Cluster Randomized Trials with a Pretest and Posttest: Equivalence of Three-, Two- and One-Level Analyses, and Sample Size Calculation

In a cluster randomized trial clusters of persons, for instance, schools or health centers, are assigned to treatments, and all persons in the same cluster get the same treatment. Although less powerful than individual randomization, cluster randomization is a good alternative if individual randomization is impossible or leads to severe treatment contamination (carry-over). Focusing on cluster randomized trials with a pretest and post-test of a quantitative outcome, this paper shows the equivalence of four methods of analysis: a three-level mixed (multilevel) regression for repeated measures with as levels cluster, person, and time, and allowing for unstructured between-cluster and within-cluster covariance matrices; a two-level mixed regression with as levels cluster and person, using change from baseline as outcome; a two-level mixed regression with as levels cluster and time, using cluster means as data; a one-level analysis of cluster means of change from baseline. Subsequently, similar equivalences are shown between a constrained mixed model and methods using the pretest as covariate. All methods are also compared on a cluster randomized trial on mental health in children. From these equivalences follows a simple method to calculate the sample size for a cluster randomized trial with baseline measurement, which is demonstrated step-by-step.

Special Issue PRO-Analysis of Clinically Meaningful Change on Patient-Reported Outcomes: Renewed Insights About Covariate Adjustment

Determining clinically meaningful change (CMC) in a patient-reported (PRO) measure is central to its existence in gauging how patients feel and function, especially for evaluating a treatment effect. Anchor-based approaches are recommended to estimate a CMC threshold on a PRO measure. Determination of CMC involves linking changes or differences in the target PRO measure to that in an external (anchor) measure that is easier to interpret than and appreciably associated with the PRO measure. One type of anchor-based approach for CMC is the "mean change method" where the mean change in score of the target PRO measure within a particular anchor transition level (e.g. one-category improvement) is subtracted from the mean change in score of within an adjacent anchor category (e.g. no change category). In the literature, the mean change method has been applied with and without an adjustment for the baseline scores for the PRO of interest. This article provides the analytic rationale and conceptual justification for keeping the analysis unadjusted and not controlling for baseline PRO scores. Two illustrative examples are highlighted. The current research is essentially a variation of Lord's paradox (where whether to adjust for a baseline variable depends on the research question) placed in a new context. Once the adjustment is made, the resulting CMC estimate reflects an artificial case where the anchor transition levels are forced to have the same average baseline PRO score. The unadjusted estimate acknowledges that the anchor transition levels are naturally occurring (not randomized) groups and thus maintains external validity.

Bias for Treatment Effect by Measurement Error in Pretest in ANCOVA Analysis

This paper investigated consequences of measurement error in the pretest on the estimate of the treatment effect in a pretest-posttest design with the analysis of covariance (ANCOVA) model, focusing on both the direction and magnitude of its bias. Some prior studies have examined the magnitude of the bias due to measurement error and suggested ways to correct it. However, none of them clarified how the direction of bias is affected by measurement error. This study analytically derived a formula for the asymptotic bias for the treatment effect. The derived formula is a function of the reliability of the pretest, the standardized population group mean difference for the pretest, and the correlation between pretest and posttest true scores. It revealed a concerning consequence of ignoring measurement errors in pretest scores: treatment effects could be overestimated or underestimated, and positive treatment effects can be estimated as negative effects in certain conditions. A simulation study was also conducted to verify the derived bias formula.

Design and Validation of an Augmented Reality Teaching System for Primary Logic Programming Education

Programming is a skill that requires high levels of logical thinking and problem-solving abilities. According to the Curriculum Guidelines for the 12-Year Basic Education currently implemented in Taiwan, programming has been included in the mandatory courses of middle and high schools. Nevertheless, the guidelines simply recommend that elementary schools conduct fundamental instructions in related fields during alternative learning periods. This may result in the problem of a rough transition in programming learning for middle school freshmen. To alleviate this problem, this study proposes an augmented reality (AR) logic programming teaching system that combines AR technologies and game-based teaching material designs on the basis of the fundamental concepts for seventh-grade structured programming. This system can serve as an articulation curriculum for logic programming in primary education. Thus, students are able to develop basic programming logic concepts through AR technologies by performing simple command programming. This study conducted an experiment using the factor-based quasi-experimental research design and questionnaire survey method, with 42 fifth and sixth graders enrolled as the experimental subjects. The statistical analysis showed the following results: In terms of learning effectiveness, both AR-based and traditional learning groups displayed a significant performance. However, of the two groups, the former achieved more significant effectiveness in the posttest results. Regarding learning motivation, according to the evaluation results of the Attention, Relevance, Confidence, and Satisfaction (ARCS) motivation model, the AR-based learning group manifested significantly higher levels of learning motivation than the traditional learning group, with particularly significant differences observed in the dimension of Attention. Therefore, the experimental results validate that the proposed AR-based logic programming teaching system has significant positive effects on enhancing students’ learning effectiveness and motivation.

