Converting art to a single bitmap inside CorelDRAW
300 DPI is OK for images, but if text is flattened along with the image for something like business cards, I'd give 600 DPI a try. The reason: the integrity of the fonts on small text is going to be better. It takes a little longer to RIP at their end, but as you can see below, I think the extra 10-20 seconds is well worth it. Most professional color lasers output at 2400 DPI these days, so IMO there's no reason not to send these devices something better. The large companies gang-run lots of different cards at once, I suspect at 2450 DPI.
The most professional way is to preserve fonts as vector shapes (whether pure fonts or as curves). Thoughtfully flattening bitmap items under any text curves can dramatically simplify the file and prevent surprises. Then text or other curves come out at the highest resolution of the device.
On the other hand, I know that pre-press is a thankless task. If a job outputs well, no one says much. If there is even a tiny problem, the world collapses on our heads. This is why I can see why some shops request one flattened bitmap of the project, for sheer output reliability.
Users sometimes complain about visible rasterization, and why it can exist on vector shapes. They don't know that those problems are usually related to CorelDraw's Rendering Resolution for certain effects such as Tranparencies, Lenses, and Drop Shadows. That's another story, for another movie. These issues are different, and are more or less noticeable based on the dimensions of a project.