Surrogate Estimates Research Articles

24 Measuring the value of endoscopic retrograde cholangiopancreatography (ERCP) activity: an opportunity to stratify endoscopists based on their value

Background The value in healthcare can be defined as patient health outcomes achieved per monetary unit spent. Aim To characterise the value of ERCP performance of four gastroenterologists. Method Medical records of patients undergoing ERCP between September 2014 and September 2016 in an academic medical centre were reviewed; all procedures were performed by one of four experienced gastroenterologists, all of whom have performed at least 1000 ERCPs. Procedure value was defined as the quality of procedure (Q) divided by the duration of procedure (T) adjusted for complexity level (C), that is, Q/(T/C). In those patients undergoing multiple ERCPs during the study period, only the index procedure was considered for analysis. ERCP quality and complexity were both graded on a 1 to 4 Likert scale based on American Society for Gastrointestinal Endoscopy (ASGE) criteria; time was recorded (in minutes) from intubation to extubation. Although individual components of procedure cost (eg, ERCP accessories, patient sedation, etc.) were not itemised, the procedure duration (ie, the cost of endoscopist’s time) was considered to be a reliable surrogate estimate of overall procedure cost. Results In total, index procedures on 465 patients were performed over 24 months; mean of 116 index ERCPs per endoscopist. Mean quality varied from 2.25 to 2.53 while adjusted mean duration (T/C) varied from 22.13 to 28.66 min per procedure. Value measurements varied from 8.1 to 10.7. Conclusions There was a 32% variation in the value of endoscopist activity. As healthcare costs are scrutinised more closely, such value measurements are likely to become more relevant.

Resistance of Australian sugarcane clones to moth and weevil borers in Papua New Guinea

Papua New Guinea (PNG) has several moth borers (Lepidoptera) that damage commercial crops of sugarcane in the Ramu Valley. These borers are not present in Australia but would almost certainly cause very serious losses if introduced. To prepare for a possible incursion, Australian cultivars and promising clones were tested for their susceptibility to these borers in field trials at Ramu Agri-Industries, PNG. Three field trials were planted, one in each of 2010 (32 clones), 2011 (33 clones) and 2013 (32 clones), and sampled for borers in the plant and first ratoon crops. There were significant differences among clones for internode damage from Sesamia grisescens Warren (Noctuidae) and Chilo terrenellus Pagenstecher (Crambidae) and meristem damage from Scirpophaga excerptalis (Walker) (Pyralidae), as well as internode damage from the weevil borer Rhabdoscelus obscurus (Boisduval) (Coleoptera: Curculionidae). Generally, the range of damage among clones in each trial was less for C. terrenellus than for the other borers. The proportion of damaged internodes was strongly correlated with the proportion of damaged stalks for all borers and so the proportion of damaged stalks could be used as a surrogate estimate of internal stalk damage for rapid sampling. Levels of damage were correlated, positively or negatively, between species of borers in some trials. Damage from R. obscurus was positively correlated with damage from the two stalk borers, S. grisescens and C. terrenellus, which agrees with its status as a mainly secondary pest; its damage was not correlated with damage from Sc. excerptalis which affects only the stalk tops and the apical meristem. Damage correlations among the moth borers could reflect similarities or differences between resistance profiles of clones against different species, or could be due to direct or indirect interactions among species. Thirteen standard clones were included in all three trials. Among data sets, results for the standard clones were least consistent for C. terrenellus and most consistent for Sc. excerptalis. Some evidence was obtained that antixenosis may play a role in influencing resistance of clones to Sc. excerptalis and, perhaps, S. grisescens, but more research would be needed to confirm this.

Assessing retinal ganglion cell damage.

Retinal ganglion cell (RGC) loss is the hallmark of optic neuropathies, including glaucoma, where damage to RGC axons occurs at the level of the optic nerve head. In experimental glaucoma, damage is assessed at the axon level (in the retinal nerve fibre layer and optic nerve head) or at the soma level (in the retina). In clinical glaucoma where measurements are generally limited to non-invasive techniques, structural measurements of the retinal nerve fibre layer and optic nerve head, or functional measurements with perimetry provide surrogate estimates of RGC integrity. These surrogate measurements, while clinically useful, are several levels removed from estimating actual RGC loss. Advances in imaging, labelling techniques, and transgenic medicine are making enormous strides in experimental glaucoma, providing knowledge on the pathophysiology of glaucoma, its progression and testing new therapeutic avenues. Advances are also being made in functional imaging of RGCs. Future efforts will now be directed towards translating these advances to clinical care.

