Estimating Equations (estimating + equation)

Distribution by Scientific Domains

Kinds of Estimating Equations

  • generalized estimating equation

  • Terms modified by Estimating Equations

  • estimating equation approach
  • estimating equation models

  • Selected Abstracts


    Longitudinal analysis of inpatient care utilization among people with intellectual disabilities: 1999,2002

    JOURNAL OF INTELLECTUAL DISABILITY RESEARCH, Issue 2 2007
    C.-H. Loh
    Abstract Background There has been no longitudinal study in Taiwan to identify the nature and the scale of medical care utilization of people with intellectual disabilities (IDs) up to the present. The aim of this study is to describe inpatient utilization among people under ID care in institutions in order to identify the pattern of medical care needs and the factors affecting utilization in Taiwan. Method The subject cohort was 168 individuals with ID who were cared for by a large public disability institution from 1999 to 2002 in Taipei, Taiwan. Results On the examination of the inpatient care that these persons underwent, it was found that these individuals had a heightened need (inpatient rate: 10.1,14.9%) for inpatient care compared with the general population with disabilities (9.37%) in Taiwan. The main reasons for hospitalization were pneumonia, gastrointestinal disorders, cellulites, orthopaedic problems, epilepsy and bronchitis. Using the full model of Generalized Estimating Equations for inpatient care utilization, the factors including low income family, living in an institution, being a subject with cerebral palsy and being a high outpatient user all influenced the use of inpatient care. Conclusions This study highlights that health authorities need to promote health planning more in order to ensure an excellent quality of health monitoring and health promotion among people with ID cared for by institutions. [source]


    A Note on the Use of Unbiased Estimating Equations to Estimate Correlation in Analysis of Longitudinal Trials

    BIOMETRICAL JOURNAL, Issue 1 2009
    Wenguang Sun
    Abstract Longitudinal trials can yield outcomes that are continuous, binary (yes/no), or are realizations of counts. In this setting we compare three approaches that have been proposed for estimation of the correlation in the framework of generalized estimating equations (GEE): quasi-least squares (QLS), pseudo-likelihood (PL), and an approach we refer to as Wang,Carey (WC). We prove that WC and QLS are identical for the first-order autoregressive AR(1) correlation structure. Using simulations, we then develop guidelines for selection of an appropriate method for analysis of data from a longitudinal trial. In particular, we argue that no method is uniformly superior for analysis of unbalanced and unequally spaced data with a Markov correlation structure. Choice of the best approach will depend on the degree of imbalance and variability in the temporal spacing of measurements, value of the correlation, and type of outcome, e.g. binary or continuous. Finally, we contrast the methods in analysis of a longitudinal study of obesity following renal transplantation in children (© 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim) [source]


    Conditional Generalized Estimating Equations for the Analysis of Clustered and Longitudinal Data

    BIOMETRICS, Issue 3 2008
    Sylvie Goetgeluk
    Summary A common and important problem in clustered sampling designs is that the effect of within-cluster exposures (i.e., exposures that vary within clusters) on outcome may be confounded by both measured and unmeasured cluster-level factors (i.e., measurements that do not vary within clusters). When some of these are ill/not accounted for, estimation of this effect through population-averaged models or random-effects models may introduce bias. We accommodate this by developing a general theory for the analysis of clustered data, which enables consistent and asymptotically normal estimation of the effects of within-cluster exposures in the presence of cluster-level confounders. Semiparametric efficient estimators are obtained by solving so-called conditional generalized estimating equations. We compare this approach with a popular proposal by Neuhaus and Kalbfleisch (1998, Biometrics54, 638,645) who separate the exposure effect into a within- and a between-cluster component within a random intercept model. We find that the latter approach yields consistent and efficient estimators when the model is linear, but is less flexible in terms of model specification. Under nonlinear models, this approach may yield inconsistent and inefficient estimators, though with little bias in most practical settings. [source]


    Extended Generalized Estimating Equations for Binary Familial Data with Incomplete Families

    BIOMETRICS, Issue 4 2002
    Patrick E. B. FitzGerald
    Summary. In this article, we assess the performance of two standard, but naive, methods for handling incomplete familial data in GEE2 analyses when the outcome is binary. We also propose a new method for analyzing such data using GEE2 when explanatory variables are discrete. Unlike the naive methods, the new method does not require the missing data process to be ignorable. We illustrate our method with an example that examines the familial aggregation of obesity. [source]


    Model Selection in Estimating Equations

    BIOMETRICS, Issue 2 2001
    Wei Pan
    Summary. Model selection is a necessary step in many practical regression analyses. But for methods based on estimating equations, such as the quasi-likelihood and generalized estimating equation (GEE) approaches, there seem to be few well-studied model selection techniques. In this article, we propose a new model selection criterion that minimizes the expected predictive bias (EPB) of estimating equations. A bootstrap smoothed cross-validation (BCV) estimate of EPB is presented and its performance is assessed via simulation for overdispersed generalized linear models. For illustration, the method is applied to a real data set taken from a study of the development of ewe embryos. [source]


