Predicting Depression among Patients with Diabetes Using Longitudinal Data

2015 ◽  
Vol 54 (06) ◽  
pp. 553-559 ◽  
Author(s):  
H. Jin ◽  
I. Vidyanti ◽  
P. Di Capua ◽  
B. Wu ◽  
S. Wu
Keyword(s):  
Major Depression ◽  
Regression Model ◽  
Predictive Accuracy ◽  
Patient Records ◽  
Historical Records ◽  
Multilevel Regression ◽  
Population Average ◽  
Subject Specific ◽  
Patients With Diabetes ◽  
Multilevel Regression Model

SummaryIntroduction: This article is part of the Focus Theme of Methods of Information in Medicine on “Big Data and Analytics in Healthcare”.Background: Depression is a common and often undiagnosed condition for patients with diabetes. It is also a condition that significantly impacts healthcare outcomes, use, and cost as well as elevating suicide risk. Therefore, a model to predict depression among diabetes patients is a promising and valuable tool for providers to proactively assess depressive symptoms and identify those with depression.Objectives: This study seeks to develop a generalized multilevel regression model, using a longitudinal data set from a recent large-scale clinical trial, to predict depression severity and presence of major depression among patients with diabetes.Methods: Severity of depression was measured by the Patient Health Questionnaire PHQ-9 score. Predictors were selected from 29 candidate factors to develop a 2-level Poisson regression model that can make population-average predictions for all patients and subject-specific predictions for individual patients with historical records. Newly obtained patient records can be incorporated with historical records to update the prediction model. Root-mean-square errors (RMSE) were used to evaluate predictive accuracy of PHQ-9 scores. The study also evaluated the classification ability of using the predicted PHQ-9 scores to classify patients as having major depression.Results: Two time-invariant and 10 time-varying predictors were selected for the model. Incorporating historical records and using them to update the model may improve both predictive accuracy of PHQ-9 scores and classification ability of the predicted scores. Subject-specific predictions (for individual patients with historical records) achieved RMSE about 4 and areas under the receiver operating characteristic (ROC) curve about 0.9 and are better than population-average predictions.Conclusions: The study developed a generalized multilevel regression model to predict depression and demonstrated that using generalized multilevel regression based on longitudinal patient records can achieve high predictive ability.

scholarly journals Development of a Pressure–Precipitation Transmitter

2019 ◽  
Vol 58 (11) ◽  
pp. 2453-2468
Author(s):  
Masaru Inatsu ◽  
Tamaki Suematsu ◽  
Yuta Tamaki ◽  
Naoto Nakano ◽  
Kao Mizushima ◽  
...  
Keyword(s):  
Regression Model ◽  
Daily Precipitation ◽  
Reanalysis Data ◽  
Precipitation Pattern ◽  
Multilevel Regression ◽  
Extreme Statistics ◽  
Spatial Continuity ◽  
Daily Precipitation Data ◽  
Multilevel Regression Model

AbstractA novel method is proposed to create very long term daily precipitation data for the extreme statistics by computing very long term daily sea level pressure (SLP) with the SLP emulator (a statistical multilevel regression model) and then converting the SLP into precipitation by combining statistical downscaling methods of the analog ensemble and singular value decomposition (SVD). After a review of the SLP emulator, we present a multilevel regression model constructed for each month that is based on a time series of 1000 principal components of SLPs on global reanalysis data. Simple integration of the SLP emulator provides 100-yr daily SLP data, which are temporally interpolated into a 6-h interval. Next, the pressure–precipitation transmitter (PPT) is developed to convert 6-hourly SLP to daily precipitation. The PPT makes its first-guess estimate from a composite of time frames with analogous SLP transition patterns in the learning period. The departure of SLPs from the analog ensemble is then corrected with an SVD relationship between SLPs and precipitation. The final product showed a fairly realistic precipitation pattern, displaying temporal and spatial continuity. The annual-maximum precipitation of the estimated 100-yr data extended the tail of probability distribution of the 8-yr learning data.

scholarly journals A Multilevel Regression Model for Geographical Studies in Sets of Non-Adjacent Cities

PLoS ONE ◽  
2015 ◽  
Vol 10 (8) ◽  
pp. e0133649 ◽  
Author(s):  
Marc Marí-Dell’Olmo ◽  
Miguel Ángel Martínez-Beneito
Keyword(s):  
Regression Model ◽  
Multilevel Regression ◽  
Multilevel Regression Model

scholarly journals Relative Deprivation, Absolute Deprivation, and Homicide: Testing an Interaction Between Income Inequality and Disadvantage

2018 ◽  
Vol 23 (1) ◽  
pp. 3-19
Author(s):  
Bert Burraston ◽  
Stephen J. Watts ◽  
James C. McCutcheon ◽  
Karli Province
Keyword(s):  
Income Inequality ◽  
Regression Model ◽  
Significant Interaction ◽  
Relative Deprivation ◽  
Multilevel Regression ◽  
Multilevel Regression Model

Both relative and absolute deprivation have effects on crime. These two concepts may be complementary, but much scholarship has treated them as separate. The present study assesses whether the effects of relative and absolute deprivation, measured as income inequality and disadvantage, respectively, interact in their effect on known homicide counts in U.S. counties. A multilevel regression model shows that there is a significant interaction between income inequality and disadvantage predicting homicide counts known to police. The plot of this interaction shows that when disadvantage is extremely high, increasing income inequality does not increase known homicides. The less disadvantage there is, the greater the effect of increasing income inequality on homicide counts in U.S. counties. This finding suggests that the effect of relative deprivation on known homicide is contingent on levels of absolute deprivation and vice versa. The implication of this finding is discussed.

