We did spend a fair bit looking at the pros and cons of implementing PBL in medical education in the first term (50% of which I have forgotten tbh). The problem with PBL is that people have very strong views about its "effectiveness" and this is the case even in the literature. There are die hard researches like Schmidt who are in favour of PBL. He wrote an awesome paper in 1993 on the educational theory behind it (see attached). There are also people like Colliver, who think PBL is a pile of crap and in the early 2000s (I think), published a paper and just tore Schmidt's paper apart (I can't seem to find a link for this paper). Sweller, in the 90's, also wrote a paper on cognitive load theory and how that basically screws up the theory behind PBL, but that goes into a fair bit of detail. You are right though in that most of the research re PBL tends to be for it, but there are still a few others who think that its either to early to call the "better" or more "effective" one, or that traditional courses are better (though many studies on the latter case tend to be based in Australia or the US).
If I'm being totally honest and neutral, the theory behind most of medical education is relatively weak in general as its only just fully branched out from Education. Hardcore educationalists don't like Medical Educationalists because they haven't been around for as long as the educationalists have and they say that medical educationalist theory is much thinner than theirs. Fair point to be honest, as most of the theory in Medical Education is literally just taken from Education and applied to medicine (research into Medical Education only really kicked off in the 70's).
This is why when someone says X is better than Y because it produces doctors with "better clinical reasoning", you really need to think about what it is they are using to measure this - most of the time, its the results of some random test students sit, or the number of clinical errors they recorded in a diary or something random like that. It's very hard to quantify anything related to education because there a ridiculous amount of social, psychological, biological etc. factors that affect one's education so its very hard to see the effect of one thing on another in education and quantify it (hence the many qualitative papers in the literature). There are some good quality quantitative papers of education (and medical education) out there, but there are always major assumptions I think that they have made.