Search

The development of next-generation sequencing (NGS) opens opportunities for new applications such as liquid biopsy, in which tumor mutation genotypes can be determined by sequencing circulating tumor DNA after blood draws. However, with highly diluted samples like those obtained with liquid biopsy, NGS invariably introduces a certain level of misclassification, even with improved technology. Recently, there has been a high demand to use mutation genotypes as biomarkers for predicting prognosis and treatment selection. Many methods have also been proposed to build classifiers based on multiple loci with machine learning algorithms as biomarkers. How the higher misclassification rate introduced by liquid biopsy will affect the performance of these biomarkers has not been thoroughly investigated. In this paper, we report the results from a simulation study focused on the clinical utility of biomarkers when misclassification is present due to the current technological limit of NGS in the liquid biopsy setting. The simulation covers a range of performance profiles for current NGS platforms with different machine learning algorithms and uses actual patient genotypes. Our results show that, at the high end of the performance spectrum, the misclassification introduced by NGS had very little effect on the clinical utility of the biomarker. However, in more challenging applications with lower accuracy, misclassification could have a notable effect on clinical utility. The pattern of this effect can be complex, especially for machine learning-based classifiers. Our results show that simulation can be an effective tool for assessing different scenarios of misclassification. [ABSTRACT FROM AUTHOR]

Background: Alcohol use contributes to the national burden of morbidity and mortality in the United States. Arrest, as a unique form of criminal justice system involvement, may be related to alcohol use from adolescence to adulthood. This study investigates the relationship between arrest and alcohol use across race/ethnicity-gender (R/E-G) status (e.g., Black, Latinx, and White men and women) as youth age. Methods: Data from 17 waves (1997-2015) of the National Longitudinal Survey of Youth, 1997 cohort (N = 8901) were used to explore how variation in R/E-G moderates the relationship between arrest history and alcohol use trajectories from 13 to 30 years old. Multilevel zero-inflated Poisson and Poisson regression were used to assess R/E-G variation in the relationship between arrest history and days of alcohol consumption, drinks per drinking occasion, and days of binge drinking after accounting for covariates, including incarceration. Results: The findings indicate that an arrest history is associated with alcohol use, and these results varied by R/E-G status, age, and alcohol use outcome. Those with an arrest history reported more days of drinking than their counterparts without an arrest; yet, the magnitude and direction of average drinks per occasion and binge drinking days varied by R/E-G status and age. Paradoxically, Black men, Black women, and Latinx men with an arrest history reported fewer days of binge drinking as they aged than their counterparts without an arrest. Conclusions: A history of arrest is important for alcohol use from adolescence to adulthood and varies by R/E-G status, age, and alcohol use outcome. This work confirms previous scholarship showing that arrest and alcohol use are socially patterned and R/E-G status is an essential consideration in understanding the relationship. Future work should include additional identities and health behaviors and the consequences related to alcohol use outcomes. [ABSTRACT FROM AUTHOR]

Human Immunodeficiency Virus (HIV)-positive individuals lost to follow-up from particular clinics may not be lost to care (LTC). After linking Vanderbilt's Comprehensive Care Clinic cohort to Tennessee's statewide HIV surveillance database, LTC decreased from 48.4% to 35.0% at 10 years. Routine surveillance linkage by domestic HIV clinics would improve LTC and retention measure accuracy. [ABSTRACT FROM AUTHOR]

Latent class analysis (LCA) is a powerful tool for detecting unobservable subgroups within a population. When a large number of covariates (features) are considered, an LCA faces great challenges in terms of both classification accuracy and computational efficiency. In this paper, we propose a novel feature screening procedure that eliminates most irrelevant features before an LCA is conducted. The proposed method is built on an EM-based hybrid hard-soft thresholding update (HHS-EM) of the latent class parameters, which naturally accounts for the joint effects between features. We show that the HHS-EM enjoys the sure screening property and leads to a refined LCA that is effective and consistent for high-dimensional classification. The performance of the proposed method is illustrated by means of simulation studies and a real-data example. [ABSTRACT FROM AUTHOR]