    Akaike's Information Criterion in Generalized Estimating Equations

    BIOMETRICS, Issue 1 2001
    Wei Pan
    Summary. Correlated response data are common in biomedical studies. Regression analysis based on the generalized estimating equations (GEE) is an increasingly important method for such data. However, there seem to be few model-selection criteria available in GEE. The well-known Akaike Information Criterion (AIC) cannot be directly applied since AIC is based on maximum likelihood estimation while GEE is nonlikelihood based. We propose a modification to AIC, where the likelihood is replaced by the quasi-likelihood and a proper adjustment is made for the penalty term. Its performance is investigated through simulation studies. For illustration, the method is applied to a real data set. [source]


    Menopausal transition and the risk of urinary incontinence: results from a British prospective cohort

    BJU INTERNATIONAL, Issue 8 2010
    Gita D. Mishra
    Study Type , Aetiology (inception cohort) Level of Evidence 2b OBJECTIVE To investigate the effect of menopausal transition and age on symptoms of urinary incontinence in midlife. SUBJECTS AND METHODS The study included a nationally representative cohort of 1211 women followed up since their birth in 1946 and annually from 48,54 years; their menopausal transition status and symptoms of stress, urge, and severe urinary incontinence (UI) at 7 consecutive years from ages 48,54 were assessed. RESULTS From Generalized Estimating Equations, women who became perimenopausal (,pre-peri') or those experiencing perimenopause for >1 year (,peri-peri') were more likely to have symptoms of stress UI than were postmenopausal women; the odds ratio (95% confidence interval) was; pre-peri 1.39 (1.11,1.73); and peri-peri 1.39 (1.4,1.71). Menopausal transition status was not associated with urge or severe UI. These relationships were not explained by age, childhood enuresis, reproductive factors, previous health status, body mass index and educational qualifications. CONCLUSION This study is unique in being able to disentangle the effects of age, menopausal transitions, and other life-long risk factors on UI. Menopausal transition was only related to stress UI, while increasing age was related to both stress and urge UI. This study suggests that there are both shared and distinct aetiological pathways leading to each type of UI. [source]


    Influence of private practice setting and physician characteristics on the use of breast cancer adjuvant chemotherapy for elderly women,,

    CANCER, Issue 17 2009
    Dawn L. Hershman MD
    Abstract BACKGROUND: Although >70% of younger women with nonmetastatic breast cancer (BC) received adjuvant chemotherapy, only approximately 15% to 20% of elderly women with BC received chemotherapy. The decision to treat may be associated with nonmedical factors, such as patient, physician, or practice characteristics. In the current study, the association between oncologist characteristics and the receipt of chemotherapy in elderly women with BC was evaluated. METHODS: Women aged >65 years who were diagnosed with American Joint Committee on Cancer stages I to III BC between 1991 and 2002 were identified in the Surveillance, Epidemiology, and End Results-Medicare database. The Physician Unique Identification Number was linked to the American Medical Association Masterfile to obtain information on oncologists. Investigated was the association between demographic, tumor, and oncologist-related factors and the receipt of chemotherapy, using Generalized Estimating Equations to control for clustering. Patients were defined as low risk (estrogen/progesterone receptor positive, stage I/II disease) and high risk (estrogen/progesterone receptor-negative, stage II/III disease). RESULTS: Of 42,544 women identified, 8714 (20%) were treated with adjuvant chemotherapy. In a hierarchical analysis, women who underwent chemotherapy were more likely be treated by oncologists primarily employed in a private practice (odds ratio [OR], 1.40; 95% confidence interval [95% CI], 1.23-1.59) and who graduated after 1975 (OR, 1.12; 95% CI, 1.01-1.26) and were less likely to have an oncologist trained in the United States (OR, 0.83; 95% CI, 0.74-0.93). The association between a private practice setting and the receipt of chemotherapy was found to be similar for patients at high risk (OR, 1.55) and low risk (OR, 1.35) for cancer recurrence. CONCLUSIONS: Elderly women with BC treated by oncologists who were employed in a private practice were more likely to receive chemotherapy. Efforts to determine whether these associations reflected experience, practice setting, insurance type, or other economic incentives are warranted. Cancer 2009. Published 2009 by the American Cancer Society. [source]


    Estimating the effect of treatment in a proportional hazards model in the presence of non-compliance and contamination

    JOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES B (STATISTICAL METHODOLOGY), Issue 4 2007
    Jack Cuzick
    Summary., Methods for adjusting for non-compliance and contamination, which respect the randomization, are extended from binary outcomes to time-to-event analyses by using a proportional hazards model. A simple non-iterative method is developed when there are no covariates, which is a generalization of the Mantel,Haenszel estimator. More generally, a ,partial likelihood' is developed which accommodates covariates under the assumption that they are independent of compliance. A key feature is that the proportion of contaminators and non-compliers in the risk set is updated at each failure time. When covariates are not independent of compliance, a full likelihood is developed and explored, but this leads to a complex estimator. Estimating equations and information matrices are derived for these estimators and they are evaluated by simulation studies. [source]


    Rate of detoxification service use and its impact among a cohort of supervised injecting facility users

