Asymptotics for Marginal Generalized Linear Models With Sparse Correlations

Marginal generalized linear models can be used for clustered and longitudinal data by fitting a model as if the data were independent and using an empirical estimator of parameter standard errors. We extend this approach to data where the number of observations correlated with a given one grows with sample size and show that parameter estimates are consistent and asymptotically Normal with a slower convergence rate than for independent data, and that an information sandwich variance estimator is consistent. We present two problems that motivated this work, the modelling of patterns of HIV genetic variation and the behavior of clustered data estimators when clusters are large.

Microhardness and marginal adaptation of pre-warmed composites

Objectives: The aim of this study was to examine the effect of pre-warmed composite on the microhardness and marginal adaptation. Methods: Ninety six identical class II cavities were prepared in extracted human molars and filled/cured in three 2 mm increments using a metal matrix. Two composites (Tetric Evo Ceram (IvoclarVivadent) and ELS(Saremco)) were cured with a LED curing unit (Bluephase (IvoclarVivadent)) using curing cycles of 20 and 40 seconds. The composite was used at room temperature or pre-warmed at 54.5ºC (Calset(AdDent)). Twelve teeth were filled for every composite-curing time-composite temperature combination. The teeth were thermocycled (1000 cycles at 5º and 55ºC) and then stored at 37° C for seven days . Dye penetration (basic fuchsine 5% for 8 hours) was measured using a score scale. Knoop microhardness was determined 100, 200, 500, 1000, 1500, 2500, 3500, 4500 and 5500µm from the occlusal surface at a distance of 150 and 1000µm from the metal matrix. The total degree of polymerization of a composite specimen was determined by calculating the area under the hardness curve. Results: Statistical analyses showed no difference in marginal adaptation (p>0.05). Hardness values at 150µm from the matrix were lower than those at 1000µm. There was an increase of the microhardness at the top of each increment and decrease towards the bottom of each increment. Longer curing times resulted in harder composite samples. Multiple linear regression showed that only the curing time (p<0.001) and composite material (p<0.001) had a significant association with the degree of polymerization. The degree of polymerization was not influenced by pre-warming the composite at a temperature of 54.5ºC (p=4.86). Conclusion: Polymerization time can not be reduced by pre-warming the composite on a temperature of 54.5ºC. The marginal adaptation is not compromised by pre-warming the composite.

Marginal Structural Models for Partial Exposure Regimes

Intensive care unit (ICU) patients are ell known to be highly susceptible for nosocomial (i.e. hospital-acquired) infections due to their poor health and many invasive therapeutic treatments. The effects of acquiring such infections in ICU on mortality are however ill understood. Our goal is to quantify these effects using data from the National Surveillance Study of Nosocomial Infections in Intensive Care Units (Belgium). This is a challenging problem because of the presence of time-dependent confounders (such as exposure to mechanical ventilation)which lie on the causal path from infection to mortality. Standard statistical analyses may be severely misleading in such settings and have shown contradicting results. While inverse probability weighting for marginal structural models can be used to accommodate time-dependent confounders, inference for the effect of ?ICU acquired infections on mortality under such models is further complicated (a) by the fact that marginal structural models infer the effect of acquiring infection on a given, fixed day ?in ICU?, which is not well defined when ICU discharge comes prior to that day; (b) by informative censoring of the survival time due to hospital discharge; and (c) by the instability of the inverse weighting estimation procedure. We accommodate these problems by developing inference under a new class of marginal structural models which describe the hazard of death for patients if, possibly contrary to fact, they stayed in the ICU for at least a given number of days s and acquired infection or not on that day. Using these models we estimate that, if patients stayed in the ICU for at least s days, the effect of acquiring infection on day s would be to multiply the subsequent hazard of death by 2.74 (95 per cent conservative CI 1.48; 5.09).

On the Behaviour of Marginal and Conditional Akaike Information Criteria in Linear Mixed Models

In linear mixed models, model selection frequently includes the selection of random effects. Two versions of the Akaike information criterion (AIC) have been used, based either on the marginal or on the conditional distribution. We show that the marginal AIC is no longer an asymptotically unbiased estimator of the Akaike information, and in fact favours smaller models without random effects. For the conditional AIC, we show that ignoring estimation uncertainty in the random effects covariance matrix, as is common practice, induces a bias that leads to the selection of any random effect not predicted to be exactly zero. We derive an analytic representation of a corrected version of the conditional AIC, which avoids the high computational cost and imprecision of available numerical approximations. An implementation in an R package is provided. All theoretical results are illustrated in simulation studies, and their impact in practice is investigated in an analysis of childhood malnutrition in Zambia.

