I did my MSc in a CAMPEP accredited medical physics program and am starting up an MD-PhD in a few weeks, so I'll put in my two cents.
CAMPEP programs will introduce you to many aspects of clinical physics work and many are heavily integrated with the clinical setting, so you can get a good feel for how a physicist is involved in treatment planning and patient care. You'll hardly ever meet a patient directly as a physicist in North America, save for some rare complex cases, although you would be involved in patient management, dosimetry, planning, QA, commissioning equipment, and possibly radiation safety and some teaching in academic centers.
For me, I thought the career was limited in it's scope, many students do research tailored to a clinical physics environment. It's also a difficult field in which to make an academic career in as well, as most med phys grads end up doing 90% clinical work.
If you want to practice medicine, medical physics is not a substitute, it's a very different profession. You can make an excellent living and work reasonable hours as a physicist, but if you're starting at this stage of the game, there will soon be a clause that you need to have completed an accredited residency program in addition to either an MSc or PhD in medical physics to write the ABR certification exams. (I believe the cutoff is 2012, but could be wrong.) Which means you can tack on 2 years of medical physics residency after the PhD.
If you want to be a heavily research oriented clinical physicist, you will need the PhD (~4-6 years). Expect to also do a postdoc of anywhere from 3-5 years either before or after the residency (in the past it's typically been one or the other, I doubt that will be the case in 5 years, although this depends on the demands of the workforce). If you want to work mainly as a clinical physicist, you can still find jobs with an MSc alone (2yrs) in many parts of Canada and the US (excluding some major cities) or an MSc and residency (4yrs), (although this window may be closing) and the pay difference is quite small. (Do some research on areas your interested in working)
To do real independent research as a medical physicist takes almost as much time as the MD-PhD route. (This is the general outlook in the field at the moment, it used to be a much shorter route) Getting research grants as a medical physicist with solely clinical appointments can also be extremely challenging. (Far more so than for an MD or MD/PhD)
MD-PhD will definitely give more options, it took half my MSc for me to figure out I didn't want to be a clinical medical physicist, although I still want to incorporate elements of it into my future work. I suggest you try and meet some medical physicists and see if it's right for you. It's also a very challenging undertaking in and of itself.
So, unless you are truly dedicated to becoming a medical physicist. I would stick with the MD/PhD, where you will have much more freedom when it comes to doing research. Also much more flexibility in determining the nature of your clinical interactions and the scope of your research.