Model-based clustering is a commonly-used technique to partition heterogeneous data into homogeneous groups. When the analysis is to be conducted with a large number of features, analysts face simultaneous challenges in model interpretability, clustering accuracy, and computational efficiency. Several Bayesian and penalization methods have been proposed to select important features for model-based clustering. However, the performance of those methods relies on a careful algorithmic tuning, which can be time-consuming for high-dimensional cases. In this paper, we propose a new sparse clustering method based on alternating hard-thresholding. The new method is conceptually simple and tuning-free. With a user-specified sparsity level, it efficiently detects a set of key features by eliminating a large number of features that are less useful for clustering. Based on the selected key features, one can readily obtain an effective clustering of the original high-dimensional data under a general sparse covariance structure. Under mild conditions, we show that the new method leads to clusters with a misclassification rate consistent to the optimal rate as if the underlying true model were used. The promising performance of the new method is supported by both simulated and real data examples. [ABSTRACT FROM AUTHOR]

Classification using linear discriminant analysis (LDA) is challenging when the number of variables is large relative to the number of observations. Algorithms such as LDA require the computation of the feature vector's precision matrices. In a high-dimension setting, due to the singularity of the covariance matrix, it is not possible to estimate the maximum likelihood estimator of the precision matrix. In this paper, we employ the Stein-type shrinkage estimation of Ledoit and Wolf for high-dimensional data classification. The proposed approach's efficiency is numerically compared to existing methods, including LDA, cross-validation, gLasso, and SVM. We use the misclassification error criterion for comparison. [ABSTRACT FROM AUTHOR]

The binary indicator of collusion is the key ingredient in estimating overcharges from bid-rigging with a regression-based approach. We develop a method for examining the effects of misclassification error in the indicator of bid-rigging status on estimates of damages from collusion. We derive partial identification of the regression model of winning bids in public procurement auctions and provide informative bounds on the price effects of bid-rigging. We find that the bounds are tight when placing a plausible restriction on the extent of measurement errors. Our findings show that relaxing the nondifferential assumption about misclassification errors leads to wider bounds. [ABSTRACT FROM AUTHOR]

Abstract Background When Service Provision Assessment (SPA) surveys on primary health service delivery are combined with the nationally representative household survey—Demographic and Health Survey (DHS), they can provide key information on the access, utilization, and equity of health service availability in low- and middle-income countries. However, existing linkage methods have been established only at aggregate levels due to known limitations of the survey datasets. Methods For the linkage of two data sets at a disaggregated level, we developed a geostatistical approach where SPA limitations are explicitly accounted for by identifying the sites where health facilities might be present but not included in SPA surveys. Using the knowledge gained from SPA surveys related to the contextual information around facilities and their spatial structure, we made an inference on the service environment of unsampled health facilities. The geostatistical linkage results on the availability of health service were validated using two criteria—prediction accuracy and classification error. We also assessed the effect of displacement of DHS clusters on the linkage results using simulation. Results The performance evaluation of the geostatistical linkage method, demonstrated using information on the general service readiness of sampled health facilities in Tanzania, showed that the proposed methods exceeded the performance of the existing methods in terms of both prediction accuracy and classification error. We also found that the geostatistical linkage methods are more robust than existing methods with respect to the displacement of DHS clusters. Conclusions The proposed geospatial approach minimizes the methodological issues and has potential to be used in various public health research applications where facility and population-based data need to be combined at fine spatial scale.

Individual and population level inference about risk and burden of MDD, particularly maternal MDD, is often made using case‐finding tools that are imperfect and prone to misclassification error (i.e. false positives and negatives). These errors or biases are rarely accounted for and lead to inappropriate clinical decisions, inefficient allocation of scarce resources, and poor planning of maternal MDD prevention and treatment interventions. The argument that the use of existing maternal MDD case‐finding instruments results in misclassification errors is not new; in fact, it has been argued for decades, but by and large its implications and particularly how to correct for these errors for valid inference is unexplored. Correction of the estimates of maternal MDD prevalence, case‐finding tool sensitivity and specificity is possible and should be done to inform valid individual and population‐level inferences. [ABSTRACT FROM AUTHOR]

