Canadian Provinces By Motassem Al-Arydah Jan M. Zielinski What is - - PowerPoint PPT Presentation

On the Efficiency of Radon Mitigation Program in Canadian Provinces

By

Mo’tassem Al-Arydah

CMS and MMEI 2019

Jan M. Zielinski

What is Radon?

• Radon is a radioactive gas results from

normal decay uranium, thorium, and radium in rocks and soil.

• Invisible, odorless, tasteless

Radon Mitigation: Does it worth to spend avg=1500$?

High levels ls lea eads to

• lung can

ancer

Rela elativ ive Ris Risk

Definition:

A measure of the strength of association based on prospective studies (cohort studies).

Attributable Risk

Definition:

The amount of disease that can be attributed to a certain exposure.

• Useful in determining how much disease can be prevented

Attri tributable le Ris Risk for

• r an

an Ex Expo posed Grou

• up

Co Comp mput uting Pop

• pul

ulation Attrib tributable Ri Risk for r Ra Rado don Expo Exposure

( ) ( ) (0) ( ) ( )

m m

R p d R PAR R p d      

 

  



{ ( ) 1} ( ) ( ) ( )

m m

RR p d RR p d      

 

  



RR R R ( ) ( ) ( )   

Method: Life-table Analysis

R h e h h e S i e q h e h e

i i i i i i i i i k k k i i

( ) ( ) ( , , ) [ exp( )] exp( ).

*

      

  

 

1 1 1

1 1 1 110

the probability of surviving year i for an individual with exposure given that the individual survived up to year i-1

S i e q e

i k k k i

( , , ) ( ) 1

1 1



 



q e h h e

i i i

( ) exp[ ( )]

*

  

all causes mortality rate lung cancer mortality rate

* i

h 

i

h 

Ex Exce cess Rela elativ ive Risk Risk: BEIR EIR VI VI Mod

• dels

ls

e i i w K

i wl

         ( ) ( ) ( )

*

Concentration model Duration model

Par aram ameter valu alues

Canada: Observed Radon Data for the period 2010-2011

Rad adon

• n Distrib

ibutio ion – Lo Lognormal l App pproxim imatio ion

AM ASD GM GSD Mdn Max Newfoundland and Labrador 460 60.1 134.1 25.0 3.2 21 1548 Prince Edward Island 83 40.2 61.4 20.6 2.9 18 352 Nova Scotia 385 66.1 146.9 27.3 3.3 23 1376 New Brunswick 469 157.4 332.0 64.1 3.9 72 5590 Quebec 690 78.6 172.3 33.1 3.4 30 2923 Ontario 2464 74.6 156.4 38.8 3.0 36 5657 Manitoba 757 132.5 160.4 74.8 3.1 87 1583 Saskatchewan 793 119.3 144.8 80.3 2.5 88 2165 Alberta 676 86.0 96.7 58.1 2.5 65 1391 British Columbia 1119 63.6 151.6 25.7 3.4 21 2941 Yukon 136 130.1 236.0 71.2 2.9 79 2357 Northwest Territories 132 56.4 75.1 31.1 2.9 30 528 Nunavut 67 10.6 6.9 9.5 1.5 8 41 Total 8231 87.3 168.1 41.6 3.3 38 5657

Mean Radon Concentrations in Canadian Homes by Province / Teritorry (2010 - 2011 Survey) First Floor, Rn Originally Monitored

Province / Teritorries Participants N Rn Concentrations (Bq/m )

Moti tivatio ion: Co Comparin ing LC C in n MN an and PE PE pr provinces (CA (CA)

0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03 5.00E-03 6.00E-03 7.00E-03 20 40 60 80 100 Rate of LC death Attained age (years)

LC mortality rate MN

Males Females 0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03 5.00E-03 6.00E-03 7.00E-03 20 40 60 80 100 Rate of LC death Attained age (years)

LC mortality rate PE

Males Females

Moti

• tivation
• n: Co

Comp mparing ng Sm Smok

• king in

n MN N and and PE PE

10 20 30 40 50 60 70 50 100 Percentages % Age (years)

Ever Smoking MN (2009)

Ever smoking (males) Ever smoking (females)

10 20 30 40 50 60 70 20 40 60 80 100 Percentage % Age (years)

Ever Smoking PE (2009)

Ever smoking (male) Ever smoking (female)

Mot

• tiv

ivation and and Res esearch Qu Questions

• Estimate the lung cancer risks from residential radon exposures in

Canada and its provinces.

• Differences in PAR estimations between miners’ models and

Residential model

• Sensitivity of PAR to the choice relative risk model and radon
• distribution. How big is the range of variability in PAR results?
• Possible probability density distribution for radon
• How sensitive is PAR to Radon action and target levels?