Obese Subjects without Eating Disorders Experience Binge Episodes Also Independently of Emotional Eating and Personality Traits among University Students of Southern Italy

It is widely acknowledged that obesity is a growing public clinical issue involving both physical and psychological well-being. Nevertheless, the relationship between psychological features and weight gain is still unclear. Although emotional eating (EE) and personality traits are considered significant predictors of eating disorders, their role in obesity without eating disorders (OB-wed) is far from proven. The present study aimed at investigating the cumulative effect of EE and personality traits on overeating behavior in a sample of 266 university students (169 female; mean age = 21.85, SD = 2.39) stratified based on their body mass index (BMI; normal weight, overweight, obese). They were enrolled during free screening days promoted by the Human Dietetic and Sport Service of a Southern Italian university. The results show a psychological pattern of increasing overeating behavior and lower Self-Directedness combined with higher Sadness and Anger. However, OB-wed subjects overate regardless of this emotional/personological configuration.

Propensity score weighting for covariate adjustment in randomized clinical trials

Chance imbalance in baseline characteristics is common in randomized clinical trials. Regression adjustment such as the analysis of covariance (ANCOVA) is often used to account for imbalance and increase precision of the treatment effect estimate. An objective alternative is through inverse probability weighting (IPW) of the propensity scores. Although IPW and ANCOVA are asymptotically equivalent, the former may demonstrate inferior performance in finite samples. In this article, we point out that IPW is a special case of the general class of balancing weights, and advocate to use overlap weighting (OW) for covariate adjustment. The OW method has a unique advantage of completely removing chance imbalance when the propensity score is estimated by logistic regression. We show that the OW estimator attains the same semiparametric variance lower bound as the most efficient ANCOVA estimator and the IPW estimator for a continuous outcome, and derive closed-form variance estimators for OW when estimating additive and ratio estimands. Through extensive simulations, we demonstrate OW consistently outperforms IPW in finite samples and improves the efficiency over ANCOVA and augmented IPW when the degree of treatment effect heterogeneity is moderate or when the outcome model is incorrectly specified. We apply the proposed OW estimator to the Best Apnea Interventions for Research (BestAIR) randomized trial to evaluate the effect of continuous positive airway pressure on patient health outcomes. All the discussed propensity score weighting methods are implemented in the R package PSweight.

Application of student's t-test, analysis of variance, and covariance

Student's t test (t test), analysis of variance (ANOVA), and analysis of covariance (ANCOVA) are statistical methods used in the testing of hypothesis for comparison of means between the groups. The Student's t test is used to compare the means between two groups, whereas ANOVA is used to compare the means among three or more groups. In ANOVA, first gets a common P value. A significant P value of the ANOVA test indicates for at least one pair, between which the mean difference was statistically significant. To identify that significant pair(s), we use multiple comparisons. In ANOVA, when using one categorical independent variable, it is called one-way ANOVA, whereas for two categorical independent variables, it is called two-way ANOVA. When using at least one covariate to adjust with dependent variable, ANOVA becomes ANCOVA. When the size of the sample is small, mean is very much affected by the outliers, so it is necessary to keep sufficient sample size while using these methods.

Impact Evaluation Using Analysis of Covariance With Error-Prone Covariates That Violate Surrogacy

BACKGROUND

Analysis of covariance (ANCOVA) is commonly used to adjust for potential confounders in observational studies of intervention effects. Measurement error in the covariates used in ANCOVA models can lead to inconsistent estimators of intervention effects. While errors-in-variables (EIV) regression can restore consistency, it requires surrogacy assumptions for the error-prone covariates that may be violated in practical settings.

OBJECTIVES

The objectives of this article are (1) to derive asymptotic results for ANCOVA using EIV regression when measurement errors may not satisfy the standard surrogacy assumptions and (2) to demonstrate how these results can be used to explore the potential bias from ANCOVA models that either ignore measurement error by using ordinary least squares (OLS) regression or use EIV regression when its required assumptions do not hold.

RESULTS

The article derives asymptotic results for ANCOVA with error-prone covariates that cover a variety of cases relevant to applications. It then uses the results in a case study of choosing among ANCOVA model specifications for estimating teacher effects using longitudinal data from a large urban school system. It finds evidence that estimates of teacher effects computed using EIV regression may have smaller bias than estimates computed using OLS regression when the data available for adjusting for students' prior achievement are limited.

Longitudinal analysis of pre- and post-treatment measurements with equal baseline assumptions in randomized trials

For continuous variables of randomized controlled trials, recently, longitudinal analysis of pre‐ and posttreatment measurements as bivariate responses is one of analytical methods to compare two treatment groups. Under random allocation, means and variances of pretreatment measurements are expected to be equal between groups, but covariances and posttreatment variances are not. Under random allocation with unequal covariances and posttreatment variances, we compared asymptotic variances of the treatment effect estimators in three longitudinal models. The data‐generating model has equal baseline means and variances, and unequal covariances and posttreatment variances. The model with equal baseline means and unequal variance–covariance matrices has a redundant parameter. In large sample sizes, these two models keep a nominal type I error rate and have high efficiency. The model with equal baseline means and equal variance–covariance matrices wrongly assumes equal covariances and posttreatment variances. Only under equal sample sizes, this model keeps a nominal type I error rate. This model has the same high efficiency with the data‐generating model under equal sample sizes. In conclusion, longitudinal analysis with equal baseline means performed well in large sample sizes. We also compared asymptotic properties of longitudinal models with those of the analysis of covariance (ANCOVA) and t‐test.