    ADDICTION, Issue 6 2007
    Evan Wood
    ABSTRACT Background Vancouver, Canada recently opened a medically supervised injecting facility (SIF) where injection drug users (IDU) can inject pre-obtained illicit drugs. Critics suggest that the facility does not help IDU to reduce their drug use. Methods We conducted retrospective and prospective database linkages with residential detoxification facilities and used generalized estimating equation (GEE) methods to examine the rate of detoxification service use among SIF participants in the year before versus the year after the SIF opened. In secondary analyses, we used Cox regression to examine if having been enrolled in detoxification was associated with enrolling in methadone or other forms of addiction treatment. We also evaluated the impact of detoxification use on the frequency of SIF use. Results Among 1031 IDU, there was a statistically significant increase in the uptake of detoxification services the year after the SIF opened. [odds ratio: 1.32 (95% CI, 1.11,1.58); P = 0.002]. In turn, detoxification was associated independently with elevated rates of methadone initiation [relative hazard = 1.56 (95% CI, 1.04,2.34); P = 0.031] and elevated initiation of other addiction treatment [relative hazard = 3.73 (95% CI, 2.57,5.39); P < 0.001]. Use of the SIF declined when the rate of SIF use in the month before enrolment into detoxification was compared to the rate of SIF use in the month after discharge (24 visits versus 19 visits; P = 0.002). Conclusions The SIF's opening was associated independently with a 30% increase in detoxification service use, and this behaviour was associated with increased rates of long-term addiction treatment initiation and reduced injecting at the SIF. [source]


    Bivariate association analyses for the mixture of continuous and binary traits with the use of extended generalized estimating equations

    GENETIC EPIDEMIOLOGY, Issue 3 2009
    Jianfeng Liu
    Abstract Genome-wide association (GWA) study is becoming a powerful tool in deciphering genetic basis of complex human diseases/traits. Currently, the univariate analysis is the most commonly used method to identify genes associated with a certain disease/phenotype under study. A major limitation with the univariate analysis is that it may not make use of the information of multiple correlated phenotypes, which are usually measured and collected in practical studies. The multivariate analysis has proven to be a powerful approach in linkage studies of complex diseases/traits, but it has received little attention in GWA. In this study, we aim to develop a bivariate analytical method for GWA study, which can be used for a complex situation in which continuous trait and a binary trait are measured under study. Based on the modified extended generalized estimating equation (EGEE) method we proposed herein, we assessed the performance of our bivariate analyses through extensive simulations as well as real data analyses. In the study, to develop an EGEE approach for bivariate genetic analyses, we combined two different generalized linear models corresponding to phenotypic variables using a seemingly unrelated regression model. The simulation results demonstrated that our EGEE-based bivariate analytical method outperforms univariate analyses in increasing statistical power under a variety of simulation scenarios. Notably, EGEE-based bivariate analyses have consistent advantages over univariate analyses whether or not there exists a phenotypic correlation between the two traits. Our study has practical importance, as one can always use multivariate analyses as a screening tool when multiple phenotypes are available, without extra costs of statistical power and false-positive rate. Analyses on empirical GWA data further affirm the advantages of our bivariate analytical method. Genet. Epidemiol. 2009. © 2008 Wiley-Liss, Inc. [source]


    Multipoint analysis using affected sib pairs: Incorporating linkage evidence from unlinked regions

    GENETIC EPIDEMIOLOGY, Issue 2 2001
    Kung-Yee Liang
    Abstract In this paper, we proposed a multipoint method to assess evidence of linkage to one region by incorporating linkage evidence from another region. This approach uses affected sib pairs in which the number of alleles shared identical by descent (IBD) is the primary statistic. This generalized estimating equation (GEE) approach is robust in that no assumption about the mode of inheritance is required, other than assuming the two regions being considered are unlinked and that there is no more than one susceptibility gene in each region. The method proposed here uses data from all available families to simultaneously test the hypothesis of statistical interaction between regions and to estimate the location of the susceptibility gene in the target region. As an illustration, we have applied this GEE method to an asthma sib pair study (Wjst et al. [1999] Genomics 58:1,8), which earlier reported evidence of linkage to chromosome 6 but showed no evidence for chromosome 20. Our results yield strong evidence to chromosome 20 (P value = 0.0001) after incorporating linkage information from chromosome 6. Furthermore, it estimates with 95% certainty that the map location of the susceptibility gene is flanked by markers D20S186 and D20S101, which are approximately 16.3 cM apart. Genet. Epidemiol. 21:105,122, 2001. © 2001 Wiley-Liss, Inc. [source]


    Doing Better to Do Good: The Impact of Strategic Adaptation on Nursing Home Performance

    HEALTH SERVICES RESEARCH, Issue 3p1 2007
    Jacqueline S. Zinn
    Objective. To test the hypothesis that a greater commitment to strategic adaptation, as exhibited by more extensive implementation of a subacute/rehabilitation care strategy in nursing homes, will be associated with superior performance. Data Sources. Online Survey, Certification, and Reporting (OSCAR) data from 1997 to 2004, and the area resource file (ARF). Study Design. The extent of strategic adaptation was measured by an aggregate weighted implementation score. Nursing home performance was measured by occupancy rate and two measures of payer mix. We conducted multivariate regression analyses using a cross-sectional time series generalized estimating equation (GEE) model to examine the effect of nursing home strategic implementation on each of the three performance measures, controlling for market and organizational characteristics that could influence nursing home performance. Data Collection/Abstraction Methods. OSCAR data was merged with relevant ARF data. Principal Findings. The results of our analysis provide strong support for the hypothesis. Conclusions. From a theoretical perspective, our findings confirm that organizations that adjust strategies and structures to better fit environmental demands achieve superior performance. From a managerial perspective, these results support the importance of proactive strategic leadership in the nursing home industry. [source]