Chronic kidney disease and obesity bias surrogate estimates of insulin sensitivity compared with the hyperinsulinemic euglycemic clamp

Insulin sensitivity can be measured by procedures such as the hyperinsulinemic euglycemic clamp or by using surrogate indices. Chronic kidney disease (CKD) and obesity may differentially affect these measurements because of changes in insulin kinetics and organ-specific effects on insulin sensitivity. In a cross-sectional study of 59 subjects with nondiabetic CKD [estimated glomerular filtration rate: (GFR) <60 ml·min-1·1.73 m2] and 39 matched healthy controls, we quantified insulin sensitivity by clamp (SIclamp), oral glucose tolerance test, and fasting glucose and insulin. We compared surrogate insulin sensitivity indices to SIclamp using descriptive statistics, graphical analyses, correlation coefficients, and linear regression. Mean age was 62.6 yr; 48% of the participants were female, and 77% were Caucasian. Insulin sensitivity indices were 8-38% lower in participants with vs. without CKD and 13-59% lower in obese compared with nonobese participants. Correlations of surrogate indices with SIclamp did not differ significantly by CKD or obesity status. Adjusting for SIclamp in addition to demographic factors, Matsuda index was 15% lower in participants with vs. without CKD (P = 0.09) and 36% lower in participants with vs. without obesity (P = 0.0001), whereas 1/HOMA-IR was 23% lower in participants with vs. without CKD (P = 0.02) and 46% lower in participants with vs. without obesity (P < 0.0001). We conclude that CKD and obesity do not significantly alter correlations of surrogate insulin sensitivity indices with SIclamp, but they do bias surrogate measurements of insulin sensitivity toward lower values. This bias may be due to differences in insulin kinetics or organ-specific responses to insulin.

Comparison of two surrogate estimates of insulin resistance to predict cardiovascular disease in apparently healthy individuals

Background and aimsInsulin resistance is associated with a cluster of abnormalities that increase cardiovascular disease (CVD). Several indices have been proposed to identify individuals who are insulin resistant, and thereby at increased CVD risk. The aim of this study was to compare the abilities of 3 indices to accomplish that goal: 1) plasma triglyceride × glucose index (TG × G); 2) plasma triglyceride/high-density lipoprotein cholesterol ratio (TG/HDL-C); and 3) Metabolic Syndrome (MetS). Methods and resultsIn a population sample of 723 individuals (486 women and 237 men, 50 ± 16 and 51 ± 16 years old, respectively), baseline demographic and metabolic variables known to increase CVD risk and incident CVD were compared among individuals defined as high vs. low risk by: TG × G; TG/HDL-C; or MetS. CVD risk profiles appeared comparable in high risk subjects, irrespective of criteria. Crude incidence of CVD events was increased in high risk subjects: 12.2 vs. 5.3% subjects/10 years, p = 0.005 defined by TG/HDL-C; 13.4 vs. 5.3% subjects/10 years, p = 0.002 defined by TG × G; and 13.4% vs. 4.5% of subjects/10 years, p < 0.001 in subjects with the MetS. The area under the ROC curves to predict CVD were similar, 0.66 vs. 0.67 for TG/HDL-C and TG × G, respectively. However, when adjusted by age, sex and multiple covariates, hazard ratios for incident CVD were significantly increased in high risk patients classified by either TG/HDL-C ratio (2.18, p = 0.021) or MetS (1.93, p = 0.037), but not by TG × G index (1.72, p = 0.087). ConclusionAlthough the 3 indices identify CVD risk comparably, the TG × G index seems somewhat less effective at predicting CVD.

Nonparametric Quantile Estimation Based on Surrogate Models

Nonparametric estimation of a quantile $q_{m(X),\alpha }$ of a random variable $m(X)$ is considered, where $m: \mathbb {R}^{d}\rightarrow \mathbb {R}$ is a function, which is costly to compute and $X$ is an $ \mathbb {R}^{d}$ -valued random variable with known distribution. Monte Carlo surrogate quantile estimates are considered, where in a first step, the function $m$ is estimated by some estimate (surrogate) $m_{n}$ , and then, the quantile $q_{m(X),\alpha }$ is estimated by a Monte Carlo estimate of the quantile $q_{m_{n}(X),\alpha }$ . A general error bound on the error of this quantile estimate is derived, which depends on the local error of the function estimate $m_{n}$ , and the rates of convergence of the corresponding Monte Carlo surrogate quantile estimates are analyzed for two different function estimates. The finite sample size behavior of the estimates is investigated in simulations.