Abstract Individual and population level inference about risk and burden of MDD, particularly maternal MDD, is often made using case‐finding tools that are imperfect and prone to misclassification error (i.e. false positives and negatives). These errors or biases are rarely accounted for and lead to inappropriate clinical decisions, inefficient allocation of scarce resources, and poor planning of maternal MDD prevention and treatment interventions. The argument that the use of existing maternal MDD case‐finding instruments results in misclassification errors is not new; in fact, it has been argued for decades, but by and large its implications and particularly how to correct for these errors for valid inference is unexplored. Correction of the estimates of maternal MDD prevalence, case‐finding tool sensitivity and specificity is possible and should be done to inform valid individual and population‐level inferences.

Abstract Background Accurate data on hypertension is essential to inform decision-making. Hypertension prevalence may be underestimated by population-based surveys due to misclassification of health status by participants. Therefore, adjustment for misclassification bias is required when relying on self-reports. This study aims to quantify misclassification bias in self-reported hypertension prevalence and prevalence ratios in the Portuguese component of the European Health Interview Survey (INS2014), and illustrate application of multiple imputation (MIME) for bias correction using measured high blood pressure data from the first Portuguese health examination survey (INSEF). Methods We assumed that objectively measured hypertension status was missing for INS2014 participants (n = 13,937) and imputed it using INSEF (n = 4910) as auxiliary data. Self-reported, objectively measured and MIME-corrected hypertension prevalence and prevalence ratios (PR) by sex, age group and education were estimated. Bias in self-reported and MIME-corrected estimates were computed using objectively measured INSEF data as a gold-standard. Results Self-reported INS2014 data underestimated hypertension prevalence in all population subgroups, with misclassification bias ranging from 5.2 to 18.6 percentage points (pp). After MIME-correction, prevalence estimates increased and became closer to objectively measured ones, with bias reduction to 0 pp - 5.7 pp. Compared to objectively measured INSEF, self-reported INS2014 data considerably underestimated prevalence ratio by sex (PR = 0.8, 95CI = [0.7, 0.9] vs. PR = 1.2, 95CI = [1.1, 1.4]). MIME successfully corrected direction of association with sex in bivariate (PR = 1.1, 95CI = [1.0, 1.3]) and multivariate analyses (PR = 1.2, 95CI = [1.0, 1.3]). Misclassification bias in hypertension prevalence ratios by education and age group were less pronounced and did not require correction in multivariate analyses. Conclusions Our results highlight the importance of misclassification bias analysis in self-reported hypertension. Multiple imputation is a feasible approach to adjust for misclassification bias in prevalence estimates and exposure-outcomes associations in survey data.

When the control charts for the ratio of two Poisson distributions need to be constructed, a situation may require controlling the ratio rather than a single parameter. Chakraborty and Khurshid [6] constructed Shewhart control chart and Chakraborty and Khurshid [7] studied measurement error effect on control chart for the ratio of two Poisson distributions, respectively. The effects of misclassification on the performance of control charts have been investigated by several authors. Measurement error variability has uncertainty that can arise from several sources. In this paper, we study the effect of the two sources of variability on the power characteristics of control chart under misclassification error for the ratio of two Poisson distributions as studied by Sahai and Khurshid [36]. Probabilities of misclassification of conforming and non-conforming units for grid of values are provided.

Background: When Service Provision Assessment (SPA) surveys on primary health service delivery are combined with the nationally representative household survey—Demographic and Health Survey (DHS), they can provide key information on the access, utilization, and equity of health service availability in low- and middle-income countries. However, existing linkage methods have been established only at aggregate levels due to known limitations of the survey datasets. Methods: For the linkage of two data sets at a disaggregated level, we developed a geostatistical approach where SPA limitations are explicitly accounted for by identifying the sites where health facilities might be present but not included in SPA surveys. Using the knowledge gained from SPA surveys related to the contextual information around facilities and their spatial structure, we made an inference on the service environment of unsampled health facilities. The geostatistical linkage results on the availability of health service were validated using two criteria—prediction accuracy and classification error. We also assessed the effect of displacement of DHS clusters on the linkage results using simulation. Results: The performance evaluation of the geostatistical linkage method, demonstrated using information on the general service readiness of sampled health facilities in Tanzania, showed that the proposed methods exceeded the performance of the existing methods in terms of both prediction accuracy and classification error. We also found that the geostatistical linkage methods are more robust than existing methods with respect to the displacement of DHS clusters. Conclusions: The proposed geospatial approach minimizes the methodological issues and has potential to be used in various public health research applications where facility and population-based data need to be combined at fine spatial scale. [ABSTRACT FROM AUTHOR]

