If you have two particles with an interparticle potential of $$V(\vec{r}_1,\vec{r_2})=\frac{1}{2}k(\vec{r}_1-\vec{r_2})\cdot(\vec{r}_1-\vec{r_2})$$ then you can find solutions by separation of variables for the center of mass (as a free particle) as for the relative separation. The relative separation solution will be a SHO solution. If one of the particles is much more massive, then you can interpret the position of the relative position as the location of the lighter particle.
In practice a uniform potential just shifts the energy which doesn't affect much and if a potential has a minimum, then that minimum might locally look similar to a upward opening quadratic, so the lowest energy solutions to that potential might look somewhat similar to a SHO.
But really physicists study it because: 1) they can, it's a solvable problem an d2) it's a model problem in that techniques to solve it can be used in other situations.