More on STD Findings

A colleague sent me the New York Times article explaining more of the study, and something caught my eye. The article says the data is from the CDC National Health and Nutrition Examination Survey (NHANES). You can read about that data here. The data uses a sampling method of households that would make all inference from the sample representative of the population sampled. Approximately 7000 people are in the survey, which in its most recent manifestation, has been conducted annually since 1999. Here the journalist writes:

The centers conducts the annual study, which asks a representative sample of the household population a wide range of health questions. The analysis was based on information collected in the 2003-4 survey.

Extrapolating from the findings, Dr. Forhan said 3.2 million teenage women were infected with at least one of the four diseases.

The 838 participants in the study were chosen at random with standard statistical techniques. Of the women asked, 96 percent agreed to submit vaginal swabs for testing.

The findings and specific treatment recommendations were available to the participants calling a password-protected telephone line. Three reminders were sent to participants who did not call.

Of that 7000, 838 apparently were randomly picked to participate in the STD section. Of the 838, around 804-805 agreed. So very high response rate. We can therefore conclude, as best as I can tell, that the sample estimation is statistically accurate, given some margin of error.

I wonder if what is going on here is that some of the infections studied are asymptomatic. For instance, chlamydia produces symptoms that often go unnoticed, and therefore the STD can be undetected for long periods of time. The population the researchers study, though, are sexually active groups - 15-19 year olders, for instance. I want to say that the median age for female sexual debut is 15.5 years, but don't quote me on that. I think it's between 15-16, at least, and is slightly older for White females. So, these adolescents are mostly sexually active. If there is an asymptomatic STD in the population of sexually active, non-married groups of females, then it would seem like it could grow into the high rates of prevalence observed by these researchers. The window of time when a person is infected with a sexually transmitted infection and sexually active is critical, which is why you want to get infected people treated as soon as possible (particularly ones who are high risk, like commercial sex workers. They have so many partners, that their one exposure can can growth rates to exponentially increase).

But, with this demographic, if the STDs have been in the population for years, then growth can happen easily because the STDs are relatively undetected. HPV and chlamydia both low detectability. I bet so do the others. So what this is saying, if I had to venture a guess, is that this is a true externality problem. They are not well-informed about their own health status, and thus neither is their partner. Plus, their adolescents, and adolescents take too many risks anyway, including having unprotected intercourse.

Here, incidentally, is the source of the study. This is not, importantly, based on surveillance data, but as with the chlamydia study, is based on a nutritional study which was randomized. Here's the description of the data:

Genital herpes prevalence data come from the New York City Health and Nutrition Examination Survey, which is modeled after the National Health and Nutrition Examination Survey (NHANES). The survey was conducted in 2004 among a representative sample of New York City adults 20 or older. Of the 1,999 people enrolled in the survey, 1,784 were tested for HSV-2.