While little has been done to estimate the actual prevalence of B. pertussis (Whooping Cough) in America, there is data available which could give us some insight.At its peak in 1922, there were over 100,000 identified cases of Whooping Cough. In 1976, that dropped to a little over 1,000. Since then, there has been a resurgence in the number of cases, but almost 33,000 cases reported in 2014. Even still, is still far below the 1920s level, especially taking into account the increase in population size (CDC). However, the current B. pertussis vaccines are not very effective at preventing the spread of infection, only the severity of symptoms (Warfel, Zimmerman, and Merkel 2013). The study also looks at the potential reason why the vaccine may prevent symptoms but not colonization or transmission. Specifically, aP induces a mismatched immune response to natural infection. This inability to prevent transmission may result in a large number of unreported asymptomatic infections.
Unfortunately little work has been conducted in the United States to identify such cases. However, from July-September 2011, a Chinese team studied the prevalence of asymptomatic infections from B. pertussis and B. parapertussus, a closely related pathogen that can also cause whooping cough. The study was limited to school aged children in China, ranging from ages 7 – 15. Two methodologies were used to identify an infection. The first method, culture testing, often results in more false negatives, while the second method, PCR testing, may be more likely to result in false positives, but can be used to identify older infections (Loeffelholz et. al. 1999). The optimal timing for culture tests is within the first two weeks of the onset of cough, while PCR has an optimal timing of up to three weeks, but may work within the first four weeks (CDC). 629 individuals were tested. Culture tests were positive for B. pertussis and B. parapertussis in 0.3% and 0.2% of the test population respectively, while PCR tests were positive in 4.8% and 2.1% respectively (Zhang et. al. 2014).
Additional testing has since been conducted. In 2014 a serological test was conducted with a total of 1707 accepted data points (He et. al. 2016). Roughly 9% of the sample tested positive for B. pertussis. While these are the results in China, vaccination rates for B. pertussis is roughly 95%, and has been for about 10 years. In contrast, the immunization rate in America is roughly 84% (CDC). While this should result in slightly fewer asymptomic carriers, it is reasonable to question just how many infections there currently might be, in the United States, if the rate of asymptomatic infection is anything like it is in China.
But what number should we use, and is it reasonable to assume China’s results will carry over reasonably to the United States? Culture tests are so limited that they result in almost no positives. serological tests offer the widest range of testing, but are more likely to give a false positive. However, if we use a 99.9% confidence interval for the PCR tests, on the low side, we get 2% and 0.22% for B. pertussis and B. parapertussis respectively. Meanwhile the serological test gives us an estimate of 6.72% for B. pertussis. Let’s stick with the PCR test results for a moment, and assume half of this figure for the US population. The halving ensures that we are getting closer to a lower limit guesstimate.
China is roughly comparable in DTP vaccination rates. The percentage of people who have received at least 3 doses of DTP (DTP3) for China, was 99% while the United States DTP3 rate in 2016 was 95% (WHO). While there is a slightly lower vaccination rate, which would potentially reduce the asymptomatic to symptomatic infection ratio, the two are close: off by only 4 percentage points. It is therefore not unreasonable to use the Chinese data, especially when the lower range estimate for the rate of asymptomatic infections is used.
Now, this technically isn’t even per year, but an estimate within the time frame of the study, but let’s assume that it’s for an entire year. For the United States we have 318M people and a 1.11% incidence rate, or roughly 3.5 million infections per year. That’s almost 100 times greater than the number of reported infections in 2014! More disturbing, it is higher than the number of estimated infections during the peak of the Whooping Cough epidemic in 1922! Even if we assume that only 80% of all infections were symptomatic in 1922, and that only 20% of all cases were diagnosed, the estimated number of total infection in 1922, adjusting for population size, would have stood at 2 million.
But why does any of this matter? If most infections are asymptomatic, then we’re better off anyway, right? There are a few problems with this idea. First, not everyone can be vaccinated. The larger the pool of infected individuals, the greater the risk to those who cannot be vaccinated. Second, a shortage of the vaccine would result in a large number of susceptible individuals. This combined with the last point could easily cause a massive outbreak. Third, and perhaps most importantly, the pathogen can evolve at a faster rate, if the pool of infected people is larger. Related to these points, not recognizing the prevalence of asymptomatic infections could lead to a conclusion that the disease has been eradicated or that the cause of infections is due to some other population when it is really due to the asymptomatic carriers.
Additionally, if the rate of contact by those who do not know that they are infectious is sufficiently high—symptoms tend to cause people to avoid contact with others—it could override the limited efficacy of the vaccine at preventing transmission and actually increase the rate of infection. Finally, if the number of infections grows sufficiently large, it could outweigh the impact of the vaccine and result in more overall symptomatic cases. It should be noted that there are issues with the whole cell vaccine as well. It may indeed be a bit better at preventing colonization and transmission, but it needs to be shown. In at least one case study, asymptomatic infections were found among fully vaccinated individuals. (Srugo et. al. 2000)