A, Unadjusted trends from 2001 to 2013 in visits with NPs and PAs as a percentage of all outpatient visits to medical and surgical specialist physicians.B, The same trends as a percentage of all “new” and “return” patient visits, respectively. C, The same trends by reason for visit. The P value for linear growth by bivariate survey-weighted logistic regression is .001 or less for all subcategories except the “preventive” visit reason, where P = .01. Error bars indicate 95% CIs incorporating survey weights and clustering from the multistage sampling design of National Ambulatory Medical Care Survey (NAMCS).