Lower motor neuron dysfunction in ALS.

In the motor system there is a complex interplay between cortical structures and spinal cord lower motor neurons (LMN). In this system both inhibitory and excitatory neurons have relevant roles. LMN loss is a marker of motor neuron disease/amyotrophic lateral sclerosis (MND/ALS). Conventional needle electromyography (EMG) does not allow LMN loss to be quantified. Measurement of compound muscle action potential (CMAP) amplitude or area, and the neurophysiological index, provide a surrogate estimate of the number of functional motor units. Increased motor neuronal excitability is a neurophysiological marker of ALS in the context of a suspected clinical and electrophysiological diagnosis. In the LMN system, fasciculation potentials (FPs) are the earliest changes observed in affected muscles, a feature of LMN hyperexcitability. Reinnervation is best investigated by needle EMG although other methods can be explored. Moreover needle EMG give information about the temporal profile of the reinnervation process, important ancillary data. Quantitative motor unit potential analysis is a valuable method of evaluating reinnervation. The importance of FPs has been recognized in the Awaji criteria for the electrodiagnosis of ALS, criteria that are a sensitive adjunct to the revised El Escorial criteria. Finally, functionality of LMN's, and perhaps excitability studies in motor nerves, aids understanding of the disease process, allowing measurement of potential treatment effects in clinical trials. Other investigational techniques, such as electrical impedance myography, muscle and nerve ultrasound, and spinal cord imaging methods may prove useful in future.

Metabolic markers associated with insulin resistance predict type 2 diabetes in Koreans with normal blood pressure or prehypertension.

BackgroundQuestions remain as to the association between essential hypertension and increased incidence of type 2 diabetes (T2DM). The premise of this analysis is that insulin resistance/compensatory hyperinsulinemia is a major predictor of T2DM, and the greater the prevalence of insulin resistance within any population, normotensive or hypertensive, the more likely T2DM will develop. The hypothesis to be tested is that surrogate estimates of insulin resistance will predict incident T2DM to a significant degree in persons with normal blood pressure or prehypertension.MethodsAnalysis of data from a population-based survey of 10, 038 inhabitants of rural and urban areas of Korea, ≥40 years-old, initiated in 2001, with measures of demographic and metabolic characteristics at baseline and 8-years later. Participants were classified as having normal blood pressure or prehypertension, and three simple manifestations of insulin resistance related to the pathophysiology of T2DM used to predict incident T2DM: (1) glycemia (plasma glucose concentration 2-hour after 75 g oral glucose challenge = 2-hour PG); (2) hyperinsulinemia (plasma insulin concentration 2-hour after 75 g oral glucose challenge = 2-hour PI); and (3) dyslipidemia (ratio of fasting plasma triglyceride/high/density lipoprotein cholesterol concentration = TG/HDL-C ratio).ResultsFully adjusted hazard ratios (HR, 95 % CI) for incident T2DM were highest (P < 0.001) in the quartile of individuals with the highest 2-hour PG concentrations, ranging from 5.84 (3.37–10.1) in women with prehypertension to 12.2 (7.12–21.00) in men with normal blood pressure. T2DM also developed to a significantly greater degree in subjects within the highest quartile of TG/HDL-C ratios, with HRs varying from 2.91 (1.63–2.58) in women with prehypertension (P < 0.001) to 1.77 (1.12–2.81, P < 0.05) in men with prehypertension. The least predictive index of insulin resistance was the 2-hour PI concentration. Subjects with normal blood pressure in the highest quartile of 2-hour PI concentrations were significantly associated with incident T2DM, with HRs of 1.5 (1.02–2.20, P = 0.25) and 2.02 (1.35–3.02, P < 0.001), in men and women, respectively. Finally, incidence of T2DM in the highest quartile was somewhat greater in patients with prehypertension, irrespective of predictor.ConclusionsMetabolic variables associated with insulin resistance (glycemia, insulinemia, and dyslipidemia) predict the development of T2DM in patients with either normal blood pressure or prehypertension.Electronic supplementary materialThe online version of this article (doi:10.1186/s12933-016-0368-7) contains supplementary material, which is available to authorized users.

