The aim of the following analysis is to test different approaches to estimate whether heat waves occuring at different timing during the summer may have different effect on mortality.

Heat wave variable was created according to the Meteo France definition for extreme weather events. Heat waves were defined as periods where the mean daily temperature exceeded the 97.5th percentile of the local (within each urban agglomeration) mean temperature distribution over the entire year (initial threshold) for at least three consecutive days, and for which at least one day recorded a mean daily temperature above the 99.5th percentile of the local temperature distribution for the entire year (peak threshold). For references, see Soubeyroux et al. 2016 (DOI: 10.4267/2042/60704) and Pascal et al. 2021 (DOI: https://doi.org/10.1016/j.atmosenv.2021.118249).

We defined heat waves episodes in 18 different cities in French durgin the period going from 2000 to 2015.

Number of Heat Waves per city and per Month, 2000-2015
city	June	July	August
BM	0	18	12
bordeaux	1	12	13
clermont	0	9	15
dijon	3	17	11
grenoble	5	14	17
lehavre	1	16	6
lille	0	18	12
lyon	4	14	14
marseille	0	19	16
montpellier	6	17	5
nancy	2	15	15
nantes	1	13	10
nice	0	6	20
paris	0	14	13
rennes	1	13	18
rouen	0	12	13
strasbourg	6	15	17
toulouse	1	7	21

There are no heat wave episodes during the months of may and september, and in some cities also during the month of june (Bassin minier, Clermont-Ferrand, Lille, Marseille, Nice, Paris, Rouen). I exclude the month of may and september

First Model strategy

We retain only the months from june to august and in the model. This first version of the model includes one categorical variable for the year, one categorical variable for month (the month of june is the category of reference), an indicator variable for heat wave and an interaction term between the month of the summer and the heat wave indicator.

`%notin%` <- Negate(`%in%`)

## exclude the months may and semptember
for (i in 1:length(villes_s)){
  villes_s[[i]]<-villes_s[[i]][which(villes_s[[i]]$mois %notin% c("5","9")),]
}

m.outcome<-list()
m.outcome2<-list()
v.coeff_july<-c()
v.coeff_aout<-c()
v.sterr_july<-c()
v.sterr_aout<-c()

for (i in 1:length(villes_s)){
m.outcome[[i]]<-glm(nocc_tot~heat_wave+annee+mois+mois*heat_wave+Jours+Vacances,data=villes_s[[i]],family=poisson)
}

## Without BM, Clermont Ferrand, lille, Marseille, Nice, Paris, ROuen
villes_s2<-villes_s[-c(1,3,7,9,13,14,16)]
for (i in 1:length(villes_s2)){
m.outcome2[[i]]<-glm(nocc_tot~heat_wave+annee+mois+mois*heat_wave+Jours+Vacances,data=villes_s2[[i]],family=poisson)
v.coeff_july[i]<-m.outcome2[[i]]$coefficients[27]
v.sterr_july[i]<-summary(m.outcome2[[i]])$coefficients[27, 2]
v.coeff_aout[i]<-m.outcome2[[i]]$coefficients[28]
v.sterr_aout[i]<-summary(m.outcome2[[i]])$coefficients[28, 2]
}

names(m.outcome)<-names(villes_s)
names(m.outcome2)<-names(villes_s2)

Results per city

summ(m.outcome[[1]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	44.361
Pseudo-R² (Cragg-Uhler)	0.030
Pseudo-R² (McFadden)	0.006
AIC	7276.766
BIC	7419.696