Mor

• re Data

a ar are e ne needed

• All Cases Mortality Rates 2006-2009

different data for different gender

• LC Mortality Rates 2006-2009:

different data for different gender

• Smoking Data: 2009

different data for different gender Conclusion!! ------Grouping results by Gender and Smoking Status

Es Estim timatin ting:Probability De Density funct function 𝒒𝒏(𝒙)

Canada: Result 1

Model Gender CA CA(smoothed) CA CA(smoothed) Ever Smokers Never Smokers

BEIR VI models 1) Exposure Age Concentration(original radon data) male 0.160 0.174 0.318 0.340 1) Exposure Age Concentration(original radon data) female 0.170 0.184 0.330 0.353 2) Exposure Age Duration (original radon data) male 0.119 0.127 0.247 0.261 2) Exposure Age Duration (original radon data) female 0.126 0.134 0.256 0.271 3) 2003 EPA Scaled Concentration model (smoothed_phi) A) original radon data male 0.135 0.146 0.274 0.294 A) original radon data female 0.143 0.155 0.285 0.305 B) lognormal (as in Jing, 2012) male 0.144 0.156 0.294 0.314 B) lognormal (as in Jing, 2012) female 0.153 0.165 0.306 0.306 4) Monte Carlo for EPA scaled concentration model (smoothed_phi) 0.284 0.292 A) original radon data male mu 0.134 0.146 0.008 0.008 A) original radon data sigma 0.004 0.005 A) original radon data fem mu 0.142 0.154 0.007 0.008 A) original radon data sigma 0.004 0.005 B) lognormal male mu 0.144 0.156 0.294 0.314 B) lognormal sigma 0.006 0.006 0.011 0.011 B) lognormal fem mu 0.153 0.165 0.305 0.325 B) lognormal sigma 0.006 0.007 0.011 0.011

Res esult lts 1 - Cana anada

Model Gender CA CA(smoothed) CA CA(smoothed) Ever Smokers Never Smokers

5) Etiologic Fraction (EPA 2003) A) original radon data male 0.175 0.193 0.378 0.416 A) original radon data female 0.184 0.203 0.401 0.440 B) lognormal male 0.194 0.213 0.418 0.460 B) lognormal female 0.204 0.224 0.443 0.487 6) CRR (miners) or BEIRVI (beta=0.0117) A) original radon data No smoking correction male 0.176 0.172 0.196 0.192 No smoking correction female 0.179 0.175 0.195 0.190 smoking correction male 0.162 0.158 0.327 0.321 smoking correction female 0.165 0.161 0.325 0.318 B) lognormal No smoking correction male 0.186 0.182 0.212 0.207 No smoking correction female 0.212 0.207 0.211 0.205 smoking correction male 0.172 0.168 0.348 0.342 smoking correction female 0.175 0.171 0.346 0.339 7) CRR (potency from Meta Analysis of all studies) A) original radon data beta=0.01 male 0.155 0.152 0.173 0.169 beta=0.01 female 0.158 0.154 0.171 0.167 B) lognormal beta=0.01 male 0.165 0.162 0.187 0.183 beta=0.01 female 0.187 0.183 0.186 0.181 c) Gaussian Kernel estimation beta=0.01 male 0.167 0.163 0.186 0.181 beta=0.01 female 0.170 0.166 0.184 0.180

Result lts 2 2 – Canada a and nd Provin

• vinces

Model Gender CA QC ON MB BC NS AB NB SK NL PE NWT (CA smk) YU (CA smk) Nu (CA smk) Ever Smokers

BEIR VI models 1) Exposure Age Concentration(original radon data) male 0.160 0.116 0.137 0.228 0.111 0.120 0.163 0.236 0.223 0.115 0.084 0.121 0.227 0.022 1) Exposure Age Concentration(original radon data) female 0.170 0.140 0.143 0.244 0.115 0.127 0.175 0.248 0.225 0.121 0.090 0.089 0.212 0.026 2) Exposure Age Duration (original radon data) male 0.119 0.085 0.101 0.174 0.082 0.087 0.121 0.181 0.167 0.083 0.061 0.069 0.170 0.015 2) Exposure Age Duration (original radon data) female 0.126 0.101 0.105 0.184 0.085 0.093 0.129 0.190 0.168 0.087 0.065 0.086 0.160 0.017 3) 2003 EPA Scaled Concentration model (smoothed_phi) A) original radon data male 0.135 0.097 0.115 0.193 0.093 0.100 0.137 0.202 0.188 0.095 0.070 0.079 0.196 0.018 A) original radon data female 0.143 0.116 0.119 0.207 0.096 0.106 0.147 0.214 0.190 0.102 0.074 0.103 0.177 0.021 B) lognormal (as in Jing, 2012) male 0.144 0.115 0.121 0.210 0.095 0.101 0.147 0.216 0.203 0.094 0.068 0.080 0.208 0.019 B) lognormal (as in Jing, 2012) female 0.153 0.138 0.126 0.224 0.098 0.107 0.157 0.230 0.204 0.101 0.072 0.104 0.188 0.023 4) Monte Carlo for EPA scaled concentration model (smoothed_phi) A) original radon data male mu 0.134 0.096 0.114 0.192 0.092 0.098 0.136 0.201 0.187 0.094 0.067 0.077 0.191 0.017 A) original radon data sigma 0.004 0.003 0.003 0.005 0.004 0.004 0.003 0.006 0.004 0.004 0.003 0.003 0.005 0.000 A) original radon data fem mu 0.142 0.115 0.119 0.206 0.095 0.104 0.146 0.212 0.189 0.100 0.071 0.100 0.172 0.020 A) original radon data sigma 0.004 0.003 0.004 0.005 0.004 0.005 0.003 0.007 0.004 0.004 0.003 0.003 0.005 0.000 B) lognormal male mu 0.144 0.115 0.121 0.210 0.095 0.101 0.146 0.216 0.204 0.094 0.068 0.080 0.208 0.019 B) lognormal sigma 0.006 0.005 0.004 0.007 0.005 0.004 0.004 0.008 0.005 0.004 0.003 0.003 0.007 0.000 B) lognormal fem mu 0.153 0.138 0.126 0.224 0.098 0.107 0.157 0.230 0.204 0.100 0.072 0.104 0.188 0.023 B) lognormal sigma 0.006 0.006 0.005 0.007 0.005 0.005 0.004 0.009 0.005 0.004 0.003 0.004 0.006 0.000

