Statistical modelling of groundwater contamination monitoring data: A comparison of spatial and spatiotemporal methods.

Field monitoring of groundwater contamination plumes is an important component of managing risks for downgradient receptors and remedial strategies that rely on monitored natural attenuation. Collection of groundwater quality data can however take a considerable effort and be associated with high cost. Here, we investigated the relative merits of analyzing groundwater quality data using spatial compared to spatiotemporal statistical modelling and assessed the accuracy of both methods and implications for data collection requirements. The aim of this was to determine whether the quantity of data collected can be reduced, while retaining the same level of estimation accuracy, by analyzing groundwater contamination data using a spatiotemporal model which "borrows strength" across time, rather than a spatial model for individual sampling events. To capture the variability encountered under field conditions, we used three hypothetical groundwater contamination plumes with increasing complexity, and site data for a large groundwater gasoline additive plume. The results show that spatiotemporal methods can increase efficiency markedly so that, in comparison with repeated spatial analysis, spatiotemporal methods can achieve the same level of performance but with smaller sample sizes.

Assessing the outcome of orthognathic surgery by three-dimensional soft tissue analysis.

Studies of orthognathic surgery often focus on pre-surgical versus post-surgical changes in facial shape. In contrast, this study provides an innovative comparison between post-surgical and control shape. Forty orthognathic surgery patients were included, who underwent three different types of surgical correction: Le Fort I maxillary advancement, bilateral sagittal split mandibular advancement, and bimaxillary advancement surgery. Control facial images were captured from volunteers from local communities in Glasgow, with patterns of age, sex, and ethnic background that matched those of the surgical patients. Facial models were fitted and Procrustes registration and principal components analysis used to allow quantitative analysis, including the comparison of group mean shape and mean asymmetry. The primary characteristic of the difference in shape was found to be residual mandibular prognathism in the group of female patients who underwent Le Fort I maxillary advancement. Individual cases were assessed against this type of shape difference, using a quantitative scale to aid clinical audit. Analysis of the combined surgical groups provided strong evidence that surgery reduces asymmetry in some parts of the face such as the upper lip region. No evidence was found that mean asymmetry in post-surgical patients is greater than that in controls.

Efficient and automatic methods for flexible regression on spatiotemporal data, with applications to groundwater monitoring.

Fitting statistical models to spatiotemporal data requires finding the right balance between imposing smoothness and following the data. In the context of P-splines, we propose a Bayesian framework for choosing the smoothing parameter, which allows the construction of fully automatic data-driven methods for fitting flexible models to spatiotemporal data. An implementation, which is highly computationally efficient and exploits the sparsity of the design and penalty matrices, is proposed. The findings are illustrated using a simulation study and two examples, all concerned with the modelling of contaminants in groundwater. This suggests that the proposed strategy is more stable that competing methods based on the use of criteria such as generalised cross-validation and Akaike's Information Criterion. © 2015 The Authors. Environmetrics Published by John Wiley Sons Ltd.

Validation and comparison of geostatistical and spline models for spatial stream networks.

Scientists need appropriate spatial-statistical models to account for the unique features of stream network data. Recent advances provide a growing methodological toolbox for modelling these data, but general-purpose statistical software has only recently emerged, with little information about when to use different approaches. We implemented a simulation study to evaluate and validate geostatistical models that use continuous distances, and penalised spline models that use a finite discrete approximation for stream networks. Data were simulated from the geostatistical model, with performance measured by empirical prediction and fixed effects estimation. We found that both models were comparable in terms of squared error, with a slight advantage for the geostatistical models. Generally, both methods were unbiased and had valid confidence intervals. The most marked differences were found for confidence intervals on fixed-effect parameter estimates, where, for small sample sizes, the spline models underestimated variance. However, the penalised spline models were always more computationally efficient, which may be important for real-time prediction and estimation. Thus, decisions about which method to use must be influenced by the size and format of the data set, in addition to the characteristics of the environmental process and the modelling goals. ©2015 The Authors. Environmetrics published by John Wiley & Sons, Ltd.

Spatiotemporal statistical modelling of long-term change in river nutrient concentrations in England & Wales.

