Subgroup Membership Research Articles

Kinematic foot types in youth with pes planovalgus secondary to cerebral palsy

BackgroundKinematic variability of the foot and ankle segments exists during ambulation among individuals with pes planovalgus (PPV) secondary to cerebral palsy (CP). Clinicians have previously recognized such variability through classification schemes to identify subgroups of individuals, but have been unable to identify kinematic foot types. Research questionThe purpose of this work was to identify kinematic foot types among children with PPV secondary to CP using 3-dimensional multi-segment foot and ankle kinematics during gait as inputs for principal component analysis (PCA) and K-means cluster analysis. MethodsIn a single assessment session, multi-segment foot and ankle kinematics using the Milwaukee Foot Model (MFM) were collected in 31 children/adolescents with pes planovalgus (49 feet) and 16 typically developing (TD) children/adolescents (31 feet). PCA was used as a data reduction technique on 34 kinematic variables. K-means cluster analysis was performed on the identified principal components (PCs) and one-way analyses of variance (ANOVA) was done to determine the effect of subgroup membership on PC scores. ResultsThe PCA reduced the kinematic variables to seven PCs which accounted for 91% of the total variance. Six distinct kinematic foot types were identified by the cluster analysis. The foot types showed unique kinematic characteristics in both the hindfoot and forefoot. SignificanceThis study provides further evidence of kinematic variability in the foot and ankle during ambulation associated with pes planovalgus secondary to CP. The specific contributions of the hindfoot and forefoot would not have been detected using a single segment foot model. The identification of kinematic foot types with unique foot and ankle characteristics has the potential to improve treatment since patients within a foot type are likely to benefit from similar intervention(s).

Impacts of enquiry-based science teaching on achievement gap between high-and-low SES students: findings from PISA 2015

ABSTRACTThe purpose of this study was to empirically investigate, in a large-scale assessment, the prevalence of enquiry-based teaching, as defined by PISA, implemented in science lessons, and the extent to which it is related to students from different socioeconomic status (SES) family backgrounds in the U.S. The Programme of International Student Assessment (PISA) 2015 data including 5,146 students in the U.S. were used. A Latent Profile Analysis (LPA) was conducted to classify students into different science teaching subgroups. Mixture regression analysis was conducted to examine whether there the association between SES and achievement varied across science teaching subgroups. Four science teaching subgroups were identified, varying in the frequency of and type of teaching practices used. Students’ attitudes, gender, science instruction time, and reading achievement are significant predictors of the subgroup membership. SES-achievement gap also varied across the science teaching subgroups. Discussions for the results and suggestions for future study were presented.

Cryptanalysis of the CLT13 Multilinear Map

In this paper, we describe a polynomial time cryptanalysis of the (approximate) multilinear map proposed by Coron, Lepoint, and Tibouchi in Crypto13 (CLT13). This scheme includes a zero-testing functionality that determines whether the message of a given encoding is zero or not. This functionality is useful for designing several of its applications, but it leaks unexpected values, such as linear combinations of the secret elements. By collecting the outputs of the zero-testing algorithm, we construct a matrix containing the hidden information as eigenvalues, and then recover all the secret elements of the CLT13 scheme via diagonalization of the matrix. In addition, we provide polynomial time algorithms to directly break the security assumptions of many applications based on the CLT13 scheme. These algorithms include solving subgroup membership, decision linear, and graded external Diffie–Hellman problems. These algorithms mainly rely on the computation of the determinants of the matrices and their greatest common divisor, instead of performing their diagonalization.

Burdened or Efficacious? Subgroups of Chinese American Language Brokers, Predictors, and Long-Term Outcomes.

