Biased bivariate correlations in combined survey data measured with different instruments

Uncalibrated Arterial Pulse Cardiac Output Measurements in Patients With Moderately Abnormal Left Ventricular Function

ABSTRACTThe purpose of this study was to identify differentially functioning items on operational administrations of the Graduate Management Admission Test (GMAT) through the use of the Mantel‐Haenszel statistic. Retrospective analyses of data collected over a three‐year period are reported for Black/White and female/male comparisons. Two tests are included in the analyses–Verbal and Quantitative. Data concerning the number of items identified as differentially difficult, the characteristics of the test questions associated with differential item difficulty, and the relationship of item difficulty to differential item functioning are discussed. In general, one percent to six percent of the items were identified as being differentially difficult per comparison with a greater number of items flagged in the female/male analyses than in the Black/White analyses. There was little overlap among the items flagged in the two analyses. Although the analyses suggested some content characteristics that may be related to differential item functioning, these findings about the GMAT items should be considered tentative since only a small number of items was studied and all investigations were post hoc analyses. Correlations between item difficulty and differential item functioning were generally low, with the exception of Quantitative items in the Black/White analyses. For these items, a moderately positive relationship existed between item difficulty and the differential item functioning statistic, showing that Black examinees performed differentially better than matched White examinees as item difficulty increased.

In educational testing, differential item functioning (DIF) examines whether item parameters differ across examinee characteristics. In this study, DIF is viewed not as bias but as item parameter heterogeneity reflecting differential engagement with related subskills. Using recursive partitioning based on parametric logistic item response theory (PL IRTree) models, this study investigated item functioning heterogeneity in a high-stakes reading comprehension test regarding vocabulary, grammar, and gender based on responses of 14,936 examinees. The comparison of Rasch tree, 2PL, 3PL, and 4PL IRTrees showed that the Rasch tree could better capture the structure of the data. The Educational Testing Service (ETS) classification scheme using the Mantel–Haenszel log odds ratio was then used to quantify effect sizes. The Rasch tree yielded 11 nodes, indicating subgroup differences in item difficulty, with seven items showing moderate parameter variation across five nodes. Except for one vocabulary-focused item showing systematic sensitivity to vocabulary knowledge, item parameter variation associated with lexico-grammatical knowledge was small, suggesting limited heterogeneity attributable to these subskills. Gender-related differences in item functioning appeared only within a subgroup of examinees with relatively high grammar-vocabulary scores. PL IRTree models can reveal item-level heterogeneity and the role of correlated subskills in shaping reading comprehension item functioning.

This study investigated the statistical and practical significance of context effects for items randomized within testlets for administration during a series of computerized non-adaptive tests. One hundred and twenty-five items from four primary school reading tests were studied. Logistic regression analyses identified from one to four items for each test as having statistically significant differences in item difficulty when administered in different positions in the testlet. Subsequent evaluation indicated that the effect sizes for most of these items were in the small category, and no consistent characteristic was identified to support a causative factor for these items. In addition, the few items flagged at each grade and their small effect sizes would argue against any effect on students' total scores. These results not only extended the study of context effects to primary school grades and computerized tests, but also supported testlets as useful to balance psychometric and practical testing needs. Further research into context effects due to item randomization within testlets should increase the number of items per testlet and the content areas studied.

