by the AFHSC (A. Eick, pers. comm.) and the US Department of Defense Mortality Surveillance Division (L. Pearse, pers. comm.), a review of Army data from various sources published before the current surveillance program began gave us a reliable count of fatal cases during 1997-2005. This, with the Army data cited above (1,2), enabled us to compare CFRs between historical periods. During 1964-1981, which was the last year of recording until 1997, the CFR in the Army was 6.9%. During 1997-2010, the CFR was 16%. This difference was not signifi cant (p = 0.14).

Discussion
Although there have been vaccine failures (disease occurring 1-48 months after vaccination), universal vaccination continues to have a protective effect against meningococcal disease in the US military. Analyses indicate that during 2006-2010, even as incidence in the general population was declining, N. meningitidis serogroup, clinical syndrome, mortality rates, and overall disease incidence among members of the military were not statistically different from those for the US general population except in the 17-to 19-year-old age group, for which the military rate was higher.
A composite of 100 years of meningococcal disease incidence data for the US Army and general population demonstrates the stark difference in incidences until 1971, when the fi rst polysaccharide N. meningitidis serogroup C vaccine was introduced into the military (Figure 1 the 3 most prevalent N. meningitidis serogroups (the third being serogroup B). From 1990 on, <0.60 cases per 100,000 person-years have been reported. Vaccination against serogroup C may be seen as particularly critical given the history of outbreaks associated with serogroup C in the general population (6). Possible reasons for the drop in cases since 1990 are changes in hygiene policies and an increase in routine use of benzathine G penicillin as prophylaxis against group A streptococcal infections in military recruit camps starting in 1991 (J. Brundage, pers. comm.). Previous studies have demonstrated a relationship between meningococcal infection rates and respiratory disease rates, which have also fallen in the military population over the same period, although the association is not entirely consistent (9)(10)(11)(12).
During 2000-2010, the differences in overall incidence between the Army, the military as a whole, and the general US population were small (incidences of 0.40, 0.49, and 0.44, respectively). Although lower incidence may be expected within the universally vaccinated military population, several factors put military members at higher risk, such as a high percentage of relatively young personnel originating from disparate regions and populations with differing carriage rates, as well as challenging operating environments and crowded living conditions. The risk is evident in the military's elevated rates of other infectious diseases, such as respiratory infections (9-12); however, Artenstein et al. (13) suggested that elevated respiratory disease rates were not likely to be relevant. The current infection rates among the military may represent a limit of vaccine effectiveness in this highly vulnerable population.
The year-to-year proportion of N. meningitidis serogroups among meningococcal infections has varied in the US general population (review of 2002-2009 annual Active Bacterial Core Surveillance reports; 5) and the military population ( Figure 2). As expected, the percentage of serogroup C infections was reduced dramatically after the introduction of the serogroup C vaccine, and consequently, so was the total number of meningococcal disease cases. Data from 1981-2005 are not available, but it appears that the quadrivalent polysaccharide vaccine was successful in reducing infection rates for all serogroups even further, perhaps by as much as 80% (Figures 1, 2).
During 2007-2008, the military gradually switched from the polysaccharide vaccine (MPSV-4) to the conjugate vaccine (MCV-4). For the cases in our study from 2006-2007 for which vaccine information was available, all but 1 person was vaccinated with the MPSV-4. A spike in cases occurred in 2008, the fi rst year that MCV-4 was universally adopted, but only half of those infected had been vaccinated with MCV-4. It is too soon to determine whether the change to MCV-4 has had an effect on the overall incidence of meningococcal disease in the military population.
Of the 15 vaccine failures (disease occurrence 1-48 months after vaccination), 12 occurred <12 months after vaccination and 2 occurred 1-4 years after vaccination. The unique conditions of the fi rst year of the military-training environment may lead to relatively high rates of observed vaccine failures. The number of recruits entering the military each year has been relatively constant; thus, so has the number of vaccinations. No trend was seen for vaccine failures (4, 1, 3, 1, 4 for the years 2006-2010, respectively), and no data suggest a different rate of vaccine failure for the 2 vaccines. The expectation with the newer MCV-4 was that disease incidence would diminish, but so far, that does not appear to be the case.
