scholarly journals Bayesian models for data missing not at random in health examination surveys

2017 ◽  
Vol 18 (2) ◽  
pp. 113-128 ◽  
Author(s):  
Juho Kopra ◽  
Juha Karvanen ◽  
Tommi Härkänen
Keyword(s):  
Data Augmentation ◽  
Missing At Random ◽  
Bias Reduction ◽  
Nonresponse Bias ◽  
Health Examination ◽  
Missing Not At Random ◽  
Additional Information ◽  
Epidemiological Surveys ◽  
Data Missing

In epidemiological surveys, data missing not at random (MNAR) due to survey nonresponse may potentially lead to a bias in the risk factor estimates. We propose an approach based on Bayesian data augmentation and survival modelling to reduce the nonresponse bias. The approach requires additional information based on follow-up data. We present a case study of smoking prevalence using FINRISK data collected between 1972 and 2007 with a follow-up to the end of 2012 and compare it to other commonly applied missing at random (MAR) imputation approaches. A simulation experiment is carried out to study the validity of the approaches. Our approach appears to reduce the nonresponse bias substantially, whereas MAR imputation was not successful in bias reduction.

Targeting Key Survey Variables at the Unit Nonresponse Treatment Stage

2020 ◽  
Author(s):  
David Haziza ◽  
Sixia Chen ◽  
Yimeng Gao
Keyword(s):  
Propensity Score ◽  
Simulation Study ◽  
Auxiliary Information ◽  
Missing At Random ◽  
Bias Reduction ◽  
Nonresponse Bias ◽  
Unit Nonresponse ◽  
Treatment Stage ◽  
Key Variables ◽  
Not Missing At Random

Abstract In the presence of nonresponse, unadjusted estimators are vulnerable to nonresponse bias when the characteristics of the respondents differ from those of the nonrespondents. To reduce the bias, it is common practice to postulate a nonresponse model linking the response indicators and a set of fully observed variables. Estimated response probabilities are obtained by fitting the selected model, which are then used to adjust the base weights. The resulting estimator, referred to as the propensity score-adjusted estimator, is consistent provided the nonresponse model is correctly specified. In this article, we propose a weighting procedure that may improve the efficiency of propensity score estimators for survey variables identified as key variables by making a more extensive use of the auxiliary information available at the nonresponse treatment stage. Results from a simulation study suggest that the proposed procedure performs well in terms of efficiency when the data are missing at random and also achieves an efficient bias reduction when the data are not missing at random. We further apply our proposed methods to 2017–2018 National Health Nutrition and Examination Survey.

scholarly journals Association between surgery with anesthesia and cognitive decline in older adults: Analysis using shared parameter models for informative dropout

2020 ◽  
pp. 1-7
Author(s):  
Katrina L. Devick ◽  
Juraj Sprung ◽  
Michelle Mielke ◽  
Ronald C. Petersen ◽  
Phillip J. Schulte
Keyword(s):  
Older Adults ◽  
Cognitive Function ◽  
Cognitive Decline ◽  
Missing At Random ◽  
Recent Analysis ◽  
Informative Dropout ◽  
Study Population ◽  
Data Missing ◽  
Clinic Study

Abstract Objectives/Goals: The association between surgery with general anesthesia (exposure) and cognition (outcome) among older adults has been studied with mixed conclusions. We revisited a recent analysis to provide missing data education and discuss implications of biostatistical methodology for informative dropout following dementia diagnosis. Methods/study population: We used data from the Mayo Clinic Study of Aging, a longitudinal study of prevalence, incidence, and risk factors for mild cognitive impairment (MCI) and dementia. We fit linear mixed effects models (LMMs) to assess the association between anesthesia exposure and subsequent trajectories of cognitive z-scores assuming data missing at random, hypothesizing that exposure is associated with greater decline in cognitive function. Additionally, we used shared parameter models for informative dropout assuming data missing not at random. Results: A total of 1948 non-demented participants were included. Median age was 79 years, 49% were female, and 16% had MCI at enrollment. Among median follow-up of 4 study visits over 6.6 years, 172 subjects developed dementia, 270 died, and 594 participants underwent anesthesia. In LMMs, exposure to anesthesia was associated with decline in cognitive function over time (change in annual cognitive z-score slope = −0.063, 95% CI: (−0.080, −0.046), p < 0.001). Accounting for informative dropout using shared parameter models, exposure was associated with greater cognitive decline (change in annual slope = −0.081, 95% CI: (−0.137, −0.026), p = 0.004). Discussion: We revisited prior work by our group with a focus on informative dropout. Although the conclusions are similar, we demonstrated the potential impact of novel biostatistics methodology in longitudinal clinical research.

