The researchers conclude that any large-scale government measures to target the mosquitoes that carry Zika are unlikely now to contain the current epidemic in South America.
Using all existing data on Zika transmission in South America, together with data on similar viruses such as dengue, the team compiled a mathematical model to represent the current epidemic and predict future waves of transmission.
Reporting in the journal Science, the researchers - from Imperial College London in the United Kingdom - also predict Zika is unlikely to resurge on an epidemic scale again in Latin America for 10 years or more, although there could be smaller-scale outbreaks in that period.
Lead author Neil Ferguson, a professor in Imperial's School of Public Health, explains that:
"This study uses all available data to provide an understanding of how the disease will unfold - and allows us to gauge the threat in the imminent future."
The model predicts the current epidemic will be over in 2-3 years for the main reason that people who have already been infected with Zika are unlikely to be infected with it again.
Prof. Ferguson says the reason is because of something called "herd immunity." When a person is infected with a virus like Zika, their immune system makes antibodies against the virus, which protects them against infection next time they are exposed to the virus.
Eventually, there are too few people left to infect for transmission to be sustained, he explains.
The virus cannot resurge until there is a new generation of people who have never been infected, and the model predicts this will not happen for at least 10 years, Prof. Ferguson notes, adding:
"This mirrors other epidemics, such as chikungunia - a similar virus to Zika - where we have seen explosive epidemics followed by long periods with few new cases."
Control measures imposed too late to contain Zika epidemic
The researchers conclude that any large-scale government measures to target the Aedes aegypti mosquitoes that carry Zika are unlikely to contain the epidemic.
- Many people infected with Zika will have no symptoms or mild ones that last up to a week
- Zika can be passed from a pregnant woman to her fetus
- Infection during pregnancy can cause a serious birth defect called microcephaly and other severe brain defects.
Zika is very similar to dengue, which is also spread through the bite of Aedes aegypti. Experience of trying to control the spread of dengue shows it is very difficult.
Prof. Ferguson says to stand a chance of containing such an epidemic, you have to start implementing control measures very early - but in the case of Zika, "by the time we realised the scale of the problem, it was too late," he notes.
In fact, there is a good chance that trying to slow Zika spread at this late stage could prolong rather than curtail the epidemic, as Prof. Ferguson explains:
"Slowing transmission between people means the population will take longer to reach the level of herd immunity needed for transmission to stop. It might also mean that the window between epidemics - which we predict may be over a decade - could actually get shorter."
Problems for vaccine development
While having the end of the epidemic in sight is good news, it raises problems for vaccine development. The model predicts Zika cases will have dropped substantially by the end of 2017. This means by the time vaccines are ready to test, there will not be enough uninfected people left for trials to be viable.
Prof. Ferguson suggests one way to overcome this is to be ready with "sleeper sites" around the world. Having already sought and been granted the legal and ethical approvals involved in running vaccine trials - a lengthy and laborious process - sleeper sites would be ready to launch a vaccine trial immediately when a new Zika epidemic breaks out in their area.
A question that still puzzles the team is why the Zika virus affected Latin America in such an explosive way.
One suggestion is the influence of climate - the outbreak coincided with an El Niño event. Another suggestion is that genetic mutation of the virus played a role, although early data suggests not, says Prof. Ferguson.
Prof. Ferguson says while their model makes these predictions, there are many caveats and unanswered questions - and he calls for more research. This should assess past exposure to Zika by testing for antibodies in at-risk populations. He says his and other groups have already started working on this.
"One research priority is to fully understand the extent of Zika transmission, and what proportion of people in Latin America - and across the globe have been infected."
"There are currently more questions surrounding Zika than answers - and only through a coordinated global research effort will we find the answers we desperately need."
Prof. Neil Ferguson