When does measurement error in covariates impact causal effect estimates? Analytic derivations of different scenarios and an empirical illustration

The average causal treatment effect (ATE) can be estimated from observational data based on covariate adjustment. Even if all confounding covariates are observed, they might not necessarily be reliably measured and may fail to obtain an unbiased ATE estimate. Instead of fallible covariates, the respective latent covariates can be used for covariate adjustment. But is it always necessary to use latent covariates? How well do analysis of covariance (ANCOVA) or propensity score (PS) methods estimate the ATE when latent covariates are used? We first analytically delineate the conditions under which latent instead of fallible covariates are necessary to obtain the ATE. Then we empirically examine the difference between ATE estimates when adjusting for fallible or latent covariates in an applied example. We discuss the issue of fallible covariates within a stochastic theory of causal effects and analyse data of a within-study comparison with recently developed ANCOVA and PS procedures that allow for latent covariates. We show that fallible covariates do not necessarily bias ATE estimates, but point out different scenarios in which adjusting for latent covariates is required. In our empirical application, we demonstrate how latent covariates can be incorporated for ATE estimation in ANCOVA and in PS analysis.

Small-sample performance and underlying assumptions of a bootstrap-based inference method for a general analysis of covariance model with possibly heteroskedastic and nonnormal errors

It is well known that the standard F test is severely affected by heteroskedasticity in unbalanced analysis of covariance models. Currently available potential remedies for such a scenario are based on heteroskedasticity-consistent covariance matrix estimation (HCCME). However, the HCCME approach tends to be liberal in small samples. Therefore, in the present paper, we propose a combination of HCCME and a wild bootstrap technique, with the aim of improving the small-sample performance. We precisely state a set of assumptions for the general analysis of covariance model and discuss their practical interpretation in detail, since this issue may have been somewhat neglected in applied research so far. We prove that these assumptions are sufficient to ensure the asymptotic validity of the combined HCCME-wild bootstrap analysis of covariance. The results of our simulation study indicate that our proposed test remedies the problems of the analysis of covariance F test and its heteroskedasticity-consistent alternatives in small to moderate sample size scenarios. Our test only requires very mild conditions, thus being applicable in a broad range of real-life settings, as illustrated by the detailed discussion of a dataset from preclinical research on spinal cord injury. Our proposed method is ready-to-use and allows for valid hypothesis testing in frequently encountered settings (e.g., comparing group means while adjusting for baseline measurements in a randomized controlled clinical trial).

Emotion experience and frailty in a sample of Italian community-dwelling older adults

Frailty increases individual vulnerability to external stressors and involves high risk for adverse geriatric outcomes. To date, few studies have addressed the role of emotion perception and its association with frailty in aged populations. This cross-sectional study aimed to explore whether a significant association between frailty and emotional experience exists in a sample of Italian community-dwelling older adults. Our sample consisted of 104 older adults (age 76±8 years; 59.6% women) living in Piedmont, Italy. Frailty was measured using the Italian version of the Tilburg Frailty Indicator (TFI), and emotion perception was measured with the Positive and Negative Affect Schedule (PANAS). The Mini-Mental State Examination was used as a screening tool for cognitive functions (people with a score ≤20 points were excluded). One-way analysis of covariance (ANCOVA), adjusted for interesting variables, and post hoc tests were performed where appropriate. According to the TFI, 57.7% of participants resulted as frail. Analysis showed a significant greater severity of frailty in the low positive affect (PA) group compared to the high PA group. Similarly, those with high negative affect (NA) showed significantly higher levels of frailty than the low NA group. As expected, significant differences for frailty were also found among the groups composed of 1) people with high PA and low NA, 2) people with low PA or high NA, and 3) people with low PA and high NA. Post hoc tests showed a greater severity of frailty in the second and in the third groups compared to the first one. Lastly, robust participants aged >75 years showed higher levels of PA than the group aged between 60 and 75 years. These findings demonstrate that both PA and NA may influence frailty, giving new insights for the evaluation and prevention of frailty in older adults.

Statistical Interactions in a Clinical Trial

New statisticians entering industry tend to "test statistical interactions" whenever there is a need. However, in many real-world applications, especially in clinical development of new drugs, most interactions need to be estimated, instead of tested. In this manuscript, the distinction between hypothesis testing and estimation will be articulated, and the use of statistical interactions in clinical development programs will be discussed. According to ICH E-9, the treatment by subgroup interaction should not be included in the prespecified primary statistical analysis model. The reasons behind this ICH E-9 recommendation are also clarified in this manuscript.