    Familial aggregation in the night eating syndrome

    INTERNATIONAL JOURNAL OF EATING DISORDERS, Issue 6 2006
    Jennifer D. Lundgren PhD
    Abstract Objective: This study examined the extent to which the night eating syndrome (NES) affects first-degree relatives of NES and control probands. Method: NES participants and controls were assessed with the Night Eating Questionnaire (NEQ), the Night Eating Syndrome History and Inventory (NESHI), 10 day sleep and food records, the Eating Disorder Examination (EDE), the Structured Clinical Interview for DSM IV Axis I Disorders (SCID I), and a Family History Questionnaire (FHQ) to assess the presence of NES among first-degree relatives. A proband predictive model, using logistic regression analyses and the generalized estimating equation to control for correlation among observations within families was used to assess familial aggregation. Results: The odds of an NES proband having an affected first-degree relative were significantly greater than that of a control proband (odds ratio = 4.9, p < .001). A number of covariates were included in the model: proband body mass index (BMI) (kg/m2), proband gender, proband age, proband ethnicity, first-degree relative gender, relationship to proband (i.e., mother, father, or sibling), and the interaction between relationship to proband and proband status (night eater or control); none was statistically significant (p > .05). Conclusion: The study showed a strong aggregation of NES in families. © 2006 by Wiley Periodicals, Inc. Int J Eat Disord 2006 [source]


    Effect of an Educational Intervention on Optimizing Antibiotic Prescribing in Long-Term Care Facilities

    JOURNAL OF AMERICAN GERIATRICS SOCIETY, Issue 8 2007
    (See Editorial Comments by Dr. Lona Mody on pp 130, 1302)
    OBJECTIVE: To assess the effect of an educational intervention aimed at optimizing antibiotic prescribing in long-term care (LTC) facilities. DESIGN: Cluster randomized, controlled trial. SETTING: Eight public LTC facilities in the Montreal area. PARTICIPANTS: Thirty-six physicians. INTERVENTION: The educational intervention consisted of mailing an antibiotic guide to physicians along with their antibiotic prescribing profile covering the previous 3 months. Targeted infections were urinary tract, lower respiratory tract, skin and soft tissues, and septicemia of unknown origin. In the prescribing profile, each antibiotic was classified as adherent or nonadherent to the guide. Physicians in the experimental group received the intervention twice, 4 months apart, whereas physicians in the control group provided usual care. MEASUREMENTS: Data on antibiotic prescriptions were collected over four 3-month periods: preintervention, postintervention I, postintervention II, and follow-up. A generalized estimating equation (GEE) model was used to compare the proportion of nonadherent antibiotic prescriptions of the experimental and control groups. RESULTS: By the end of the study, nonadherent antibiotic prescriptions decreased by 20.5% in the experimental group, compared with 5.1% in the control group. Based on the GEE model, during postintervention II, physicians in the experimental group were 64% less likely to prescribe nonadherent antibiotics than those in the control group (odds ratio=0.36, 95% confidence interval=0.18,0.73). CONCLUSION: An educational intervention combining an antibiotic guide and a prescribing profile was effective in decreasing nonadherent antibiotic prescriptions. Repetition of the intervention at regular intervals may be necessary to maintain its effectiveness. [source]


    Regression modelling of correlated data in ecology: subject-specific and population averaged response patterns

    JOURNAL OF APPLIED ECOLOGY, Issue 5 2009
    John Fieberg
    Summary 1.,Statistical methods that assume independence among observations result in optimistic estimates of uncertainty when applied to correlated data, which are ubiquitous in applied ecological research. Mixed effects models offer a potential solution and rely on the assumption that latent or unobserved characteristics of individuals (i.e. random effects) induce correlation among repeated measurements. However, careful consideration must be given to the interpretation of parameters when using a nonlinear link function (e.g. logit). Mixed model regression parameters reflect the change in the expected response within an individual associated with a change in that individual's covariates [i.e. a subject-specific (SS) interpretation], which may not address a relevant scientific question. In particular, a SS interpretation is not natural for covariates that do not vary within individuals (e.g. gender). 2.,An alternative approach combines the solution to an unbiased estimating equation with robust measures of uncertainty to make inferences regarding predictor,outcome relationships. Regression parameters describe changes in the average response among groups of individuals differing in their covariates [i.e. a population-averaged (PA) interpretation]. 3.,We compare these two approaches [mixed models and generalized estimating equations (GEE)] with illustrative examples from a 3-year study of mallard (Anas platyrhynchos) nest structures. We observe that PA and SS responses differ when modelling binary data, with PA parameters behaving like attenuated versions of SS parameters. Differences between SS and PA parameters increase with the size of among-subject heterogeneity captured by the random effects variance component. Lastly, we illustrate how PA inferences can be derived (post hoc) from fitted generalized and nonlinear-mixed models. 4.,Synthesis and applications. Mixed effects models and GEE offer two viable approaches to modelling correlated data. The preferred method should depend primarily on the research question (i.e. desired parameter interpretation), although operating characteristics of the associated estimation procedures should also be considered. Many applied questions in ecology, wildlife management and conservation biology (including the current illustrative examples) focus on population performance measures (e.g. mean survival or nest success rates) as a function of general landscape features, for which the PA model interpretation, not the more commonly used SS model interpretation may be more natural. [source]