This paper studies the binary classification problem associated with a family of Lipschitz convex loss functions called large-margin unified machines (LUMs), which offers a natural bridge between distribution-based likelihood approaches and margin-based approaches. LUMs can overcome the so-called data piling issue of support vector machine in the high-dimension and low-sample size setting, while their theoretical analysis from the perspective of learning theory is still lacking. In this paper, we establish some new comparison theorems for all LUM loss functions which play a key role in the error analysis of large-margin learning algorithms. Based on the obtained comparison theorems, we further derive learning rates for regularized LUMs schemes associated with varying Gaussian kernels, which maybe of independent interest. [ABSTRACT FROM AUTHOR]

Classification using linear discriminant analysis (LDA) is challenging when the number of variables is large relative to the number of observations. Algorithms such as LDA require the computation of the feature vector’s precision matrices. In a high-dimension setting, due to the singularity of the covariance matrix, it is not possible to estimate the maximum likelihood estimator of the precision matrix. In this paper, we employ the Stein-type shrinkage estimation of Ledoit and Wolf for high-dimensional data classification. The proposed approach’s efficiency is numerically compared to existing methods, including LDA, cross-validation, gLasso, and SVM. We use the misclassification error criterion for comparison.

We evaluated whether randomly sampling and testing a set number of individuals for coronavirus disease 2019 (COVID-19) while adjusting for misclassification error captures the true prevalence. We also quantified the impact of misclassification error bias on publicly reported case data in Maryland. Using a stratified random sampling approach, 50,000 individuals were selected from a simulated Maryland population to estimate the prevalence of COVID-19. We examined the situation when the true prevalence is low (0.07%–2%), medium (2%–5%), and high (6%–10%). Bayesian models informed by published validity estimates were used to account for misclassification error when estimating COVID-19 prevalence. Adjustment for misclassification error captured the true prevalence 100% of the time, irrespective of the true prevalence level. When adjustment for misclassification error was not done, the results highly varied depending on the population's underlying true prevalence and the type of diagnostic test used. Generally, the prevalence estimates without adjustment for misclassification error worsened as the true prevalence level increased. Adjustment for misclassification error for publicly reported Maryland data led to a minimal but not significant increase in the estimated average daily cases. Random sampling and testing of COVID-19 are needed with adjustment for misclassification error to improve COVID-19 prevalence estimates. [ABSTRACT FROM AUTHOR]

Sparse additive machines (SAMs) have attracted increasing attention in high dimensional classification due to their representation flexibility and interpretability. However, most of existing methods are formulated under Tikhonov regularization scheme with the hinge loss, which are susceptible to outliers. To circumvent this problem, we propose a sparse additive machine with ramp loss (called ramp-SAM) to tackle classification and variable selection simultaneously. Misclassification error bound is established for ramp-SAM with the help of detailed error decomposition and constructive hypothesis error analysis. To solve the nonsmooth and nonconvex ramp-SAM, a proximal block coordinate descent method is presented with convergence guarantees. The empirical effectiveness of our model is confirmed on simulated and benchmark datasets. [ABSTRACT FROM AUTHOR]

