Assessment of Optimal Selected Prognostic Factors

The identification and assessment of prognostic factors is one of the major tasks in clinical research. The assessment of one single prognostic factor can be done by recently established methods for using optimal cutpoints. Here, we suggest a method to consider an optimal selected prognostic factor from a set of prognostic factors of interest. This can be viewed as a variable selection method and is the underlying decision problem at each node of various tree building algorithms. We propose to use maximally selected statistics where the selection is defined over the set of prognostic factors and over all cutpoints in each prognostic factor. We demonstrate that it is feasible to compute the approximate null distribution. We illustrate the new variable selection test with data of the German Breast Cancer Study Group and of a small study on patients with diffuse large B‐cell lymphoma. Using the null distribution for a p‐value adjusted regression trees algorithm, we adjust for the number of variables analysed at each node as well. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Tracheostomy Tube Failure in Obstructive Sleep Apnea

In our institutions we routinely do posttracheostomy sleep studies on patients being treated for obstructive sleep apnea. We have identified several patients who failed to show objective evidence of improvement after tracheostomy. From our studies we have found that both mechanical obstruction and concomitant respiratory control dysfunction caused this failure. A unique tracheostomy tube was constructed to treat the subset of patients with internal collapse of the tracheostomy tube.     

Natal philopatry and recruitment of willow ptarmigan in north central and northwestern Canada

Summary Natal philopatry and recruitment were measured in two populations of willow ptarmigan; one near Churchill, Manitoba and the other in northwestern British Columbia. We examined the return of tagged offspring in subsequent years with respect to geographical area, annual variation, their age when tagged, their sex, their body weight, age and number of their parents, and time of hatch (first nest or renest). Most chicks were tagged before they fledged, but chicks tagged after that had the highest rate of return. We also observed a strong positive relationship between fledging success of broods and offspring return in following years.Patterns of offspring return were similar in both populations except that male offspring in Manitoba settled closer to their natal sites than those in British Columbia and more yearling captured in Manitoba had been tagged as chicks. Return of offspring did not vary with year, their body weights shortly after hatch, or with the age or number of parents raising them. However, a significantly higher proportion of offspring hatched from first nests (first-initiated clutches) returned compared to those hatched from renests (replacement clutches). The low return of chicks hatched from renests may due to low survival, low philopatry, or both. We observed no differences in the mating status (recruitment) of returning offspring with respect to the time they hatched or the number of parents that raised them.     

Aspirin and acetaminophen use by pregnant women and subsequent child IQ and attention decrements

In a longitudinal prospective study of 1,529 women pregnant in 1974-1975, aspirin and acetaminophen were the two medications most frequently taken during the first half of pregnancy (46 and 41%, respectively). In a selected cohort of 421 offspring of these women, examined at 4 years of age, maternal aspirin use during the first half of pregnancy was significantly related to IQ and attention decrements in the exposed children. Multiple regression analyses were used to statistically adjust for a variety of potentially confounding factors including demographic characteristics, child characteristics, other exposures, and lifestyle/environmental variables. Continuous dose-response and step-function parameterizations of aspirin exposure were both statistically significant and not clearly distinguishable from each other. The estimated aspirin effect is significantly greater for girls than boys. Aspirin effects on offspring function were found in the absence of effects on physical size both at birth and at 4 years. Maternal acetaminophen use was not significantly related to child IQ or attention. As this exploratory research originated from observations of a data set gathered for other purposes, it would be desirable to have these findings replicated in other studies. Further follow-up of the children at a later age is planned.