Concentrations of nutrient nitrogen (N) and phosphorus (P) are elevated in rivers across large areas of Europe (European Nitrogen Assessment (ENA), Sutton et al., 2011). Environmental policies have been implemented over the past 20 years with the aim of reducing nitrogen inputs to surface waters. However, environmental and ecological status is still below set targets (ENA, Sutton et al., 2011). Identification of patterns in long-term change for nutrient trends in hydrological catchments in England & Wales is required to assess impacts of nutrient management policy and provide better evidence for future policy. Such information could provide essential evidence for supporting policy by combining information from the wider catchment, rather than relying on the analysis of data from individual sites. Surface water quality is subject to considerable spatial and short-period temporal variability, reflecting variability in loading and dilution. This makes it difficult to determine temporal trends at individual monitoring sites with relatively sparse sampling. Here we apply spatiotemporal statistical additive models for both nitrogen and phosphorus in river networks across England & Wales to investigate the overall pattern of nutrient concentrations in these river surface waters over the past 20-40 years. Concentrations of Orthophosphate (OP) have generally decreased over time for many of the Large Hydrological Areas with a seasonal pattern highlighting one peak in the summer months. Over the past ten years, Total Oxidised Nitrogen (Nitrate+Nitrite, TON) concentrations have generally been slowly decreasing or fairly constant. However, prior to 2000, concentrations were generally on an upward trend. The seasonal pattern highlights one trough in the summer months. The highest levels for OP and TON broadly occur in the same general areas across England & Wales. On average, over time, the lowest values are evident in the north-west and south-west (particularly for OP) and highest values are evident in the Midlands, Anglian and Southern regions.

The effect of transmucosal 0.2 mg/kg midazolam premedication on dental anxiety, anaesthetic induction and psychological morbidity in children undergoing general anaesthesia for tooth extraction.

BACKGROUND: The project aims were to evaluate the benefit of transmucosal midazolam 0.2 mg/kg pre-medication on anxiety, induction behaviour and psychological morbidity in children undergoing general anaesthesia (GA) extractions. METHOD: One hundred and seventy-nine children aged 5-10 years (mean 6.53 years) participated in this randomised, double blind, placebo-controlled trial. Ninety children had midazolam placed in the buccal pouch. Dental anxiety was recorded preoperatively and 48 hours later using a child reported MCDAS-FIS scale. Behaviour at anaesthetic induction was recorded and psychological morbidity was scored by the parent using the Rutter Scale preoperatively and again one week later. Subsequent dental attendance was recorded at one, three and six months after GA. RESULTS: While levels of dental anxiety did not reduce overall, the most anxious patients demonstrated a reduction in anxiety after receiving midazolam premedication (p = 0.01). Neither induction behaviour nor psychological morbidity improved. Irrespective of group, parents reported less hyperactive (p = 0.002) and more pro-social behaviour (p = 0.002) after the procedure; older children improved most (p = 0.048). Post-GA dental attendance was poor and unaffected by premedication. CONCLUSION: 0.2 mg/kg buccal midazolam provided some evidence for reducing anxiety in the most dentally anxious patients. However, induction behaviour, psychological morbidity and subsequent dental attendance were not found to alter.

A randomised placebo-controlled trial of the effects of midazolam premedication on children's postoperative cognition.

This randomised, placebo-controlled study assessed the effects of midazolam premedication on children's postoperative cognition and physical morbidity. In all, 179 children aged 5-10 years were randomly assigned to receive buccal midazolam (0.2 mg x kg(-1)) or placebo before sevoflurane-nitrous oxide anaesthesia for multiple dental extractions. They performed tests of choice reaction time, attention, psychomotor co-ordination and memory pre-operatively (baseline), before discharge and at 48 h. The reaction time of both groups was significantly slower before discharge compared to baseline, with the midazolam group being significantly slower than placebo. Psychomotor co-ordination was also significantly impaired postoperatively after midazolam. Performance on both tests had recovered to baseline by 48 h. Midazolam was also associated with significant anterograde amnesia, both postoperatively and at 48 h, for information presented in the interval between premedication and surgery. The results show significant short-term impairment of children's cognitive function and amnesia enduring for 48 h after low-dose midazolam premedication.

The effects of brief sevoflurane-nitrous oxide anaesthesia upon children's postoperative cognition and behaviour.

This study assessed the effects of brief sevoflurane-nitrous oxide anaesthesia on children's postoperative cognition, behaviour and physical morbidity. Forty-eight children aged 5-10 years undergoing anaesthesia without premedication for multiple dental extractions, and 48 control children, performed tests of choice reaction time, attention, psychomotor co-ordination and memory pre-operatively (baseline), prior to discharge and at 48 h (anaesthesia group only). Physical and psychological morbidity were recorded at 1 week. Mean choice reaction time and psychomotor co-ordination were significantly impaired postoperatively but had recovered at 48 h. However, measures of performance variability suggested the presence of residual impairment. Profound retrograde amnesia affected postoperative and 48-h recall of pictorial stimuli presented prior to anaesthesia, but recognition memory was unimpaired. Attention-seeking, tantrums, crying and nightmares were occurring more frequently in some 8-20% of children 1 week after the procedure.