Utilizing global data to estimate analytical performance on the Sigma scale: A global comparative analysis of methods, instruments, and manufacturers through external quality assurance and proficiency testing programs

ObjectiveTo assess the analytical performance of instruments and methods through external quality assessment and proficiency testing data on the Sigma scale. Design and methodsA representative report from five different EQA/PT programs around the world (2 US, 1 Canadian, 1 UK, and 1 Australasian) was accessed. The instrument group standard deviations were used as surrogate estimates of instrument imprecision. Performance specifications from the US CLIA proficiency testing criteria were used to establish a common quality goal. Then Sigma-metrics were calculated to grade the analytical performance. ResultsDifferent methods have different Sigma-metrics for each analyte reviewed. Summary Sigma-metrics estimate the percentage of the chemistry analytes that are expected to perform above Five Sigma, which is where optimized QC design can be implemented. The range of performance varies from 37% to 88%, exhibiting significant differentiation between instruments and manufacturers. Median Sigmas for the different manufacturers in three analytes (albumin, glucose, sodium) showed significant differentiation. ConclusionsChemistry tests are not commodities. Quality varies significantly from manufacturer to manufacturer, instrument to instrument, and method to method. The Sigma-assessments from multiple EQA/PT programs provide more insight into the performance of methods and instruments than any single program by itself. It is possible to produce a ranking of performance by manufacturer, instrument and individual method. Laboratories seeking optimal instrumentation would do well to consult this data as part of their decision-making process. To confirm that these assessments are stable and reliable, a longer term study should be conducted that examines more results over a longer time period.

Relationship between insulin sensitivity and insulin secretion rate: not necessarily hyperbolic.

There is general acceptance that the physiological relationship between insulin sensitivity and insulin secretion is hyperbolic. This conclusion has evolved from studies in which one test assessed both variables, and changes in plasma insulin concentration were used as a surrogate measure for insulin secretion rate. The aim of this study was to see if a hyperbolic relationship would also emerge when separate and direct measures were used to quantify both insulin sensitivity and insulin secretion rate. Steady-state plasma glucose (SSPG) was determined in 146 individuals without diabetes using the insulin suppression test, with 1/SSPG used to quantify insulin sensitivity. The graded-glucose infusion test was used to quantify insulin secretion rate. Plasma glucose and insulin concentrations obtained during an oral glucose tolerance test (OGTT) were used to calculate surrogate estimates of insulin action and insulin secretion rate. A hyperbolic relationship was assumed if the β coefficient was near -1 using the following model: log (insulin secretion measure) = constant + β × log (insulin sensitivity measure). OGTT calculations of insulin sensitivity (Matsuda) and plasma insulin response [ratio of insulin/glucose area-under-the-curve (AUC) or insulin total AUC] provided the expected hyperbolic relationship [β = -0.95, 95% CI (-1.09, -0.82); -1.06 (-1.14, -0.98)]. Direct measurements of insulin sensitivity and insulin secretion rate did not yield the same curvilinear relationship [β = -1.97 (-3.19, -1.36)]. These findings demonstrate that the physiological relationship between insulin sensitivity and insulin secretion rate is not necessarily hyperbolic, but will vary with the method(s) by which it is determined.

The Atherogenic Dyslipidemia Ratio Log (Tg)/Hdl-C Was Not Associated with Urinary Albumin Excretion Rate (Uaer) and Increased Cardiovascular Risk in Black Patients with Type 2 Diabetes

