Your browser doesn't support javascript.
loading
Show: 20 | 50 | 100
Results 1 - 2 de 2
Filter
Add more filters










Database
Language
Publication year range
1.
Stat Med ; 35(9): 1488-501, 2016 Apr 30.
Article in English | MEDLINE | ID: mdl-26626135

ABSTRACT

The Generalised linear mixed model (GLMM) is widely used for modelling environmental data. However, such data are prone to influential observations, which can distort the estimated exposure-response curve particularly in regions of high exposure. Deletion diagnostics for iterative estimation schemes commonly derive the deleted estimates based on a single iteration of the full system holding certain pivotal quantities such as the information matrix to be constant. In this paper, we present an approximate formula for the deleted estimates and Cook's distance for the GLMM, which does not assume that the estimates of variance parameters are unaffected by deletion. The procedure allows the user to calculate standardised DFBETAs for mean as well as variance parameters. In certain cases such as when using the GLMM as a device for smoothing, such residuals for the variance parameters are interesting in their own right. In general, the procedure leads to deleted estimates of mean parameters, which are corrected for the effect of deletion on variance components as estimation of the two sets of parameters is interdependent. The probabilistic behaviour of these residuals is investigated and a simulation based procedure suggested for their standardisation. The method is used to identify influential individuals in an occupational cohort exposed to silica. The results show that failure to conduct post model fitting diagnostics for variance components can lead to erroneous conclusions about the fitted curve and unstable confidence intervals.


Subject(s)
Linear Models , Data Interpretation, Statistical , Datasets as Topic , Environmental Exposure/statistics & numerical data , Humans , Models, Statistical
2.
Ergonomics ; 55(4): 396-414, 2012.
Article in English | MEDLINE | ID: mdl-22397385

ABSTRACT

A cohort of 536 workers was enrolled from 10 diverse manufacturing facilities and was followed monthly for six years. Job physical exposures were individually measured. Worker demographics, medical history, psychosocial factors, current musculoskeletal disorders (MSDs) and nerve conduction studies (NCS) were obtained. Point and lifetime prevalence of carpal tunnel syndrome (CTS) at baseline (symptoms + abnormal NCS) were 10.3% and 19.8%. During follow-up, there were 35 new CTS cases (left, right or both hands). Factors predicting development of CTS included: job physical exposure (American conference of governmental industrial hygienists Threshold Limit Value (ACGIH TLV) for Hand Activity Level (HAL) and the Strain Index (SI)), age, BMI, other MSDs, inflammatory arthritis, gardening outside of work and feelings of depression. In the adjusted models, the TLV for HAL and the SI were both significant per unit increase in exposure with hazard ratios (HR) increasing up to a maximum of 5.4 (p = 0.05) and 5.3 (p = 0.03), respectively; however, similar to other reports, both suggested lower risk at higher exposures. Data suggest that the TLV for HAL and the SI are useful metrics for estimating exposure to biomechanical stressors. PRACTITIONER SUMMARY: This study was conducted to determine how well the TLV for HAL and the SI predict risk of CTS using a prospective cohort design with survival analysis. Both the TLV for HAL and the SI were found to predict risk of CTS when adjusted for relevant covariates.


Subject(s)
Carpal Tunnel Syndrome/physiopathology , Hand/physiopathology , Occupational Diseases/physiopathology , Threshold Limit Values , Adult , Biomechanical Phenomena , Cohort Studies , Confounding Factors, Epidemiologic , Effect Modifier, Epidemiologic , Female , Humans , Male , Models, Statistical , Neural Conduction , Occupational Health , Survival Analysis
SELECTION OF CITATIONS
SEARCH DETAIL
...