Smoothed quantile regression analysis of competing risks

Censored quantile regression models, which offer great flexibility in assessing covariate effects on event times, have attracted considerable research interest. In this study, we consider flexible estimation and inference procedures for competing risks quantile regression, which not only provides meaningful interpretations by using cumulative incidence quantiles but also extends the conventional accelerated failure time model by relaxing some of the stringent model assumptions, such as global linearity and unconditional independence. Current method for censored quantile regressions often involves the minimization of the L₁‐type convex function or solving the nonsmoothed estimating equations. This approach could lead to multiple roots in practical settings, particularly with multiple covariates. Moreover, variance estimation involves an unknown error distribution and most methods rely on computationally intensive resampling techniques such as bootstrapping. We consider the induced smoothing procedure for censored quantile regressions to the competing risks setting. The proposed procedure permits the fast and accurate computation of quantile regression parameter estimates and standard variances by using conventional numerical methods such as the Newton–Raphson algorithm. Numerical studies show that the proposed estimators perform well and the resulting inference is reliable in practical settings. The method is finally applied to data from a soft tissue sarcoma study.     

A Q-Q plot aids interpretation of the false discovery rate

False discovery rates are routinely controlled by application of the Benjamini–Hochberg step-up procedure to a set of p-values. A method is demonstrated for representing the values so obtained (the BH-FDRs) on a quantile–quantile (Q-Q) plot of the p-values transformed to the negative-logarithmic scale. Recognition of this connection between the BH-FDR and the Q-Q plot facilitates both understanding of the meaning of the BH-FDR and interpretation of the BH-FDR in a particular data set.     

Regression on Quantile Residual Life

Summary A time‐specific log‐linear regression method on quantile residual lifetime is proposed. Under the proposed regression model, any quantile of a time‐to‐event distribution among survivors beyond a certain time point is associated with selected covariates under right censoring. Consistency and asymptotic normality of the regression estimator are established. An asymptotic test statistic is proposed to evaluate the covariate effects on the quantile residual lifetimes at a specific time point. Evaluation of the test statistic does not require estimation of the variance–covariance matrix of the regression estimators, which involves the probability density function of the survival distribution with censoring. Simulation studies are performed to assess finite sample properties of the regression parameter estimator and test statistic. The new regression method is applied to a breast cancer data set with long‐term follow‐up to estimate the patients' median residual lifetimes, adjusting for important prognostic factors.     

Generalizing Double and Triple Sampling for Repeated Surveys and Partial Verification

Population composition is often estimated by double sampling in which the value of a covariate is noted on each of a large number of randomly selected units and the value of the covariate and the exact class to which the unit belongs is noted for a smaller sample. The cross‐classified sample can be used to estimate the classification rates and these, in turn, can be used in conjunction with the estimated distribution of the covariate to obtain an improved estimate of the population composition over that obtained by direct observation of the identity of the individuals in a small sample. There are two approaches to this problem characterized by the way in which the classification rates are defined. The simplest approach uses estimates of the probability P(i | j) that the unit is actually in class i given that the covariate is in class j. The more complicated approach uses estimates of the probability Pi | j) that the covariate falls in class j given that the unit is actually in class i. The latter approach involves estimating more parameters than the former but avoids the necessity for the two samples to be drawn from the same population. We show the two approaches can be combined when there are multiple surveys. For example, one might conduct a disease survey for several years; in each year the accurate and/or error‐prone techniques may be applied to samples. The sensitivities and specificities of the error‐prone test are assumed constant across surveys. Generalizations allow for more than one error‐prone classifier and partial verification (estimation of misclassification rates by application of the accurate technique to fixed subsamples from each error‐prone category). The general approach is illustrated by considering a repeated survey for malaria.     

A method for estimating natural survival rate and mean fecundity of an adult insect population by dissecting the female reproductive organs

Several problems were discussed in relation toMacDonald' s method (MacDonald , 1957) for estimating the survival rate of a natural population of adults with varying survival rates and unstable age structure.