Despite growing research on youth language brokering in immigrant families, evidence regarding its developmental outcomes remains mixed. This study took a person-centered approach, exploring subgroups of language brokers and identifying predictors and long-term outcomes of the subgroup membership. Participants were Chinese American adolescents (N = 350 at Time 1; Mage = 17.04; SD = 0.72; 59% female) followed over two waves spaced four years apart (longitudinal N = 291). Two distinct subgroups of adolescent language brokers were identified using latent profile analyses on language brokering feelings: efficacious and burdened brokers. Adolescents proficient in both English and Chinese were more likely to be efficacious brokers. Furthermore, burdened brokers reported higher parent-child alienation, and in turn, more depressive symptoms in emerging adulthood, compared to efficacious brokers and non-language-brokers. The current findings inform future interventions that burdened language brokers may be most at risk and that improving parent-child relationships may be one way to promote the well-being of young brokers.

Distinct attentional function profiles in older adults receiving cancer chemotherapy.

While attentional function is an extremely important patient outcome for older adults, research on changes in function in this group is extremely limited. The purposes of this study were to: identify subgroups of older patients (i.e., latent growth classes) based on changes in their level of self-reported attentional function; determine which demographic and clinical characteristics were associated with subgroup membership; and determine if these subgroups differed on quality of life (QOL) outcomes. Older oncology outpatients (n = 365) who were assessed for changes in attention and working memory using the Attentional Function Index a total of six times over two cycles of chemotherapy (CTX). QOL was assessed using the Medical Outcomes Study-Short Form 12 and the QOL-Patient Version Scale. Latent profile analysis (LPA) was used to identify subgroups of older adults with distinct attentional function profiles. Three distinct attentional functional profiles were identified (i.e., low, moderate, and high attentional function). Compared to the high class, older adults in the low and moderate attentional function classes had lower functional status scores, a worse comorbidity profile and were more likely to be diagnosed with depression. In addition, QOL scores followed an expected pattern (low class < moderate class < high attentional function class). Three distinct attentional function profiles were identified among a relatively large sample of older adults undergoing CTX. The phenotypic characteristics associated with membership in the low and moderate latent classes can be used by clinicians to identify high risk patients.

Adolescent Psychological Functioning and Membership in Latent Adolescent-Parent Communication Dual Trajectory Classes.

The primary goal of this study was to examine the relationship between adolescents' psychological functioning (as indicated by depressive symptoms) and substance use (alcohol and drug use) and membership in adolescent-parent communication trajectory subgroups in a large, diverse community sample of adolescents from the United States (n=1,057; 53% female; 51% Caucasian; Age: M=16.15, SD=.75). Adolescents completed questionnaires at three annual assessments. Fit indices from parallel process growth mixture models suggested three dual trajectory classes: (1) Average communication with both parents (Average-Both); (2) Good adolescent-mother and poor adolescent-father communication (Good-Mom/Poor-Dad); and (3) Poor adolescent-mother and good adolescent-father communication (Poor-Mom/Good-Dad). The trajectory classes differed by gender. In addition, psychological functioning and substance use were differentially related to the trajectory classes.

Toward a Typology of Technology-Using Teachers in the “New Digital Divide”: A Latent Class Analysis of the NCES Fast Response Survey System Teachers’ Use of Educational Technology in U.S. Public Schools, 2009 (FRSS 95)

Background Recently, policy makers and school leaders have heavily invested in the promise that educational technology could catalyze systemic school change. Yet some critics note that the conversation surrounding technology in schools is a red herring that has not produced char, definitive, and equitable results across different school settings. In order to address this concern, prior research has mainly focused on understanding how and why teachers use technology. Still, we argue that an understudied third perspective— examining what types of technology-using teachers exist— could provide innovative and impactful insights to shape research, policy, and practice in instructional technology. Purpose of the Study We investigate the extent to which there is a typology of teachers who use technology, as well as to what extent school- and teacher-level variables predict membership in the different subgroups in the typology, by analyzing a nationally generalizable sample (2, 764 teachers) from the Teachers’ Use of Educational Technology in U.S. Public Schools, 2009 Fast Response Survey System dataset, collected by the National Center for Education Statistics. Research Design We used a three-step, one-level latent class analysis (LCA) with nationally generalizable data that identify significantly different types of technology-using teachers, as well as what covariates predict membership in the identified subgroups. Findings We find that there are four statistically significant subgroups of technology-using teachers: Dexterous (24.4%), Evaders (22.2%), Assessors (28.4%), and Presenters (24.8%). We also find that several covariates, such as student socioeconomic status, school type, enrollment, years of teacher experience, and total number of school computers, predicted teachers’ membership in these four subgroups of technology-using teachers. Conclusions Our findings reiterate the notion that technology-using teachers are not a monolithic group or are randomly distributed across school settings, finding that low-income schools are more likely to have teachers who use technology in less meaningful ways. As a quantitative phenomenology, this study provides one of the first empirically based, nationally generalizable depictions of technology use in schools, which could inform school leaders and policy makers as they evaluate new digital tools, design professional learning for teachers, and tackle inequalities in technology access, teacher knowledge, and technology-mediated learning experiences and outcomes for students.