Echocardiographic characterization of fundamental mechanisms of abnormal diastolic filling in diabetic rats with a parameterized diastolic filling formalism.

Abnormalities of diastolic function (DF) precede systolic dysfunction in diabetic cardiomyopathy. Transmitral Doppler flow analysis is the primary method for noninvasively assessing DF. We used model-based Doppler E-wave analysis to evaluate diastolic function differences between normal and diabetic rat hearts. Control rats and those with diabetes underwent echocardiography with analysis by traditional Doppler indexes and by the parameterized diastolic filling (PDF) formalism, generating 3 parameters, x0, c, and k, that uniquely characterize each E-wave. Significant intergroup differences in the E/A ratios (P <.01), isovolumic relaxation times (P <.01), and the modeling parameter c (P <.05) were found. There were no significant differences in shortening fraction, deceleration time, myocardial collagen content, or the parameters x0 and k between diabetic and control rats. These results indicate that differences in diastolic function may be noninvasively quantified and that diabetic hearts may exhibit defects in uncoupling of the contractile apparatus without concomitant increases in chamber stiffness.

Chamber properties from transmitral flow: prediction of average and passive left ventricular diastolic stiffness.

A chamber stiffness (K(LV))-transmitral flow (E-wave) deceleration time relation has been invasively validated in dogs with the use of average stiffness [(DeltaP/DeltaV)(avg)]. K(LV) is equivalent to k(E), the (E-wave) stiffness of the parameterized diastolic filling model. Prediction and validation of 1) (DeltaP/DeltaV)(avg) in terms of k(E), 2) early rapid-filling stiffness [(DeltaP/DeltaV)(E)] in terms of k(E), and 3) passive (postdiastasis) chamber stiffness [(DeltaP/DeltaV)(PD)] from A waves in terms of the stiffness parameter for the Doppler A wave (k(A)) have not been achieved. Simultaneous micromanometric left ventricular (LV) pressure (LVP) and transmitral flow from 131 subjects were analyzed. (DeltaP)(avg) and (DeltaV)(avg) utilized the minimum LVP-LV end-diastolic pressure interval. (DeltaP/DeltaV)(E) utilized DeltaP and DeltaV from minimum LVP to E-wave termination. (DeltaP/DeltaV)(PD) utilized atrial systolic DeltaP and DeltaV. E- and A-wave analysis generated k(E) and k(A). For all subjects, noninvasive-invasive relations yielded the following equations: k(E) = 1,401. (DeltaP/DeltaV)(avg) + 59.2 (r = 0.84) and k(E) = 229.0. (DeltaP/DeltaV)(E) + 112 (r = 0.80). For subjects with diastasis (n = 113), k(A) = 1,640. (DeltaP/DeltaV)(PD) - 8.40 (r = 0.89). As predicted, k(A) showed excellent correlation with (DeltaP/DeltaV)(PD); k(E) correlated highly with (DeltaP/DeltaV)(avg). In vivo validation of average, early, and passive chamber stiffness facilitates quantitative, noninvasive diastolic function assessment from transmitral flow.

On the use of the variogram in checking for independence in spatial data.

The variogram is a standard tool in the analysis of spatial data, and its shape provides useful information on the form of spatial correlation that may be present. However, it is also useful to be able to assess the evidence for the presence of any spatial correlation. A method of doing this, based on an assessment of whether the true function underlying the variogram is constant, is proposed. Nonparametric smoothing of the squared differences of the observed variables, on a suitably transformed scale, is used to estimate variogram shape. A statistic based on a ratio of quadratic forms is proposed and the test is constructed by investigating the distributional properties of this statistic under the assumption of an independent Gaussian process. The power of the test is investigated. Reference bands are proposed as a graphical follow-up. An example is discussed.

Graphical exploration of covariate effects on survival data through nonparametric quantile curves.

Kaplan-Meier curves provide an effective means of presenting the distributional pattern in a sample of survival data. However, in order to assess the effect of a covariate, a standard scatterplot is often difficult to interpret because of the presence of censored observations. Several authors have proposed a running median as an effective way of indicating the effect of a covariate. This article proposes a form of kernel estimation, employing double smoothing, that can be applied in a simple and efficient manner to construct an estimator of a percentile of the survival distribution as a function of one or two covariates. Permutations and bootstrap samples can be used to construct reference bands that help identify whether particular features of the estimates indicate real features of the underlying curve or whether this may be due simply to random variation. The techniques are illustrated on data from a study of kidney transplant patients.