Collecting accurate and reliable nutritional data from adolescent populations is challenging, with current methods providing significant under-reporting. Therefore, the aim of the study was to determine the accuracy of a combined dietary data collection method (self-reported weighed food diary, supplemented with a 24-h recall) when compared to researcher observed energy intake in male adolescent soccer players. Twelve Academy players from an English Football League club participated in the study. Players attended a 12 h period in the laboratory (08:00 h–20:00 h), during which food and drink items were available and were consumed ad libitum. Food was also provided to consume at home between 20:00 h and 08:00 h the following morning under free-living conditions. To calculate the participant reported energy intake, food and drink items were weighed and recorded in a food diary by each participant, which was supplemented with information provided through a 24-h recall interview the following morning. Linear regression, limits of agreement (LOA) and typical error (coefficient of variation; CV) were used to quantify agreement between observer and participant reported 24-h energy intake. Difference between methods was assessed using a paired samples t-test. Participants systematically under-reported energy intake in comparison to that observed (p < 0.01) but the magnitude of this bias was small and consistent (mean bias = −88 kcal·day−1, 95% CI for bias = −146 to −29 kcal·day−1). For random error, the 95% LOA between methods ranged between −1.11 to 0.37 MJ·day−1 (−256 to 88 kcal·day−1). The standard error of the estimate was low, with a typical error between measurements of 3.1%. These data suggest that the combined dietary data collection method could be used interchangeably with the gold standard observed food intake technique in the population studied providing that appropriate adjustment is made for the systematic under-reporting common to such methods.

In this study we examined various measures, including the concordance correlation (CC) coefficient, for determining the goodness of fit of forest models estimated by nonlinear mixed-model (NLMM) methods. Based on the volume–age data for black spruce, we analyzed the use of CC and other traditional goodness-of-fit measures such as coefficient of determination (R2), mean bias, percent bias, root mean square error, and graphic techniques on both the population and subject-specific levels within the NLMM framework. We also examined the relationship between goodness-of-fit measures and the number of observations per subject. We found that the standard overall goodness-of-fit measures commonly reported on combined data from different subjects were generally insufficient in determining the goodness of fitted models. We recommend that CC and other selected goodness-of-fit measures be calculated for individual subjects, and that the frequency distributions of the calculated values be examined and used as the principal criteria for determining the goodness of fit of forest models estimated by NLMM methods and for comparing alternative models and covariance structures. We also emphasized the importance of using pertinent graphic techniques to assess the appropriateness of NLMMs, especially at the subject-specific level, wherein lies the main interest of NLMMs.

Background and Aims Intradialytic hypotension (IDH) is the most common complication of hemodialysis (HD), which is often related to ultrafiltration. Due to the high prevalence of cardiovascular disease (CVD), compensatory mechanisms often fail. In clinical practice, BP is measured 2x/hour, while additional readings are only performed when symptoms arise. However, since IDH defined by a nadir systolic BP (SBP)&amp;lt;90 mmHg or &amp;lt;100 mmHg (depending on pre-dialysis SBP), regardless of symptoms, is most strongly associated with mortality, current BP measurement frequency may severely underestimate the true incidence of IDH. Continuous BP-monitoring may offer a solution to this issue. The Clearsight device (CSD) is a non-invasive continuous BP-monitoring system that measures arterial pressure and waveforms through a finger cuff by photoplethysmography. Yet, a compromised peripheral circulation may hamper its reliability in HD patients. This study aims to assess the agreement between intradialytic CSD-monitoring and standard oscillometric brachial arterial pressure (oBAP) assessment. Methods 40 Chronic HD patients were cross-over randomized to 4 modalities, each for 2 weeks: [standard HD (S-HD), cool HD (C-HD), low volume hemodiafiltration (LV-HDF) and high-volume HDF (HV-HDF)]. During one session of each second week, BP and heart rate (HR) were continuously recorded with the CSD (BMEYE/Edwards Lifesciences) and every 15 minutes with a brachial arterial cuff (contralateral of the fistula or graft, if applicable). Simultaneous measurements of SBP, diastolic BP (DBP) and HR were compared with descriptive statistics (mean difference [bias, BAP minus CSD], standard deviation (SD) and range) and Bland-Altman analysis. Bias was defined as the mean difference between oBAP and CSD, and 95% limits of agreement (LA) as the mean difference ±1.96 SD. Post-hoc, patients were stratified into tertiles of unexplained missing CSD-readings, to explore whether specific patient characteristics explain failed readings. Results 75% of patients was male, with a mean age 69.7 (±13.5 years) and median dialysis vintage 3.0 years (IQR 1.0-5.8). Diabetes mellitus and CVD were present in 48% and 73%, resp. Of the 2720 paired BP-measurements, 42% CSD and 6% of oBAP-readings were missing, leaving 36% of absent CSD-readings unexplained. Bland-Altman analysis, comparing oBAP and CSD-measurements, revealed a mean bias ± LA of 13.3 ± 44.5 mmHg for SBP, a mean bias ± LA of 13.3 ± 25.5 for DBP, and a mean bias ± LA of −0.03 ± 22.9 for HR. Surprisingly, the group with a high (&amp;gt;38%) proportion of unexplained missing CSD-values was significantly younger compared to the low (&amp;lt;21%) and intermediate group (21-38%) (p = 0.015), whereas no significant differences were observed with respect to dialysis vintage, diabetes mellitus or CVD. Conclusions 1) Compared to the oBAP method, finger cuff-monitoring shows poor precision and accuracy in BP-measurements, but good accuracy and poor precision in HR-readings; 2) 1/3 of readings failed for unknown reasons. Thus, CSD-monitoring is unsuitable for intradialytic hemodynamic monitoring in chronic HD patients.

