Clinic Based Surveillance for Hazards: Questionnaire, Expert, and Expert System Approaches

Details

• Personal Author:
  Bailey K ; Brown C ; Harber P ; Howard K ; Levin M ; McCoy JM
• Description:
  Clinic based surveillance for occupational hazards requires that the patient or health care worker identify the possibly causative agents. Three approaches are described in this pilot study. The Occupational Questionnaire (00) involved a standard interviewer administered questionnaire, asking patients to list "significant" exposures. The Occupational Expert Interview (OE) was performed by an experienced industrial hygienist in three stages: 1. Ask patient to list jobs, industries, and possible exposures; 2. Based on "expert" knowledge, suggest additional exposure possibilities; 3. Ask patient to confirm or deny the suspected exposures. The Expert System (ES) approach applied an artificial intelligence expert system to the basic interviewer data from OQ to determine possible paths from job or industry to exposure to disease. The system is a C-based expert system including a knowledge base of pairwise relationships and an inference engine (for search, predicate calculus, and selection of relevant paths). 30 successive unselected pulmonary patients participated in OE: 55 chemical, 10 product, and 41 activity items were added by the OE to those stated by the patient; confirmation by the patient was in 56%, 80%, and 56% of instances respectively. 36 other pulmonary patients (mainly asthmatics) were assessed by OQ and ES. The OE results above and the table show: 1. highly standardized questionnaires elicit fewer possible exposure agents than less structured human interview (stage 1 of OE); 2. expert knowledge can add significantly, inferring from job title to exposures; 3. artificial intelligence (ES) can define a large number of possible paths from job to possible disease. ES may therefore provide occupational health expertise to primary clinicians. Clinic based surveillance can help raise the possibility of exposures for clinical or public-health investigation. [Description provided by NIOSH]
• Subjects:
  Diagnosis Hazardous Substances Occupational Health Safety Software
• Keywords:
  Artificial Intelligence Computer Applications Diagnostic Techniques Hazard Recognition Occupational Medicine Surveillance
• ISSN:
  0003-0805
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Abstract or Summary
• Place as Subject:
  California ; OSHA Region 9
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Volume:
  141
• Issue:
  4
• NIOSHTIC Number:
  nn:20058584
• Citation:
  Am Rev Respir Dis 1990 Apr; 141(4)(Pt 2):A594
• Federal Fiscal Year:
  1990
• Performing Organization:
  University of California Los Angeles, Los Angeles, California
• Peer Reviewed:
  False
• Part Number:
  2
• Start Date:
  19870501
• Source Full Name:
  American Review of Respiratory Disease
• End Date:
  19930630
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:18ff8d7d866c0275da2c63d77bd5b37abb49ca3beafd5d4966e155d41da19de68d4cba0d985be8a5b40368abea9315d6a3651249a49a2ba6a9bba07cfc9112dd
• Download URL:
  https://stacks.cdc.gov/view/cdc/221643/cdc_221643_DS1.pdf
• File Type:
  [PDF - 545.09 KB ]