1. Random samplings with a fixed sampling ratio and an appropriate census interval is pre-requisite during the occurrence of the adults.
2. At each sampling, female adults are dissected to know the ratio, p_i, of nulliparous females in ith sample (i=0, 1, 2, 3, …,). The Σn_ip_i/Σn_i gives an estimate of the ratio, Fα/F, of nulliparous females in the population where n_i refers to the population size on ith census date. If a constant daily survival rate is assumed, the daily survival rate is estimated from equation (4′). When the survival rate is not constant over the period of adult occurrence, e. g. before and after the initiation of oviposition, the survival rate during pre-ovipositional period is estimated by equation (4″).
3. Decision of an economic census interval to obtain a reliable estimate of the ratio, Fα/F, is depending on the form of emergence curve, particularly on its duration and the length of pre-ovipositional period. If the normal distribution can be assumed for the emergence curve, an interval less than one third of the emergence period is recommended. Concerning with insects having a long pre-ovipositional period, a census interval which exceeds one third of the emergence period still gives a good estimate of Fα/F.
4. The mean realized fecundity of some kind of insects can be estimated by equation (5′ or 5″) using the estimates obtained by the present method.

     

The effect of environmental factors on floating fresh marsh end-of-season biomass

We used both stepwise and quantile regression to determine the sources of environmental variation that explained the observed inter-annual variation in end-of-season freshwater floating marsh aboveground biomass over an 18-year period. The vegetation at our study site had high species diversity with an average of 20 species recorded from 10 0.25 m⁻² plots. However, Panicum hemitomon was clearly the dominant contributing 74% of the total biomass. Only three other species (Solidago sempervirens, Vigna luteola, and Thelypteris palustris) were so common that they were sampled in all years. We expected that the most important factors controlling interannual variation in aboveground biomass are temperature and nitrogen availability. We also expected that nitrogen availability to the plants is affected by water movement through and under the mat driven by precipitation (lower N), evaporation (transportation of higher N waters to roots), and local runoff (higher N). Stepwise regression analysis indicated that P. hemitomon average biomass was negatively related to average water level and positively related to maximum water level and had a curvilinear response with TKN. Using quantile regression the best fit for P. hemitomon maximum-biomass with two parameters was obtained using hot days (positive relationship) and maximum water level (negative relationship). Both analytical methods showed maximum water level (negative relationship) and cold front passage (positive relationship) to be the environmental parameters that best explained interannual variation in S. sempervirens biomass. V. luteola biomass was positively related to temperature. Stepwise regression added chloride concentration as an additional positive parameter explaining V. luteola biomass, while quantile regression identified nitrogen as an important positive parameter. Both analytical methods identified pH, TKN, and water level as environmental parameters that were negatively correlated with T. palustris biomass. The overall negative effect of water level on all species was unexpected in this floating mat system. We initially assumed that higher water levels were due to higher runoff which should have a positive effect on biomass. However, higher water levels may also be related to a higher retention time in this fresh-water tidal system, which decreases water exchange and nutrient replenishment.     

Plasma Levels of Adiponectin,a Novel Adipocyte‐Derived Hormone,in Sleep Apnea**

Objective: Obstructive sleep apnea (OSA) is associated with obesity, sympathetic activation, systemic inflammation, and cardiovascular morbidity. Obesity, β‐adrenergic agonists, and inflammation are linked to decreased expression and/or secretion of an adipose tissue‐derived antiatherogenic hormone, adiponectin. The purpose of the study was to investigate whether OSA affected plasma levels of adiponectin, which might help explain OSA‐associated cardiovascular morbidity. Research Methods and Procedures: We randomly selected 68 otherwise healthy male subjects, either with moderate/severe OSA [apnea‐hypopnea index (AHI) ≥ 20; n = 35] or without OSA (AHI ≤ 5; n = 33). The diagnosis of OSA was made based on prospective full polysomnography. Adiponectin was measured before polysomnography between 8 and 10 PM . Results: AHI was higher in the OSA group (49.5 ± 4.4 vs. 2.9 ± 0.4 events/h; p < 0.001). OSA subjects were also more obese, with greater BMI (33 ± 1 vs. 30 ± 1; p = 0.016) and percentage body fat (29 ± 1% vs. 26 ± 1%; p = 0.030). Adiponectin levels were 7.67 ± 0.73 and 6.33 ± 0.51 μg/mL in the OSA and non‐OSA groups, respectively, and this difference was significant in covariate analysis (taking into account age, hemodynamic characteristics, measures of body fat, and OSA severity) (p = 0.009). After excluding from both groups the subjects with extreme BMI, such that the OSA and non‐OSA study cohorts had similar BMI and percentage body fat, subjects with OSA had significantly higher plasma adiponectin (8.49 ± 0.92 vs. 6.32 ± 0.55 μg/mL; p = 0.042), differences also evident in covariate analysis (p = 0.017). Discussion: Plasma adiponectin levels are elevated in otherwise healthy subjects with OSA. Therefore, low adiponectin is unlikely to explain the association between OSA and cardiovascular disease.     

