Industrial Hygiene Monitoring How Much is Enough? When Can It Stop??

1
Industrial Hygiene MonitoringHow Much is
Enough?When Can It Stop??
Presented by Jonathan Haas, CSP, CIHOrange
Park, Florida 32073 [phone] STAR
Consultants, Inc.580 Bellerive Dr. Unit
5B Annapolis, MD 21409(410) 349-9713 (410)
757-0524 FAX www.starconsultants.net
Presented at AIHA Florida Annual Region IV
Conference Tampa, FL Friday March 24, 2006
You may download this presentation by going to
www.starconsultants.net
2
What I Typically See

• Industrial hygiene at a company location is
  handled by a safety manager
• The safety manager typically has been on the job
  for less than two years
• Industrial hygiene monitoring is conducted by a
  consultant
• Reports from prior years have not been read by
  the current safety manager

3
What I Typically See

• Monitoring is conducted to a list of contaminants
  that was prepared some time ago sometimes many
  years ago
• Monitoring is repeated this year because it was
  on the list last year and has always been
  monitored. The reason why is not known.
• Employees who are monitored are notified of the
  results
• Employees similarly exposed are not

4
What I Typically See

• Medical services, whether on site or contract
  occupational, have not been provided with
  summaries of data
• Exposure monitoring results, if provided to
  Medical services occasionally are filed in the
  employees jacket but usually are filed
  somewhere else
• Data summaries for similar exposure groups are
  not available anywhere on site

5
What I Typically See

• IH Consultants rarely are contracted to review
  all prior data and provide summaries and
  recommendations
• IH Consultants rarely are contracted to fully
  document all materials by similar exposure group,
  the toxicity, usage and controls resulting in a
  fully documented QEA
• Safety staff rarely have a process to review new
  materials to make a determination if monitoring
  is needed before full production

6
Good HSMS Programs Help

• Employee complaints (coughing, irritation, etc.)
  are signs that one or more Health and Safety
  Management Systems are not working properly.
• IH Consultant observations for defective
  ventilation, unacceptable use of PPE, or improper
  handling of materials is another sign that HSMS
  programs are defective
• What one HSMS program misses, another is supposed
  to catch see next page

7
30 Elements
WORKSITE HAZARD ANALYSIS
HAZARD PREVENTION CONTROL
SAFETY HEALTH TRAINING
MANAGEMENT LEADERSHIP
Management Commitment
Employee Involvement

• Baseline Inventories
• Routine Hazard Analysis (JHA Ergo)
• Change (Pre-use) Hazard Analysis
• IH Program
• Routine Self-Inspections
• (required, general, specific)
• Reporting System
• Investigations
• Trend Analysis

• Certified Professional Resources
• Hazard Elimination and Control Methods
• Engineering
• Admin
• PPE
• Work Rules/Practices/ Discipline
• Occupational Health Care
• Preventive Maintenance
• Tracking of Corrections
• Emergency Preparedness

• Managers
• Supervisors
• Workers
• Contractors
• Committees
• Emergencies
• PPE

• Policy
• Goals, Objectives
• Planning
• Visible Top
• Management Leadership
• Responsibility and Authority
• Line Accountability
• Resources
• Contract Worker Coverage
• Written SH Management System
• Annual Self-Evaluations

• Encouragement
• Participation
• (Committees)

1/26/89 3/25/03
8
Clients Need

• A process for determining monitoring needed based
  on a standardized qualitative exposure assessment
  process that results in a number
• A process for taking initial measurements or
  contracting for same that determines if
  additional monitoring is needed
• A process, when initial measurements indicate a
  continuing need for monitoring, for determining
  the number of samples needed to make the next
  decision

9
Hypothetical Case 1

• Client has determined that a new material should
  be monitored
• Discussions with employees reveal that certain
  situations results in worse exposure, based on
  their knowledge of the job
• Monitoring of those worst case exposures using a
  validated method of collection with analysis in a
  AIHA accredited lab shows all samples Below
  Detection Limit
• Do you recommend additional tests?

10
Hypothetical Case 2

• Client has determined that a new material should
  be monitored
• Discussions with employees reveal that certain
  situations results in worse exposure, based on
  their knowledge of the job
• Monitoring of those worst case exposures using a
  validated method of collection with analysis in a
  AIHA accredited lab shows all samples above the
  Exposure Limit
• Do you recommend additional tests?

11
Hypothetical Case 3

• Client has determined that a new material should
  be monitored
• Discussions with employees reveal that certain
  situations results in worse exposure, based on
  their knowledge of the job
• Monitoring of those worst case exposures using a
  validated method of collection with analysis in a
  AIHA accredited lab shows 50 of the samples
  above the Exposure Limit
• Do you recommend additional tests?

12
Hypothetical Case 4

• Client has determined that a new material should
  be monitored
• Discussions with employees reveal that certain
  situations results in worse exposure, based on
  their knowledge of the job
• Monitoring of those worst case exposures using a
  validated method of collection with analysis in a
  AIHA accredited lab shows none of the samples
  above the Exposure Limit
• Do you recommend additional tests?