    Effects of correlation and missing data on sample size estimation in longitudinal clinical trials

    PHARMACEUTICAL STATISTICS: THE JOURNAL OF APPLIED STATISTICS IN THE PHARMACEUTICAL INDUSTRY, Issue 1 2010
    Song Zhang
    Abstract In longitudinal clinical trials, a common objective is to compare the rates of changes in an outcome variable between two treatment groups. Generalized estimating equation (GEE) has been widely used to examine if the rates of changes are significantly different between treatment groups due to its robustness to misspecification of the true correlation structure and randomly missing data. The sample size formula for repeated outcomes is based on the assumption of missing completely at random and a large sample approximation. A simulation study is conducted to investigate the performance of GEE sample size formula with small sample sizes, damped exponential family of correlation structure and non-ignorable missing data. Copyright © 2008 John Wiley & Sons, Ltd. [source]


    Modified weights based generalized quasilikelihood inferences in incomplete longitudinal binary models

    THE CANADIAN JOURNAL OF STATISTICS, Issue 2 2010
    Brajendra C. Sutradhar
    Abstract In an incomplete longitudinal set up, a small number of repeated responses subject to an appropriate missing mechanism along with a set of covariates are collected from a large number of independent individuals over a small period of time. In this set up, the regression effects of the covariates are routinely estimated by solving certain inverse weights based generalized estimating equations. These inverse weights are introduced to make the estimating equation unbiased so that a consistent estimate of the regression parameter vector may be obtained. In the existing studies, these weights are in general formulated conditional on the past responses. Since the past responses follow a correlation structure, the present study reveals that if the longitudinal data subject to missing mechanism are generated by accommodating the longitudinal correlation structure, the conditional weights based on past correlated responses may yield biased and hence inconsistent regression estimates. The bias appears to get larger as the correlation increases. As a remedy, in this paper the authors proposed a modification to the formulation of the existing weights so that weights are not affected directly or indirectly by the correlations. They have then exploited these modified weights to form a weighted generalized quasi-likelihood estimating equation that yields unbiased and hence consistent estimates for the regression effects irrespective of the magnitude of correlation. The efficiencies of the regression estimates follow due to the use of the true correlation structure as a separate longitudinal weights matrix in the estimating equation. The Canadian Journal of Statistics © 2010 Statistical Society of Canada Dans un cadre de données longitudinales incomplètes, nous observons un petit nombre de réponses répétées sujettes à un mécanisme de valeurs manquantes approprié avec un ensemble de covariables provenant d'un grand nombre d'individus indépendants observés sur une petite période de temps. Dans ce cadre, les composantes de régression des covariables sont habituellement estimées en résolvant certains poids inverses obtenus à partir d'équations d'estimation généralisées. Ces poids inverses sont utilisés afin de rendre les équations d'estimation sans biais et ainsi permettre d'obtenir des estimateurs cohérents pour le vecteur des paramètres de régressions. Dans les études déjà existantes, ces poids sont généralement formulés conditionnement aux réponses passées. Puisque les réponses passées possèdent une structure de corrélation, cet article révèle que si les données longitudinales, soumises à un mécanisme de valeurs manquantes, sont générées en adaptant la structure de corrélation longitudinale, alors les poids conditionnels basés sur les réponses corrélées passées peuvent mener à des estimations biaisées, et conséquemment non cohérentes, des composantes de régression. Ce biais semble augmenter lorsque la corrélation augmente. Pour remédier à cette situation, les auteurs proposent dans cet article, une modification aux poids déjà existants afin que ceux-ci ne soient plus affectés directement ou indirectement par les corrélations. Par la suite, ils ont exploité ces poids modifiés pour obtenir une équation d'estimation généralisée pondérée basée sur la quasi-vraisemblance qui conduit à des estimateurs sans biais, et ainsi cohérents, pour les composantes de régression sans égard à l'ampleur de la corrélation. L'efficacité de ces estimateurs est attribuable à l'utilisation de la vraie structure de corrélation comme matrice de poids longitudinale à part dans l'équation d'estimation. La revue canadienne de statistique © 2010 Société statistique du Canada [source]


    Are There Enough Doctors in My Rural Community?

