Dose-Response Curves for NO2 and Non-accidental Mortality for 4 different periods

Descriptive Statistics for NO2
Ville Minimum X1st.Qu Median Mean X3rd.Qu Max NA.s
bordeaux 2.657306 20.53435 29.10077 30.22419 38.71460 94.13857 894
clermont 3.858328 19.95917 28.17745 31.14686 39.09285 124.32815 731
grenoble 5.125000 30.25481 39.69271 40.52026 49.70832 105.72996 767
lehavre 3.979180 19.06250 26.48889 28.59517 36.27969 91.26041 750
lille 0.000000 20.68042 30.03292 32.35433 41.75202 121.25834 1512
lyon 3.983750 33.43843 45.58635 47.10957 59.04313 158.40278 756
marseille 11.314822 35.43359 46.14023 47.08797 57.91884 102.81252 762
montpellier 3.432801 24.15611 33.23611 34.39422 42.89583 90.07251 751
nancy 5.725694 23.98847 33.00000 35.05250 44.10787 111.58334 767
nantes 2.738333 19.42448 26.93958 28.79082 36.46004 95.91659 760
nice -1.000000 34.65278 43.45833 44.72345 53.10416 116.66668 753
paris 16.597227 45.11439 55.05466 56.03320 65.26600 165.52762 731
rennes 5.622031 22.97876 30.44995 31.42495 38.45833 90.08332 731
rouen 7.048129 28.97918 37.97917 39.21817 47.69227 114.56250 760
strasbourg 7.656250 31.33905 40.08896 41.07317 49.74792 148.61667 770
toulouse 5.157343 24.57810 32.54802 34.09613 42.28371 90.61159 752

Descriptive Statistics for NO2 by period

Meta-Regression results

## Call:  mvmeta(formula = y0003 ~ 1, S = S0003, method = "ml")
## 
## Multivariate random-effects meta-analysis
## Dimension: 3
## Estimation method: ML
## 
## Fixed-effects coefficients
##     Estimate  Std. Error        z  Pr(>|z|)  95%ci.lb  95%ci.ub   
## y1   -0.0029      0.0153  -0.1878    0.8510   -0.0328    0.0271   
## y2   -0.0250      0.0338  -0.7393    0.4597   -0.0912    0.0412   
## y3   -0.0110      0.0263  -0.4172    0.6765   -0.0626    0.0406   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 
## 
## Between-study random-effects (co)variance components
## Structure: General positive-definite
##     Std. Dev  Corr    
## y1    0.0366    y1  y2
## y2    0.0500     1    
## y3    0.0398     1   1
## 
## Multivariate Cochran Q-test for heterogeneity:
## Q = 66.1426 (df = 45), p-value = 0.0217
## I-square statistic = 32.0%
## 
## 16 studies, 48 observations, 3 fixed and 6 random-effects parameters
## (2 studies removed due to missingness)
##   logLik       AIC       BIC  
##  47.1943  -76.3886  -59.5478
## Call:  mvmeta(formula = y0407 ~ 1, S = S0407, method = "ml")
## 
## Multivariate random-effects meta-analysis
## Dimension: 3
## Estimation method: ML
## 
## Fixed-effects coefficients
##     Estimate  Std. Error        z  Pr(>|z|)  95%ci.lb  95%ci.ub   
## y1   -0.0009      0.0160  -0.0536    0.9573   -0.0323    0.0306   
## y2    0.0322      0.0344   0.9366    0.3490   -0.0352    0.0996   
## y3    0.0252      0.0242   1.0445    0.2963   -0.0221    0.0726   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 
## 
## Between-study random-effects (co)variance components
## Structure: General positive-definite
##     Std. Dev  Corr    
## y1    0.0409    y1  y2
## y2    0.0681     1    
## y3    0.0300     1   1
## 
## Multivariate Cochran Q-test for heterogeneity:
## Q = 57.7209 (df = 45), p-value = 0.0967
## I-square statistic = 22.0%
## 
## 16 studies, 48 observations, 3 fixed and 6 random-effects parameters
## (2 studies removed due to missingness)
##   logLik       AIC       BIC  
##  53.4075  -88.8150  -71.9742
## Call:  mvmeta(formula = y0811 ~ 1, S = S0811, method = "ml")
## 
## Multivariate random-effects meta-analysis
## Dimension: 3
## Estimation method: ML
## 
## Fixed-effects coefficients
##     Estimate  Std. Error       z  Pr(>|z|)  95%ci.lb  95%ci.ub    
## y1    0.0308      0.0111  2.7621    0.0057    0.0089    0.0526  **
## y2    0.0879      0.0328  2.6800    0.0074    0.0236    0.1522  **
## y3    0.0153      0.0234  0.6515    0.5147   -0.0306    0.0612    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 
## 
## Between-study random-effects (co)variance components
## Structure: General positive-definite
##     Std. Dev  Corr    
## y1    0.0068    y1  y2
## y2    0.0470     1    
## y3    0.0187     1   1
## 
## Multivariate Cochran Q-test for heterogeneity:
## Q = 42.0846 (df = 45), p-value = 0.5962
## I-square statistic = 1.0%
## 
## 16 studies, 48 observations, 3 fixed and 6 random-effects parameters
## (2 studies removed due to missingness)
##   logLik       AIC       BIC  
##  56.4622  -94.9245  -78.0837
## Call:  mvmeta(formula = y1215 ~ 1, S = S1215, method = "ml")
## 
## Multivariate random-effects meta-analysis
## Dimension: 3
## Estimation method: ML
## 
## Fixed-effects coefficients
##     Estimate  Std. Error       z  Pr(>|z|)  95%ci.lb  95%ci.ub   
## y1    0.0071      0.0121  0.5890    0.5558   -0.0166    0.0309   
## y2    0.0634      0.0362  1.7512    0.0799   -0.0076    0.1343  .
## y3    0.0249      0.0265  0.9395    0.3475   -0.0270    0.0767   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 
## 
## Between-study random-effects (co)variance components
## Structure: General positive-definite
##     Std. Dev  Corr    
## y1    0.0208    y1  y2
## y2    0.0799     1    
## y3    0.0583     1   1
## 
## Multivariate Cochran Q-test for heterogeneity:
## Q = 51.6630 (df = 45), p-value = 0.2297
## I-square statistic = 12.9%
## 
## 16 studies, 48 observations, 3 fixed and 6 random-effects parameters
## (2 studies removed due to missingness)
##   logLik       AIC       BIC  
##  57.8654  -97.7308  -80.8900

Pooled Dose-Responses Curves

Doses-Response Curves by cities

Bordeaux

Clermont-Ferrand

Grenoble

Le Havre

Lille

Lyon

Marseille

Montpellier

Nancy

Nantes

Nice

Paris

Rennes

Rouen

Strasbourg

Toulouse