Early graft function and patient survival following cadaveric renal transplantation.

BACKGROUND: The influence of events that occur early following renal transplantation such as delayed graft function (DGF) and acute rejection on long-term graft survival has been widely reported, but its association with patient survival has received less attention. METHODS: We studied 589 patients who received their first cadaveric transplants between 1984 and 1993, all of whom received cyclosporine-based immunosuppression and who had a median follow-up of seven years. The following factors were identified, and both univariate and multivariate analyses were used to determine their association with long-term patient and graft survival: age, sex, duration of pretransplant dialysis, primary renal disease, immediate graft function (IGF), DGF, primary nonfunction (PNF), acute rejection, and serum creatinine at 3, 6, and 12 months. RESULTS: Patients with PNF had a poorer survival than those with DGF and IGF (P = 0.01), but there was no difference in survival between DGF and IGF (P = 0.54). Good graft function (serum creatinine of less than 200 mumol/liter) at three months was predictive of better long-term patient survival (P = 0.03). Other factors associated with poor patient outcome were older age, diabetes, adult polycystic kidney disease, male gender, and acute rejection. Cardiovascular disease was the most common cause of death (51.8%). Good graft function at three months (P < 0.001) and an absence of rejection episodes (P = 0.01) were associated with better graft survival. CONCLUSION: Patients with poor levels of early graft function (but not DGF) and those with either acute rejection episodes or early graft loss are at an increased risk of early death. These high-risk groups should be targeted for interventional studies in an attempt to improve patient survival.

An ultrasound study of gestational and postural changes in the deep venous system of the leg in pregnancy.

OBJECTIVE: To investigate gestational and postural changes in diameter and blood flow in the proximal deep leg veins during pregnancy. DESIGN: A longitudinal, prospective observational study. SETTING: The ultrasound department of a teaching maternity hospital. POPULATION: Twenty-four healthy women with uncomplicated singleton pregnancies. METHODS: Real-time and duplex Doppler ultrasound assessments of the vessel diameter, flow velocity and respiratory flow fluctuation in the proximal deep leg veins of women serially measured from the first trimester of pregnancy to six weeks postnatally. MAIN OUTCOME MEASURES: The effects of increasing gestation and the adoption of the left lateral position on the above parameters. RESULTS: An increase in vessel diameter and a fall in flow velocity with increasing gestation was observed. However, no change in venous flow variation was observed. Delivery had reverse effects. Flow velocity was slower in the left than right legs, but on adoption of the left lateral position an increase in flow velocity and venous flow variation was observed in both legs during pregnancy. CONCLUSIONS: These data are consistent with the observed increase in incidence and pattern of deep venous thrombosis in pregnancy and may aid interpretation of duplex Doppler ultrasound examinations for deep venous thrombosis in pregnancy. Postural changes should be part of this evaluation. The gravid uterus may not be the sole cause for postural changes in deep venous flow velocity.

Intrauterine growth restriction with absent end-diastolic flow velocity in the umbilical artery is associated with maldevelopment of the placental terminal villous tree.

OBJECTIVE: Our purpose was to evaluate the structure of placental terminal villi and their capillaries in pregnancies complicated by intrauterine growth restriction with absent end-diastolic flow velocity in the umbilical artery. STUDY DESIGN: Glutaraldehyde-perfusion-fixed villous tissue and a plastic cast of the vessels in at least two cotyledons were prepared from 10 cases with intrauterine growth restriction and 9 gestational age-matched control placentas. The structure and dimensions of 20 terminal capillary loops per cast were determined by scanning electron microscopic examination, and their appearances were correlated with the peripheral villi of the perfusion-fixed villous tissue. RESULTS: Capillary loops in the growth-restricted cases were sparse in number and significantly longer than in the control cases (218 microns [72] vs 137 microns [30], mean and SD, p < 0.05). They exhibited fewer branches (4.0 [1.9] per loop vs 6.1 [2.2], p < 0.05) and a majority of loops were uncoiled (79% vs 18%, p < 0.05). The villous tissues from the growth-restricted cases demonstrated elongated villi, consistent with the cast findings. The trophoblast surface was wrinkled and in some areas covered by fibrin plaques. CONCLUSIONS: The terminal villous compartment of the placenta appears to be maldeveloped in preterm intrauterine growth restriction pregnancies where absent end-diastolic flow velocity is demonstrated in the umbilical artery before delivery. These findings are consistent with an increase in fetoplacental vascular impedance at the capillary level and may account for the impaired gas and nutrient transfer in this disorder.

