http://www.medscape.com/viewarticle/810692
This paper has sources linked for almost every major paper I know of on the topic. Just look in the sources. Here's an excerpt of this meta-analysis of the evidence, which points out the weaknesses in all of these papers:
"Methodological Issues
Although
all the reviewers were nurses, the investigator team included 2 experts in the evaluation of heath care quality and effectiveness and a physician with extensive experience conducting systematic reviews. Articles included in the review were published in peer-reviewed medical (n = 12),[33,37,39,45,48,51,52,56,57,59,61,64] interprofessional (n = 10)[34,36,38,43,46,47,49,53,55,60] and nursing (n = 15) journals.[35,40–42,44,50,54,58,62,63,65–69] A draft of the report was reviewed by 2 independent panels of technical experts: 1 panel comprised a consumer, a statistician, and a physician leader;
the other included highly respected NPs. Written comments and recommendations from these reviewers were addressed by the authors.
Limitations in the body of research reviewed should be considered when interpreting the results of this systematic review. Heterogeneity of study designs and measures, multiple time points for measuring outcomes, limited number of randomized designs, and inadequate statistical data for meta-analysis were among the methodological limitations encountered. Diffusion of treatment because of inclusion of MDs in both experimental and usual care groups was also a potential problem in some studies.[33,34,58,60,66]
In addition, the failure to fully describe the nature of the NP roles and responsibilities and the relationships of team members, including frequency and qualities of collaboration with MDs, limits the ability to replicate the models of care employed."
Basically, NPs have been shown to work as well as a physician
when they are working with a physician. That's pretty useless evidence of their abilities to me, as the physician is likely often picking up the slack in the relationship.