Background: Accurate data on hypertension is essential to inform decision-making. Hypertension prevalence may be underestimated by population-based surveys due to misclassification of health status by participants. Therefore, adjustment for misclassification bias is required when relying on self-reports. This study aims to quantify misclassification bias in self-reported hypertension prevalence and prevalence ratios in the Portuguese component of the European Health Interview Survey (INS2014), and illustrate application of multiple imputation (MIME) for bias correction using measured high blood pressure data from the first Portuguese health examination survey (INSEF).Methods: We assumed that objectively measured hypertension status was missing for INS2014 participants (n = 13,937) and imputed it using INSEF (n = 4910) as auxiliary data. Self-reported, objectively measured and MIME-corrected hypertension prevalence and prevalence ratios (PR) by sex, age group and education were estimated. Bias in self-reported and MIME-corrected estimates were computed using objectively measured INSEF data as a gold-standard.Results: Self-reported INS2014 data underestimated hypertension prevalence in all population subgroups, with misclassification bias ranging from 5.2 to 18.6 percentage points (pp). After MIME-correction, prevalence estimates increased and became closer to objectively measured ones, with bias reduction to 0 pp - 5.7 pp. Compared to objectively measured INSEF, self-reported INS2014 data considerably underestimated prevalence ratio by sex (PR = 0.8, 95CI = [0.7, 0.9] vs. PR = 1.2, 95CI = [1.1, 1.4]). MIME successfully corrected direction of association with sex in bivariate (PR = 1.1, 95CI = [1.0, 1.3]) and multivariate analyses (PR = 1.2, 95CI = [1.0, 1.3]). Misclassification bias in hypertension prevalence ratios by education and age group were less pronounced and did not require correction in multivariate analyses.Conclusions: Our results highlight the importance of misclassification bias analysis in self-reported hypertension. Multiple imputation is a feasible approach to adjust for misclassification bias in prevalence estimates and exposure-outcomes associations in survey data. [ABSTRACT FROM AUTHOR]

Administrative health databases have been used to monitor trends in infective endocarditis hospitalization related to nonprescription injection drug use (IDU) using International Classification of Diseases (ICD) code algorithms. Because no ICD code for IDU exists, drug dependence and hepatitis C virus (HCV) have been used as surrogate measures for IDU, making misclassification error (ME) a threat to the accuracy of existing estimates. In a serial cross-sectional analysis, we compared the unadjusted and ME-adjusted prevalences of IDU among 70,899 unweighted endocarditis hospitalizations in the 2007–2016 National Inpatient Sample. The unadjusted prevalence of IDU was estimated with a drug algorithm, an HCV algorithm, and a combination algorithm (drug and HCV). Bayesian latent class models were used to estimate the median IDU prevalence and 95% Bayesian credible intervals and ICD algorithm sensitivity and specificity. Sex- and age group-stratified IDU prevalences were also estimated. Compared with the misclassification-adjusted prevalence, unadjusted estimates were lower using the drug algorithm and higher using the combination algorithm. The median ME-adjusted IDU prevalence increased from 9.7% (95% Bayesian credible interval (BCI): 6.3, 14.8) in 2008 to 32.5% (95% BCI: 26.5, 38.2) in 2016. Among persons aged 18–34 years, IDU prevalence was higher in females than in males. ME adjustment in ICD-based studies of injection-related endocarditis is recommended. [ABSTRACT FROM AUTHOR]

Many research domains use data elicited from 'citizen scientists' when a direct measure of a process is expensive or infeasible. However, participants may report incorrect estimates or classifications due to their lack of skill. We demonstrate how Bayesian hierarchical models can be used to learn about latent variables of interest, while accounting for the participants' abilities. The model is described in the context of an ecological application that involves crowdsourced classifications of georeferenced coral‐reef images from the Great Barrier Reef, Australia. The latent variable of interest is the proportion of coral cover, which is a common indicator of coral reef health. The participants' abilities are expressed in terms of sensitivity and specificity of a correctly classified set of points on the images. The model also incorporates a spatial component, which allows prediction of the latent variable in locations that have not been surveyed. We show that the model outperforms traditional weighted‐regression approaches used to account for uncertainty in citizen science data. Our approach produces more accurate regression coefficients and provides a better characterisation of the latent process of interest. This new method is implemented in the probabilistic programming language Stan and can be applied to a wide number of problems that rely on uncertain citizen science data. [ABSTRACT FROM AUTHOR]