CNN-based fire detection method on autonomous ships using composite channels composed of RGB and IR data

In most randomized controlled trials (RCT), data is primarily and often only available for individuals who have agreed to be randomized, with little, if any, consideration for those who elected not to participate. This study evaluated the value of including a concurrent observational cohort of patients who declined randomization in the Foraminotomy ACDF Cost-Effectiveness Trial (FACET-RCT) but still underwent anterior or posterior cervical surgery. The goal was to determine if the FACET-RCT results could be generalized by comparing baseline characteristics and clinical outcomes between the randomized trial and observational cohort. A nationwide RCT with a parallel observational cohort recruiting patients from routine care. Between January 2016 and May 2020, 389 patients with cervical radiculopathy were screened, and 358 were eligible. Of these, 265 (74%) were randomized in the FACET-RCT for either posterior or anterior cervical surgery, while 80 (22%) opted out of randomization and were followed in an observational cohort. Only 13 (4%) patients declined participation in both FACET-RCT and cohort. Demographic data was collected, and primary outcomes included treatment success, evaluated using the Odom criteria as well as reduction in arm pain, assessed with a Visual Analogue Scale (VAS) at 6 weeks, and every 6 months up to 2 years postsurgery. Secondary outcomes included VAS for neck pain, neck disability, work ability, quality of life, treatment satisfaction, and need for revision surgeries. Baseline characteristics were compared between the FACET-RCT and cohort using logistic regression. Primary and secondary outcomes were analyzed for differences between study designs using mixed-model analyses adjusted for confounders. The primary noninferiority endpoint of the FACET-RCT was evaluated in both the cohort and combined data from both cohort and FACET-RCT at 2 years of follow-up. Patients in the cohort were slightly younger than those in the FACET-RCT (mean age of 48.4 versus 51.2 years; mean difference [MD], -2.5; 95% confidence interval [CI], -4.8 to -0.2; p=.04). In sub-analyses stratified by surgical approach (anterior vs posterior surgery), fewer patients in the observational cohort who underwent posterior surgery reported severe neck pain at baseline compared to their counterparts in the FACET-RCT (OR, 0.38; 95% CI: 0.14 to 0.92; p=.04). No other significant baseline differences were found. No significant differences in treatment success (OR, 1.3; 95% CI: 0.3 to 6.0; p=.75) and arm pain reduction (MD, -3.9; 95% CI: -9.2 to 1.5; p=.16) were observed between study designs. The primary noninferiority endpoint was achieved in the combined data from both the cohort and FACET-RCT, with a narrower CI compared to the FACET-RCT alone, indicating a more robust result. Secondary outcomes were comparable between groups. Randomization did not influence clinical outcomes for cervical surgery patients. Combining RCT with the observational cohort increased statistical power, external validity and robustness. Our findings support the value of observational methods as a complement to RCTs, especially when a large number of patients refuse RCT participation and high dropout and crossover rates are expected.