Heterogeneity of adolescent health risk behaviors in rural western China: A latent class analysis.

BackgroundAdolescent health risk behaviors are a public health priority given their prevalence and their associations with chronic diseases and life quality in adulthood. This study examined the heterogeneity of adolescent health risk behaviors and the associations between demographic characteristics and subgroup membership in rural western China.MethodsIn fall 2015, 2805 students from rural middle schools in Sichuan Province were surveyed using the Health-Related Behavior Questionnaire for Adolescents. Latent class analysis (LCA) was used to identify subgroups of adolescents with distinct patterns of health risk behaviors. Differences in class membership related to selected demographic characteristics were examined using multinomial logistic regression analysis.ResultsA four-class model emerged: (1) high-risk group (n = 108, 4.0%), (2) high-physical-inactivity and suicide-risk group (n = 340, 12.1%), (3) moderate-risk group (n = 897, 32.0%), and (4) low-risk group (n = 1460, 52.1%). The multinomial logistic regression analysis revealed that boys and adolescents with poor parental relationships and high allowances (spending money) were significantly more likely to be in the high-risk group than the low-risk group.ConclusionsAdolescents in rural western China are a heterogeneous population requiring different tailored and effective interventions.

Plasma metabolites reveal distinct profiles associating with different metabolic risk factors in monozygotic twin pairs.

Obesity is related to a myriad of cardiometabolic outcomes, each of which may have a specific metabolomic signature and a genetic basis. We identified plasma metabolites associating with different cardiometabolic risk factors (adiposity, cholesterol, insulin resistance, and inflammation) in monozygotic (MZ) twins. Additionally, we assessed if metabolite profiling can identify subgroups differing by cardiometabolic risk factors. We quantified 111 plasma metabolites (Acquity UPLC-triple quadrupole mass spectrometry), and measured blood lipids, HOMA index, CRP, and adiposity (BMI, %bodyfat by DEXA, fat distribution by MRI) in 40 MZ twin pairs (mean BMI 27.9 kg/m2, age 30.7). We determined associations among individuals (via linear regression) between metabolites and clinical phenotypes, and assessed, with within-twin pair analysis, if these associations were free from genetic confounding. We also performed cluster analysis to identify distinct subgroups based on subjects' metabolite profiles. We identified 42 metabolite-phenotype associations (FDR < 0.05), 19 remained significant after controlling for shared factors within the twin pairs. Aspartate, propionylcarnitine, tyrosine hexanoylcarnitine, and deoxycytidine associated positively with two or more adiposity measures. HDL cholesterol (HDL-C) associated negatively and BMI positively with the most numbers of metabolites; 12 were unique for HDL-C and 3 for BMI. Metabolites associating with HDL-C had the strongest effect size. Metabolite profiling revealed two distinct subgroups of individuals, differing by 32 metabolites (p < 0.05), and by total and LDL cholesterol(LDL-C). Forty-two metabolites predicted subgroup membership in correlation with total cholesterol and 45 metabolites predicted subgroup membership in correlation with LDL-C. Different fat depots share metabolites associating with general adiposity. BMI and HDL-C associated with the most pronounced and specific metabolomic signature. Metabolomics profiling can be used to identify distinct subgroups of individuals that differ by cholesterol measures. Most of the observed metabolite-phenotype associations are free of confounding by genetics and environmental factors shared by the co-twins.