Objective: The objective is to assess the relationship of Log (TG)/HDL-c as surrogate estimate of atherogenic dyslipidemia with urinary albumin excretion rate and cardiovascular risk among black patients with type 2 diabetes. Patients and methods: A post-hoc analysis of data from 181 type 2 diabetes patients enrolled in a cross-sectional study of urinary albumin excretion rate seen at a tertiary healthcare. Microalbuminuria and macroalbuminuria were defined as ACR 30 - 299.9 mg/g and ACR ≥ 300 mg/g, respectively. Quartiles of Log (TG)/HDL-c were used as surrogate estimates of atherogenic dyslipidemia. Cardiovascular risk was assessed using WHO chart for estimation of CV risk in low and middle income countries. Comparisons across Log (TG)/HDL-c quartiles were performed using one way ANOVA and Chi square for trend as appropriate. P < 0.05 defined the level of statistical significance. Results: A high prevalence (69%) of atherogenic dyslipidemia (AD) was observed in the present case series of Black Africans with type 2 diabetes. Average total cholesterol levels showed significant (p = 0.010) trends towards lower values across quartiles of Log (TG)/ HDL-c. No significant trends were observed for average UAER and cardiovascular risk across quartiles of Log (TG)/HDL-c. Conclusion: Log (TG)/HDL-c as a surrogate estimate of atherogenic failed to predict cardiovascular risk in the present case series of black patients with type 2 diabetes.

A comparative study of expected improvement-assisted global optimization with different surrogates

Efficient global optimization (EGO) uses the surrogate uncertainty estimator called expected improvement (EI) to guide the selection of the next sampling candidates. Theoretically, any modelling methods can be integrated with the EI criterion. To improve the convergence ratio, a multi-surrogate efficient global optimization (MSEGO) was suggested. In practice, the EI-based optimization methods with different surrogates show widely divergent characteristics. Therefore, it is important to choose the most suitable algorithm for a certain problem. For this purpose, four single-surrogate efficient global optimizations (SSEGOs) and an MSEGO involving four surrogates are investigated. According to numerical tests, both the SSEGOs and the MSEGO are feasible for weak nonlinear problems. However, they are not robust for strong nonlinear problems, especially for multimodal and high-dimensional problems. Moreover, to investigate the feasibility of EGO in practice, a material identification benchmark is designed to demonstrate the performance of EGO methods. According to the tests in this study, the kriging EGO is generally the most robust method.

The effect of framing on surrogate optimism bias: A simulation study

PurposeTo explore the effect of emotion priming and physician communication behaviors on optimism bias. Materials and MethodsWe conducted a 5 × 2 between-subject randomized factorial experiment using a Web-based interactive video designed to simulate a family meeting for a critically ill spouse/parent. Eligibility included age at least 35 years and self-identifying as the surrogate for a spouse/parent. The primary outcome was the surrogate's election of code status. We defined optimism bias as the surrogate's estimate of prognosis with cardiopulmonary resuscitation (CPR) > their recollection of the physician's estimate. ResultsOf 373 respondents, 256 (69%) logged in and were randomized and 220 (86%) had nonmissing data for prognosis. Sixty-seven (30%) of 220 overall and 56 of (32%) 173 with an accurate recollection of the physician's estimate had optimism bias. Optimism bias correlated with choosing CPR (P < .001). Emotion priming (P = .397), physician attention to emotion (P = .537), and framing of CPR as the social norm (P = .884) did not affect optimism bias. Framing the decision as the patient's vs the surrogate's (25% vs 36%, P = .066) and describing the alternative to CPR as “allow natural death” instead of “do not resuscitate” (25% vs 37%, P = .035) decreased optimism bias. ConclusionsFraming of CPR choice during code status conversations may influence surrogates' optimism bias.

Global optimization of general constrained grey-box models: new method and its application to constrained PDEs for pressure swing adsorption

This paper introduces a novel methodology for the global optimization of general constrained grey-box problems. A grey-box problem may contain a combination of black-box constraints and constraints with a known functional form. The novel features of this work include (i) the selection of initial samples through a subset selection optimization problem from a large number of faster low-fidelity model samples (when a low-fidelity model is available), (ii) the exploration of a diverse set of interpolating and non-interpolating functional forms for representing the objective function and each of the constraints, (iii) the global optimization of the parameter estimation of surrogate functions and the global optimization of the constrained grey-box formulation, and (iv) the updating of variable bounds based on a clustering technique. The performance of the algorithm is presented for a set of case studies representing an expensive non-linear algebraic partial differential equation simulation of a pressure swing adsorption system for $$\hbox {CO}_{2}$$CO2. We address three significant sources of variability and their effects on the consistency and reliability of the algorithm: (i) the initial sampling variability, (ii) the type of surrogate function, and (iii) global versus local optimization of the surrogate function parameter estimation and overall surrogate constrained grey-box problem. It is shown that globally optimizing the parameters in the parameter estimation model, and globally optimizing the constrained grey-box formulation has a significant impact on the performance. The effect of sampling variability is mitigated by a two-stage sampling approach which exploits information from reduced-order models. Finally, the proposed global optimization approach is compared to existing constrained derivative-free optimization algorithms.