Vision and Reading Deficits in Post-Concussion Patients: A Retrospective Analysis

2015 ◽  
pp. 206-213
Keyword(s):  
Retrospective Chart Review ◽  
Health Examination ◽  
Common Symptom ◽  
Reading Deficits ◽  
Grade Levels ◽  
Convergence Insufficiency ◽  
Return To Learn ◽  
Vision Deficits ◽  
Accommodative Facility

The prevalence of vision deficits in the pediatric/young adult concussion population in the private optometric practice setting remains unknown. Thus, a retrospective chart review in this area was conducted in the practice of the first author. Twenty-five consecutive patients with a medical diagnosis of concussion received a comprehensive vision and ocular health examination, which also included an objectively-based Visagraph reading assessment and clinical vergence/accommodative facility testing. Three primary categories of oculomotor-based deficits were found: convergence insufficiency (56%), accommodative insufficiency (76%), and oculomotor-based reading dysfunctions (68-82%). The most common symptom was headaches (84%), with 25% of the symptoms related to reading. 68% (15/22) were categorized as reading at least 2 grade levels below their current school grade level for reading eye movements based on the Visagraph findings. These overall findings are consistent with the general oculomotor-based/reading findings in the concussion/mTBI literature. The present results have important practical ramifications regarding the importance of preconcussion baseline oculomotor and Visagraph testing, as well as post-concussion follow-up testing, to help assess a student’s ability to return-to-learn (RTL).

scholarly journals Mean Empirical Likelihood Inference for Response Mean with Data Missing at Random

2020 ◽  
Vol 2020 ◽  
pp. 1-12
Author(s):  
Hanji He ◽  
Guangming Deng
Keyword(s):  
Empirical Likelihood ◽  
Missing At Random ◽  
Confidence Regions ◽  
Likelihood Inference ◽  
High Dimensional ◽  
Finite Sample ◽  
The Mean ◽  
Data Missing ◽  
Consistency And Asymptotic Normality ◽  
The Impact

We extend the mean empirical likelihood inference for response mean with data missing at random. The empirical likelihood ratio confidence regions are poor when the response is missing at random, especially when the covariate is high-dimensional and the sample size is small. Hence, we develop three bias-corrected mean empirical likelihood approaches to obtain efficient inference for response mean. As to three bias-corrected estimating equations, we get a new set by producing a pairwise-mean dataset. The method can increase the size of the sample for estimation and reduce the impact of the dimensional curse. Consistency and asymptotic normality of the maximum mean empirical likelihood estimators are established. The finite sample performance of the proposed estimators is presented through simulation, and an application to the Boston Housing dataset is shown.

scholarly journals Hemoperitoneum due to Splenic Laceration Caused by Colonoscopy: A Rare and Catastrophic Complication

2014 ◽  
Vol 2014 ◽  
pp. 1-3 ◽  
Author(s):  
Shiao-Han Chen ◽  
Jiann-Ruey Ong ◽  
Hon-Ping Ma ◽  
Po-Shen Chen
Keyword(s):  
Emergency Department ◽  
Abdominal Pain ◽  
Splenic Flexure ◽  
Splenic Injury ◽  
Underlying Disease ◽  
Sudden Onset ◽  
Health Examination ◽  
Asymptomatic Patients ◽  
Grade Ii

Numerous studies suggest that in asymptomatic patients, routine follow-up CT is not indicated due to the insignificant findings found on these patients. A 53-year-old man, who denied any underlying disease before, underwent colonoscopy for routine health examination. Sudden onset of abdominal pain around left upper quarter was mentioned at our emergency department. Grade II spleen laceration was found on CT scan. Splenic injury was found few hours later on the day of colonoscopy. It might result from the extra tension between the spleen and splenic flexure which varies from different positions of patients.

scholarly journals Use of static and dynamic 18F-FDopa PET parameters for detecting patients with glioma recurrence or progression