Influence of beveling and ultrasound application on marginal adaptation of box-only Class II (slot) resin composite restorations

A laboratory study was performed to assess the influence of beveling the margins of cavities and the effects on marginal adaptation of the application of ultrasound during setting and initial light curing. After minimal access cavities had been prepared with an 80 microm diamond bur, 80 box-only Class II cavities were prepared mesially and distally in 40 extracted human molars using four different oscillating diamond coated instruments: (A) a U-shaped PCS insert as the non-beveled control (EMS), (B) Bevelshape (Intensiv), (C) SonicSys (KaVo) and (D) SuperPrep (KaVo). In groups B-D, the time taken for additional bevel finishing was measured. The cavities were filled with a hybrid composite material in three increments. Ultrasound was also applied to one cavity per tooth before and during initial light curing (10 seconds). The specimens were subjected to thermomechanical stress in a computer-controlled masticator device. Marginal quality was assessed by scanning electron microscopy and the results were compared statistically. The additional time required for finishing was B > D > C (p < or = 0.05). In all groups, thermomechanical loading resulted in a decrease in marginal quality. Beveling resulted in higher values for "continuous" margins compared with that of the unbeveled controls. The latter showed better marginal quality at the axial walls when ultrasound was used. Beveling seems essential for good marginal adaptation but requires more preparation time. The use of ultrasonic vibrations may improve the marginal quality of unbeveled fillings and warrants further investigation.

Marginal fit of cemented and screw-retained crowns incorporated on the Straumann (ITI) Dental Implant System: an in vitro study

AIM: The aim of this study was to assess the marginal fit of crowns on the Straumann (ITI) Dental Implant System with special consideration of different casting dental materials. MATERIAL AND METHODS: Sixty porcelain-fused-to-metal crowns were fabricated: 18 crowns on standard cone abutments with an impression cylinder, partially prefabricated analogs, no coping and screw-retained (A); 18 crowns on solid abutments without an impression device, no analogs, no coping and cemented (B); and 18 crowns on solid abutments using an impression transfer cap, an analog with a shoulder, no coping and cemented (C). In each group, six crowns were made on epoxy mastercasts (Bluestar), six on synthetic plaster (Moldasynt) and six on super hard stone (Fujirock). Six additional crowns were fabricated with the transversal screw retention system onto the Octa system with impression transfer caps, metal analogs, gold copings and screw-retained (D). Impregum was used as impression material. Crowns of B and C were cemented with KetacCem. Crowns of A and D were fixed with an occlusal screw torqued at 15 N cm. Crowns were embedded, cut and polished. Under a light microscope using a magnification of x 100, the distance between the crown margin (CM) and the shoulder (marginal gap, MG) and the distance between the CM and the end of the shoulder (crown length, CL) was measured. RESULTS: MGs were 15.4+/-13.2 microm (A), 21.2+/-23.1 microm (B), 11+/-12.1 microm (C) and 10.4+/-9.3 microm (D). No statistically significantly differences using either of the casting materials were observed. CLs were -21.3+/-24.8 microm (A), 3+/-28.9 microm (B), 0.5+/-22 microm (C) and 0.1+/-15.8 microm (D). Crowns were shorter on synthetic casting materials compared with stone casts (P<0.005). CONCLUSIONS: CMs fit precisely with both cemented and screw-retained versions as well as when using no, partial or full analogs.

The NF-{kappa}B negative regulator TNFAIP3 (A20) is inactivated by somatic mutations and genomic deletions in marginal zone lymphomas

Unique and shared cytogenetic abnormalities have been documented for marginal zone lymphomas (MZLs) arising at different sites. Recently, homozygous deletions of the chromosomal band 6q23, involving the tumor necrosis factor alpha-induced protein 3 (TNFAIP3, A20) gene, a negative regulator of NF-kappaB, were described in ocular adnexal MZL, suggesting a role for A20 as a tumor suppressor in this disease. Here, we investigated inactivation of A20 by DNA mutations or deletions in a panel of extranodal MZL (EMZL), nodal MZL (NMZL), and splenic MZL (SMZL). Inactivating mutations encoding truncated A20 proteins were identified in 6 (19%) of 32 MZLs, including 2 (18%) of 11 EMZLs, 3 (33%) of 9 NMZLs, and 1 (8%) of 12 SMZLs. Two additional unmutated nonsplenic MZLs also showed monoallelic or biallelic A20 deletions by fluorescent in situ hybridization (FISH) and/or SNP-arrays. Thus, A20 inactivation by either somatic mutation and/or deletion represents a common genetic aberration across all MZL subtypes, which may contribute to lymphomagenesis by inducing constitutive NF-kappaB activation.

Predictive Margins and Marginal Effects in Stata

Tables of estimated regression coefficients, usually accompanied by additional information such as standard errors, t-statistics, p-values, confidence intervals or significance stars, have long been the preferred way of communicating results from statistical models. In recent years, however, the limits of this form of exposition have been increasingly recognized. For example, interpretation of regression tables can be very challenging in the presence of complications such as interaction effects, categorical variables, or nonlinear functional forms. Furthermore, while these issues might still be manageable in the case of linear regression, interpretational difficulties can be overwhelming in nonlinear models such as logistic regression. To facilitate sensible interpretation of such models it is often necessary to compute additional results such as marginal effects, predictive margins, or contrasts. Moreover, smart graphical displays of results can be very valuable in making complex relations accessible. A number of helpful commands geared at supporting these tasks have been recently introduced in Stata, making elaborate interpretation and communication of regression results possible without much extra effort. Examples of such commands are -margins-, -contrasts-, and -marginsplot-. In my talk, I will discuss the capabilities of these commands and present a range of examples illustrating their use.