Although studies have demonstrated the waning of correlates of protection over time after initial meningococcal vaccination (14,15), most military personnel do not receive a booster during their military careers (4). Boosters are effective (16) and are given 5 years after vaccination to personnel traveling to high-risk areas. A study to gather serologic data to address the question of long-term vaccination coverage is underway.
During 2006-2010, N. meningitidis serogroup distribution in the military was similar to that for the age-matched general population (general population data from CDC, unpub. data), although the military has seen more serogroup Y infections during the past 3 years when compared with the historical predominance of serogroup C. Serogroup distribution varies year-to-year in the general population; a large increase in the percentage of serogroup Y infections was seen during 1989-1991 (17). The increase in serogroup Y infections in military personnel is contemporaneous with the changeover to the MCV-4 vaccine, but half of those cases were in persons who had been vaccinated with MPSV-4. Our data cover only the 2-3 years after the adoption of MCV-4 in the military and are insuffi cient to support claims about the effect of vaccine change on serogroup Y incidence; thus, we were unable to identify any vaccine-related trends in cases related to particular serogroups. All of the serogroup Y infections have occurred in persons <25 years of age (Table 2).
N. meningitidis serogroup B is not included in the vaccines, but this absence has not led to its dominance over other circulating strains that are included in the vaccine. In fact, serogroup B accounts for fewer infections than serogroups C and Y, which suggests an unknown factor is preventing an increase in the proportion of serogroup B infections in the military.
As in the general US population, the highest incidence of meningococcal disease in the military was among young adults, but the incidence among 17-to 19-year-old military personnel exceeded that among the same age group in the general population (general-population rate from CDC, unpub. data). However, rates for the military should be compared with those in similar general settings, such as college dormitories, where rates as high as 5.1 have been reported (18).
In the Army, the 1997-2010 CFR of 16% for meningococcal disease appears strikingly higher than the CFR of 7% for 1964-1981, a period spanning the use of 3 different vaccines (1). Given that the Army had only 1, nonfatal case during 2008-2010, it is reasonable to suggest that the high CFR during 1997-2010 is an anomaly.
In this study, while constituting only 14% of the total military population, the Marine Corps accounted for half the total cases of meningococcal disease in the military. In particular, Marine Corps recruits contributed a disproportionate 7 of 11 cases among the 17-to 19-year-old age group. Given that most cases across all military branches were among basic trainees (mainly the 17-to 19-year-old age group), among whom the rate of 1.18 is signifi cantly higher than for enlisted and offi cer personnel, and that most of the basic trainee cases were from the Marine Corps, the exceptional rate of meningococcal disease in the Marine Corps is thus a function of the Marine Corps basic training rate. One possible reason for the high rate of infection among Marine Corps basic trainees is that the basic training period is 4 weeks longer than that of other branches, although only 2 of 8 Marine Corps cases occurred during the last 4 weeks of training. In addition, data from Brundage et al. (1) and AFHSC (Angelia Eick, pers. comm.) show that the Marine Corps rates over previous 5-and 10-year periods were 0.34 and 0.47, respectively. Thus, the current 5-year rate may be a short-term anomaly.
In summary, while meningococcal disease has historically been associated with substantial illness among members of the US military, disease incidence during the past 5 years is at a historic low and is comparable to the rate among the age-adjusted US population. The current rates in the military have been maintained since the 1990s and can be attributed to effective meningococcal vaccines. While some observed trends suggest covariation between length of basic training, age, and service branch, surveillance has not shown consistent trends by which military rates meaningfully differ from those in the general population. Covariation was seen, however, between the higher rate among basic trainees, age group, and Marine Corps versus the rest of the military.
In the military setting, and especially in the basic training setting, epidemic meningococcal disease remains controlled due to a robust immunization program. To date, no evidence of a change in overall disease rates has been associated with the switch to MCV-4 vaccine. However, the increase of serogroup Y infections among the military population warrants further observation.