Robustness of Selection Procedures

In the article Bechhofers Indifference-zone formulation for selecting the t populations with the t highest means is considered in a set of non-normal distributions. Selection rules based on the sample mean, the 10% and the 20% trimmed means, two estimators proposed by Tiku (1981) for valuating the smallest and highest accepted sample values higher, the sample median and a linear combination of quantile estimators, two adaptive procedures and a ranksum procedure are investigated in a large scale simulation experiment in respect of their robustness against deviations from an assumed distribution. Robustness is understood as a small percentage of the difference β_A-β between the actual probability of incorrect selection β_A and the nominal β-value. We obtained a relatively good robustness for the classical sample mean selection rule and useful derivations for the employment of other selection rules in an area of practical importance.     

Prioritizing covariates in the planning of future studies in the meta‐analytic framework

Science can be seen as a sequential process where each new study augments evidence to the existing knowledge. To have the best prospects to make an impact in this process, a new study should be designed optimally taking into account the previous studies and other prior information. We propose a formal approach for the covariate prioritization, that is the decision about the covariates to be measured in a new study. The decision criteria can be based on conditional power, change of the p‐value, change in lower confidence limit, Kullback–Leibler divergence, Bayes factors, Bayesian false discovery rate or difference between prior and posterior expectation. The criteria can be also used for decisions on the sample size. As an illustration, we consider covariate prioritization based on genome‐wide association studies for C‐reactive protein levels and make suggestions on the genes to be studied further.     

Formation of cis-p-Menthan-1-ol from p-Menthane by Microbial Hydroxylation

In the course of investigation of p-menthane-utilizing microorganism, it was found that a strain assumed as the genus Pseudomonas produced p-menthan-1-ol from p-menthane into the medium.

The microbial formation of p-menthan-1-ol and its chemical properties are described. The configuration of p-menthan-1-ol was identified to be cis-form by chemical synthesis.     

Weight Status in Childhood as a Predictor of Becoming Overweight or Hypertensive in Early Adulthood**

Objective: To assess the extent to which weight status in childhood or adolescence predicts becoming overweight or hypertensive by young adulthood. Research Methods and Procedures: We conducted a prospective study of 314 children, who were 8 to 15 years old at baseline, and were followed up 8 to 12 years later. Weight, height, and blood pressure were measured by trained research staff. Incident overweight was defined as BMI ≥ 25 kg/m² among participants who had not been overweight as children. Results: More male subjects (48.3%) than female subjects (23.5%) became overweight or obese between their first childhood visit and the young adult follow‐up (p < 0.001). Being in the upper one half of the normal weight range (i.e., BMI between the 50th and 84th percentiles for age and gender in childhood) was a good predictor of becoming overweight as a young adult. Compared with children with a BMI <50th percentile, girls and boys between the 50th and 74th percentiles of BMI were ～5 times more likely [boys, odds ratio (OR) = 5.3, p = 0.002; girls, OR = 4.8, p = 0.07] and those with a BMI between the 75th and 84th percentiles were up to 20 times more likely (boys, OR = 4.3, p = 0.02; girls, OR = 20.2, p = 0.001) to become overweight. The incidence of high blood pressure was greater among the male subjects (12.3% vs. 1.9%). Compared with boys who had childhood BMI below the 75th percentile, boys between the 75th and 85th percentiles of BMI as children were four times more likely (OR = 3.6) and those at above the 85th percentile were five times more likely (OR = 5.1) to become hypertensive. Discussion: High normal weight status in childhood predicted becoming overweight or obese as an adult. Also, among the boys, elevated BMI in childhood predicted risk of hypertension in young adulthood.     