2020 ◽  
Author(s):  
Timothée Zaragori ◽  
Merwan Ginet ◽  
Pierre-Yves Marie ◽  
Veronique Roch ◽  
Rachel Grignon ◽  
...  
Keyword(s):  
Progression Free Survival ◽  
Metabolic Tumor Volume ◽  
Diagnostic Value ◽  
Normal Brain ◽  
Dynamic Parameters ◽  
Free Survival ◽  
Additional Information ◽  
Time To Peak ◽  
Glioma Recurrence

Abstract Background: Static 18 F-FDopa PET images are currently used for identifying patients with glioma recurrence/progression after treatment, although the additional diagnostic value of dynamic parameters remains unknown in this setting. The aim of the present study was to evaluate the performances of static and dynamic 18 F-FDopa PET parameters for detecting patients with glioma recurrence/progression as well as to assess further relationships with patient outcome. Fifty-one consecutive patients who underwent an 18 F-FDopa PET for a suspected glioma recurrence/progression at post-resection MRI, were retrospectively included. Static parameters including mean and maximum tumor-to-normal-brain (TBR), tumor-to-striatum (TSR) ratios, and metabolic tumor volume (MTV), as well as dynamic parameters with time-to-peak (TTP) values and curve slope, were tested for predicting: 1) glioma recurrence/progression at 6-months after the PET exam and 2) survival on longer follow-up. Results: All static parameters were significant predictors of a glioma recurrence/progression (accuracy≥94%) with all parameters also associated with mean progression-free survival (PFS) in the overall population (all p<0.001, 29.7 vs. 0.4 months for TBR max , TSR max and MTV). The curve slope was the sole dynamic PET predictor of glioma recurrence/progression (accuracy=76.5%) and was also associated with the mean PFS (p<0.001, 18.0 vs. 0.4 months). However, no additional information was provided relative to static parameters in multivariate analysis. Conclusion: Although patients with glioma recurrence/progression can be detected by both static and dynamic 18 F-FDopa PET parameters, most of this diagnostic information can be achieved by conventional static parameters.

scholarly journals ‘Pediatric Bipolar Disorder’ Rates are Still Lower than Claimed – A Re-examination of Eight Epidemiological Surveys used by an Updated Meta-analysis

2020 ◽  
Author(s):  
Peter Parry ◽  
Stephen Allison ◽  
Tarun Bastiampillai
Keyword(s):  
Bipolar Disorder ◽  
Narrative Analysis ◽  
Meta Analysis ◽  
Pediatric Bipolar Disorder ◽  
Bipolar Spectrum ◽  
Community Surveys ◽  
Bipolar Spectrum Disorder ◽  
Epidemiological Surveys ◽  
Highly Cited

Abstract Background: ‘Pediatric bipolar disorder’ (PBD) is a controversial diagnosis with varying rates of clinical diagnosis. A highly cited meta-analysis (Van Meter et al. 2011) of a dozen epidemiological surveys suggested a global community prevalence of 1.8%. This was further updated to 3.9% with eight additional surveys (Van Meter et al. 2019a). A narrative analysis (Parry et al. 2018) of the original 12 surveys concluded rates of PBD were substantially lower than 1.8% and led to a nine-article debate on the validity, overdiagnosis and iatrogenic aspects of the PBD diagnosis (e.g. Carlson and Dubicka 2019). This article extends the narrative analysis to include the eight newer community surveys.Methods: In terms of the Cochrane Handbook for Systematic Reviews of Interventions, the heterogenous community surveys were arguably unsuitable for statistical meta-analysis and warranted a narrative analysis.Results: Across all twenty surveys there was significant variation in methodologies and reported prevalence rates. Of the eight newer surveys, five (two Brazilian, one English, one Turkish, one United States) provided information of pre-adolescent rates of bipolar spectrum disorder. These pre-adolescent rates were zero or close to zero. Rates of adolescent hypomania/mania were higher, but follow-up data suggested most hypomania did not progress to adult bipolar disorder.Limitations: Methods in the original surveys vary and criteria used for various bipolar diagnoses were not always fully described. This limitation applies to a narrative analysis but also to a statistical meta-analysis.Conclusion: Bipolar disorder is very rare in childhood and rare in adolescence. PBD as a diagnostic construct fails to correlate with adult bipolar disorder and the term should be abandoned. Hypomanic syndromes in adolescence may not progress to adult bipolar disorder. Early diagnosis of bipolar disorder is important, but over-diagnosis risks adverse iatrogenic consequences.

Sign in / Sign up

Export Citation Format

Share Document