13
Professional Judgment

• If you are called to testify in a court case
  involving your monitoring, your exposure
  assessment determination process will be attacked
• If you used a validated sampling method and
  accredited lab, that is an expected question
• What will you say when a statistical analysis of
  the samples you took, which were all below the
  exposure limit, shows that 5 overexposures are
  predicted?? Did you advise the client?

14
Document The Process

• Your client needs to understand how you make
  recommendations for added sampling, fewer
  samples, or the termination of sampling
• Pulling the cloak over your head and saying
  Trust me I know what Im doing makes you
  either Obi Wan or the Evil Sith depending on who
  is asked
• Showing your client a strategy based on
  statistics and probability, with graphs of
  examples, gains you credibility
• You will not be stammering when attacked as a
  witness

15
Industrial Hygiene Data

• Before we can debate how much is needed and when
  can we stop, we need to debate how much predicted
  overexposure is acceptable
• The debate is NOT about how much actual
  overexposure is acceptable, but is about how much
  predicted overexposure is acceptable

16
How Much Exceedance is OK?

• Does the acceptable percentage predicted to
  exceed the limit depend on what the materials
  toxicity the reason for the limit?
• Does the acceptable percentage predicted to
  exceed the limit depend on who is asked?
• Employees
• Management
• The industrial hygienist

Yes
17
What Predicted Exceedance is OK?
Situation A
Situation B

• Reproductive toxin
• Adequate controls
• Good PM for controls
• Knowledgeable people
• Good concern handling
• Great supervision
• Good mgmt of change
• Your daughter works it

• Reproductive Toxin
• Iffy controls
• Poor PM for controls
• MSDS available
• Poor concern handling
• Weak supervision
• No mgmt of change
• Your daughter works it

Do controls change the answer?
18
The Wall Street Journal Test

• How would management react to having a Wall
  Street Journal article describe how they were
  knowingly permitting employees to be potentially
  overexposed to a reproductive toxin, based on
  available data?
• And describing how one out of five children have
  observable birth defects and that data predicts
  that your plant averages one day a year over the
  exposure limit? (0.4 of 250 days)

With anger, right?
19
What Predicted Exceedance is OK?
Situation A
Situation B

• Carcinogen
• Adequate controls
• Good PM for controls
• Knowledgeable people
• Good concern handling
• Great supervision
• Good mgmt of change
• Your daughter works it

• Carcinogen
• Iffy controls
• Poor PM for controls
• MSDS available
• Poor concern handling
• Weak supervision
• No mgmt of change
• Your daughter works it

Do controls change the answer?
20
What Predicted Exceedance is OK?
Situation A
Situation B

• Lung damage
• Adequate controls
• Good PM for controls
• Knowledgeable people
• Good concern handling
• Great supervision
• Good mgmt of change
• Your daughter works it

• Lung damage
• Iffy controls
• Poor PM for controls
• MSDS available
• Poor concern handling
• Weak supervision
• No mgmt of change
• Your daughter works it

Do controls matter, really?
21
What Predicted Exceedance is OK?
Situation A
Situation B

• Irritant only
• Adequate controls
• Good PM for controls
• Knowledgeable people
• Good concern handling
• Great supervision
• Good mgmt of change
• Your daughter works it

• Irritant only
• Iffy controls
• Poor PM for controls
• MSDS available
• Poor concern handling
• Weak supervision
• No mgmt of change
• Your daughter works it

Do controls matter, really?
22
Available Software

• Really easy to use Ihstat.xls
• Determines whether samples are normally
  distributed or log normally distributed
• Determines predicted to exceed limit
• Really terrific to use Minitab
• Same as above, plus terrific graphs
• Used by many for Six Sigma projects
• Obsolete Logan
• Unusable NIOSH Decision Logic graphs

23
(No Transcript)
24
Here is a Minitab graph of the same data
25
Monitoring Decisions