Identification of potential biomarkers of sepsis using bioinformatics analysis

Sepsis is defined as the systemic inflammatory response to infection and is one of the leading causes of mortality in critically ill patients. The goal of the present study is to elucidate the molecular mechanism of sepsis. Transcription profile data (GSE12624) were downloaded that had a total of 70 samples (36 sepsis samples and 34 non-sepsis samples) from the Gene Expression Omnibus database. Protein-protein interaction network analysis was conducted in order to comprehensively understand the interactions of genes in all samples. Hierarchical clustering and analysis of covariance (ANCOVA) global test were performed to identify the differentially expressed clusters in the networks, followed by function and pathway enrichment analyses. Finally, a support vector machine (SVM) was performed to classify the clusters, and 10-fold cross-validation method was performed to evaluate the classification results. A total of 7,672 genes were obtained after preprocessing of the mRNA expression profile data. The PPI network of genes under sepsis and non-sepsis status collected 1,996/2,147 genes and 2,645/2,783 interactions. Moreover, following the ANCOVA global test (P<0.05), 24 differentially expressed clusters with 12 clusters in septic and 12 clusters in non-septic samples were identified. Finally, 207 biomarker genes, including CDC42, CSF3R, GCA, HMGB2, RHOG, SERPINB1, TYROBP SERPINA1, FCER1 G and S100P in the top six clusters, were collected using the SVM method. The SERPINA1, FCER1 G and S100P genes are thought to be potential biomarkers. Furthermore, Gene oncology terms, including the intracellular signaling cascade, regulation of programmed cell death, regulation of cell death, regulation of apoptosis and leukocyte activation may participate in sepsis.

Explorations in statistics: the analysis of change

Learning about statistics is a lot like learning about science: the learning is more meaningful if you can actively explore. This tenth installment of Explorations in Statistics explores the analysis of a potential change in some physiological response. As researchers, we often express absolute change as percent change so we can account for different initial values of the response. But this creates a problem: percent change is really just a ratio, and a ratio is infamous for its ability to mislead. This means we may fail to find a group difference that does exist, or we may find a group difference that does not exist. What kind of an approach to science is that? In contrast, analysis of covariance is versatile: it can accommodate an analysis of the relationship between absolute change and initial value when percent change is useless.

Evaluation of the effectiveness of X-ray protective aprons in experimental and practical fields

Few practical evaluation studies have been conducted on X-ray protective aprons in workplaces. We examined the effects of exchanging the protective apron type with regard to exposure reduction in experimental and practical fields, and discuss the effectiveness of X-ray protective aprons. Experimental field evaluations were performed by the measurement of the X-ray transmission rates of protective aprons. Practical field evaluations were performed by the estimation of the differences in the transit doses before and after the apron exchange. A 0.50-mm lead-equivalent-thick non-lead apron had the lowest transmission rate among the 7 protective aprons, but weighed 10.9 kg and was too heavy. The 0.25 and 0.35-mm lead-equivalent-thick non-lead aprons differed little in the practical field of interventional radiology. The 0.35-mm lead apron had lower X-ray transmission rates and transit doses than the 0.25-mm lead-equivalent-thick non-lead apron, and each of these differences exceeded 8% in the experimental field and approximately 0.15 mSv/month in the practical field of computed tomography (p < 0.01). Therefore, we concluded that the 0.25-mm lead-equivalent-thick aprons and 0.35-mm lead apron are effective for interventional radiology operators and computed tomography nurses, respectively.

Inflammatory cytokines and chemokines, skeletal muscle and polycystic ovary syndrome: effects of pioglitazone and metformin treatment

OBJECTIVE

Chronic low-grade inflammation is a common feature of insulin resistant states, including obesity and type 2 diabetes. Less is known about inflammation in Polycystic Ovary Syndrome (PCOS). Thus we evaluated the impact of PCOS on circulating cytokine levels and the effects of anti-diabetic therapies on insulin action, cytokine and chemokine levels and inflammatory signaling in skeletal muscle.

METHODS

Twenty subjects with PCOS and 12 healthy normal cycling (NC) subjects of similar body mass index were studied. PCOS subjects received oral placebo or pioglitazone, 45 mg/d, for 6 months. All PCOS subjects then had metformin, 2 g/day, added to their treatment. Circulating levels of cytokines, chemokines, and adiponectin, skeletal muscle markers of inflammation and phosphorylation of signaling proteins, insulin action evaluated by the hyperinsulinemic/euglycemic clamp procedure and Homeostasis Model Assessment of Insulin Resistance were measured.

RESULTS

Circulating levels of a number of cytokines and chemokines were generally similar between PCOS and NC subjects. Levels in PCOS subjects were not altered by pioglitazone or metformin treatment, even though whole body insulin action and adiponectin levels increased with pioglitazone. In spite of the lack of change in levels of cytokines and chemokines, several markers of inflammation in skeletal muscle were improved with Pio treatment.

CONCLUSIONS

PCOS may represent a state of elevated sensitivity of inflammatory cells in skeletal muscle to cytokines and chemokines, a property that could be reversed by pioglitazone treatment together with improved insulin action.

