Document Detail

Inside the black box: starting to uncover the underlying decision rules used in a one-by-one expert assessment of occupational exposure in case-control studies.
MedLine Citation:
PMID:  23155187     Owner:  NLM     Status:  MEDLINE    
OBJECTIVES: Evaluating occupational exposures in population-based case-control studies often requires exposure assessors to review each study participant's reported occupational information job-by-job to derive exposure estimates. Although such assessments likely have underlying decision rules, they usually lack transparency, are time consuming and have uncertain reliability and validity. We aimed to identify the underlying rules to enable documentation, review and future use of these expert-based exposure decisions.
METHODS: Classification and regression trees (CART, predictions from a single tree) and random forests (predictions from many trees) were used to identify the underlying rules from the questionnaire responses, and an expert's exposure assignments for occupational diesel exhaust exposure for several metrics: binary exposure probability and ordinal exposure probability, intensity and frequency. Data were split into training (n=10 488 jobs), testing (n=2247) and validation (n=2248) datasets.
RESULTS: The CART and random forest models' predictions agreed with 92-94% of the expert's binary probability assignments. For ordinal probability, intensity and frequency metrics, the two models extracted decision rules more successfully for unexposed and highly exposed jobs (86-90% and 57-85%, respectively) than for low or medium exposed jobs (7-71%).
CONCLUSIONS: CART and random forest models extracted decision rules and accurately predicted an expert's exposure decisions for the majority of jobs, and identified questionnaire response patterns that would require further expert review if the rules were applied to other jobs in the same or different study. This approach makes the exposure assessment process in case-control studies more transparent, and creates a mechanism to efficiently replicate exposure decisions in future studies.
David C Wheeler; Igor Burstyn; Roel Vermeulen; Kai Yu; Susan M Shortreed; Anjoeka Pronk; Patricia A Stewart; Joanne S Colt; Dalsu Baris; Margaret R Karagas; Molly Schwenn; Alison Johnson; Debra T Silverman; Melissa C Friesen
Related Documents :
23349287 - Identification of the long-sought common-envelope events.
9663897 - Clinical validation of pd adequest software: modeling error assessment.
23852837 - Predicting developmental disorder in infants using an artificial neural network.
23314707 - A comparison of statistical methods for deriving freshwater quality criteria for the pr...
7235547 - How much lift in a facelift?
24262197 - Re-thinking the biology of essential tremor: from models to morphology.
Publication Detail:
Type:  Journal Article; Research Support, N.I.H., Intramural; Validation Studies     Date:  2012-11-15
Journal Detail:
Title:  Occupational and environmental medicine     Volume:  70     ISSN:  1470-7926     ISO Abbreviation:  Occup Environ Med     Publication Date:  2013 Mar 
Date Detail:
Created Date:  2013-02-11     Completed Date:  2013-04-04     Revised Date:  2014-04-10    
Medline Journal Info:
Nlm Unique ID:  9422759     Medline TA:  Occup Environ Med     Country:  England    
Other Details:
Languages:  eng     Pagination:  203-10     Citation Subset:  IM    
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Case-Control Studies
Decision Making*
Middle Aged
Models, Statistical
Occupational Exposure / analysis*
Reproducibility of Results
Research Design / standards*
Vehicle Emissions*
Grant Support
Reg. No./Substance:
0/Vehicle Emissions

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Previous Document:  Brentuximab vedotin.
Next Document:  Pattern and predictors for respiratory illnesses and symptoms and lung function among textile worker...