How do scientists know that Ebola is only contagious once someone shows symptoms?

The Wall Street Journal asks How do scientists know that Ebola is only contagious once someone shows symptoms? 

Their answer:

"Scientists have a great deal of experience with Ebola from previous outbreaks in Africa. The virus first emerged in 1976, and there had been more than 2,000 cases between then and 2012. This latest outbreak has resulted in more than 4,000 deaths. Health officials are certain that a person infected with Ebola isn’t contagious until they show symptoms, such as fever, severe headache, vomiting, diarrhea, stomach pain and muscle pain."

The gist of the answer is that thousands of people have been infected with Ebola virus (EV) since 1976, and for each patient, medical officials try to trace back the sequence of events that led to the infection. For many (most) cases, they can pinpoint the infection incident to direct contact with a symptomatic individual. There is little evidence (although it is probably harder to trace) of an infection being caused by direct contact with an asymptomatic individual who only later became symptomatic.

A second argument (not in the article) arises from a consideration of the numbers.

The worse your symptoms, the more virus you have in your system and your bodily fluids. If you do not have any symptoms, it is likely that your virus count is low. Indeed an infected asymptomatic person may not even test positive for Ebola virus.

As I described in my "Direct Contact" post, there is a greater likelihood of becoming infected if (1) the bodily fluid source contains a higher level of virus, and (2) the contact point into your body is larger.

Finally, the fast onset of severe symptoms indicates that Ebola virus can replicate very quickly once it overwhelms the immune system, and so when a patient goes from asymptomatic to symptomatic there is likely to be a dramatic increase in viral load.

Thus, a symptomatic person is much more contagious than an asymptomatic person because of the much higher virus count. That does not mean it is impossible to become infected from someone without symptoms. For example, if you transfuse yourself with the blood of an asymptomatic EV-infected individual, then it is likely you will get Ebola virus disease. On the other hand, the blood of someone who has successfully recovered from an EV infection (e.g. Dr. Kent Brantly) contains no detectable live virus, and instead contains antibodies against EV which are likely to help an infected patient.

Figure 1. This will not happen with Ebola virus disease.