Differences in surrogate threshold effect estimates between original and simplified correlation-based validation approaches.

Surrogate endpoint validation has been well established by the meta-analytical correlation-based approach as outlined in the seminal work of Buyse et al. (Biostatistics, 2000). Surrogacy can be assumed if strong associations on individual and study levels can be demonstrated. Alternatively, if an effect on a true endpoint is to be predicted from a surrogate endpoint in a new study, the surrogate threshold effect (STE, Burzykowski and Buyse, Pharmaceutical Statistics, 2006) can be used. In practice, as individual patient data (IPD) are hard to obtain, some authors use only aggregate data and perform simplified regression analyses. We are interested in to what extent such simplified analyses are biased compared with the ones from a full model with IPD. To this end, we conduct a simulation study with IPD and compute STEs from full and simplified analyses for varying data situations in terms of number of studies, correlations, variances and so on. In the scenarios considered, we show that, for normally distributed patient data, STEs derived from ordinary (weighted) linear regression generally underestimate STEs derived from the original model, whereas meta-regression often results in overestimation. Therefore, if individual data cannot be obtained, STEs from meta-regression may be used as conservative alternatives, but ordinary (weighted) linear regression should not be used for surrogate endpoint validation.

Blood pressure monitoring during exercise: Comparison of pulse transit time and volume clamp methods

During physical exercise, pulse transit time (PTT), expressed as the interval between ventricular electrical activity and peripheral pulse wave, may provide a surrogate estimate for blood pressure by the use of specific calibration procedures. The objective of this study was to determine systolic blood pressure (SBP) values derived from the PTT method and from an established method of non-invasive continuous blood pressure measurement based on the volume clamp technique, and to compare their agreement with sphygmomanometry during exercise tests. In 18 subjects, electrocardiogram (ECG) and finger-photoplethysmography were continuously recorded during maximal cycle exercise tests. Intermittent and continuous blood pressure measurements were simultaneously taken using automated sphygmomanometry and a Portapres Model-2 device, respectively. PTT was calculated for each ECG R-wave and the corresponding steepest upstroke slope in the photoplethysmogram, and was transformed to a continuous blood pressure estimate using multipoint nonlinear regression calibration based on the individual subject's sphygmomanometer readings. Bland–Altman limits of agreement between PTT-derived SBP estimates and sphygmomanometer values were –24.7 to 24.1 mmHg, and between Portapres and sphygmomanometer SBP values were –42.0 to 70.1 mmHg. For beat-to-beat SBP estimation during exercise, PTT measurement combined with multipoint nonlinear regression calibration based on intermittent sphygmomanometry may be an alternative to volume clamp devices.

Quality of Communication in the ICU and Surrogate’s Understanding of Prognosis*

Although misperceptions about prognosis by surrogates in ICUs are common and influence treatment decisions, there is no validated, practical way to measure the effectiveness of prognostic communication. Surrogates' subjective ratings of quality of communication have been used in other domains as markers of effectiveness of communication. We sought to determine whether surrogates' subjective ratings of the quality of prognostic communication predict accurate expectation about prognosis by surrogates. We performed a cross-sectional cohort study. Surrogates rated the quality of prognostic communication by survey. Physicians and surrogates gave their percentage estimate of patient survival on ICU day 3 on a 0-100 probability scale. We defined discordance about prognosis as a difference in the physician's and surrogate's estimates of greater than or equal to ±20%. We used multilevel logistic regression modeling to account for clustering under physicians and patients and adjust for confounders. Medical-surgical, trauma, cardiac, and neurologic ICUs of five U.S. academic medical centers located in California, Pennsylvania, Washington, North Carolina, and Massachusetts. Two hundred seventy-five patients with acute respiratory distress syndrome at high risk of death or severe functional impairment, their 546 surrogate decision makers, and their 150 physicians. None. There was no predictive utility of surrogates' ratings of the quality of communication about prognosis to identify inaccurate expectations about prognosis (odds ratio, 1.04 ± 0.07; p = 0.54). Surrogates' subjective ratings of the quality of communication about prognosis were high, as assessed with a variety of questions. Discordant prognostic estimates were present in 63.5% (95% CI, 59.0-67.9) of physician-surrogate pairs. Although most surrogates rate the quality of prognostic communication high, inaccurate expectations about prognosis are common among surrogates. Surrogates' ratings of the quality of prognostic communication do not reliably predict an accurate expectation about prognosis.

