If it ain't broke, don't fix it. Some stuff to think about:

• Is there a compelling reason for the calling code to know about implementation details?
• Is there a compelling run time issue the change would fix?
• Is there a compelling maintenance issue the change would fix?
• Are you confident you won't introduce bugs making the change?
• Does all the affected code have unit tests?
• Is this the most important thing you can be doing right now?
• Does this change advance your carer or improve your coding ability?

Not all considerations are completely serious, but decide which are important to you, or if there are others I've missed, and think about those criteria to make your decision.