Refining the Exposure Assessment Decision Making Using Quantitative Data

1
Refining the Exposure Assessment Decision
Making Using Quantitative Data

• S. Smallets Jr., MS, CIH
• NAVMEDCEN Portsmouth

2
Exposure Assessments Central Role
A Strategy for Assessing and Managing
Occupational Exposures 1998 Fig 1.1 pg 3
3
AIHAExposure Assessment Process
OPNAVINST 5100.23F Chapter 8
A Strategy for Assessing and Managing
Occupational Exposures 1998
4
Figure 5.1
Defining and Judging Exposure Profiles
Basic Characterization
Environmental
Workplace
Workforce
Agent
Information
Information
Information
Establish Similar Exposure Groups
Exposure
Define Exposure Profile
Select/DefineOELs
Assessment
Compare
Exposure Profile
OEL and its Uncertainty
and its
Uncertainty
Acceptable
Uncertain
Unacceptable
A Strategy for Assessing and Managing
Occupational Exposures 1998
rjoe\expopros.ppt
5
Define Exposure Profile

• Characterization of the day to day variability of
  exposure levels for a SEG
• Requires an estimate of exposure levels and
  variability
• Quantitative (6-10 samples)
• Point estimates i.e. means, standard deviation,
  percentiles confidence limits
• Qualitative
• Knowledge, experience

pdf
6
Judging Exposures

• Following the definition of the exposure profile
  and selection of OEL
• SEG must be judged acceptable, unacceptable
• Resolved by comparing profile with OEL

A Strategy for Assessing and Managing
Occupational Exposures 1998
7
Focus on Upper-percentile or Mean ?
95ile
UTL
0.014
AM and UCL
0.012
OEL
LTA-OEL
0.01
0.008
0.006
0.004
0.002
0
0
50
100
150
200
250
300
Concentration
8
Criteria for assessing exposures (NIOSH 77-173
Occupational Exposure Sampling Manual)

• Employer should have high confidence that a high
  percentage of exposures are below the standard
• The employer should try to attain 95 confidence
  that no more than 5 of employee days are over
  the standard
• Philosophy driving development of 50 Action
  Level (AL)

9
Focus on upper tail --95th percentile and UTL
95,95
95th Percentile Point Estimate
0.014
UTL95,95
0.012
OEL
0.01
0.008
0.006
0.004
95 Certain that 95 of the exposures are less
than this concentration
0.002
0
0
50
100
150
200
250
300
Concentration
10
IHFOM Exposure ClassificationsAcceptable-
UTL95,95 ltOEL
UTL95,95
0.014
95ile
0.012
OEL
0.01
0.008
0.006
0.004
0.002
0
0
50
100
150
200
250
300
Concentration
11
IHFOM Exposure Classifications Unacceptable-
UTL95,95 gt OEL
95ile
UTL95,95
0.014
OEL
0.012
0.01
0.008
0.006
0.004
0.002
0
0
50
100
150
200
250
300
Concentration
12
IHFOM Exposure Classifications Unacceptable-
UTL95,95 gt OEL but 95 ilelt OEL
0.014
95ile
UTL95,95
0.012
OEL
0.01
0.008
0.006
0.004
0.002
0
0
50
100
150
200
250
300
Concentration
13
Examples 95th tile lt OEL and UTL gt OEL
- UG
14
Calculation of 95th Percentile and UTL 95,95
15
Tolerance Limits

• Tuggle, R. M. Assessment of Occupational
  exposure using one-sided tolerance limits Am.
  Ind. Hyg. Assoc. J (43) 338-346 (1982)
• Proposed EA scheme using UTL95,95
• Proposed how to proceed when UTL95,95 gt OEL but
  95 ilelt OEL

16
Comparison of Z and K

• Z
• 95th percentile1.645

• K
• Depends on sample size
• N 4 K 7.655
• N 6 K 3.707
• N 9 K 3.031
• N10 K 2.911
• N50 K 2.065

17
Test Statistic
Test Statistic represents how many sample
standard deviations the sample mean is from the
OEL
18
Comparison of Z to K
Test Statistic represents how many sample
standard deviations the sample mean is from the
OEL
Tuggle, R. M. Assessment of Occupational
exposure using one-sided tolerance limits Am.
Ind. Hyg. Assoc. J (43) 338-346 (1982)
19
Decision Criteria
A- Acceptable B- Unacceptable C- Future
Acceptable D- Acceptable
Tuggle, R. M. Assessment of Occupational
exposure using one-sided tolerance limits Am.
Ind. Hyg. Assoc. J (43) 338-346 (1982)
20
One Side Tolerance Limits Decision Scheme
Tuggle, R. M. Assessment of Occupational
exposure using one-sided tolerance limits Am.
Ind. Hyg. Assoc. J (43) 338-346 (1982)
21
How many more samples?

• Sanding SrCrO4
• OEL 0.5
• N6
• GM0.05
• GSD2.119
• 95th 0.18
• UTL0.9

N 9 K 3.031 N 10 K 2.911 N50 K 2.065
22
Applied to all examples
ug
23
Suggestions

• Use 95th ile as decision point
• Use UTL as a measure of certainty in point
  estimate
• When 95thile ltOEL and UTL gt OEL
• Base decision on 95th percentile rather than UTL
  and classify as acceptable
• Determine how many samples to get TSgtK
• If reasonable consider obtaining the samples
• Use of UTL may require more than 6-10 samples