Paul Switzer

1
Paul Switzer
MPEPM by H. Ö.

• aggregation of micro-environments
• temporal spatial variability of concentrations
  in ME studies
• estimate percentiles of population exposure
• how affected by source mitigation
• how to assess indoor / outdoor relation
• time aggregation issues personal
• population stratification vs ambient
• 60 of indoor is outdoor (vs 25)?
• deposition K proportional to concentration?
• more details for Monte Carlo study
• student data - lognormal?
• over time
• over students
• m.e. exposure time variability affects estimates
  of source mitigation efficacy

2
Jerry Sacks
MODELS (1) PRE-SELECTION OF DATA Example - O3
Respiratory Hospital Admissions O3 is much
higher in summer. Respiratory Hospital Admissions
peak in January February
3
Jerry Sacks
MODELS (2) VARIATION IN THE SUSCEPTIBLE
POOL (RISK-WEIGHTING)
4
Jerry Sacks
MODELS (3) EXAMPLES - - HEAT O3 - PM10
WINTER CONDITIONS
5
Jerry Sacks
MODELS (4) Effect of HEAT O3 greater than
either alone Effect of PM10 exposure greater in
winter (because susceptible pool is greater) than
at other times.
6
Jerry Sacks
MODELS (5) It follows that there will be
inconsistency in the effect of a given pollution
increase. Evidence of inconsistency is not
evidence against a causal inference.
7
Jerry Sacks
MODELS (6) Evidence of an Association getting
stronger when the exposure is more accurately
measured, is evidence in favor of a causal
inference.
8
Allan Marcus
Berkson errors can look like classical errors if
exposure-response is nonlinear
Response
Exposure
True for non-cooker
Assigned from Stationary Monitor
True for cooker
True for smoker
9
Allan Marcus
Models for Microenvironment Measurement Error
a ventilation rate (air exchanges per hour) P
penetration fraction for ME (P _at_ 0.9 for PM2.5,
P _at_ 0.7 for PM10) K deposition rate (per
hour) F random variation in Cambient adjusting
for wind direction, local sources, (PM) filter
errors and flow rate errors Qsource ME source
emission rate (mg PM per hour) V volume of
indoor / personal space
10
Allan Marcus
Recent Results on EPA Personal Monitoring for
Riverside CA (PTEAM)
a ventilation, K deposition, t fraction of
time spent indoors Estimated indoor exposure to
ambient penetrating indoors Within-subject
(longitudinal) Epersonal shows little
correlation with Coutdoors in some
subjects, high correlation in others.
is almost uncorrelated with Coutdoors.
Coutdoors, visit j, is highly correlated with
.
Is the ME model adequate for indoor source
variability?
11
Adrian Raftery
Discussion of Merlise Clydes paper
Goal Towards causal inference from observational
env. data
Obstacles
OR
(time ordering)
Z
OR
OR
What should Z be? R.A. Fisher Make your model
as big as an elephant. Here dim(Z) 200.
.
12
Adrian Raftery
(very) inefficient
But include all Zs

• Statistical variable selection
• Standard method (S, SAS,) stepwise condition
  on selected model
• (frequentist) properties unknown
• underestimates uncertainty
• overstates significance
• understates SEs, CIs.
• can be VERY misleading (Freedman 83).
• BMA
• accounts for model uncertainty
• optimal predictive performance (MR 94)
• tests minimize total (frequentist) error rates
  ( Type IType II) (Jeffreys 61).

.
13
Adrian Raftery

• Designing a REALISTIC simulation study
• We analyzed the 49 relevant case-control
  studies in Amer. J. Epi. in 1996
• and based our simulation study on these.
• IQRs of reported ORs

• Simulation design
• 32 X-variables
• 10 of these associated with Y
• 22 of these indep. of Y

.
14
Results Standard approaches
Adrian Raftery
BMA
.
Point estimation MSE of
is lower for BMA.
Best BMAlt 2-stage lt stepwise worst.
15
Adrian Raftery

• Merlise
• BMA over large q
• Presentation of results
• Many cute tricks

Orthogonalization Recall BMA Û inference for
the full model
with prior point mass on
Justified by prior beliefs / approx. prior mass
on
1st version of paper orthogonalize X. Very fast,
but what prior does it correspond to?
.
16
Adrian Raftery
E.g. X1 PM X2 temperature Then PCs
model
BMA prior puts point mass on
Different example X1 PMt X2 PMt-1 Then W1
general PM level W2 increase in PM
since yesterday Then the hypotheses may make
sense.
.
17
Adrian Raftery
Þ Orthogonalization may make sense for groups of
variables measuring the same thing. Better
alternative? Measurement model LISREL (Joreskog)
?
X
Y
(latent)
(latent)
meas. error
(obs.)
(obs.)
meas. error
18
Larry Cox
Workshop Summary

• Objectives
• immerse statisticians in the scientific
  framework / problems for PM
• illustrate role of statistical methods /
  analysis towards their solution
• make connections
• Objectives met? I think so.
• Perhaps nothing entirely new - but new to you.

19
Larry Cox
Study Design, Sampling Modeling

• combining studies
• understanding differences in covariates / lags
  between studies

20
Larry Cox

• Estimating Status and Trend
• proper removal of trend and seasonal effects
• inference robust to smoothing method
• investigate effects of combined (eg PCA)
  pollutants instead of one at a time
• how to define trend / regional trend

• Spatio - Temporal Modeling
• accounting for bias due to network design
• validating atmospheric models using ambient
  measurements
• spatio - temporal models of ambient exposure
• use of short term / mobile monitoring
• simulating regulatory outcomes

• Health Effects
• specificity of effects
• pollutant mixtures
• get dose in context

21
Larry Cox

• Aggregation and Scale
• aggregating short term measurements to (ANOVA)
  averages
• imputing for unobserved subgroups /
  microenvironments
• predict aggregate or aggregate predictions?
• (consistency)
• what is appropriate scale(s) on which to link
  exposure and health effects?
• tradeoff between temporal and spatial
  resolution
• role of (daily) observations in estimating
  (hourly) exposures
• stratification
• aggregating microenvironments
• simulating individual activities over multiple
  days
• Aggregate when similar disaggregate when
  different. P.S.

22
Larry Cox

• Synergies
• Atmospheric and spatio - temporal models
• Atmospheric and receptor models
• Spatio - temporal and receptor models

• Personal Lessons Learned
• Particles are cunning little devils. They
• grow
• bounce
• pig in poke?
• but
• occasionally are coarse
• We breathe our colleagues air
• Computers dont have unions
• We need more data!

• Workshop Outcomes
• workshop summary videos etc
• research

• NRCSE

• others

• Connections
• Þ More Work for Everyone.