Results 3 – Canada and Provinces

Model Gender CA QC ON MB BC NS AB NB SK NL PE NWT (CA smk) YU (CA smk) Nu (CA smk) Ever Smokers

5) Etiologic Fraction (EPA 2003) A) original radon data male 0.175 0.127 0.148 0.277 0.115 0.128 0.178 0.310 0.256 0.119 0.082 0.094 0.272 0.022 A) original radon data female 0.184 0.146 0.152 0.296 0.119 0.134 0.189 0.314 0.266 0.122 0.087 0.130 0.253 0.027 B) lognormal male 0.194 0.158 0.159 0.317 0.118 0.129 0.195 0.358 0.284 0.117 0.080 0.096 0.298 0.024 B) lognormal female 0.204 0.182 0.163 0.338 0.122 0.134 0.207 0.363 0.296 0.121 0.085 0.133 0.277 0.029 6) CRR (miners) or BEIRVI (beta=0.0117) A) original radon data No smoking correction male 0.176 0.131 0.155 0.248 0.130 0.130 0.186 0.252 0.237 0.120 0.098 0.138 0.238 0.022 No smoking correction female 0.179 0.136 0.159 0.254 0.132 0.134 0.189 0.268 0.233 0.125 0.099 0.130 0.229 0.022 smoking correction male 0.162 0.120 0.142 0.230 0.119 0.120 0.171 0.235 0.219 0.110 0.090 0.127 0.220 0.020 smoking correction female 0.165 0.125 0.146 0.236 0.120 0.122 0.174 0.249 0.215 0.114 0.090 0.119 0.212 0.020 B) lognormal No smoking correction male 0.186 0.151 0.162 0.264 0.132 0.131 0.197 0.265 0.253 0.118 0.096 0.140 0.140 0.025 No smoking correction female 0.212 0.160 0.166 0.272 0.134 0.134 0.201 0.283 0.248 0.123 0.096 0.131 0.241 0.024 smoking correction male 0.172 0.139 0.149 0.246 0.121 0.120 0.182 0.248 0.234 0.108 0.087 0.128 0.128 0.022 smoking correction female 0.175 0.147 0.153 0.253 0.122 0.123 0.185 0.265 0.230 0.113 0.088 0.120 0.224 0.022 7) CRR (potency from Meta Analysis of all studies) A) original radon data beta=0.01 male 0.155 0.115 0.136 0.222 0.114 0.115 0.164 0.227 0.211 0.105 0.086 0.121 0.212 0.019 beta=0.01 female 0.158 0.119 0.140 0.227 0.115 0.117 0.167 0.240 0.207 0.109 0.086 0.114 0.205 0.019 B) lognormal beta=0.01 male 0.165 0.133 0.143 0.237 0.116 0.115 0.175 0.240 0.226 0.104 0.084 0.123 0.225 0.021 beta=0.01 female 0.187 0.141 0.147 0.244 0.117 0.118 0.178 0.256 0.221 0.108 0.084 0.115 0.216 0.021 c) Gaussian Kernel estimation beta=0.01 male 0.167 0.146 0.144 0.237 0.128 0.126 0.169 0.257 0.217 0.116 0.091 0.136 0.228 0.021 beta=0.01 female 0.170 0.153 0.148 0.243 0.129 0.129 0.172 0.272 0.213 0.120 0.092 0.127 0.220 0.021

Distribution of PAR : MN

The PAR for Ontario as a function of the variable radon action level. Here radon is reduced to the outside level 15 Bq/m3. The PAR is estimated using empirical, log-normal and Gaussian kernel density estimator distributions.

The PAR for Ontario as a function of the variable radon target or reached level for a fixed action level 200 Bq/m3. The PAR is estimated using empirical, log-normal and Gaussian kernel density estimator distributions.

Conclusions

• Agreement between miner models and residential models Results
• The estimated risk in provinces with high levels of residential radon can as much as 50%

larger than in provinces with low levels of residential radon.

• Population Attributable Risk is sensitive to: Model and distribution choices
• Population Attributable Risk is very sensitive to: low Radon action level and

Radon target level

• The results can help in improving Radon mitigation programs

Th Thank You

• u!