[2,600 men and women, with an average age of 64 years old] were surveyed about their physical activity over their lifetime. As part of the study, researchers took X-ray images to evaluate signs of arthritis in their knee joints.
The study can not prove cause and effect, given it was an observational study that assessed osteoarthritis at one point in time.
Credit where it's due, the editor has written a headline which actually comports with the merits of the study, not overstating the benefits of cycling on elderly knee arthritis. And the author takes care to do the same. The article also discusses the risks specific to elderly cyclists, and identifies the aspects of cycling which are low-impact.
Overall, an informative read.