Multidimensional Patterns of Sexual Risk Behavior and Psychiatric Disorders in Men with Substance Use Disorders.

Previous evidence links substance use disorders (SUD) to STI/HIV risk and suggests that comorbid psychiatric disorders increase the probability to engage in sexual risk behaviors. This study had two aims: (1) to identify subgroups based on sexual risk behavior using a person-centered approach in a sample of substance users and (2) to measure the association of psychiatric and SUD with subgroup membership. We assessed 402 male adults with SUD, reporting sexual intercourse in the previous 12months using the HIV-Risk Behavior Scale and the Mini International Neuropsychiatric Interview. Latent class analysis was performed to determine multidimensional patterns of sexual risk behaviors and multinomial logistic regression was utilized to associate classes with disorders. The three-class model showed the best fit, and the classes were labeled: Relationship-Based (31.34% of the sample), Condom-Based (39.55%), and Multiple Risks (29.10%). Controlling for age and marital status, major depressive disorders, antisocial personality disorder, and any psychiatric disorder were associated with the Multiple Risks class. Results stress the importance of developing a personalized assessment and counseling for sexual risk behaviors in individuals with SUD, particularly when they endorse criteria for comorbid psychiatric disorders. Future studies should focus on evaluating differential response to preventive interventions.

Cryptanalysis of Morillo–Obrador polynomial delegation schemes

Verifiable computation (VC) allows a client to outsource (delegate) the computation of a function f on an input x to a server and then verify the server's results with substantially less time than computing f(x) from scratch. The security of VC requires no efficient adversary can persuade the client to accept any wrong results. Morillo and Obrador (PST 2013) proposed three VC schemes for outsourcing the computation of polynomial functions and claimed that all schemes are secure under the decisional subgroup membership assumption. The authors show a simple attack against the security of their first scheme and then extend the attack to the other two schemes. Morillo and Obrador (PST 2013) also claimed that their third scheme keeps the client's input private under the square root assumption. The authors show that this is not true under the standard definition of input privacy. In particular, a curious server can extract the client's input x, if the x is not too large. The authors' results show that Morillo-Obrador schemes cannot be used in the polynomial delegation.

Patterns of heart failure symptoms are associated with self-care behaviors over 6 months.

Both heart failure symptoms and self-care are associated with patient outcomes. Although it is thought that symptoms drive self-care, there is limited evidence to support this assumption over time. To determine whether patterns of physical symptoms are significantly associated with heart failure self-care over time. Latent mixture analysis was used to identify subgroups based on physical symptoms of dyspnea, sleepiness and edema (using the heart failure somatic perception and Epworth sleepiness scales). Growth modeling was used to determine if symptom subgroups were associated with self-care behaviors (using the self-care in heart failure index) over 6 months. Sociodemographic and clinical variables predicting the likelihood of subgroup membership were identified using logistic regression. The sample ( n=146) was on average 57 years old, 70% were men and 59% had class III/IV heart failure. Two symptom subgroups were identified (entropy 0.91): a high symptom group ( n=24; 16%) with no significant change in symptoms over time (high sustained), and a low symptom group ( n=122; 84%) with no significant change in symptoms over time (low sustained). The high sustained group was associated significantly with better self-care behaviors at baseline and over 6 months. Women (odds ratio (OR) 3.67, P=0.023) and patients with more depressive symptoms (OR 1.16, P=0.015) were more likely to be in the high sustained symptom group. Those treated with a renin-angiotensin-aldosterone system agent were less likely to be in the high symptom group (OR 0.17, P=0.015). Patients bothered more by symptoms are consistently more engaged in self-care behaviors over time. The results of this study support symptoms as an important driver of self-care behaviors.

