Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring,...
Transcript of Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring,...
![Page 1: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/1.jpg)
Industry Perspectives on
Environmental Epidemiology
November 21, 2017
Carol Burns, MPH, PhD
Burns Epidemiology Consulting
![Page 2: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/2.jpg)
Toxicology
approach
Glass is half empty
Page 54, Figure 4. EFSA 2017 Scientific Opinion
![Page 3: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/3.jpg)
Epidemiology
approach
Glass is half full
Page 54, Figure 4, 2017 EFSA Scientific Opinion
![Page 4: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/4.jpg)
How do we fill the glass?
![Page 5: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/5.jpg)
ECPA Funding exposure validation study (IOM)
– Builds from US Farm Family Exposure Study
CLI workshop (2016)
– Better studies yield better decisions
ILSI HESI (Government, Industry, Academia)
– Evaluating uncertainty in epidemiology
• Workshop and publication 2012
– Improving application of epidemiology
• In progress, multiple focus groups for 2018
Industry is involved
![Page 6: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/6.jpg)
Glow – areas of agreement
Grow – areas for discussion
The weeds – points to ponder
Outline*
* 2017: Scientific Opinion of the PPR Panel on the follow-up of the findings of the External Scientific Report “Literature review of epidemiological studies linking exposure to pesticides and health effects”
![Page 7: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/7.jpg)
Quality is important
Statistical significance is subjective
– More than yes/no
– Effect size magnification occurs
Direction of misclassification bias is
unpredictable
Evidence integration should rely on best data
“GLOW”
Areas of agreement
![Page 8: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/8.jpg)
Exposure – adequate assessment, preferably to
allow for dose-response
Outcome – well defined clinical entities (or
validated surrogates
Analysis – adequate accounting for
confounding
Analysis – subgroups by gender, age, ethnicity
“GLOW”
Certain epidemiology parameters should be
considered for quality and relevance
![Page 9: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/9.jpg)
Gold standard rarely possible
No minimum requirements for study quality
– Can more detail be provided?
– What if only low-quality studies are available?
No explicit guidance for systematic
review/meta-analyses
– Only as good as the contributing studies and
selected findings
No recommendations for evidence integration
– With toxicity, mechanistic and epidemiology data
“GROW”
Areas for discussion
![Page 10: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/10.jpg)
Distinguish hypothesis-generating vs. hypothesis-
testing studies
– Is study done for regulatory purpose?
Define biologically relevant effects
– Clinical diagnosis vs. mean difference?
– Is statistical significance adequate?
Discuss dose-response more fully
– How would you use an Odds Ratio?
– Is there concordance of exposure-outcome findings
“GROW”
Specific Issues to develop
![Page 11: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/11.jpg)
What does quality assessment look like?
– A case study of pyrethroid epidemiology studies
(Burns and Pastoor, Crit Rev Tox, in review)
Example
Adverse Outcomes
![Page 12: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/12.jpg)
Quality Outcome Evaluation
OUTCOME - EFSA/OPP OUTCOME
Example
TIER 1
Valid and reliable outcome
assessment
Medical Record or
diagnosis confirmed
TIER 2 Standardized outcome, not
validated in population, or
screening tool; or, medical record
non-confirmed.
Self -reported disease,
symptoms,
test scores, screening tool.
Multiple biological samples
TIER 3 Non-standardized and non-
validated health outcome.
Single sample of a short-
lived chemical or hormone
Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al.,
2014)
![Page 13: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/13.jpg)
Outcome quality varies by effect (examples from pyrethroid epidemiology)
Birth weight
Developmental delay
Sperm quality
0 0
8
Outcome
0
7
1
Outcome
![Page 14: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/14.jpg)
“Outcomes under study should be well-defined”
– What is adverse? An IQ point? Mean difference?
– What is biologically relevant? (EFSA 2017a)
– Are there differences in screening vs. diagnostic?
“Use should be made of biological markers…”
– How can they be validated?
– Should multiple collections be taken?
Example: Adverse outcome (8.1c)
Can we have more detail?
![Page 15: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/15.jpg)
Hazard identification
– Relies upon transparency and complete reporting
Exposure assessment
– Impacted by quality, validity, and reliability
Dose-response assessment
– Requires more than low vs. high
Risk characterization
– Is epidemiology for reassurance or “what if”?
Concluding…
Risk Assessment Process
![Page 16: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/16.jpg)
The weeds (1 of 3)
Synchronization of results
How can EFSA
connect the TIMING
between review
process and the output
of epidemiological
studies?
Can data analyses of
pesticide X in the AHS
occur before (re)-
registration of pesticide
X?
Page 9, 2017 EFSA Scientific Opinion
![Page 17: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/17.jpg)
The weeds
Transparency of results
How can null data be viewed
as important?
Can dose-response be used
to identify a no-effect level?
– When does a null study confirm
a NOEL (or BMD)?
Can we use existing models
from other disciplines?
– Can results (or data) be shared
online before peer review?
![Page 18: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/18.jpg)
“propose potential refinements for future
epidemiological studies to increase the quality,
relevance and reliability of the findings and how
they may impact pesticide risk assessment”
Are there incentives for investigators?
• “Discovery” vs. “regulatory” science
• “Our” problems are not “their” problems
Can barriers be reduced?
• Quality can be expensive, time consuming
The weeds
Can EFSA influence change?
![Page 19: Industry Perspectives on Environmental Epidemiology · 2017-12-05 · Approach of Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (LaKind et al., 2014) Outcome](https://reader034.fdocuments.in/reader034/viewer/2022050406/5f83892b5478776f2e67a9b5/html5/thumbnails/19.jpg)
Questions
Mouse + Man: Fill the glass
with best data