I must say I was quite fascinated by the phenomenon when I first heard about it.
It seems to plague the bleeding edge research areas of science especially where getting the results you need as fast as you need them seems to be a priority. Lets think psychology, medicine (pharmaceuticals), genetics etc. It seems more prevalent in the high stakes, "need to get those papers published for my grant" areas.
When science is about money or getting the next wonder drug released it seems a few corners are being cut off the scientific method.
The "decline effect" works this way:
Initial studies come out with splendid results but the more those results are attempted to be duplicated the less spectacular the results become with time. Hence initial measured effect declines.
This however should not be confused with large areas of well established scientific disciplines and theory's.
These are alive and well and by contrast the more experiments that are conducted to further research in these areas the more previous results are confirmed again and again.
This confirms that new science is only as good as the rigor to which it has been subjected.
Cutting edge stuff really needs to mature before we should go betting the farm on it. Insufficiently large sample sizes, publication bias and just the sheer desire to publish that which is fresh, new, exciting and sensational to get that scoop should make us exercise a little reservation in some frontiers.
This is not a licence to become a general science septic, as the author state in his follow up article:
One of the sad ironies of scientific denialism is that we tend to be skeptical of precisely the wrong kind of scientific claims. Natural selection and climate change have been verified in thousands of different ways by thousands of different scientists working in many different fields. (This doesn’t mean, of course, that such theories won’t change or get modified—the strength of science is that nothing is settled.) Instead of wasting public debate on solid theories, I wish we’d spend more time considering the value of second-generation antipsychotics or the verity of the latest gene-association study.
You can read more here
That initial article got a little flak, so a follow-up was published here