Stress reactivity and adiposity of youth

Objective: The relationship between stress reactivity and total or central adiposity in children has not been widely studied. Data from two studies were combined to determine the relationship between reactivity to interpersonal stress and the adiposity of children. Research Methods and Procedures: Stress reactivity to an interpersonal stressor (speech) was measured in 36 boys (9.8 ± 1.3 years of age) and 27 girls (9.3 ± 1.3 years of age). Total adiposity (percentage body fat) was estimated from skinfolds and central adiposity from the abdominal girth. Multiple regression was used to establish the associations of change in perceived stress and heart rate reactivity with adiposity. Age, sex, ethnicity, and baseline perceived stress and heart rate served as covariates for total adiposity. Fat mass was included as an additional covariate for the prediction of log abdominal girth (central adiposity). Results: Based on adjusted β‐weights, change in perceived stress (β = 1.13, p ≤ 0.001) and heart rate reactivity (β = 0.14, p ≤ 0.03) independently predicted percentage body fat. Heart rate reactivity (β = 0.002, p ≤ 0.04) independently predicted log abdominal girth. Discussion: Reactivity to psychological stress may initiate the antecedents of cardiovascular disease before adolescence by increasing total and central adiposity. Future studies should determine whether stress reactivity increases the adiposity of youth by increasing their consumption of energy‐dense snack foods and decreasing their willingness to be physically active.     

p-Coumaric acid — a monomer in the sporopollenin skeleton

Sporopollenin obtained from wings of Pinus mugo (Turra) pollen was analysed by pyrolysis mass spectrometry. In the spectrum, mass peaks which are characteristic for p-coumaric acid were dominant. p-Coumaric acid was the main degradation compound when the wing material was treated by a gentle method using AII₃, and also when the remaining residue of the treated sporopollenin material was saponified. It is therefore assumed that p-coumaric acid is a genuine structural unit in the sporopollenin skeleton. In addition, the effects of AII₃ treatment indicate that the p-coumaric acid might be bound by ether linkages.Abbreviations HPLC high-performance liquid chromatography - MS mass spectrometry - TLC thin-layer chromatography     

Modeling continuous auxiliary covariate data in generalized linear mixed models using the kernel smoother

Auxiliary covariate data are often collected in biomedical studies when the primary exposure variable is only assessed on a subset of the study subjects. In this study, we investigate a semiparametric‐estimated likelihood estimation for the generalized linear mixed models (GLMM) in the presence of a continuous auxiliary variable. We use a kernel smoother to handle continuous auxiliary data. The method can be used to deal with missing or mismeasured covariate data problems in a variety of applications when an auxiliary variable is available and cluster sizes are not too small. Simulation study results show that the proposed method performs better than that which ignores the random effects in GLMM and that which only uses data in the validation data set. We illustrate the proposed method with a real data set from a recent environmental epidemiology study on the maternal serum 1,1‐dichloro‐2,2‐bis(p‐chlorophenyl) ethylene level in relationship to preterm births.     

Sequential Classification Based on Regression

Let us consider m general populations π₁, …,π_m. Each object belonging to these populations is represented by (p ± 1) characteristics x₁, x₂,…,x_p,y. A certain object, which is an element of one of the m general populations π₁,…,π_m has to be classified into the correct population. It will be assumed that knowledge of the value of the characteristic y would permit its correct classification, but that the observation of this characteristic is expensive, difficult or dangerous, as e.g. in medical applications. y is correlated with a set of p characteristics x₁,x₂,…,x_p, which are observed sequentially. The classification procedure is based on the division of the space of the observed value of characteristics x₁,x₂,…,x_p into nonintersecting areas determined so as to minimize the value of BAYES' risk given by equation (3).     