Development of methods to accurately estimate human immunodeficiency virus (HIV) incidence rate remains a challenge. Ideally, one would follow a random sample of HIV‐negative individuals under a longitudinal study design and identify incident cases as they arise. Such designs can be prohibitively resource intensive and therefore alternative designs may be preferable. We propose such a simple, less resource‐intensive study design and develop a weighted log likelihood approach which simultaneously accounts for selection bias and outcome misclassification error. The design is based on a cross‐sectional survey which queries individuals' time since last HIV‐negative test, validates their test results with formal documentation whenever possible, and tests all persons who do not have documentation of being HIV‐positive. To gain efficiency, we update the weighted log likelihood function with potentially misclassified self‐reports from individuals who could not produce documentation of a prior HIV‐negative test and investigate large sample properties of validated sub‐sample only versus pooled sample estimators through extensive Monte Carlo simulations. We illustrate our method by estimating incidence rate for individuals who tested HIV‐negative within 1.5 and 5 years prior to Botswana Combination Prevention Project enrolment. This article establishes that accurate estimates of HIV incidence rate can be obtained from individuals' history of testing in a cross‐sectional cohort study design by appropriately accounting for selection bias and misclassification error. Moreover, this approach is notably less resource‐intensive compared to longitudinal and laboratory‐based methods. [ABSTRACT FROM AUTHOR]

Scoring rules serve to quantify predictive performance. A scoring rule is proper if truth telling is an optimal strategy in expectation. Subject to customary regularity conditions, every scoring rule can be made proper, by applying a special case of the Bayes act construction studied by Grünwald and Dawid (Ann Stat 32:1367–1433, 2004) and Dawid (Ann Inst Stat Math 59:77–93, 2007), to which we refer as properization. We discuss examples from the recent literature and apply the construction to create new types, and reinterpret existing forms, of proper scoring rules and consistent scoring functions. In an abstract setting, we formulate sufficient conditions under which Bayes acts exist and scoring rules can be made proper. [ABSTRACT FROM AUTHOR]

We study the effect of imperfect training data labels on the performance of classification methods. In a general setting, where the probability that an observation in the training dataset is mislabelled may depend on both the feature vector and the true label, we bound the excess risk of an arbitrary classifier trained with imperfect labels in terms of its excess risk for predicting a noisy label. This reveals conditions under which a classifier trained with imperfect labels remains consistent for classifying uncorrupted test data points. Furthermore, under stronger conditions, we derive detailed asymptotic properties for the popular |$k$| -nearest neighbour, support vector machine and linear discriminant analysis classifiers. One consequence of these results is that the |$k$| -nearest neighbour and support vector machine classifiers are robust to imperfect training labels, in the sense that the rate of convergence of the excess risk of these classifiers remains unchanged; in fact, our theoretical and empirical results even show that in some cases, imperfect labels may improve the performance of these methods. The linear discriminant analysis classifier is shown to be typically inconsistent in the presence of label noise unless the prior probabilities of the classes are equal. Our theoretical results are supported by a simulation study. [ABSTRACT FROM AUTHOR]

In this paper, we developed both crisp and fuzzy EOQ models with proportionate discount (discount increases when number of defects decrease in each lot) for items with imperfect quality. First, we construct an optimal order quantity of crisp case. Next, proposed three different fuzzy inventory models where in the first case the defective rate is fuzzified, in the next case, both defective rate and annual demand rate are fuzzified and finally in the case of the third model all costs, defective rate and annual demand are taken to be fuzzy. Lastly, we developed the model for items with imperfect quality with inspection errors, as the inspector may commit errors while screening the lot. The probability of misclassification errors is assumed to be known. The inspection process may consist of three costs: (a) cost of inspection (b) cost of Type I errors and (c) cost of Type II errors. The defective items, classified by the inspector and the buyer, would be salvaged as a single batch that is sold at a discounted price. The objective is to find the optimal lot size for models to maximize the total profit (both for crisp and fuzzy models) and used fuzzy numbers for defective items, demand rate and/or all types of costs (exclusively for fuzzy models). We considered the triangular fuzzy numbers to represent the fuzziness of all types of costs, defective items and annual demand. Finally, the optimum order quantity is obtained using the signed distance method. A numerical example is provided to illustrate the results of the proposed models and the sensitivity analysis is conducted to know the effect of changes made for the values of different parameters on the actual lot size and the profit respectively. [ABSTRACT FROM AUTHOR]