The Conjugacy Problem in Free Solvable Groups and Wreath Products of Abelian Groups is in TC0

We show that the conjugacy problem in a wreath product A ≀ B is uniform-TC0-Turing-reducible to the conjugacy problem in the factors A and B and the power problem in B. If B is torsion free, the power problem in B can be replaced by the slightly weaker cyclic submonoid membership problem in B. Moreover, if A is abelian, the cyclic subgroup membership problem suffices, which itself is uniform-AC0-many-one-reducible to the conjugacy problem in A ≀ B. Furthermore, under certain natural conditions, we give a uniform TC0 Turing reduction from the power problem in A ≀ B to the power problems of A and B. Together with our first result, this yields a uniform TC0 solution to the conjugacy problem in iterated wreath products of abelian groups – and, by the Magnus embedding, also in free solvable groups.

Global Online Subculture Surrounding School Shootings

This study is grounded in extensive online ethnographic fieldwork and interviews with 22 people who expressed a deep interest in school shootings. Such people form a global online subculture; they share common interests and find the same cultural objects important. Media accounts of school shootings have fueled this subculture; its members participate in the re-creation and circulation of online media content and give new meanings to that content. We found that people deeply interested in school shootings do not form a homogenous group, and they are divided into four subgroups within the subculture based on members’ focus and interest: researchers, fan girls, Columbiners, and copycats. Out of these, copycats are the only subgroup explicitly interested in replicating the acts, although subgroup membership can overlap, and members can move from one subgroup to another. Beyond copycats, other subgroups also participate in giving perpetrators fame and circulate reasons for the shootings. These accounts may influence future perpetrators.

BLAST: Bayesian Latent Subgroup Design for Basket Trials Accounting for Patient Heterogeneity

SummaryThe basket trial refers to a new type of phase II cancer trial that evaluates the therapeutic effect of a targeted agent simultaneously in patients with different types of cancer that involve the same genetic or molecular aberration. Although patients who are enrolled in the basket trial have the same molecular aberration, it is common for the targeted agent to be effective for patients with some types of cancer, but not others. We propose a Bayesian latent subgroup trial (BLAST) design to accommodate such treatment heterogeneity across cancer types. We assume that a cancer type may belong to the sensitive subgroup, which is responsive to the treatment, or the insensitive subgroup, which is not responsive to the treatment. Conditionally on the latent subgroup membership of the cancer type, we jointly model the binary treatment response and the longitudinal biomarker measurement that represents the biological activity of the targeted agent. The BLAST design makes the interim go–no-go treatment decision in a group sequential fashion for each cancer type on the basis of accumulating data. The simulation study shows that the BLAST design outperforms existing trial designs. It yields high power to detect the treatment effect for sensitive cancer types that are responsive to the treatment and maintains a reasonable type I error rate for insensitive cancer types that are not responsive to the treatment.

Predicting disease Risk by Transformation Models in the Presence of Unspecified Subgroup Membership.

Some biomedical studies lead to mixture data. When a discrete covariate defining subgroup membership is missing for some of the subjects in a study, the distribution of the outcome follows a mixture distribution of the subgroup-specific distributions. Taking into account the uncertain distribution of the group membership and the covariates, we model the relation between the disease onset time and the covariates through transformation models in each sub-population, and develop a nonparametric maximum likelihood based estimation implemented through EM algorithm along with its inference procedure. We further propose methods to identify the covariates that have different effects or common effects in distinct populations, which enables parsimonious modeling and better understanding of the difference across populations. The methods are illustrated through extensive simulation studies and a real data example.

Dose-escalation strategies which use subgroup information.

Dose‐escalation trials commonly assume a homogeneous trial population to identify a single recommended dose of the experimental treatment for use in future trials. Wrongly assuming a homogeneous population can lead to a diluted treatment effect. Equally, exclusion of a subgroup that could in fact benefit from the treatment can cause a beneficial treatment effect to be missed. Accounting for a potential subgroup effect (ie, difference in reaction to the treatment between subgroups) in dose‐escalation can increase the chance of finding the treatment to be efficacious in a larger patient population.A standard Bayesian model‐based method of dose‐escalation is extended to account for a subgroup effect by including covariates for subgroup membership in the dose‐toxicity model. A stratified design performs well but uses available data inefficiently and makes no inferences concerning presence of a subgroup effect. A hypothesis test could potentially rectify this problem but the small sample sizes result in a low‐powered test. As an alternative, the use of spike and slab priors for variable selection is proposed. This method continually assesses the presence of a subgroup effect, enabling efficient use of the available trial data throughout escalation and in identifying the recommended dose(s). A simulation study, based on real trial data, was conducted and this design was found to be both promising and feasible.

