You-Gan Wang

Factors Potentiating the Risk of Sudden Infant Death Syndrome Associated with the Prone Position

BACKGROUND In several studies the sudden infant death syndrome (SIDS) has been significantly associated with sleeping in the prone position. It is not known how the prone position increases the risk of SIDS. METHODS We analyzed data from a case-control study (58 infants with SIDS and 120 control infants) and a prospective cohort study (22 infants with SIDS and 213 control infants) in Tasmania. Interactions were examined in matched analyses with a multiplicative model of interaction. RESULTS In the case-control study, SIDS was significantly associated with sleeping in the prone position, as compared with other positions (unadjusted odds ratio, 4.5; 95 percent confidence interval, 2.1 to 9.6). The strength of this association was increased among infants who slept on natural-fiber mattresses (P = 0.05), infants who were swaddled (P = 0.09), infants who slept in heated rooms (P = 0.006), and infants who had had a recent illness (P = 0.02). These variables had no significant effect on infants who did not sleep in the prone position. A history of recent illness was significantly associated with SIDS among infants who slept prone (odds ratio, 5.7; 95 percent confidence interval, 1.8 to 19) but not among infants who slept in other positions (odds ratio, 0.83). In the cohort study, the risk of SIDS was greater among infants who slept prone on natural-fiber mattresses (odds ratio, 6.6; 95 percent confidence interval, 1.3 to 33) than among infants who slept prone on other types of mattresses (odds ratio, 1.8). CONCLUSIONS When infants sleep prone, the elevated risk of SIDS is increased by each of four factors: the use of natural-fiber mattresses, swaddling, recent illness, and the use of heating in bedrooms.

The impact of global positioning systems and plotters on fishing power in the northern prawn fishery, Australia

The impact of global positioning systems (GPS) and plotter systems on the relative fishing power of the northern prawn fishery fleet on tiger prawns (Penaeus esculentus Haswell, 1879, and P. semisulcatus de Haan, 1850) was investigated from commercial catch data. A generalized linear model was used to account for differences in fishing power between boats and changes in prawn abundance. It was found that boats that used a GPS alone had 4% greater fishing power than boats without a CPS. The addition of a plotter raised the power by 7% over boats without the equipment. For each year between the first to third that a fisher has been working with plotters, there is an additional 2 or 3% increase. It appears that when all boats have a GPS and plotter for at least 3 years, the fishing power of the fleet will increase by 12%. Management controls have reduced the efficiency of each boat and lowered the number of days available to fish, but this may not have been sufficient to counteract the increases. Further limits will be needed to maintain the desired levels of mortality.

Working‐correlation‐structure identification in generalized estimating equations

Selecting an appropriate working correlation structure is pertinent to clustered data analysis using generalized estimating equations (GEE) because an inappropriate choice will lead to inefficient parameter estimation. We investigate the well-known criterion of QIC for selecting a working correlation structure, and have found that performance of the QIC is deteriorated by a term that is theoretically independent of the correlation structures but has to be estimated with an error. This leads us to propose a correlation information criterion (CIC) that substantially improves the QIC performance. Extensive simulation studies indicate that the CIC has remarkable improvement in selecting the correct correlation structures. We also illustrate our findings using a data set from the Madras Longitudinal Schizophrenia Study.

Isotonic designs for phase I trials

The purpose of a phase I trial in cancer is to determine the level (dose) of the treatment under study that has an acceptable level of adverse effects. Although substantial progress has recently been made in this area using parametric approaches, the method that is widely used is based on treating small cohorts of patients at escalating doses until the frequency of toxicities seen at a dose exceeds a predefined tolerable toxicity rate. This method is popular because of its simplicity and freedom from parametric assumptions. In this paper, we consider cases in which it is undesirable to assume a parametric dose-toxicity relationship. We propose a simple model-free approach by modifying the method that is in common use. The approach assumes toxicity is nondecreasing with dose and fits an isotonic regression to accumulated data. At any point in a trial, the dose given is that with estimated toxicity deemed closest to the maximum tolerable toxicity. Simulations indicate that this approach performs substantially better than the commonly used method and it compares favorably with other phase I designs. Control Clin Trials 2001;22:126-138

Unbiased Estimating Equations From Working Correlation Models for Irregularly Timed Repeated Measures

The method of generalized estimating equations (GEEs) has been criticized recently for a failure to protect against misspecification of working correlation models, which in some cases leads to loss of efficiency or infeasibility of solutions. However, the feasibility and efficiency of GEE methods can be enhanced considerably by using flexible families of working correlation models. We propose two ways of constructing unbiased estimating equations from general correlation models for irregularly timed repeated measures to supplement and enhance GEE. The supplementary estimating equations are obtained by differentiation of the Cholesky decomposition of the working correlation, or as score equations for decoupled Gaussian pseudolikelihood. The estimating equations are solved with computational effort equivalent to that required for a first-order GEE. Full details and analytic expressions are developed for a generalized Markovian model that was evaluated through simulation. Large-sample “sandwich” standard errors for working correlation parameter estimates are derived and shown to have good performance. The proposed estimating functions are further illustrated in an analysis of repeated measures of pulmonary function in children.