TWO APPLICATIONS OF PERMUTATION TESTS IN BIOSTASTICS

We show two examples of how we answer biological questions by converting them into statistical hypothesis testing problems. We consider gene abundance data, and apply permutation tests. Though these tests are simple, they allow us to test biologically relevant hypotheses. Here we present the analysis of data rising from two studies on Type 1 Diabetes. In the first study [3] are interested in comparing the gut bacterial biodiversity in children at risk and not at risk of developing diabetes. In the second study, [4] compare the gut bacterial biodiversity of children in six different sites in USA and Europe. The statistical analyses presented here are parts of the "statistical methods" in two papers mentioned above. Here we offer a detailed explanation of the "Statistical Methods" addressed to readers with a statistics background.

Acute prenatal exposure to a moderate dose of valproic acid increases social behavior and alters gene expression in rats

Prenatal exposure to moderate doses of valproic acid (VPA) produces brainstem abnormalities, while higher doses of this teratogen elicit social deficits in the rat. In this pilot study, we examined effects of prenatal exposure to a moderate dose of VPA on behavior and on transcriptomic expression in three brain regions that mediate social behavior. Pregnant Long Evans rats were injected with 350 mg/kg VPA or saline on gestational day 13. A modified social interaction test was used to assess social behavior and social preference/avoidance during early and late adolescence and in adulthood. VPA-exposed animals demonstrated more social investigation and play fighting than control animals. Social investigation, play fighting, and contact behavior also differed as a function of age; the frequency of these behaviors increased in late adolescence. Social preference and locomotor activity under social circumstances were unaffected by treatment or age. Thus, a moderate prenatal dose of VPA produces behavioral alterations that are substantially different from the outcomes that occur following exposure to a higher dose. At adulthood, VPA-exposed subjects exhibited transcriptomic abnormalities in three brain regions: anterior amygdala, cerebellar vermis, and orbitofrontal cortex. A common feature among the proteins encoded by the dysregulated genes was their ability to be modulated by acetylation. Analysis of the expression of individual exons also revealed that genes involved in post-translational modification and epigenetic regulation had particular isoforms that were ubiquitously dysregulated across brain regions. The vulnerability of these genes to the epigenetic effects of VPA may highlight potential mechanisms by which prenatal VPA exposure alters the development of social behavior.

Explorations in statistics: the analysis of ratios and normalized data

Learning about statistics is a lot like learning about science: the learning is more meaningful if you can actively explore. This ninth installment of Explorations in Statistics explores the analysis of ratios and normalized-or standardized-data. As researchers, we compute a ratio-a numerator divided by a denominator-to compute a proportion for some biological response or to derive some standardized variable. In each situation, we want to control for differences in the denominator when the thing we really care about is the numerator. But there is peril lurking in a ratio: only if the relationship between numerator and denominator is a straight line through the origin will the ratio be meaningful. If not, the ratio will misrepresent the true relationship between numerator and denominator. In contrast, regression techniques-these include analysis of covariance-are versatile: they can accommodate an analysis of the relationship between numerator and denominator when a ratio is useless.

Iron deficiency alters expression of dopamine-related genes in the ventral midbrain in mice

A clear link exists between iron deficiency (ID) and nigrostriatal dopamine malfunction. This link appears to play an important role in at least restless legs syndrome (RLS) if not several other neurological diseases. Yet, the underlying mechanisms remain unclear. The effects of ID on gene expression in the brain have not been studied extensively. Here, to better understand how exactly ID alters dopamine functioning, we investigated the effects of ID on gene expression in the brain, seeking to identify any potential transcription-based mechanisms. We used six strains of recombinant inbred mice (BXD type) known to differ in susceptibility to ID in the brain. Upon weaning, we subjected mice from each strain to either an iron-deficient or iron-adequate diet. After 100 days of dietary treatment, we measured the effects of ID on gene expression in the ventral midbrain, a region containing the substantia nigra. The substantia nigra is the base of the nigrostriatal dopamine pathway and a region particularly affected by iron loss in RLS. We screened for ID-induced changes in expression, including changes in that of both iron-regulating and dopamine-related genes. Results revealed a number of expression changes occurring in ID, with large strain-dependent differences in the genes involved and number of expression changes occurring. In terms of dopamine-related genes, results revealed ID-induced expression changes in three genes with direct ties to nigrostriatal dopamine functioning, two of which have never before been implicated in an iron-dopamine pathway. These were stromal cell-derived factor 1 (Cxcl12, or SDF-1), a ferritin regulator and potent dopamine neuromodulator, and hemoglobin, beta adult chain 1 (Hbb-b1), a gene recently shown to play a functional role in dopaminergic neurons. The extent of up-regulation of these genes varied by strain. This work not only demonstrates a wide genetic variation in the transcriptional response to ID in the brain, but also reveals two novel biochemical pathways by which iron may potentially alter dopamine function.

Gender and body mass index modify the effect of increasing amounts of caffeinated coffee on postprandial glucose and insulin concentrations; a randomized, controlled, clinical trial

OBJECTIVE

To examine the effects of different coffee amounts on blood glucose and insulin concentrations of healthy volunteers, and to assess potential effect modification by sex and body mass index category.