Latent class cluster analysis of symptom ratings identifies distinct subgroups within the clinical high risk for psychosis syndrome

The clinical-high-risk for psychosis (CHR-P) syndrome is heterogeneous in terms of clinical presentation and outcomes. Identifying more homogenous subtypes of the syndrome may help clarify its etiology and improve the prediction of psychotic illness. This study applied latent class cluster analysis (LCCA) to symptom ratings from the North American Prodrome Longitudinal Studies 1 and 2 (NAPLS 1 and 2). These analyses produced evidence for three to five subgroups within the CHR-P syndrome. Differences in negative and disorganized symptoms distinguished among the subgroups. Subgroup membership was found to predict conversion to psychosis. The authors contrast the methods employed within this study with previous attempts to identify more homogenous subgroups of CHR-P individuals and discuss how these results could be tested in future samples of CHR-P individuals.

A recursive partitioning approach to investigating correlates of self-rated health: The CARDIA Study

Self-rated health (SRH) is an independent predictor of mortality; studies have investigated correlates of SRH to explain this predictive capability. However, the interplay of a broad array of factors that influence health status may not be adequately captured with parametric multivariate regression. This study investigated associations between several health determinants and SRH using recursive partitioning methods. This non-parametric analytic approach aimed to reflect the social-ecological model of health, emphasizing relationships between multiple health determinants, including biological, behavioral, and from social/physical environments. The study sample of 3648 men and women was drawn from the year 15 (2000–2001) data collection of the CARDIA Study, USA, in order to study a young adult sample. Classification tree analysis identified 15 distinct, mutually exclusive, subgroups (eight with a larger proportion of individuals with higher SRH, and seven with a larger proportion of lower SRH), and multi-domain risk and protective factors associated with subgroup membership. Health determinant profiles were not uniform between subgroups, even for those with similar health status. The subgroup with the largest proportion of higher SRH was characterized by several protective factors, whilst that with the largest proportion of lower SRH, with several negative risk factors; certain factors were associated with both higher and lower SRH subgroups. In the full sample, physical activity, education and income were highest ranked by variable importance (random forests analysis) in association with SRH. This exploratory study demonstrates the utility of recursive partitioning methods in studying the joint impact of multiple health determinants. The findings indicate that factors do not affect SRH in the same way across the whole sample. Multiple factors from different domains, and with varying relative importance, are associated with SRH in different subgroups. This has implications for developing and prioritizing appropriate interventions to target conditions and factors that improve self-rated health status.

Person-centered examination of salivary cortisol and alpha-amylase responses to psychosocial stress: Links to preadolescent behavioral functioning and coping

This study adopted a person-centered approach to identify preadolescent salivary cortisol (sC) and alpha-amylase (sAA) co-activation response patterns and examine links to behavioral functioning and coping. Children (N = 151, 51.7% male) were exposed to the Trier Social Stress Test (TSST) and one of two randomly-assigned, post-TSST coping conditions: distraction or avoidance. Multi-trajectory modeling yielded four child subgroups. Child internalizing and externalizing positively predicted High sC–High sAA relative to Low sC–Low sAA and Low sC–High sAA relative to High sC–Low sAA subgroup membership, respectively. Low sC–Low sAA children demonstrated more efficient sC recovery when primed with distraction and more protracted sC recovery when primed with avoidance. For High sC–High sAA, internalizing children, the opposite was true. Findings illustrate adjustment-linked variability in preadolescent sC–sAA co-activation response patterns that further articulates for whom effortful coping works to effectively manage stressor-induced neuroendocrine activation.