Thoracic surgery patients who have blocks performed with liposomal bupivacaine require fewer analgesics postoperatively. This may decrease complications related to poor pain control and decrease side effects related to narcotic use in our patient population.

The purpose of this cross-sectional study was to assess the 2009 prevalence of chlamydial (CT) and gonococcal (GC) infection in two populations in a mid-western city in the United States: patients at a public sexually transmitted disease (STD) clinic, and individuals seeking HIV counseling and testing services at an AIDS community organization. We characterized STD prevalence in a random sample of 592 STD clinic patients and a convenience sample of 471 individuals agreeing to STD testing through outreach efforts at the community organization. The STD clinic population was 59% male, 60% black, with 3.1 mean sex partners in the last year. The community organization population was 72% male, 19% black, with a mean of 4.3 partners in the last year. The prevalence of CT and GC combined was consistently higher in STD clinic patients than at the community organization (18% vs. 4%). CT prevalence was higher than GC prevalence in both populations (CT: 3% and 13% at the STD clinic and community organization, respectively, vs. GC: 1% and 7%, respectively). Factors significantly associated with increased odds of GC/CT at the STD clinic include unmarried status, younger age, at least six partners in the last year, and unprotected sex in the last year. At the community organization, the only factor significantly associated with increased odds of GC/CT was lower educational attainment. Our findings confirm that STD prevalence differs widely by population group. Given these differences, local approaches to STD control should also be carefully targeted to specific subgroups.

Our neural architecture search algorithm progressively searches a tree of neural network architectures. Child nodes are created by inserting new layers determined by a transition graph into a parent network up to a maximum depth and pruned when performance is worse than its parent. This increases efficiency but makes the algorithm greedy. Simpler networks are successfully found before more complex ones that can achieve benchmark performance similar to other top-performing networks.