Sensitivity and uncertainty analyses for burden of disease and risk factor estimates

Epidemiological studies report confidence or uncertainty intervals around their estimates. Estimates of the burden of diseases and risk factors are subject to a broader range of uncertainty because of the combination of multiple data sources and value choices. Sensitivity analysis can be used to examine the effects of social values that have been incorporated into the design of the disability–adjusted life year (DALY). Age weight, where a year of healthy life lived at one age is valued differently from at another age, is the most controversial value built into the DALY. The discount rate, which addresses the difference in value of current versus future health benefits, also has been criticized. The distribution of the global disease burden and rankings of various conditions are largely insensitive to alternate assumptions about the discount rate and age weighting. The major effects of discounting and age weighting are to enhance the importance of neuropsychiatric conditions and sexually transmitted infections. The Global Burden of Disease study also has been criticized for estimating mortality and disease burden for regions using incomplete and uncertain data. Including uncertain results, with uncertainty quantified to the extent possible, is preferable, however, to leaving blank cells in tables intended to provide policy makers with an overall assessment of burden of disease. No estimate is generally interpreted as no problem. Greater investment in getting the descriptive epidemiology of diseases and injuries correct in poor countries will do vastly more to reduce uncertainty in disease burden assessments than a philosophical debate about the appropriateness of social value

Disturbances in microbiota - cause or effect?

IBD are a group of complex polygenetic diseases also involving environmental factors. Evidence for a role for bacteria in IBD include an increased abundance of mucosa-associated bacteria in IBD (which occurs even where there is no intestinal inflammation), and the positive impact of antibiotics on the progress of both Crohn's disease (CD) and ulcerative colitis (UC) of the pouch - pouchitis. Bacteria are necessary for most animal models of IBD. The increased abundance of mucosal bacteria in IBD is not non-specific because while some mucosal bacteria are more abundant this is not the case for all mucosal bacteria including the very abundant Bacteroides vulgatus. On the other hand, antibiotic treatments are not curative, and the humoral immune Ig response to bacterial antigens which is more evident in CD, appears to be polyclonal. While this argues against a role for specific bacteria causing a classical infection, certain mucosal bacteria may damage the mucosal barrier. This would promote invasion by other commensal mucosal bacteria triggering an immune response. Altered adaptive, and to a lesser extent, innate immunity have been extensively studied, and genetic defects in the CARD15 (or NOD2) gene that encodes a bacterial sensing protein modulating innate and adaptive immunity are strongly associated with ileal CD. However, the penetrance of the homozygous CARD15 frameshift mutation, which is the most strongly CD-associated genotype, is very low with only 4% of humans with this developing CD. Furthermore, mice with the same defects in CARD15 do not develop spontaneous ileitis or colitis. Therefore, there have to be other aetiological factor(s). Altered permeability is a consistent finding in subclinical CD. There are other data to suggest that altered mucin is an early event in UC. We propose that the pathogenesis of IBD is multifactorial involving specific mucosal bacteria, defective barrier function and altered mucosal immunity in an aetiology triangle.

Algebraic Bethe ansatz for integrable extended Hubbard models arising from supersymmetric group solutions

Integrable extended Hubbard models arising from symmetric group solutions are examined in the framework of the graded quantum inverse scattering method. The Bethe ansatz equations for all these models are derived by using the algebraic Bethe ansatz method.