MATERIALS/METHODS

Thirty-three volunteers [16 ♀/17 ♂, 16 normal-weight and 17 overweight/obese, 27.3 ± 7.2 (19-44) y] took part in this randomized, crossover study. Ιn the morning of each experimental day volunteers received a standardized meal along with 200 mL of water or instant coffee containing either 3 or 6 mg of caffeine/kg body weight. Blood samples were obtained and analyzed for glucose and insulin concentrations in the fasting state, immediately after meal/drink consumption and at standard time points for the next 3h thereafter.

RESULTS

Coffee delayed the rise of insulin in response to the standardized meal and the fall of glucose concentrations from its maximum levels in the entire study sample. Glucose incremental area under the curve (IAUC) was significantly different between interventions (P=.009) with both coffee amounts inducing a greater area compared to water. Secondary, subgroup analysis at the nominal level showed that this might be more evident among females (PIAUC=.05) and overweight/obese participants (PIAUC=.03). Furthermore, coffee, mainly the 6 mg dose, could be lowering insulin concentrations the first 30 min after its consumption compared to water in men and overweight/obese participants.

CONCLUSIONS

Coffee exerts an acute effect on postprandial glucose and insulin concentrations. This effect may be modified by sex and overweight/obese status. Future research is necessary to elucidate underlying mechanisms.

Objectively quantified physical activity in persons with multiple sclerosis

OBJECTIVES

To investigate levels of moderate-to-vigorous physical activity (MVPA) in a large sample of persons with multiple sclerosis (MS) and controls using accelerometry as a measure of physical activity, and to compare the rates of meeting public health guidelines for MVPA (ie, 30min/d) between persons with MS and controls.

DESIGN

Secondary analysis of a combined data set of persons with MS and healthy controls from 13 previous investigations of physical activity over a 8-year period (2005-2013).

SETTING

University.

PARTICIPANTS

Participants with MS (n=800) were recruited primarily within Illinois through multiple sources, including print and e-mail flyers and an online advertisement on the National Multiple Sclerosis Society website. Healthy controls (n=137) were recruited via public e-mail postings delivered across the university community.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Levels of MVPA and meeting public health guidelines for MVPA between persons with MS and controls.

RESULTS

After controlling for covariates (ie, age, sex, education, race, income), there was a moderate (d=.68) and statistically significant (F=47.2, P<.001) difference of 13.1 minutes of MVPA per day (95% confidence interval, 9.4-16.8) between MS and controls. There was a difference in the rates of meeting public health guidelines for MVPA (χ(2)=50.7, P<.001) between MS patients (20%) and controls (47%). Among those with MS, minutes of MVPA significantly differed as a function of education, employment status, clinical course, disease duration, and disability status.

CONCLUSIONS

We provide data using an objective physical activity measure and a large sample to indicate that only a small proportion of persons with MS are achieving adequate amounts of daily MVPA.

Open-label, randomized study of the effect of rivaroxaban with or without acetylsalicylic acid on thrombus formation in a perfusion chamber

INTRODUCTION

Rivaroxaban, a direct factor Xa inhibitor, has demonstrated effectiveness for the management of both venous and arterial thrombosis. This study was designed to investigate the antithrombotic effect of rivaroxaban, with or without acetylsalicylic acid (ASA), in an ex vivo perfusion chamber at both low and high shear rates.

MATERIALS AND METHODS

Healthy subjects (N=51) were enrolled in a randomized, crossover (rivaroxaban 5, 10 or 20mg with or without ASA), and parallel-group (compared with ASA plus clopidogrel) study. Thrombi formed on pig aorta strips were measured after a 5-minute perfusion at low and high shear rates with blood from the subjects by measuring D-dimer concentration (for fibrin deposition) and P-selectin content (for platelet deposition).

RESULTS

ASA alone had no impact on thrombus D-dimer levels, whereas rivaroxaban alone at peak concentrations decreased D-dimer levels by 9%, 84% and 65% at low shear rate and 37%, 73% and 74% at high shear rate after doses of 5, 10 and 20mg, respectively. Steady-state ASA plus rivaroxaban 5mg caused a greater reduction in D-dimer levels (63%) than monotherapy at low shear rate. Co-administration of ASA with clopidogrel was associated with a 30% decrease in D-dimer levels at low shear rate and a 14% decrease at high shear rate. No conclusive effect on P-selectin content was observed across the treatment groups.

CONCLUSIONS

Rivaroxaban dose-dependently inhibited ex vivo thrombus formation under low and high shear rates. Co-administration of ASA had an additional effect on the antithrombotic action of low-dose rivaroxaban.

Effectiveness of home- and community-based rehabilitation in a large cohort of patients disabled by cerebrovascular accident: evidence of a dose-response relationship

OBJECTIVES

To (1) assess the effectiveness of home- and community-based rehabilitation (HCBR) in a large cohort of individuals with disabilities secondary to cerebrovascular accident (CVA); and (2) evaluate the responsiveness to treatment of the Mayo-Portland Adaptability Inventory (MPAI-4) to changes resulting from HCBR in this patient group.

DESIGN

Retrospective analysis of program evaluation data for treatment completers and noncompleters.

SETTING

HCBR conducted in 7 geographically distinct U.S. cities.