• Starting from no data at all

26
IH Monitoring Decisions Based on Statistical Data
Results
Take three Worst Case samples
Reduce exposure
Yes
Any sample gt full or short term limit
No
Investigate Reduce Repeat
Yes
All lt limit either mean gt50 of limit
No
Yes
All lt limit both means lt50 of limit
Start PM
(See next page)
No
Yes
All three lt 10 of limit
No more samples needed
Calculated STEL Full shift limit 0 to 1 PPM
STEL 3 times full shift limit is based on
full shift Full shift limit 2 to 10 PPM STEL
2 times exposure limit when Full
shift limit 11 to 100 PPM STEL 1.5 times
none is published Full shift limit gt100
PPM STEL 1.25 times
Worst case sampling can be personal samples or
area samples and can last for only as long as the
agent is handled that day. If sampling is just
for 15 to 30 minutes, the result is compared to a
Short Term Limit ( or calculated STEL ). The
result is also to be converted to an 8-hour time
weighted average and compared against that full
shift exposure limit.
27
Worst Case Both means gt 50 Limit Investigate,
Reduce, Repeat
28
IH Monitoring Decisions Based on Statistical Data
Results
Take three Worst Case samples
Reduce exposure
Yes
Any sample gt full or short term limit
No
Investigate Reduce Repeat
Yes
All lt limit either mean gt50 of limit
No
Yes
All lt limit both means lt50 of limit
Start PM
(See next page)
No
Yes
All three lt 10 of limit
No more samples needed
Calculated STEL Full shift limit 0 to 1 PPM
STEL 3 times full shift limit is based on
full shift Full shift limit 2 to 10 PPM STEL
2 times exposure limit when Full
shift limit 11 to 100 PPM STEL 1.5 times
none is published Full shift limit gt100
PPM STEL 1.25 times
Worst case sampling can be personal samples or
area samples and can last for only as long as the
agent is handled that day. If sampling is just
for 15 to 30 minutes, the result is compared to a
Short Term Limit ( or calculated STEL ). The
result is also to be converted to an 8-hour time
weighted average and compared against that full
shift exposure limit.
29
Worst Case Both means lt 50 LimitStart Periodic
Monitoring
30
IH Monitoring Decisions Based on Statistical Data
Results
Take three Worst Case samples
Reduce exposure
Yes
Any sample gt full or short term limit
No
Investigate Reduce Repeat
Yes
All lt limit either mean gt50 of limit
No
Yes
All lt limit both means lt50 of limit
Start PM
(See next page)
No
Yes
All three lt 10 of limit
No more samples needed
Calculated STEL Full shift limit 0 to 1 PPM
STEL 3 times full shift limit is based on
full shift Full shift limit 2 to 10 PPM STEL
2 times exposure limit when Full
shift limit 11 to 100 PPM STEL 1.5 times
none is published Full shift limit gt100
PPM STEL 1.25 times
Worst case sampling can be personal samples or
area samples and can last for only as long as the
agent is handled that day. If sampling is just
for 15 to 30 minutes, the result is compared to a
Short Term Limit ( or calculated STEL ). The
result is also to be converted to an 8-hour time
weighted average and compared against that full
shift exposure limit.
31
Worst Case Both means lt 10 LimitNo Predicted
ExceedanceStop Monitoring
32
Monitoring Decisions

• Based on Results of Periodic Monitoring

33
Periodic Monitoring Decisions based on Statistics
Take 8 full-shift samples, random
Yes
Conduct exposure assessment Worst Case
Any sample gt Limit
No

Yes
Conduct exposure assessment Worst Case
Predicted gtLimit gt 5
No
Yes
Predicted gtLimit 0.1 - 5
Take 8 random samples /year
No
Yes
Predicted gtLimit lt 0.1
Take 4 random samples / year

No
Yes
Predicted gtLimit 0
Stop monitoring for this.
34
Exceedance Predicted 1.4Continue at 8 samples /
year
35
Periodic Monitoring Decisions based on Statistics
Take 8 full-shift samples, random
Yes
Conduct exposure assessment Worst Case
Any sample gt Limit
No

Yes
Conduct exposure assessment Worst Case
Predicted gtLimit gt 5
No
Yes
Predicted gtLimit 0.1 - 5
Take 8 random samples /year
No
Yes
Predicted gtLimit lt 0.1
Take 4 random samples / year

No
Yes
Predicted gtLimit 0
Stop monitoring for this.
36
Exceedance Predicted 0.3Continue at 8 samples /
year
37
Definitely looks lognormal when plotted
38
Periodic Monitoring Decisions based on Statistics
Take 8 full-shift samples, random
Yes
Conduct exposure assessment Worst Case
Any sample gt Limit
No

Yes
Conduct exposure assessment Worst Case
Predicted gtLimit gt 5
No
Yes
Predicted gtLimit 0.1 - 5
Take 8 random samples /year
No
Yes
Predicted gtLimit lt 0.1
Take 4 random samples / year

No
Yes
Predicted gtLimit 0
Stop monitoring for this.
39
Exceedance Predicted lt0.1Reduce to 4 samples /
year
40
Periodic Monitoring Decisions based on Statistics
Take 8 full-shift samples, random
Yes
Conduct exposure assessment Worst Case
Any sample gt Limit
No

Yes
Conduct exposure assessment Worst Case
Predicted gtLimit gt 5
No
Yes
Predicted gtLimit 0.1 - 5
Take 8 random samples /year
No
Yes
Predicted gtLimit lt 0.1
Take 4 random samples / year

No
Yes
Predicted gtLimit 0
Stop monitoring for this.
41
Exceedance Predicted 0.000Communicate Stop
Sampling
42
Precautions

• Some regulations mandate monitoring
• Changes in the workplace will change QEA
• Production rate
• Changed ventilation
• Different chemical composition
• Changed methods for handling
• Even though nothing has changed, management may
  expect additional sampling just because it is
  time to confirm alls well

43
Summary

• Statistics and probability can be used to modify
  sampling frequency
• Use of graphs to communicate data to similar
  exposure groups will enhance HAZCOM
• Use of graphs will enhance Medical program
  understanding of workplace stressors
• Use of graphs for tracking and trending is an
  expectation of VPP under OSHA