Progressive lung damage causes most deaths in cystic fibrosis. Non-steroidal anti-inflammatory drugs (such as ibuprofen) may prevent progressive pulmonary deterioration and morbidity in cystic fibrosis. This is an update of a previously published review. To assess the effectiveness of treatment with oral non-steroidal anti-inflammatory drugs in cystic fibrosis. We searched the Cochrane Cystic Fibrosis and Genetic Disorders Group Trials Register comprising references identified from comprehensive electronic database searches, hand searches of relevant journals and abstract books of conference proceedings. We contacted manufacturers of non-steroidal anti-inflammatory drugs and searched online trials registries.Latest search of the Group's Trials Register: 21 November 2018. Randomized controlled trials comparing oral non-steroidal anti-inflammatory drugs, at any dose for at least two months, to placebo in people with cystic fibrosis. Two authors independently assessed trials for inclusion the review and their potential risk of bias. Two authors independently rated the quality of the evidence for each outcome using the GRADE guidelines. The searches identified 17 trials; four are included (287 participants aged five to 39 years; maximum follow-up of four years) and one is currently awaiting classification pending publication of the full trial report and two are ongoing. Three trials compared ibuprofen to placebo (two from the same center with some of the same participants); one trial assessed piroxicam versus placebo.The three ibuprofen trials were deemed to have good or adequate methodological quality, but used various outcomes and summary measures. Reviewers considered measures of lung function, nutritional status, radiological assessment of pulmonary involvement, intravenous antibiotic usage, hospital admissions, survival and adverse effects. Combined data from the two largest ibuprofen trials showed a lower annual rate of decline for lung function, % predicted forced expiratory volume in one second (FEV1), mean difference (MD) 1.32 (95% confidence interval (CI) 0.21 to 2.42) (moderate-quality evidence); forced vital capacity (FVC), MD 1.27 (95% CI 0.26 to 2.28) (moderate-quality evidence); forced expiratory flow (FEF25%-75%), MD 1.80 (95% CI 0.15 to 3.45). The post hoc analysis of data from two trials split by age showed a slower rate of annual decline of FEV1 % predicted and FVC in the ibuprofen group in younger children, MD 1.41% (95% CI 0.03 to 2.80) (moderate-quality evidence) and MD 1.32% (95% CI 0.04 to 2.60) (moderate-quality evidence) respectively. Data from four trials demonstrated the proportion of participants with at least one hospitalization may be slightly lower in the ibuprofen group compared to placebo, Peto odds ratio 0.61 (95% CI 0.37 to 1.01) (moderate-quality evidence). In one trial, long-term use of high-dose ibuprofen was associated with reduced intravenous antibiotic usage, improved nutritional and radiological pulmonary status. No major adverse effects were reported, but the power of the trials to identify clinically important differences in the incidence of adverse effects was low.We did not have any concerns with regards to risk of bias for the trial comparing piroxicam to placebo. However, the trial did not report many data in a form that we could analyze in this review. No data were available for the review's primary outcome of lung function; available data for hospital admissions showed no difference between the groups. No analyzable data were available for any other review outcome. High-dose ibuprofen can slow the progression of lung disease in people with cystic fibrosis, especially in children, which suggests that strategies to modulate lung inflammation can be beneficial for people with cystic fibrosis.