PARTICIPANTS

Individuals with CVA (n=738) who completed the prescribed course of rehabilitation (completed course of treatment [CCT]) compared with 150 individuals who were precipitously discharged (PD) before program completion.

INTERVENTION

HCBR delivered by certified professional staff on an individualized basis.

MAIN OUTCOME MEASURES

Mayo-Portland Adaptability Inventory (MPAI-4) completed by professional consensus on admission and at discharge.

RESULTS

With the use of analysis of covariance, MPAI-4 total scores at discharge for CCT participants were compared with those of PD participants, with admission MPAI-4, age, length of stay, and time since event as covariates. CCT participants showed greater improvement than PD participants (F=99.48, P<.001) with a moderate effect size (partial η(2)=.10). Group differences and effect sizes were similar for the 3 index scores: Ability (F=75.96, P<.001; partial η(2)=.08), Adjustment (F=99.67, P<.001; partial η(2)=.10), and Participation (F=69.15, P<.001; partial η(2)=.07).

CONCLUSIONS

Individuals in the CCT group who received the entire planned course of HCBR showed greater improvement on all MPAI-4 indexes than those in the PD group who were discharged before completing the prescribed program. This dose-response relationship provides evidence of a causal relationship between treatment and outcome.

Steps per day among persons with multiple sclerosis: variation by demographic, clinical, and device characteristics

OBJECTIVES

To identify steps per day in a large sample of persons with multiple sclerosis (MS) and to describe variation by demographic and clinical characteristics and device type.

DESIGN

Cross-sectional design.

SETTING

General community.

PARTICIPANTS

Convenience sample of persons with multiple sclerosis (N=645) recruited from the general community who were ambulatory and relapse free for 30 days. Mean age ± SD of the participants was 46.3 ± 10.6 years old. Participants were mostly women (85%), white (93%), and employed (64%).

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURE

Step counts measured by a motion sensor during a 7-day period.

RESULTS

The average value for the entire sample was 5,903 ± 3,185 steps per day. This value varied by demographic and clinical characteristics, but not device type, and indicated that men, participants who were unemployed, had a high school education or less, progressive MS, a longer disease duration, and higher disability were less physically active based on the metric of steps per day.

CONCLUSIONS

This study provides an expected value for average steps per day among persons with MS. Such an expected value for this population is an important first step to help researchers and clinicians interested in improving the overall health of persons with MS through physical activity promotion.

Association of abdominal circumference with serum nitric oxide concentration in healthy population

OBJECTIVES

We determined the relationship between abdominal circumference and the concentration of nitric oxide (NO), an endothelial cell product known to play an important role in the regulation of vascular tone and thrombocyte activations.

METHODS

Subjects were 177 men and 339 women aged 40 or over who were free from a history of diabetes or malignancy. Analysis of covariance was applied to examine the gender-specific and smoking-status-specific associations of abdominal fat volume measured as waist circumference, waist-hip ratio, and waist-to-stature ratio, with serum NO level represented by the concentration of NO metabolites (NOx; nitrate plus nitrite).

RESULTS

Although men showed no statistical association between abdominal fat accumulation and NOx concentration, abdominal adiposity seemed to inversely affect the serum NOx concentration of never- and current-smoking women.

CONCLUSION

Our results suggest that a reduction in NO bioactivity occurs with abdominal fat accumulation in women. The underlying biological mechanism might involve adipocytokines secreted from visceral fat, but is yet to be elucidated.

Including all individuals is not enough: lessons for intention-to-treat analysis

BACKGROUND

Intention-to-treat (ITT) analysis requires all randomised individuals to be included in the analysis in the groups to which they were randomised. However, there is confusion about how ITT analysis should be performed in the presence of missing outcome data.

PURPOSES

To explain, justify, and illustrate an ITT analysis strategy for randomised trials with incomplete outcome data.

METHODS

We consider several methods of analysis and compare their underlying assumptions, plausibility, and numbers of individuals included. We illustrate the ITT analysis strategy using data from the UK700 trial in the management of severe mental illness.

RESULTS

Depending on the assumptions made about the missing data, some methods of analysis that include all randomised individuals may be less valid than methods that do not include all randomised individuals. Furthermore, some methods of analysis that include all randomised individuals are essentially equivalent to methods that do not include all randomised individuals.

LIMITATIONS

This work assumes that the aim of analysis is to obtain an accurate estimate of the difference in outcome between randomised groups and not to obtain a conservative estimate with bias against the experimental intervention.

CONCLUSIONS

Clinical trials should employ an ITT analysis strategy, comprising a design that attempts to follow up all randomised individuals, a main analysis that is valid under a stated plausible assumption about the missing data, and sensitivity analyses that include all randomised individuals in order to explore the impact of departures from the assumption underlying the main analysis. Following this strategy recognises the extra uncertainty arising from missing outcomes and increases the incentive for researchers to minimise the extent of missing data.

Contrast ultrasound imaging of the aorta alters vascular morphology and circulating von Willebrand factor in hypercholesterolemic rabbits

OBJECTIVES

Ultrasound contrast agents (UCAs) are intravenously infused microbubbles that add definition to ultrasonic images. Ultrasound contrast agents continue to show clinical promise in cardiovascular imaging, but their biological effects are not known with confidence. We used a cholesterol-fed rabbit model to evaluate these effects when used in conjunction with ultrasound (US) to image the descending aorta.