    THE JOURNAL OF RURAL HEALTH, Issue 2 2009
    Perceptions of the Local Physician Supply
    ABSTRACT:,Purpose: To assess whether people in the rural Southeast perceive that there is an adequate number of physicians in their communities, assess how these perceptions relate to county physician-to-population (PtP) ratios, and identify other factors associated with the perception that there are enough local physicians. Methods: Adults (n = 4,879) from 150 rural counties in eight southeastern states responded through a telephone survey. Agreement or disagreement with the statement "I feel there are enough doctors in my community" constituted the principal outcome. Weighted chi-square analysis and a generalized estimating equation (GEE) assessed the strength of association between perceptions of an adequate physician workforce and county PtP ratios, individual characteristics, attitudes about and experiences with medical care, and other county characteristics. Findings: Forty-nine percent of respondents agreed there were enough doctors in their communities, 46% did not agree, and 5% were undecided. Respondents of counties with higher PtP ratios were only somewhat more likely to agree that there were enough local doctors (Pearson's correlation coefficient = 0.09, P < .001). Multivariate analyses revealed that perceiving that there were enough local physicians was more common among men, those 65 and older, whites, and those with lower regard for physician care. Perceptions that the local physician supply was inadequate were more common for those who had longer travel distances, problems with affordability, and little confidence in their physicians. Perceptions of physician shortages were more common in counties with higher poverty rates. Conclusions: County PtP ratios only partially account for rural perceptions that there are or are not enough local physicians. Perceptions of an adequate local physician workforce are also related to how much people value physicians' care and whether they face other barriers to care. [source]


    Attending and Resident Satisfaction with Feedback in the Emergency Department

    ACADEMIC EMERGENCY MEDICINE, Issue 2009
    Lalena M Yarris MD
    Abstract Objectives:, Effective feedback is critical to medical education. Little is known about emergency medicine (EM) attending and resident physician perceptions of feedback. The focus of this study was to examine perceptions of the educational feedback that attending physicians give to residents in the clinical environment of the emergency department (ED). The authors compared attending and resident satisfaction with real-time feedback and hypothesized that the two groups would report different overall satisfaction with the feedback they currently give and receive in the ED. Methods:, This observational study surveyed attending and resident physicians at 17 EM residency programs through web-based surveys. The primary outcome was overall satisfaction with feedback in the ED, ranked on a 10-point scale. Additional survey items addressed specific aspects of feedback. Responses were compared using a linear generalized estimating equation (GEE) model for overall satisfaction, a logistic GEE model for dichotomized responses, and an ordinal logistic GEE model for ordinal responses. Results:, Three hundred seventy-three of 525 (71%) attending physicians and 356 of 596 (60%) residents completed the survey. Attending physicians were more satisfied with overall feedback (mean score 5.97 vs. 5.29, p < 0.001) and with timeliness of feedback (odds ratio [OR] = 1.56, 95% confidence interval [CI] = 1.23 to 2.00; p < 0.001) than residents. Attending physicians were also more likely to rate the quality of feedback as very good or excellent for positive feedback, constructive feedback, feedback on procedures, documentation, management of ED flow, and evidence-based decision-making. Attending physicians reported time constraints as the top obstacle to giving feedback and were more likely than residents to report that feedback is usually attending initiated (OR = 7.09, 95% CI = 3.53 to 14.31; p < 0.001). Conclusions:, Attending physician satisfaction with the quality, timeliness, and frequency of feedback given is higher than resident physician satisfaction with feedback received. Attending and resident physicians have differing perceptions of who initiates feedback and how long it takes to provide effective feedback. Knowledge of these differences in perceptions about feedback may be used to direct future educational efforts to improve feedback in the ED. [source]


    Candidate Molecular Pathway Genes Related to Appetite Regulatory Neural Network, Adipocyte Homeostasis and Obesity: Results from the CARDIA Study

    ANNALS OF HUMAN GENETICS, Issue 5 2010
    Yechiel Friedlander
    Summary Appetite regulatory neural network and adipocyte homeostasis molecular pathways are critical to long-term weight maintenance. Associations between obesity-related phenotypes and four genes in these pathways , leptin (LEP), leptin receptor (LEPR), neuropeptide Y2 receptor (NPY2R) and peptide YY (PYY) were examined in CARDIA Study participants (aged 18,30 at recruitment in 1985,6). Weight, BMI and waist circumference were measured at baseline and at years 2, 5, 7, 10, 15, and 20. Genotyping was conducted using tag SNPs characterising common genetic variations in these genes. Generalized estimating equation (GEE) models estimated associations between SNPs and repeated anthropometric measurements, controlling for sex and age. False discovery rate was used to adjust for multiple testing. In African-Americans, SNPs across the LEP gene demonstrated significant overall associations with all obesity-related phenotypes. The associations between LEP rs17151919 with weight tended to strengthen with time , the difference in weight associated with each additional minor allele increased from 2.6 kg at baseline to 4.8 kg at year 20 (SNP*time interaction p = 0.0193). NPY2R gene SNPs were associated with waist circumference among African-American men (p = 0.0462). In Caucasians, LEP SNPs also tended to be associated with weight (p = 0.0471), and PYY rs11684664 was associated with obesity-related phenotypes in women only (p = 0.010,0.026). Several LEP, and NPY2R and PYY SNPs were associated with obesity-related phenotypes in young adults, particularly among African-Americans. [source]


    Adjustment for Missingness Using Auxiliary Information in Semiparametric Regression

    BIOMETRICS, Issue 1 2010
    Donglin Zeng
    Summary In this article, we study the estimation of mean response and regression coefficient in semiparametric regression problems when response variable is subject to nonrandom missingness. When the missingness is independent of the response conditional on high-dimensional auxiliary information, the parametric approach may misspecify the relationship between covariates and response while the nonparametric approach is infeasible because of the curse of dimensionality. To overcome this, we study a model-based approach to condense the auxiliary information and estimate the parameters of interest nonparametrically on the condensed covariate space. Our estimators possess the double robustness property, i.e., they are consistent whenever the model for the response given auxiliary covariates or the model for the missingness given auxiliary covariate is correct. We conduct a number of simulations to compare the numerical performance between our estimators and other existing estimators in the current missing data literature, including the propensity score approach and the inverse probability weighted estimating equation. A set of real data is used to illustrate our approach. [source]