Stock-recruitment relationships of the tiger prawns (Penaeus esculentus and Penaeus semisulcatus) in the Australian Northern Prawn Fishery

This paper investigates the stock-recruitment and equilibrium yield dynamics for the two species of tiger prawns (Penaeus esculentus and Penaeus semisulcatus) in Australias most productive prawn fishery: the Northern Prawn Fishery. Commercial trawl logbooks for 1970-93 and research surveys are used to develop population models for these prawns. A population model that incorporates continuous recruitment is developed. Annual spawning stock and recruitment indices are then estimated from the population model. Spawning stock indices represent the abundance of female prawns that are likely to spawn; recruitment indices represent the abundance of all prawns less than a certain size. The relationships between spawning stock and subsequent recruitment (SRR), between recruitment and subsequent spawning stock (RSR), and between recruitment and commercial catch were estimated through maximum-likelihood models that incorporated autoregressive terms. Yield as a function of fishing effort was estimated by constraining to equilibrium the SRR and RSR. The resulting production model was then used to determine maximum sustainable yield (MSY) and its corresponding fishing effort (fMSY). Long-term yield estimates for the two tiger prawn species range between 3700 and 5300 t. The fishing effort at present is close to the level that should produce MSY for both species of tiger prawns. However, current landings, recruitment and spawning stock are below the equilibrium values predicted by the models. This may be because of uncertainty in the spawning stock-recruitment relationships, a change in carrying capacity, biased estimates of fishing effort, unreliable catch statistics, or simplistic assumptions about stock structure. Although our predictions of tiger prawn yields are uncertain, management will soon have to consider new measures to counteract the effects of future increases in fishing effort.

Analysis of Human Immunodeficiency Virus Type 1 Drug Resistance in Children Receiving Nucleoside Analogue Reverse-Transcriptase Inhibitors plus Nevirapine, Nelfinavir, or Ritonavir (Pediatric AIDS Clinical Trials Group 377)

In Pediatric AIDS Clinical Trials Group 377, antiretroviral therapy-experienced children were randomized to 4 treatment arms that included different combinations of stavudine, lamivudine (3TC), nevirapine (Nvp), nelfinavir (Nfv), and ritonavir (Rtv). Previous treatment with zidovudine (Zdv), didanosine (ddI), or zalcitabine (ddC) was acceptable. Drug resistance mutations were assessed before study treatment (baseline) and at virologic failure. Zdv, ddI, and ddC mutations were detected frequently at baseline but were not associated with virologic failure. Children with drug resistance mutations at baseline had greater reductions in virus load over time than did children who did not. Nvp and 3TC mutations were detected frequently at virologic failure, and Nvp mutations were more common among children receiving 3-drug versus 4-drug Nvp-containing regimens. Children who were maintained on their study regimen after virologic failure accumulated additional Nvp and 3TC mutations plus Rtv and Nfv mutations. However, Rtv and Nfv mutations were detected at unexpectedly low rates.

Criteria for working-correlation-structure selection in GEE : Assessment via simulation

Efficiency of analysis using generalized estimation equations is enhanced when intracluster correlation structure is accurately modeled. We compare two existing criteria (a quasi-likelihood information criterion, and the Rotnitzky–Jewell criterion) to identify the true correlation structure via simulations with Gaussian or binomial response, covariates varying at cluster or observation level, and exchangeable or AR(1) intracluster correlation structure. Rotnitzky and Jewells approach performs better when the true intracluster correlation structure is exchangeable, while the quasi-likelihood criteria performs better for an AR(1) structure.

A Simulation Model for Evaluating Seasonal Closures in Australia's Multispecies Northern Prawn Fishery

Abstract In Australias northern prawn fishery, seasonal closures have long been used to manage the size composition of the commercial catch. Although primarily aimed at short-term economic benefits of improved yield per recruit, closures were also thought to reduce the risk of recruitment overfishing. Until recent years, seasonal closures took the form of a single closed season in late summer to protect juvenile banana prawns Penaeus merguiensis during months of peak recruitment. However, with the growing commercial importance of tiger prawns P. esculentus and P. semisulcatus and associated concern about recruitment overfishing, a second midyear closure was introduced to further protect subadult tiger prawns. In this paper, we use a multispecies simulation model of the northern prawn fishery to assess the biological and economic effectiveness of seasonal closures in relation to yield, income, net operating income, and spawning stock indices per recruit. The model incorporates factors such as seasonal tre...

A generalized estimating equations approach for analysis of the impact of new technology on a trawl fishery

The article describes a generalized estimating equations approach that was used to investigate the impact of technology on vessel performance in a trawl fishery during 1988-96, while accounting for spatial and temporal correlations in the catch-effort data. Robust estimation of parameters in the presence of several levels of clustering depended more on the choice of cluster definition than on the choice of correlation structure within the cluster. Models with smaller cluster sizes produced stable results, while models with larger cluster sizes, that may have had complex within-cluster correlation structures and that had within-cluster covariates, produced estimates sensitive to the correlation structure. The preferred model arising from this dataset assumed that catches from a vessel were correlated in the same years and the same areas, but independent in different years and areas. The model that assumed catches from a vessel were correlated in all years and areas, equivalent to a random effects term for vessel, produced spurious results. This was an unexpected finding that highlighted the need to adopt a systematic strategy for modelling. The article proposes a modelling strategy of selecting the best cluster definition first, and the working correlation structure (within clusters) second. The article discusses the selection and interpretation of the model in the light of background knowledge of the data and utility of the model, and the potential for this modelling approach to apply in similar statistical situations.