Actually, I wanted to find out what causes a bulb to blow. Wait, I know that if there is air leakage it will cause the bulb to burn right after oxygen enters. I believe that the other reason behind bulbs blowing is the same as the reason causing the bulb to glow brighter (if I am wrong, please correct me). I used to think that current is the reason why bulbs blow, but I saw on the National Geographic TV show Genius that Einstein removed some straps of coil (on a transformer) causing the voltage to decrease and the current to increase and thus he fixes the problem of bulbs blowing. So, in this sense, voltage is responsible for bulbs blowing, which is against my understanding, since I believe it should be the current.
These qualities are interdependent due to Ohm's law.
Why does a filament bulb blow?
When current flows through the filament, it heats up and this heat is also dissipated. There is a maximum temperature the filament is about to endure. If the rate of dissipation of heat catches up the heat produced well before the bulb attains that temperature, it glows, otherwise the filament splits.
The heat produced per unit time is $P =IV = I^2R = V^2/R$, which is the instantaneous power produced by the bulb.
There is a maximum power specified by the manufacturer, and whenever the above-specified quantity $P$ exceeds that (either due to increase in voltage or current, actually increase in voltage will proportionally increase the current due to Ohm's law, assuming that the Resistance remains almost constant), the bulb will produce more heat than it can sustain, and the filament will melt and split.
There are two ways to interpret the question what causes a bulb to blow.
The bulb is new and in good condition but it is connected to an inappropriate power supply.
The bulb is connected to an appropriate power source but it does not last for ever.
For 1, Archisman has given you the answer but I will expand a little. Suppose that the bulb is designed for the US $110V$ supply and should consume $110W$. The rules that Archisman gives implies that the current needs to be $1A$ and hence the resistance needs to be $110\Omega$. The designer needs to choose the length and diameter to give the correct resistance and that it will stabilise at a suitable temperature when it dissipates that power. Now suppose that the bulb is taken to another country and connected to a $220V$ supply. It will start to draw $2A$ instead of $1A$ and hence consume $440W$ rather than $110W$. Note that doubling the voltage has quadrupled the power and not just doubled it (since the current also doubled). It is unlikely that a filament designed for $110W$ could handle $440W$. It will probably very quickly vaporise.
For 2, even if the bulb is used within its design parameters, the filament will gradually vaporise and thin. Eventually, it will break.