Elaboration of stem villous vessels in growth restricted pregnancies with abnormal umbilical artery Doppler waveforms.

OBJECTIVE: To assess the elaboration of placental stem villous vessels from pregnancies complicated by intrauterine growth restriction (IUGR) with absent end-diastolic flow velocity detected prior to delivery in the umbilical artery. DESIGN: Comparison between IUGR and control groups of the distribution, in 15 microns increments of 600 randomly chosen stem vessel profiles (post-fixation diameter 10-160 microns) identified by immunohistochemical localisation of alpha-smooth muscle actin in the vessel media. SETTING: Clinical teaching hospital and university anatomy department. SUBJECTS: Paraffin-fixed blocks obtained from placentas of eight pregnancies complicated by IUGR and eight gestational age-matched controls. RESULTS: The distribution of the stem villous vessels in the IUGR placentas, as assessed by the mean vessel diameter in each case, did not differ from the controls (mean vessel diameter 31.8 microns [SD 2.4] vs 29.6 microns [2.3]; P = 0.13). In five IUGR cases alpha-smooth muscle actin positive cells (myofibroblasts) were identified within the stroma of nonmuscularised peripheral (mature intermediate and terminal) villi, but in none of the controls. CONCLUSIONS: Our data do not support the theory that IUGR with absent end-diastolic flow velocity in the umbilical artery is due to a selective loss of small stem villous vessels. The increased impedance in this condition may be conferred more distally within the nonmuscularised capillaries of the peripheral villi.

Atrial natriuretic peptide: a vasodilator of the fetoplacental circulation?

Paired maternal and fetal atrial natriuretic peptide concentrations were measured in 62 percutaneous umbilical blood samplings performed principally for the assessment and treatment of rhesus isoimmunization. Pretransfusion fetal atrial natriuretic peptide levels were significantly higher than maternal atrial natriuretic peptide levels (median 117 pg/ml vs median 32 pg/ml; p less than 0.001); paired pretransfusion fetal and maternal atrial natriuretic peptide samples showed a weak correlation with each other (R2 = 17%; p = 0.002). Fetal atrial natriuretic peptide levels correlated inversely with hematocrit (R2 = 14%; p = 0.003), but not with albumin or gestational age. Paired pretransfusion and posttransfusion (median = 134 pg/ml) fetal atrial natriuretic peptide levels (n = 38) showed a significant rise after transfusion (p less than 0.001); this rise was related to the percentage of fetoplacental blood volume transfused (R2 = 33%; p = 0.035). In a subgroup of 26 procedures, change in fetal atrial natriuretic peptide levels was weakly correlated with transient reductions in the Doppler systolic/diastolic ratio of the umbilical artery (R2 = 14%; p = 0.07). These data support work in animals that indicate a role for atrial natriuretic peptide in the human fetus, but these data do not confirm that atrial natriuretic peptide modulates fetoplacental vascular impedance in the human fetus.

Studies on the concentration of pregnancy-associated plasma protein-A during normal and complicated pregnancy.

Pregnancy-associated plasma protein-A was assayed in the blood of 347 women during pregnancy, using a new primary standard of PAPP-A as reference. The protein was assayed by antibody-antigen crossed electrophoresis with the lower limit of confident assay being 9.5 micrograms PAPP-A/ml (13 pmol/ml). PAPP-A was first detected at 14 weeks of gestation; by term it had risen to within the range 20 to 320 micrograms/ml. There was an indication that pregnancies involving a male baby had higher PAPP-A levels in blood than did those involving female babies. In 51 blood samples from 30 patients with gestational diabetes (taken between 28 weeks of pregnancy and term) there was no significant alteration in PAPP-A values compared with controls. In 35 blood samples from 15 patients with insulin-dependent diabetes, levels of PAPP-A were significantly lower than in controls or in gestational diabetes. In 43 blood samples from 35 patients with babies affected with intrauterine growth retardation (between 28 weeks and term), there was no significant difference in PAPP-A levels compared with controls. The effect of insulin on the blood levels of PAPP-A suggests that the concentration of PAPP-A is capable of altering significantly in response to certain physiological changes associated with the control of carbohydrate metabolism.