Evaluating latent class models with conditional dependence in record linkage.

Record linkage methods commonly use a traditional latent class model to classify record pairs from different sources as true matches or non-matches. This approach was first formally described by Fellegi and Sunter and assumes that the agreement in fields is independent conditional on the latent class. Consequences of violating the conditional independence assumption include bias in parameter estimates from the model. We sought to further characterize the impact of conditional dependence on the overall misclassification rate, sensitivity, and positive predictive value in the record linkage problem when the conditional independence assumption is violated. Additionally, we evaluate various methods to account for the conditional dependence. These methods include loglinear models with appropriate interaction terms identified through the correlation residual plot as well as Gaussian random effects models. The proposed models are used to link newborn screening data obtained from a health information exchange. On the basis of simulations, loglinear models with interaction terms demonstrated the best misclassification rate, although this type of model cannot accommodate other data features such as continuous measures for agreement. Results indicate that Gaussian random effects models, which can handle additional data features, perform better than assuming conditional independence and in some situations perform as well as the loglinear model with interaction terms.

CORRECTED SCORE WITH SIZABLE COVARIATE MEASUREMENT ERROR: PATHOLOGY AND REMEDY.

Corrected score (Nakamura, 1990; Stefanski, 1989) is an important consistent functional modeling method for covariate measurement error in nonlinear regression. Although its pathological behaviors are known to exacerbate with increasing error contamination, neither their nature nor severity is well understood. In this article, we conduct a detailed investigation with the log-linear model for count data in the presence of sizable measurement error. Our study reveals that multiple roots, estimate-finding failure, and skewness in distribution are common and they may persist even when the sample size is practically large. Furthermore, these pathological behaviors are attributed to a surprising fact that desirable trend of the corrected score always goes astray as the parameter space approaches extremes. A novel remedy is proposed to constrain the derivatives with additional estimating functions. The resulting trend-constrained corrected score may also substantially improve the estimation efficiency. These findings and estimation strategy shed light on the developments for other nonlinear models as well, including logistic and Cox regression models, and for nonparametric correction.

Depression, anxiety and stress, comorbidity evaluation among a large sample of general adults: results from SEPAHAN study

Depression, anxiety and stress are common psychological disorders (PDs).This study aimed to assess the odds of co-occurrence of mentioned PDs in total sample and different levels of socio-demographic characteristics, specifically among a large sample of general adults.Ina cross-sectional, community-based study conducted among 4763 Iranian adults, depression and anxiety were assessed with Hospital Anxiety and Depression Scale (HADS) and stress with General Health Questionnaire (GHQ). The loglinear analysis was applied to investigate their comorbidities. Based on selected models with pair-comorbidity of anxiety with stress, depression with stress, and anxiety with depression, the results showed the odds of comorbidity between anxiety and depression (odds ratio (OR) =12.29, 95%CI: 9.58-15.80), depression and stress (OR = 7.80, 95%CI:6.55-10.18), and stress and anxiety (OR = 4.62, 95%CI:3.71-5.75). Also, ORs of pair-comorbidities were the same, except between stress and anxiety for men compared to women (adjusted-OR = 6.47, 95%CI: 4.44-9.49 versus 3.85, 95%CI:2.95-5.00) and comorbidity between stress and depression for the participants withlower than 40 years compared to others (adjusted-OR = 9.03, 95%CI: 7.17-11.36 versus 6.41, 95%CI: 4.90-8.41), p< 0.05. Stress comorbidity with depression was higher level than other pair-comorbidities. Obvious discrepancies were also observed in terms of ORs of pair-comorbidities between three mentioned disorders in different levels of SDCs.

Reliability for Multiple Reviewers by using Loglinear Models / 예방의학회지

To guarantee the inter-reviewer reliability is very important in evaluating the quality of large number of clinical research papers by multiple reviewers. We cannot find reports on statistical methods for evaluating reliability for multiple raters in clinical research field. The purpose of this paper is to introduce the statistical methods focused on kappa statistic and five kinds of loglinear models for, which can be applied when evaluating the reliability of multiple raters. We have applied these methods to the result of a project, in which seven reviewers have evaluated the quality of 33 papers with regard to four aspects of paper contents including study hypothesis, study design, study population, study method, data analysis and interpretation. Among the five loglinear models including Symmetry model, Conditional symmetry model, Quasi-symmetry model, Independence model, and Quasi-independence model, Quasi-symmetry model shows the best model of fitting. And the level of reliability among seven reviewers revealed to be acceptable as meaningful.