Soil total nitrogen concentration explains variation in pasture response to spring nitrogen fertiliser across a single farm

Soil total N (TN) has been shown to be a useful surrogate measure of soil N supply in pasture soils. Since soil N supply can modify fertilizer N response, we hypothesized that TN would be a good predictor of the relative pasture dry matter (DM) response to applied fertilizer N across a single property. We chose a single property to exclude confounding effects from large variation in weather. Nitrogen response experiments were placed in 20 paddocks on a single farm to test this. Each experiment comprised a single application in spring of 25, 50, 100 or 200 kg N/ha (plus a nil-N control). Range of soil TN (0–75 mm) was 0.3–1.3 %. The full response to applied N was captured in the first two pasture harvests (49 days), even at the highest N rate. Nil-N yield increased with increasing TN, but the maximum yield with 200 kg N/ha was constant across sites. Consequently, the response to fertilizer N (difference between yield at 200 N and nil-N) decreased with increasing soil TN concentration, with TN explaining c. 53 % of the variation in fertilizer N response (P < 0.001); this increased to 72 % when all N rates were included. The experiment suggests that use of TN, as a surrogate estimate of soil N supply, can improve the prediction of pasture DM response to applied fertilizer N in spring to a ryegrass/clover sward. Further work is required to test this in other seasons and at other sites.

Airway Pressures as Surrogate Estimates of Intra-abdominal Pressure

Intra-abdominal pressure (IAP) measurements are essential to the diagnosis and management of patients with intra-abdominal hypertension (IAH) and abdominal compartment syndrome (ACS). Peak inspiratory pressure (PIP), plateau pressure (Pplat), and mean airway pressure (Paw) are used by some surgeons as surrogate estimates of IAP during abdominal closure. Thirty mechanically ventilated surgical/trauma patients with risk factors for IAH/ACS underwent simultaneous triplicate measurements of PIP, Pplat, Paw, and IAP. PIP, Pplat, and Paw were compared with IAP using both coefficient of determination and Bland and Altman analysis. The coefficient of determination for each airway pressure in predicting change in IAP was: PIP 5 per cent (P = 0.24), Pplat 17 per cent (P = 0.02), and Paw 15 per cent (P = 0.03). Bland and Altman analysis identified that marked variability exists between airway pressure and IAP measurements: PIP 19.3 ± 18.7 mmHg, Pplat 11.1 ± 13.7 mmHg, and Paw 2.0 ± 9.8 mmHg. Airway pressures do not accurately reflect IAP and cannot be substituted for IAP measurements in patients at risk for IAH/ACS.

A field application of a personal sensor for ultrafine particle exposure in children

BackgroundUltrafine particles (UFPs) have been associated with adverse health outcomes in children, but studies are often limited by surrogate estimates of exposure. Accurately characterizing children's personal exposure to UFP is difficult due to the high spatiotemporal variability of UFP and children's time–activity patterns. ObjectiveThe objectives of this study were to conduct a field test of a personal sensor for UFP (PUFP) by measuring UFP exposure among children and assess the sensor's capabilities and limitations. MethodsChildren wore the sensor at school, during transit periods between school and home, and in their home for 2–4h on 2 consecutive days and provided feedback regarding their experience with the sensor. The PUFP sensor recorded UFP number concentration at one second intervals and recorded GPS location allowing for comparisons of UFP exposure at homes, schools, and during transit. A mixed-effects linear model was used to compare the effect of microenvironment on personal UFP measurements. ResultsThe overall total median personal exposure to UFP was 12,900particles/cm3 (p/cm3). Median UFP exposure at homes, schools and during transit was 17,800, 11,900, and 13,600p/cm3, respectively. Results of the mixed-effects model found that riding in a car and walking were significantly associated with 1.36 (95% CI 1.33–1.39) and 2.51 (95% CI 2.44–2.57) times higher UFP concentrations compared to the home. ConclusionsThe PUFP sensor can measure near real-time exposure to UFP with high spatiotemporal resolution. Children's exposure to UFP varies by location, with increased exposure during transit to and from school.