Analysis of count data from clinical trials using mixed effect analysis has recently become widely used. However, algorithms
available for the parameter estimation, including LAPLACE and Gaussian quadrature (GQ), are associated with certain limitations,
including bias in parameter estimates and the long analysis runtime. The stochastic approximation expectation maximization
(SAEM) algorithm has proven to be a very efficient and powerful tool in the analysis of continuous data. The aim of this study
was to implement and investigate the performance of a new SAEM algorithm for application to count data. A new SAEM algorithm
was implemented in MATLAB for estimation of both, parameters and the Fisher information matrix. Stochastic Monte Carlo simulations
followed by re-estimation were performed according to scenarios used in previous studies (part I) to investigate properties
of alternative algorithms (Plan et al., 2008, Abstr 1372 []). A single scenario was used to explore six probability distribution models. For parameter estimation, the relative bias
was less than 0.92% and 4.13% for fixed and random effects, for all models studied including ones accounting for over- or
under-dispersion. Empirical and estimated relative standard errors were similar, with distance between them being <1.7% for
all explored scenarios. The longest CPU time was 95 s for parameter estimation and 56 s for SE estimation. The SAEM algorithm
was extended for analysis of count data. It provides accurate estimates of both, parameters and standard errors. The estimation
is significantly faster compared to LAPLACE and GQ. The algorithm is implemented in Monolix 3.1, (beta-version available in
July 2009). 相似文献
BackgroundRecently, several preoperative proinflammatory markers and nutritional factors such as neutrophil-to-lymphocyte ratio (NLR) and prognostic nutrition index (PNI) have been reported as significant predictor for poor prognosis of various malignant tumors. In this study, we evaluated the prognostic values of these preoperative parameters in patients with resectable pancreatic head cancer.MethodsWe retrospectively reviewed consecutive patients who underwent PD for pancreatic head cancer between 2007 and 2012. A total of 46 patients were enrolled in this analysis. Preoperative parameters such as CRP, CA19-9, NLR and PNI at the time of presentation were recorded as well as overall survival. Cancer specific survival was assessed using Kaplan–Meier method. Univariate and multivariate Cox regression models were applied to evaluate the prognostic relevance of preoperative parameters. The correlations between CA19-9 values, NLR and pathological findings, first recurrence site were respectively reviewed.ResultsIn multivariable analysis preoperative high NLR (≧2.7) and high CA19-9 (≧230) were independent prognostic factors for poor survival (P value: 0.03 and 0.025, respectively). Kaplan–Meier survival analysis demonstrated the overall 2-year survival rate in patients with high NLR or high CA19-9 were 37.5% compared with 89.9% in patients with low NLR and low CA19-9.ConclusionPreoperative NLR and serum CA19-9 offer significant prognostic information associated with overall survival following PD in the patients with pancreatic head cancer. 相似文献
We investigated the possible use of D -lactate as a predictor in the diagnosis of appendicitis. C-reactive protein level (CRP) and leukocyte counts were also evaluated. Venous blood D -lactate, CRP, and leukocyte counts were measured preoperatively in 53 patients undergoing surgery for appendicitis, as well as in 20 healthy subjects. Levels of all three parameters in the surgical patients were significantly higher than in the control group ( p < .05). Previous studies have shown that venous D -lactate is more specific to the intestine than CPR or leukocyte count. Based on our data, venous D -lactate, which had the lowest false-negative rate among these laboratory parameters, may be a useful diagnostic marker for appendicitis. None of these parameters were helpful in identifying the type of the appendicitis. 相似文献
Because nanomaterials have been increasingly developed and used in many technology and industry sectors over the last 20 years, an increasing number of workers is likely to be exposed to airborne nanoparticles. In addition, the question of the nanomaterial characteristics that should be assessed in epidemiological studies remains open. Thus, assessing occupational exposure to airborne nanoparticles will not only rely on mass concentration and chemical composition. Rather, key parameters, such as particle size, have to be included in measurement strategies.
We previously proposed a methodology to estimate the Count Median Diameter (CMD) of an aerosol based on the simultaneous size-integrated measurement of two particle concentrations, lung-deposited surface area, and number, thanks to field-portable, commercially available aerosol instruments (Nanoparticle Surface Area Monitor/Condensation Particle Counter combination).
In addition to previous work, this study investigates the case of various polydisperse metal oxides, organic oil, and salt particles with CMDs ranging from 16–410 nm. Once corrected, the CMDs derived from the NSAM/CPC agree within ±20% with regard to the reference electrical mobility equivalent diameter, regardless of aerosol composition, morphology, or geometric standard deviation (GSD).
Furthermore, the field-applicability of the method was tested through 6 sets of experimental data stemming from workplace measurement campaigns where different materials were produced and handled (TiO2, SiO2, Ag, Multi-Walled Carbon Nanotubes—MWCNT), covering a range of CMDs between 40 and 190 nm. All situations considered, the approach based on the combination of a NSAM and a CPC leads to a satisfying estimation of particle CMD, within ±20% compared to reference CMD. 相似文献
This paper investigates the effects of reducing subsidies for private health insurance on public sector expenditure for hospital care. An econometric framework using simultaneous equation models is developed to analyse the interrelated decisions on the intensity and type of health care use and private insurance. The framework is applied to the context of the mixed public–private system in Australia. The simulation projections show that reducing premium subsidies is expected to generate net cost savings. This arises because the cost savings achieved from reducing subsidies are larger than the potential increase in public expenditure on hospital care. 相似文献
To analyze the impact of timing of clamping and obstetric, biological and
socioeconomic factors on the iron stores of full-term newborns.
METHODS
Cross-sectional study between October 2011 and July 2012 in which
hematological parameters were evaluated for newborns in Viçosa, MG,
Southeastern Brazil. It involved collecting 7 mL of umbilical cord blood
from 144 full-term not underweight newborns. The parameters investigated
were complete blood count, serum iron, ferritin and C-reactive protein. The
time of umbilical cord clamping was measured using a digital timer without
interfering in the procedures of childbirth. The birth data were collected
from Live Birth Certificates and other information was obtained from the
mother through a questionnaire applied in the first month postpartum.
Analysis of multiple linear regression was then used to estimate the
influence of biological, obstetrics and socioeconomic factors on the
ferritin levels at birth.
RESULTS
The median ferritin was 130.3 µg/L (n = 129, minimum = 16.4; maximum =
420.5 µg/L), the mean serum iron was 137.9 μg/dL (n = 144, SD =
39.29) and mean hemoglobin was 14.7 g/dL (n = 144, SD = 1.47). The median
time of cord clamping was 36 seconds, ranging between 7 and 100. The
bivariate analysis detected an association between ferritin levels and color
of the child, timing clamping of 60 seconds, type of delivery, the presence
of gestational diabetes and per capita family income. In
multivariate analysis, the variables per capita income,
number of antenatal visits and length at birth accounted for 22.0% of
variation in ferritin levels.
CONCLUSIONS
Iron stores at birth were influenced by biological, obstetric and social
characteristics. Tackling anemia should involve creating policies aimed at
reducing social inequalities, improving the quality of antenatal care, as
well as implementing a criterion of delayed clamping of the umbilical cord
within the guidelines of labor. 相似文献