Is server-side reactive programming a workaround to achieve high level concurrency?

On the server side reactive programming is used to achieve high level of concurrency for io-heavy applications, because a single thread is capable of serving numerous concurrent requests as long as most of them are waiting for io operation. (to the contrary of typical multithreaded server application where each request is served in it's own thread, above a certain point OS threads will not scale well)

So is reactive programming basically an awkward and inconvenient workaround to achieve high level concurrency in programming languages that don't support light-weight threads, continuations or coroutines?