Infradian rhythmicity in milk leukocyte activity together with plasma cortisol and prolactin levels throughout the lactation period in high-yielding crossbred cows

Infradian rhythmicity in milk leukocyte activity together with plasma cortisol and prolactin levels throughout the lactation period in high-yielding crossbred cows has been studied in 10 high-yielding (milk production: 5000 l per lactation) Karan Fries crossbred (Holstein Friesian × Tharparkar) cows. Milk and blood samples were collected from all the experimental animals. Isolation of milk phagocytes (neutrophils and macrophages) and lymphocytes were done by density gradient centrifugation. In vitro phagocytic index of milk neutrophils and macrophages was performed by colorimetric NBT reductive assay. Mitogen-induced milk lymphocyte blastogenic response was estimated by colorimetric MTT (tetrazolium) assay. Total plasma cortisol and prolactin were estimated by enzyme immune assay. Highest value of plasma cortisol and prolactin was observed at calving which decreased significantly (p < 0.01) on 15th day postpartum for both prolactin and cortisol. Immune activity of milk leukocytes was highest on day 0 colostrum and decreased significantly (p < 0.01) on 7th day postpartum. A significant (p < 0.01) rise of plasma prolactin was observed around 135th and 225th days postpartum, whereas a peak level of plasma cortisol was observed at 105th, 180^th, and 270th days postpartum. Phagocytic index of milk neutrophils and macrophages remains almost in a steady state during mid-lactation period (between 100 and 200 days postpartum). A decline in increasing trend of milk phagocytic activity was observed during late lactation. Mitogen-induced milk lymphocyte blastogenic response was highest on day 0 colostrum which decreased significantly (p < 0.01) on 15th day postpartum. Con A-induced milk lymphocyte blastogenic response showed an increasing trend from 120th to 210th days postpartum. Upon correlation study, it showed that the plasma cortisol has a negative effect on milk leukocyte activity, while prolactin has a positive effect, though the effect is lactation stage specific.     

A New Dental Approach for Reducing Food Intake

Objective: A behavioral recommendation for weight loss is reduction of size of bites of food. This “proof of concept” study tested the efficacy of a new, patented, dental approach, the DDS System, for reducing food intake. This removable tool is inserted into the upper palate of the mouth, reducing the size of the oral cavity, thereby potentially reducing bite size. Research Methods and Procedures: Thirty‐two adults (18 to 65 years) with BMI between 27 and 40 were randomly assigned to the control or experimental conditions. Participants ate all meals and stayed between meals at a research center. Day 1 served as baseline for both groups. On Day 2, experimental participants utilized the tool during meals. Changes in subjective ratings of hunger and satiety were measured using visual analog scales before and after each meal. Results: Food intake difference scores were calculated for each participant (Day 2 ? Day 1). Analysis of covariance on difference scores, using baseline as a covariate, showed that the experimental group ate significantly less (p < 0.05) on the second day (M = ?659.2 kcal/d) compared with the control group (M = ?125.9 kcal/d). Analysis of covariance, with ratings on Day 1 as a covariate, revealed that the experimental and control group did not differ on visual analog scale difference scores (premeal ? postmeal) from Day 1 to Day 2. Discussion: These findings suggest that use of this tool during meals significantly reduced food intake. This reduction of food intake was not associated with changes in ratings of hunger or satiety.     

Biochemical and Clinical Improvement of Cytotoxic State by Amantadine Sulphate

  1. The main idea of the open clinical trial was to compare the income and outcome clinical picture and the evolution of the biochemical markers in the defined intervals in closed head injury group patients.2. In the group of 32 patients, mean age 40.78±15.36 years suffering from closed traumatic brain injury the following markers were measured: glycaemia, malondialdehyde (MDA) as marker of lipid peroxidation, beta-caroten, total SH groups as marker of protein oxidation in the following intervals: between the 1st and the 3rd, between the 3rd and the 7th, between the 1st and the 7th day respectively.3. Glycaemia significantly decreased since the 1st day till the 3rd day (p < 0.05) and since the 1st day till the 7th day (p < 0.05) but it was not significantly changed since the 3rd day till the 7th day (p > 0.05).4. MDA 1st × MDA 3rd p > 0.05 insignificant change, MDA 1st × MDA 7th p < 0.001—high significant decrease, MDA 3rd × MDA 7th—p < 0.0001—very high significant decrease.5. Beta-caroten the 1st × beta-caroten the 3rd day was insignificantly changed—p > 0.05, the 3rd × the 7th day beta-caroten increased significantly—p < 0.0002, the 1st day × 7th day beta-caroten significantly increased—p < 0.0001.6. We examined the SH groups only in nine patients, due to technical problems and SH groups decrease on the 3rd day (p < 0.005).7. In 18 amantadine sulphate subgroups (randomly selected), there was 5.5% lethality and mean outcome GCS (outGCS) 9.83±3.8, while lethality of the control subgroup (n=14) was 42.9%, mean outGCS 6.28±3.5.     

