Why is it that so many scientific stories seem intent on debunking and end by demonstrating that they are just rehashing opinions and have no actual data?

A recent article purported to debunk the story of the Candiru, which supposedly would be attracted to urine in the water and swim up the urethra to anchor itself into a penis to suck the host's blood.
The candiru, apparently cannot get back out of a human host, and cannot be removed because of the spines – so this little fish does horrendous damage, leading to laceration, hemorrhage, bladder destruction, penis amputation, even death!
So, here we have a story that would easily make most readers decidedly uncomfortable and the aura of a scientific study or finding that debunks this story.

Yet, what do we actually have.
This is a vastly different environment from a fish’s gill and, as Dr. Bauer states in her paper, “with no oxygen available and no room to ‘swim’ up the urethra it is unlikely that the fish survives even a few minutes.”
Yet, this doesn't debunk anything.  There is no requirement that the fish survive for the initial stories to be true.  

What is claimed to be known is:
Vandellia spp. normally ingest the blood of fish hosts. The catfish attack by entering the gill opening and anchoring near a major blood vessel. Jansen Zuanon and Ivan Sazima studied the feeding habits of several species of vampire catfish including V. cirrhosa, and report their findings in a 2004 paper in the Journal of Ichthyology and Aquatic Biology (“Vampire Catfishes Seek the Aorta, not the Jugular: Candirus of the Genus Vandellia [Trichomycteridae] Feed on Major Gill Arteries of Host Fishes”). The researchers observed that the fish do appear to use spines on their gill openings to anchor themselves in the gills of the host fish, but rather than actually sucking blood, they lacerate the host’s artery and feed on blood that pumps out through the wound. Feeding takes less than two and half minutes, and then the candiru leaves the host.
Yet, this doesn't debunk anything, because it clearly supports the fact that the fish enters narrow openings, it uses its spines to anchor itself and it feels on blood.  While it may normally leave a host in less than two and a half minutes, this would clearly be problematic if it died as indicated previously, so this still debunks nothing.

We are then treated to the following statements
Researchers have not proven how the candiru locates the gill opening of its fish host, but it is true that urea is excreted by fish through their gills, and by humans in the urine. It is also possible that the movement of liquid out through the orifice attracts the fish.
Once again, this debunks nothing, but instead adds supporting evidence as to why the fish would be attracted to a human peeing in the water.

In the end, the article concludes:
From the evidence, we know that the candiru does sometimes swim up the wrong stream, so to speak, but it appears that attacks on humans, if they occur at all, are vanishingly rare – maybe once every hundred years or so.
So, we've gone from debunking this story to providing evidence that it is plausible, to finally claiming that it is merely rare.

What we find in this article, is simply another case of scientists making a claim, having it sensationalized in an article and finally demonstrating that they are still clueless.

Yeah ... that's great science.