Survey estimation and marginal effects in Stata

Stata is a general purpose software package that has become popular among various disciplines such as epidemiology, economics, or social sciences. Users like Stata for its scientific approach, its robustness and reliability, and the ease with which its functionality can be extended by user written programs. In this talk I will first give a brief overview of the functionality of Stata and then discuss two specific features: survey estimation and predictive margins/marginal effects. Most surveys are based on complex samples that contain multiple sampling stages, are stratified or clustered, and feature unequal selection probabilities. Standard estimators can produce misleading results in such samples unless the peculiarities of the sampling plan are taken into account. Stata offers survey statistics for complex samples for a wide variety of estimators and supports several variance estimation procedures such as linearization, jackknife, and balanced repeated replication (see Kreuter and Valliant, 2007, Stata Journal 7: 1-21). In the talk I will illustrate these features using applied examples and I will also show how user written commands can be adapted to support complex samples. Complex can also be the models we fit to our data, making it difficult to interpret them, especially in case of nonlinear or non-additive models (Mood, 2010, European Sociological Review 26: 67-82). Stata provides a number of highly useful commands to make results of such models accessible by computing and displaying predictive margins and marginal effects. In my talk I will discuss these commands provide various examples demonstrating their use.

Challenging conservation agriculture on marginal slopes in Sehoul, Morocco

In Sehoul, Morocco, the use of marginal land for agriculture became a necessity for the local population due to increased poverty and the occupation of the best land by new owners. Desertification poses an additional threat to agricultural production on marginal slopes, which are often stony and degraded. In a participatory process embedded in the EU DESIRE research project, potential sustainable land management measures were selected to address land degradation and desertification. Promising experiences with no-tillage practices elsewhere in Morocco had motivated the Moroccan government to promote conservation agriculture throughout the country. This combination of crop rotation, minimal soil disturbance and soil cover maintenance, however, had not yet been tested on sloping degraded land. Field trials of grazing enclosure combined with no or minimum tillage were conducted on the plots of two farmers, and trial results were analyzed based on stakeholders’ criteria. Results suggest that increased soil cover with barley residues improved rainwater use efficiency and yields only slightly, although soil water was generally enhanced. Soil moisture measurements revealed that no-tillage was favorable mainly at soil depths of 5 cm and in connection with low-rainfall events (<20 mm); under these circumstances, moisture content was generally higher under no-tillage than under conventional tillage. Moreover, stakeholder discussion confirmed that farmers in Sehoul remain primarily interested in animal husbandry and are reluctant to change the current grazing system. Implementation of conservation agriculture is thus challenged both by the degraded, sloping and stony nature of the land, and by the socio-economic circumstances in Sehoul.

Resin composites: Modulus of elasticity and marginal quality

OBJECTIVE To investigate how the modulus of elasticity of resin composites influences marginal quality in restorations submitted to thermocyclic and mechanical loading. METHODS Charisma, Filtek Supreme XTE and Grandio were selected as they were found to possess different moduli of elasticity but quite similar polymerization contraction. MOD cavities (n=30) were prepared in extracted premolars, restored and then subjected to thermocyclic and mechanical loading. Marginal quality of the restorations before and after loading was analyzed on epoxy replicas under a scanning electron microscope. The percentage of gap-free margins and occurrence of paramarginal fractures were registered. Modulus of elasticity and polymerization contraction were analyzed with parametric and margins with nonparametric ANOVA and post hoc Tukey HSD or Wilcoxon rank-sum tests, respectively. The number of paramarginal fractures was analyzed with exact Fisher tests (α=0.05). RESULTS Grandio demonstrated significantly more gap-free enamel margins than Charisma and Filtek Supreme XTE, before and after loading (p<0.01), whereas there was no difference between Charisma and Filtek Supreme XTE (p>0.05). No significant effect of resin composite (p=0.81) on the quality of dentine margins was observed, before or after loading. Deterioration of all margins was evident after loading (p<0.0001). More paramarginal enamel fractures were observed after loading in teeth restored with Grandio when compared to Charisma (p=0.008). CONCLUSIONS The resin composite with the highest modulus of elasticity resulted in the highest number of gap-free enamel margins but with an increased incidence of paramarginal enamel fractures. CLINICAL SIGNIFICANCE The results from this study suggest that the marginal quality of restorations can be improved by the selection of a resin composite with modulus of elasticity close to that of dentine, although an increase in paramarginal enamel fractures can result as a consequence.