Satellite-based land-use data sets are providing new opportunities for land-use research. However, care must be used when working with these datasets due to misclassification error, which causes inconsistent parameter estimates in typical land-use models. Results from satellite imagery data from the Northern Great Plains indicate that ignoring misclassification will lead to biased results. Even seemingly insignificant levels of misclassification error ( e.g. , 1%) result in biased parameter estimates, which alter marginal effects enough to affect policy inference. At the levels of misclassification typical in current satellite imagery datasets ( e.g. , 35%), ignoring misclassification can lead to systematically erroneous land-use policies. [ABSTRACT FROM AUTHOR]

The objective of this study was to measure the bias introduced by misclassification error when estimating the prevalence of MDD among mothers in two community-based studies. Baseline data were collected from mothers participating in two home visitation study sites in South Central United States between 2010 and 2014. The operational definition of MDD was a Center of Epidemiological Studies-Depression - Short Form ( CESD-SF ) score of 10 or higher. Misclassification error was adjusted for using CESD-SF sensitivity and specificity priors that were either antepartum or postpartum specific or non-specific. Bias was measured as the difference between the observed and misclassification error-adjusted prevalence estimates using a Binomial Bayesian Latent Class model. The proportion of mothers in the antepartum and postpartum periods confounded the level of bias in estimating MDD prevalence. When using antepartum and postpartum specific sensitivity and specificity of the CESD-SF , misclassification error led to nearly no bias in prevalence estimates. In contrast, ignoring differences in CESD-SF sensitivity and specificity between these periods showed considerable MDD prevalence bias. The use of period of assessment (antepartum versus postpartum) specific case-finding instrument diagnostic performance values is critical to the valid estimation of MDD prevalence among mothers. Studies using other case-finding instruments are needed to support this conclusion. [ABSTRACT FROM AUTHOR]

Background and aims Studies that report the relationship between alcohol consumption and disease risk have predominantly operationalized drinking according to a single baseline measure. The resulting assumption of longitudinal stability may be simplistic and complicate interpretation of risk estimates. This study aims to describe changes to the volume of consumption during the adult life-course according to baseline categories of drinking. Design A prospective observational study. Setting United Kingdom. Participants A cohort of British civil servants totalling 6838 men and 3372 women aged 34-55 years at baseline, followed for a mean 19.1 (standard deviation = 9.5) years. Measurements The volume of weekly alcohol consumption was estimated from data concerning the frequency and number of drinks consumed. Baseline categories were defined: non-current drinkers, infrequent drinkers, 0.1-50.0 g/week, 50.1-100.0 g/week, 100.1-150.0 g/week, 150.1-250.0 g/week and >250.0 g/week. For women, the highest category was defined as > 100.0 g/week. Baseline frequency was derived as 'daily or almost daily' and 'not daily or almost daily'. Trajectories were estimated within baseline categories using growth curve models. Findings Trajectories differed between men and women, but were relatively stable within light-to-moderate categories of baseline consumption. Drinking was least stable within the highest categories of baseline consumption (men: > 250.0 g/week; women: > 100.0 g/week), declining by 47.0 [95% confidence interval (CI) = 40.7, 53.2] and 16.8 g/week (95% CI = 12.6, 21.0), respectively, per 10-year increase in age. These declines were not a consequence of sudden transitions to complete abstention. Rates of decline appear greatest in older age, with trajectories converging toward moderate volumes. Conclusion Among UK civil servants, consumption within baseline drinking categories is generally stable during the life-course, except among heavier baseline drinkers, for whom intakes decline with increasing age. This shift does not appear to be driven by transitions to non-drinking. Cohorts of older people may be at particular risk of misclassifying former heavy drinkers as moderate consumers of alcohol. [ABSTRACT FROM AUTHOR]