METHODS

Male New Zealand White rabbits (n = 41) were weaned onto an atherogenic diet containing 1% cholesterol, 10% fat, and 0.11% magnesium. At 21 days, rabbits were exposed to contrast US at 1 of 4 pressure levels using either the UCA Definity (Lantheus Medical Imaging, Inc, North Billerica, MA) or a saline control (n = 5 per group). Blood samples were collected and analyzed for lipids and von Willebrand factor (vWF), a marker of endothelial function. Animals were euthanized at 42 days, and tissues were collected for histologic analysis.

RESULTS

After adjustment for pre-exposure vWF, high-level US (in situ [at the aorta] peak rarefactional pressure of 1.4 or 2.1 MPa) resulted in significantly lower vWF 1 hour post exposure (P = .0127; P(adj) < .0762). This difference disappeared within 24 hours. Atheroma thickness in the descending aorta was lower in animals receiving the UCA compared to animals receiving saline.

CONCLUSIONS

Contrast US affected the descending aorta, as evidenced by two separate outcome measures. These results may be a first step in elucidating a previously unknown biological effect of UCAs. Further research is warranted to characterize the effects of this procedure.

Amniotic fluid volume and composition after fetal membrane resection in late-gestation sheep

The chronically catheterized fetal sheep is a widely used model for fetal physiologic and pathophysiologic investigations. Catheterization involves opening the amniochorion to gain access to the fetus. In the current study, we explored the role of the amnion and amniochorion in maintaining normal amniotic fluid volume (AFV) and composition and fetal blood-gas status after surgery. Fetal sheep were catheterized at 119.6 ± 0.3 (mean ± SE, n = 25) d gestation (term, approximately 147 d). An opening equal to approximately 5% of total membrane surface area was created by resecting a portion of the amnion or amniochorion during surgery. The uterine wall was closed in all animals. Compared with control sheep (AFV = 992 ± 153 mL, n = 11), resection of the amnion had no significant effect on AFV (745 ± 156 mL, n = 7) measured 5 d after surgery, whereas resection of the amniochorion resulted in extensive loss of amniotic fluid (AFV = 131 ± 38 mL, n = 7). This loss resulted from extensive entry of amniotic fluid into the space between the chorion and uterine wall. Amniotic fluid, fetal plasma, and urinary solute concentrations; arterial pH; oxygen tension; and carbon dioxide tension were unchanged. A small opening in the amnion has minimal effects on ovine AFV, whereas a small opening in the amniochorion results in oligohydramnios. In addition, the amnion appears to be the primary site that limits the rate of amniotic fluid absorption by the chorionic vasculature.

Bayesian designs to account for patient heterogeneity in phase II clinical trials

PURPOSE OF REVIEW

Between-patient heterogeneity is very common in clinical trials. This complicates treatment evaluation, due to known prognostic subgroup effects or potential treatment-subgroup interactions. We review two Bayesian phase II clinical trial designs that account explicitly for patient heterogeneity. The first design uses analysis of covariance to assess treatment effects in subgroups known to have different prognoses. The second design uses a hierarchical model for settings where, a priori, the experimental treatment effects in the subgroups are assumed to be exchangeable.

RECENT FINDINGS

Compared with simpler designs that ignore patient heterogeneity, each design provides substantial improvements by reducing both false positive and false negative rates and focusing resources on subgroups where an experimental treatment is more likely to provide an advance over standard therapy. In either case, accounting for potential treatment-subgroup interactions is extremely important.

SUMMARY

Due to the rapidly increasing number of potential new treatments to be evaluated clinically, increasing costs and limited resources, it is critically important to perform early phase clinical trials efficiently. The new Bayesian methods described here and other related methods provide efficient, broadly applicable tools to address these problems.

Semiparametric Estimation of Treatment Effect in a Pretest-Posttest Study with Missing Data

The pretest-posttest study is commonplace in numerous applications. Typically, subjects are randomized to two treatments, and response is measured at baseline, prior to intervention with the randomized treatment (pretest), and at prespecified follow-up time (posttest). Interest focuses on the effect of treatments on the change between mean baseline and follow-up response. Missing posttest response for some subjects is routine, and disregarding missing cases can lead to invalid inference. Despite the popularity of this design, a consensus on an appropriate analysis when no data are missing, let alone for taking into account missing follow-up, does not exist. Under a semiparametric perspective on the pretest-posttest model, in which limited distributional assumptions on pretest or posttest response are made, we show how the theory of Robins, Rotnitzky and Zhao may be used to characterize a class of consistent treatment effect estimators and to identify the efficient estimator in the class. We then describe how the theoretical results translate into practice. The development not only shows how a unified framework for inference in this setting emerges from the Robins, Rotnitzky and Zhao theory, but also provides a review and demonstration of the key aspects of this theory in a familiar context. The results are also relevant to the problem of comparing two treatment means with adjustment for baseline covariates.