Uncertainties in the numerical realization of the physical climate system in coarse-resolution climate models in the Coupled Model Intercomparison Project phase 3 (CMIP3) cause large spread in the global mean and regional response amplitude to a given anthropogenic forcing scenario, and they cause the climate models to have mean state climates different from the observed and different from each other. In a series of sensitivity simulations with an atmospheric general circulation model coupled to a Slab Ocean Model, the role of differences in the control mean sea surface temperature (SST) in simulating the global mean and regional response amplitude is explored. The model simulations are forced into the control mean state SST of 24 CMIP3 climate models, and 2xCO2 forcing experiments are started from the different control states. The differences in the SST mean state cause large differences in other climate variables, but they do not reproduce most of the large spread in the mean state climate over land and ice-covered regions found in the CMIP3 model simulations. The spread in the mean SST climatology leads to a spread in the global mean and regional response amplitude of about 10%, which is about half as much as the spread in the response of the CMIP3 climate models and is therefore of considerable size. Since the SST climatology biases are only a small part of the models’ mean state climate biases, it is likely that the climate model’s mean state climate biases are accounting for a large part of the model’s climate sensitivity spread.

<h3>Introduction</h3> The ultimate aim of neurorehabilitation for traumatic brain injury (TBI) survivors is good health-related quality of life (HRQL). Despite recommendation by the WHO that self-reporting measures of HRQL in children should be used whenever possible, paediatric TBI studies have almost all have used parent–proxy report to assess HRQL. <h3>Aims</h3> (1) To examine the relationship between parent and child-report estimations of HRQL at 1 (T1), 6 (T6) and 12 (T12) months post-TBI; (2) to investigate factors influencing parent–child agreement. <h3>Methods</h3> Prospective cohort study. HRQL was measured by Paediatric Quality of Life Inventory (PedsQL). Severity of TBI was classified by admission Glasgow Coma Score (GCS), post-traumatic amnesia (PTA) and length of hospital stay (LOS). Verbal and performance IQ in the child were assessed by the Wechsler Intelligence Scale for Children-III; emotional state by the Child Behaviour Checklist (CBCL). Burden to the main carer and family were assessed by Beck Depression Inventory (BDI) and family burden of injury inventory (FBII). <h3>Results</h3> 47 TBI subjects were studied. Mean age was 11.2 yrs. 66% were boys. 51% had severe/moderate TBI, 49% mild. Significant correlations were found between parent and child report PedsQL total, physical and psychosocial summary scores at T1 (TSS, r=0.36, p=0.020; physical, r&lt;0.001; PSS, r=0.32, p=0.036), T6 (TSS, r=0.63, p&lt;0.001; physical, r=0.69, p&lt;0.001; PSS, r=0.58, p&lt;0.001) and T12 (r=0.57, p=0.001; physical, r=0.74, p&lt;0.001; PSS, r=0.45, p=0.008) post-TBI. In contrast to other chronic disorders, parents tended to score HRQL higher than children. However, mean differences between parent and child scores were not significant for any domain at any time point, with the exception of school domain at T6 (mean bias, −7.91, SD=21.88, p=0.022) and T12 (mean bias, −7.83, SD=19.55, p=0.036). Differences between parent and child-report scores did not correlate with VIQ or PIQ, but did significantly correlate with parent and child emotional state at T1 and T6 (T6, BDI, r=−0.40, p=0.010; FBII, r=−0.56, p=0.001; CBCL total problem score, r=−0.511, p=0.001). By T12, differences only correlated with injury severity (admission GCS, r=0.44, p=0.011; PTA, r=−0.51, p=0.009; LOS, r=−0.49, p=0.008). <h3>Conclusion</h3> Parent and child-report HRQL generally agree, despite emotional state influencing assessment. Parents tend to underestimate HRQL after TBI.