scholarly journals Farm Gate Prices for Non-Varietal Wine in Argentina: A Multilevel Comparison of the Prices Paid by Cooperatives and Investor-Oriented Firms

2017 ◽  
Vol 16 (1) ◽  
Author(s):  
Agustina Malvido Perez Carletti ◽  
Markus Hanisch ◽  
Jens Rommel ◽  
Murray Fulton
Keyword(s):  
Multilevel Regression ◽  
Data Set ◽  
Unique Data ◽  
Market Shares ◽  
Price Effects ◽  
Theoretical Predictions ◽  
The One ◽  
The Cost ◽  
Price Differences ◽  
Multilevel Regression Model

AbstractIn this paper, we use a unique data set of the prices paid to farmers in Argentina for grapes to examine the prices paid by non-varietal wine processing cooperatives and investor-oriented firms (IOFs). Motivated by contrasting theoretical predictions of cooperative price effects generated by the yardstick of competition and property rights theories, we apply a multilevel regression model to identify price differences at the transaction level and the departmental level. On average, farmers selling to cooperatives receive a 3.4 % lower price than farmers selling to IOFs. However, we find cooperatives pay approximately 2.4 % more in departments where cooperatives have larger market shares. We suggest that the inability of cooperatives to pay a price equal to or greater than the one paid by IOFs can be explained by the market structure for non-varietal wine in Argentina. Specifically, there is evidence that cooperative members differ from other farmers in terms of size, assets and the cost of accessing the market. We conclude that the analysis of cooperative pricing cannot solely focus on the price differential between cooperatives and IOFs, but instead must consider other factors that are important to the members.

Relevance of Brain Regions’ Eloquence Assessment in Patients With a Large Ischemic Core Treated With Mechanical Thrombectomy

Neurology ◽  
2021 ◽  
pp. 10.1212/WNL.0000000000012863
Author(s):  
Basile Kerleroux ◽  
Joseph Benzakoun ◽  
Kévin Janot ◽  
Cyril Dargazanli ◽  
Dimitri Daly Eraya ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Random Forest ◽  
Regression Model ◽  
Logistic Regression Model ◽  
Mechanical Thrombectomy ◽  
Predictive Accuracy ◽  
Unmet Need ◽  
C Statistic ◽  
Ischemic Core ◽  
Functional Relevance

ObjectiveIndividualized patient selection for mechanical thrombectomy (MT) in patients with acute ischemic stroke (AIS) and large ischemic core (LIC) at baseline is an unmet need.We tested the hypothesis, that assessing the functional relevance of both the infarcted and hypo-perfused brain tissue, would improve the selection framework of patients with LIC for MT.MethodsMulticenter, retrospective, study of adult with LIC (ischemic core volume > 70ml on MR-DWI), with MRI perfusion, treated with MT or best medical management (BMM).Primary outcome was 3-month modified-Rankin-Scale (mRS), favourable if 0-3. Global and regional-eloquence-based core-perfusion mismatch ratios were derived. The predictive accuracy for clinical outcome of eloquent regions involvement was compared in multivariable and bootstrap-random-forest models.ResultsA total of 138 patients with baseline LIC were included (MT n=96 or BMM n=42; mean age±SD, 72.4±14.4years; 34.1% females; mRS=0-3: 45.1%). Mean core and critically-hypo-perfused volume were 100.4ml±36.3ml and 157.6±56.2ml respectively and did not differ between groups. Models considering the functional relevance of the infarct location showed a better accuracy for the prediction of mRS=0-3 with a c-Statistic of 0.76 and 0.83 for logistic regression model and bootstrap-random-forest testing sets respectively. In these models, the interaction between treatment effect of MT and the mismatch was significant (p=0.04). In comparison in the logistic regression model disregarding functional eloquence the c-Statistic was 0.67 and the interaction between MT and the mismatch was insignificant.ConclusionConsidering functional eloquence of hypo-perfused tissue in patients with a large infarct core at baseline allows for a more precise estimation of treatment expected benefit.

Neurocircuitries of Emotion Processing Affected in Major Depressive Disorder

2019 ◽  
pp. 101-114
Author(s):  
Anjali Sankar ◽  
Cynthia H.Y. Fu
Keyword(s):  
Major Depression ◽  
Predictive Accuracy ◽  
Emotion Processing ◽  
Affective State ◽  
Brain Regions ◽  
First Episode ◽  
Neuroimaging Data ◽  
Multivariate Pattern ◽  
Future Work ◽  
Hallmark Feature

Impairments in processing emotions are a hallmark feature of depression. Advances in neuroimaging techniques have rapidly improved our understanding of the pathophysiology underlying major depression. In this chapter, we provide an overview of influential neural models of emotion perception and regulation and discuss the neurocircuitries of emotion processing that are affected. Major depression is characterized by impairments in widespread brain regions that are evident in the first episode. Models have sought to distinguish the neural circuitry associated with recognition of the emotion, integration of somatic responses, and monitoring of the affective state. In particular, there has been a preponderance of research on the neurocircuitries affected during processing of mood-congruent negative emotional stimuli in depression. While neuroimaging correlates have been investigated and models proposed, these findings have had limited clinical applicability to date. Novel methods such as multivariate pattern recognition applied to neuroimaging data might enable identification of reliable, valid, and robust biomarkers with high predictive accuracy that can be applied to an individual. Last, we discuss avenues for extension and future work.

Sign in / Sign up

Export Citation Format

Share Document