They do not spread,but in general if a HDD starts to produce bad sectors it's so old that it starts to produces bad sectors...so it will continue to do so.
But this is by no means a general rule or anything,just an observation that an old HDD is more prone to faults.
Also not all bad sectors are bad sectors,a sector can be flagged because the OS had troubles reading from it without the sector being bad.
You should run the diagnostic tool you will find on the website of your disk to make sure that they are bad.
Depends on whether its a physical or a logical problem. Basically a bad sector arrives when the HDD sector is attempted to be read or wrote to, and for whatever reason it hasn't been able to.
A physical bad sector is physical damage that has occurred to the sector, I.e. the head has damaged the sector, the HDD has become damaged from a drop or hit for example. If this is the case, no it can't be fixed. However these sectors can be remapped to another - which the CHKDSK command can do automatically for example. Or you can go down CountMike's route to partition around them.
If it is logical, for example your OS found the contents of the sector didn't match what it expected (like an ECC) , or the request took too long to process and timed out, the OS marks the sector has bad - even though it might not be physically bad. If so, these can be repaired usually by running commands like CHKDSK / SFC or rewriting over the sector.
All hard disks develop bad sectors in use. As hard disks age bad sectors will appear more often. Magnetic media is not 100% reliable in every spot across the surface of a hard drive platter--this is a simple fact of data storage engineering.
Bad sectors are areas on the hard drive platters that either cannot be read from reliably (data read from the track the first time does not match the data read from the track at a later time), or cannot be read from at all.
You are not usually aware of this because hard disk drives have automatic error management built-into their firmware code. The bad sector cannot be moved from one spot on the hard drive platter to another, but the error sector can be marked to prevent its use, and the address of a "spare sector" on the drive can be substituted in place of the error sector.
This usually is a spot on the hard drive platter that is not adjacent to the error sector, so the heads have to move to a physically separate location to retrieve the data that is supposed to be in the error sector. Absent fragmentation effects, this is why hard disk drives tend to "slow down" as they age.
Once you start becoming aware of bad sectors, in normal use, the spare sector pool has been exhausted, and the sectors that are being automatically marked as bad cannot be re-mapped. From your point of view, the drive is "suddenly" producing bad sectors when, in actuality, it has been finding and hiding bad sectors the entire time. The difference is that the firmware remapping routines can no longer make the process hidden, and you start "losing" storage space to the new bad sectors cropping-up.
At that point of awareness, it's usually a very good idea to replace the drive before you lose data, or it becomes unbootable, and you lose the entire OS installation.
Bad sectors can only be bypassed. Run check disk command with switch R to repair the disk by placing data on working sectors:
Open an administrative command prompt and enter the command "chkdsk /r" without quotes.
Some drives (eg Hitachi's) support the Format Unit ATA command. This is a "low level format" operation that moves any grown defects to the primary defect list (as long as the P-list is not full). The grown defect list is then cleared, ie the reallocated sector count is now 0. Of course this operation is data destructive.