Purpose To evaluate simplified dosimetry methods for 131 I-mIBG therapy that maintain agreement with reference method while reducing clinical workflow burdens, particularly beneficial for pediatric applications. Methods In 24 patients, we implemented a hybrid protocol combining planar whole-body scans (1, 24, 48 h p.i.) with 24 h SPECT/CT for whole-body and liver dosimetry. Three liver segmentation methods were compared on SPECT/CT: whole-organ manual delineation (reference), 4 mL homogeneous sphere, and 4 mL peak sphere. To estimate time-integrated activity coefficients (TIACs) we modeled biokinetics with monoexponential using: full three-time-point data (reference), reduced combinations (1–2, 1–3, 2–3), Hänscheid single-time-point method, and population-based half-lives. Statistical analysis included linear mixed-effects modeling for segmentation comparisons and Bland-Altman analysis for TIAC method validation. Results The median absorbed dose was 0.36 mGy/MBq for liver and 0.09 mGy/MBq for whole-body. Among simplified protocols, dual-time-point 24–48 h imaging demonstrated the least bias, with liver absorbed doses showing − 0.02 mGy/MBq (limits of agreement: -0.08 to 0.04 mGy/MBq) and whole-body absorbed doses 0.001 mGy/MBq (limits of agreement: -0.004 to 0.005 mGy/MBq) compared to the full three-time-point reference results. Among single-time-point methods, for both liver and whole-body, the Hänscheid approach applied at 48 h demonstrated superior performance. The 4 mL peak sphere overestimated absorbed doses by 69% versus whole-organ delineation (0.61 vs. 0.36 mGy/MBq, p &lt; 0.001), and homogeneous spheres underestimated by 8% (0.33 vs. 0.36 mGy/MBq, p = 0.046). Conclusions Clinically feasible 131 I-mIBG dosimetry can be achieved through: (1) dual 24 h/48 h imaging (&lt; 6% mean bias), (2) single 48 h Hänscheid method (&lt; 6% mean bias), and (3) homogeneous sphere liver segmentation ( &lt; 8% mean difference from reference).

Compositional data are commonly used in chemical ecology to describe the biological role of chemical compounds in communication, defense or other behavioral modifications. Statistical analyses of compositional data, however, are challenging due to several constraints (e.g., constant sum constraint). We use an ontogenetic series of defensive gland secretions from larvae, three nymphal stages and adults of the oribatid model species Archegozetes longisetosus as a typical chemo-ecological data set to prepare a practical guide for compositional data analyses in chemical ecology. We compare various common and less common statistical and ordination methods to depict small quantitative and/or qualitative differences in compositional datasets: principal component analysis (PCA), non-metric multidimensional scaling (NMDS), multivariate statistical tests (Anderson’s permutational multivariate analyses of variance = PERMANOVA; permutational analyses of multivariate dispersions = PERMDIPS), linear discriminant analysis (LDA), the data mining algorithm Random Forests, bipartite network analysis and dynamic range boxes (dynRB). We summarize which methods are suitable for different research questions and how data needs to be structured and pre-processed. Network analyses and dynamic range boxes are promising tools for analyzing compositional data beyond the “classical” methods and provide additional information.

Otoliths are commonly used to discriminate between fish stocks, through both elemental composition and otolith shape. Typical studies also have a large number of elemental compositions and shape measures relative to the number of otolith samples, with these measures exhibiting strong mean–variance relationships. These properties make otolith composition and shape data highly suitable for use within a multivariate generalised linear model (MGLM) framework, yet MGLMs have never been applied to otolith data. Here we apply both a traditional distance based permutational multivariate analysis of variance (PERMANOVA) and MGLMs to a case study of striped snakehead (Channa striata) in India. We also introduce the Tweedie and gamma distributions as suitable error structures for the MGLMs, drawing similarities to the properties of Biomass data. We demonstrate that otolith elemental data and combined otolith elemental and shape data violate the assumption of homogeneity of variance of PERMANOVA and may give misleading results, while the assumptions of the MGLM with Tweedie and gamma distributions are shown to be satisfied and are appropriate for both otolith shape and elemental composition data. Consistent differences between three groups of C. striata were identified using otolith shape, otolith chemistry and a combined otolith shape and chemistry dataset. This suggests that future research should be conducted into whether there are demographic differences between these groups which may influence management considerations. The MGLM method is widely applicable and could be applied to any multivariate otolith shape or elemental composition dataset.