Construction of Chimeric Glucansucrases for Analyzing Substrate-binding Regions That Affect the Structure of Glucan Products

A gene that encodes dextransucrase S (dsrS) from Leuconostoc mesenteroides NRRL B-512F encodes a glucansucrase dextransucrase S (DSRS) which mainly produces water-soluble glucan (dextran), while the dsrT5 gene derived from dsrT of the B-512F strain encodes an enzyme dextransucrase T5 (DSRT5), which mainly produces water-insoluble glucan. Tyr340-Asn510 of DSRS and Tyr307-Asn477 of DSRT5 (Site 1), Lys696-Gly768 of DSRS and Lys668-Gly740 of DSRT5 (Site 2), and Asn917-Lys1131 of DSRS and Asn904-Lys1118 of DSRT5 (Site 3) were exchanged and six different chimeric enzymes were constructed. Water-soluble glucan produced by recombinant DSRS was composed of 64% 6-linked glucopyranoside (Glcp), 9% 3,6-linked Glcp, and 13% 4-linked Glcp. Water-insoluble glucan produced by recombinant DSRT5 was composed of 47% 6-linked Glcp and 43% 3-linked Glcp. All of the chimeric enzymes produced glucans different from the ones produced by their parental enzymes. Some of the glucans produced by chimeric enzymes were extremely changed. The Site 1 chimeric enzyme of DSRS (STS1) produced water-soluble glucan composed mostly of 6-linked Glcp. That of DSRT5 (TST1) produced water-insoluble glucan composed mostly of 4-linked Glcp. The Site 3 chimeric enzyme of DSRS (STS3) produced mainly water-insoluble glucan, DSRT5 (TST3) produced mainly water-soluble glucans, and all of the glucan fractions consisted of 3-Glcp, 4-Glcp, and 6-Glcp. The amounts of the three linkages in the water-soluble glucan produced by TST3 were about 1:1:1. Site 1 was assumed to be important for making or avoiding making α-1,4 linkages, while Site 3 was assumed to be important for determining the kinds of glucosyl linkages made.     

Visceral Adipose Tissue in Women: Longitudinal Study of the Effects of Fat Gain,Time, and Race

Objective: To determine the effects of fat gain, time, and race on the accumulation of visceral adipose tissue (VAT) in a group of normal‐weight premenopausal women. Research Methods and Procedures: Sixty‐five women participated in the study (32 African American and 33 white). The mean age of subjects was 34 ± 6 years (range, 22 to 47 years). Eligible subjects were women who had body mass indices <25 kg/m² at baseline and who had completed evaluations at baseline and at follow‐up year 1, without intervention. A subset of subjects was reevaluated annually for up to 4 years. Body composition was assessed by DXA, and VAT was determined from a single computed tomography scan. A linear mixed model was used to examine changes in VAT over time, with total body fat as a covariate Results: Total fat mass was not significantly different between races at baseline and increased significantly in both groups over time (p < 0.001). Time‐related increases in total body fat were greater in African‐American women (p < 0.01). VAT was significantly higher in white women at baseline (p < 0.01) and increased significantly over time in both races (p < 0.01), but remained higher in white women (p < 0.001). Increases in VAT, relative to total body fat, were greater than the increases in total body fat over time, independent of age and race (p < 0.001). Discussion: Gaining total body‐fat mass results in a higher increase in VAT, relative to total body fat, regardless of race and age, although African‐American women maintain a lower VAT levels across time.