    Marginal Mark Regression Analysis of Recurrent Marked Point Process Data

    BIOMETRICS, Issue 2 2009
    Benjamin French
    Summary Longitudinal studies typically collect information on the timing of key clinical events and on specific characteristics that describe those events. Random variables that measure qualitative or quantitative aspects associated with the occurrence of an event are known as marks. Recurrent marked point process data consist of possibly recurrent events, with the mark (and possibly exposure) measured if and only if an event occurs. Analysis choices depend on which aspect of the data is of primary scientific interest. First, factors that influence the occurrence or timing of the event may be characterized using recurrent event analysis methods. Second, if there is more than one event per subject, then the association between exposure and the mark may be quantified using repeated measures regression methods. We detail assumptions required of any time-dependent exposure process and the event time process to ensure that linear or generalized linear mixed models and generalized estimating equations provide valid estimates. We provide theoretical and empirical evidence that if these conditions are not satisfied, then an independence estimating equation should be used for consistent estimation of association. We conclude with the recommendation that analysts carefully explore both the exposure and event time processes prior to implementing a repeated measures analysis of recurrent marked point process data. [source]


    Model Selection in Estimating Equations

    BIOMETRICS, Issue 2 2001
    Wei Pan
    Summary. Model selection is a necessary step in many practical regression analyses. But for methods based on estimating equations, such as the quasi-likelihood and generalized estimating equation (GEE) approaches, there seem to be few well-studied model selection techniques. In this article, we propose a new model selection criterion that minimizes the expected predictive bias (EPB) of estimating equations. A bootstrap smoothed cross-validation (BCV) estimate of EPB is presented and its performance is assessed via simulation for overdispersed generalized linear models. For illustration, the method is applied to a real data set taken from a study of the development of ewe embryos. [source]


    Statistical methods for longitudinal research on bipolar disorders

    BIPOLAR DISORDERS, Issue 3 2003
    John Hennen
    Objectives: Outcomes research in bipolar disorders, because of complex clinical variation over-time, offers demanding research design and statistical challenges. Longitudinal studies involving relatively large samples, with outcome measures obtained repeatedly over-time, are required. In this report, statistical methods appropriate for such research are reviewed. Methods: Analytic methods appropriate for repeated measures data include: (i) endpoint analysis; (ii) endpoint analysis with last observation carried forward; (iii) summary statistic methods yielding one summary measure per subject; (iv) random effects and generalized estimating equation (GEE) regression modeling methods; and (v) time-to-event survival analyses. Results: Use and limitations of these several methods are illustrated within a randomly selected (33%) subset of data obtained in two recently completed randomized, double blind studies on acute mania. Outcome measures obtained repeatedly over 3 or 4 weeks of blinded treatment in active drug and placebo sub-groups included change-from-baseline Young Mania Rating Scale (YMRS) scores (continuous measure) and achievement of a clinical response criterion (50% YMRS reduction). Four of the methods reviewed are especially suitable for use with these repeated measures data: (i) the summary statistic method; (ii) random/mixed effects modeling; (iii) GEE regression modeling; and (iv) survival analysis. Conclusions: Outcome studies in bipolar illness ideally should be longitudinal in orientation, obtain outcomes data frequently over extended times, and employ large study samples. Missing data problems can be expected, and data analytic methods must accommodate missingness. [source]


    The effect of social networks and social support on common mental disorders following specific life events

    ACTA PSYCHIATRICA SCANDINAVICA, Issue 2 2010
    P. K. Maulik
    Maulik PK, Eaton WW, Bradshaw CP. The effect of social networks and social support on common mental disorders following specific life events. Objective:, This study examined the association between life events and common mental disorders while accounting for social networks and social supports. Method:, Participants included 1920 adults in the Baltimore Epidemiologic Catchment Area Cohort who were interviewed in 1993,1996, of whom 1071 were re-interviewed in 2004,2005. Generalized estimating equations were used to analyze the data. Results:, Social support from friends, spouse or relatives was associated with significantly reduced odds of panic disorder and psychological distress, after experiencing specific life events. Social networks or social support had no significant stress-buffering effect. Social networks and social support had almost no direct or buffering effect on major depressive disorder, and no effect on generalized anxiety disorder and alcohol abuse or dependence disorder. Conclusion:, The significant association between social support and psychological distress, rather than diagnosable mental disorders, highlights the importance of social support, especially when the severity of a mental health related problem is low. [source]


    Social functioning and communication in children with cerebral palsy: association with disease characteristics and personal and environmental factors

    DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY, Issue 5 2010
    JEANINE M VOORMAN
    Aim, The objective of this longitudinal study was to describe the course of social functioning and communication in children with cerebral palsy (CP) over a 3-year period, its difference with the normative course, and its relationship with disease characteristics and personal and environmental factors. Method, Participants in this study were 110 children with CP (70 males, 40 females) with a mean age of 11 years and 3 months (SD 1y 8mo). Social functioning and communication were measured with the Vineland Adaptive Behavior Scales. Comparisons were made with normative data; data were analysed with generalized estimating equations. According to the Gross Motor Function Classification System (GMFCS), 50 of the 110 children were categorized as GMFCS level I, 16 as level II, 13 as level III, 13 as level IV, and 18 as level V. Results, The course of social functioning over a 3-year period showed an increase in restrictions in children with CP (p<0.001). Restrictions in communication increased more in children with the most severe forms of CP (p<0.001). In addition to disease characteristics (GMFCS category, presence of epilepsy, and speech problems), personal factors (externalizing behaviour problems) and environmental factors (having no siblings, low parental level of education, and parental stress) were associated with greater restrictions in social functioning and communication. Interpretation, The results indicate that it is important to focus not only on the medical treatment of children with CP, but also on their behavioural problems and social circumstances, and to support the parents so that social functioning and communication in these children may be improved. [source]


    Methods to account for spatial autocorrelation in the analysis of species distributional data: a review

    ECOGRAPHY, Issue 5 2007
    Carsten F. Dormann
    Species distributional or trait data based on range map (extent-of-occurrence) or atlas survey data often display spatial autocorrelation, i.e. locations close to each other exhibit more similar values than those further apart. If this pattern remains present in the residuals of a statistical model based on such data, one of the key assumptions of standard statistical analyses, that residuals are independent and identically distributed (i.i.d), is violated. The violation of the assumption of i.i.d. residuals may bias parameter estimates and can increase type I error rates (falsely rejecting the null hypothesis of no effect). While this is increasingly recognised by researchers analysing species distribution data, there is, to our knowledge, no comprehensive overview of the many available spatial statistical methods to take spatial autocorrelation into account in tests of statistical significance. Here, we describe six different statistical approaches to infer correlates of species' distributions, for both presence/absence (binary response) and species abundance data (poisson or normally distributed response), while accounting for spatial autocorrelation in model residuals: autocovariate regression; spatial eigenvector mapping; generalised least squares; (conditional and simultaneous) autoregressive models and generalised estimating equations. A comprehensive comparison of the relative merits of these methods is beyond the scope of this paper. To demonstrate each method's implementation, however, we undertook preliminary tests based on simulated data. These preliminary tests verified that most of the spatial modeling techniques we examined showed good type I error control and precise parameter estimates, at least when confronted with simplistic simulated data containing spatial autocorrelation in the errors. However, we found that for presence/absence data the results and conclusions were very variable between the different methods. This is likely due to the low information content of binary maps. Also, in contrast with previous studies, we found that autocovariate methods consistently underestimated the effects of environmental controls of species distributions. Given their widespread use, in particular for the modelling of species presence/absence data (e.g. climate envelope models), we argue that this warrants further study and caution in their use. To aid other ecologists in making use of the methods described, code to implement them in freely available software is provided in an electronic appendix. [source]


    Evidence of a complex association between dose, pattern and timing of prenatal alcohol exposure and child behaviour problems

    ADDICTION, Issue 1 2010
    Colleen M. O'Leary
    ABSTRACT Background There is a lack of evidence regarding the effect of dose, pattern and timing of prenatal alcohol exposure and behaviour problems in children aged 2 years and older. Methods A 10% random sample of women delivering a live infant in Western Australia (1995,96) were invited to participate in an 8-year longitudinal survey (78% response rate n = 2224); 85% were followed-up at 2 years, 73% at 5 years and 61% at 8 years. Alcohol consumption was classified by combining the overall dose, dose per occasion and frequency to reflect realistic drinking patterns. Longitudinal analysis was conducted using generalized estimating equations (GEE) to investigate the association between child behaviour as measured by the Child Behaviour Checklist at 2, 5 and 8 years of age and prenatal alcohol exposure collected 3 months postpartum for each trimester separately, adjusting for a wide range of confounding factors. Results Low levels of prenatal alcohol were not associated with child behaviour problems. There were increased odds of internalizing behaviour problems following heavy alcohol exposure in the first trimester; anxiety/depression [adjusted odds ratio (aOR) 2.82; 95% confidence interval (CI) 1.07,7.43] and somatic complaints (aOR 2.74; 95% CI 1.47,5.12) and moderate levels of alcohol exposure increased the odds of anxiety/depression (aOR 2.24; 95% CI 1.16,4.34). Conclusions Prenatal alcohol exposure at moderate and higher levels increased the odds of child behaviour problems with the dose, pattern and timing of exposure affecting the type of behaviour problems expressed. Larger studies with more power are needed to confirm these findings. [source]


    CONTRACT ENFORCEMENT AND INTERNATIONAL TRADE

    ECONOMICS & POLITICS, Issue 2 2007
    PRIYA RANJAN
    This paper derives estimating equations from a model where individuals consume two classes of goods, and the degree of contract enforcement affects the transaction cost of trade in the two classes of goods differentially. Empirically, using Rauch's classification, internationally traded goods are classified into differentiated goods and those possessing a reference price, with the presumption that contract enforcement issues are more important for the former. It is verified that the measures of contract enforcement affect the volume of trade in both types of goods, but the impact is larger for differentiated goods. [source]