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	7.437	6.790	8.145	43.214	0.000
heat_wave1	1.306	1.084	1.575	2.801	0.005
annee2001	1.010	0.909	1.121	0.180	0.858
annee2002	1.041	0.938	1.155	0.762	0.446
annee2003	1.109	1.000	1.231	1.952	0.051
annee2004	0.980	0.882	1.089	-0.378	0.705
annee2005	0.978	0.880	1.087	-0.419	0.675
annee2006	1.046	0.941	1.162	0.834	0.404
annee2007	1.034	0.932	1.147	0.626	0.531
annee2008	1.038	0.935	1.151	0.695	0.487
annee2009	1.011	0.910	1.122	0.197	0.844
annee2010	1.043	0.940	1.157	0.795	0.426
annee2011	1.018	0.917	1.130	0.332	0.740
annee2012	1.110	1.002	1.230	2.003	0.045
annee2013	0.999	0.899	1.109	-0.027	0.978
annee2014	1.068	0.963	1.185	1.252	0.211
annee2015	1.058	0.953	1.173	1.057	0.290
mois7	0.910	0.812	1.021	-1.611	0.107
mois8	0.872	0.772	0.986	-2.185	0.029
Joursjeudi	1.036	0.967	1.110	1.001	0.317
Jourslundi	1.075	1.004	1.151	2.070	0.038
Joursmardi	1.048	0.978	1.122	1.325	0.185
Joursmercredi	1.064	0.994	1.140	1.783	0.075
Jourssamedi	0.991	0.924	1.063	-0.255	0.799
Joursvendredi	1.083	1.012	1.160	2.297	0.022
Vacances1	1.088	0.970	1.220	1.442	0.149
heat_wave1:mois7	0.793	0.616	1.021	-1.799	0.072
heat_wave1:mois8	NA	NA	NA	NA	NA
Standard errors: MLE

Bordeaux

summ(m.outcome[[2]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	138.804
Pseudo-R² (Cragg-Uhler)	0.090
Pseudo-R² (McFadden)	0.017
AIC	7935.092
BIC	8083.334

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.191	10.389	12.054	63.679	0.000
heat_wave1	1.080	0.624	1.869	0.273	0.785
annee2001	1.040	0.956	1.132	0.921	0.357
annee2002	1.017	0.934	1.107	0.394	0.693
annee2003	1.076	0.988	1.172	1.681	0.093
annee2004	0.945	0.867	1.031	-1.272	0.203
annee2005	0.930	0.852	1.014	-1.648	0.099
annee2006	0.987	0.905	1.075	-0.307	0.759
annee2007	0.940	0.862	1.025	-1.400	0.161
annee2008	0.985	0.904	1.073	-0.354	0.723
annee2009	0.955	0.876	1.041	-1.054	0.292
annee2010	1.029	0.946	1.120	0.662	0.508
annee2011	0.986	0.905	1.074	-0.323	0.747
annee2012	0.962	0.883	1.049	-0.878	0.380
annee2013	1.009	0.927	1.099	0.208	0.835
annee2014	1.031	0.948	1.122	0.713	0.476
annee2015	1.067	0.981	1.161	1.520	0.129
mois7	1.055	0.964	1.155	1.165	0.244
mois8	1.035	0.939	1.141	0.699	0.485
Joursjeudi	1.046	0.989	1.107	1.563	0.118
Jourslundi	1.045	0.987	1.105	1.520	0.129
Joursmardi	1.019	0.963	1.078	0.647	0.518
Joursmercredi	1.036	0.979	1.097	1.238	0.216
Jourssamedi	1.041	0.984	1.102	1.389	0.165
Joursvendredi	1.028	0.972	1.088	0.969	0.333
Vacances1	0.951	0.869	1.041	-1.081	0.280
heat_wave1:mois7	1.349	0.765	2.379	1.033	0.301
heat_wave1:mois8	1.603	0.917	2.805	1.655	0.098
Standard errors: MLE

Clermont-Ferrand

summ(m.outcome[[3]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1466 (6 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	69.745
Pseudo-R² (Cragg-Uhler)	0.047
Pseudo-R² (McFadden)	0.011
AIC	6531.028
BIC	6673.866

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	4.983	4.453	5.575	28.031	0.000
heat_wave1	1.753	1.457	2.108	5.950	0.000
annee2001	0.902	0.790	1.030	-1.522	0.128
annee2002	0.996	0.875	1.133	-0.065	0.948
annee2003	1.047	0.920	1.191	0.694	0.487
annee2004	0.956	0.839	1.090	-0.671	0.502
annee2005	0.957	0.840	1.091	-0.659	0.510
annee2006	0.917	0.803	1.047	-1.284	0.199
annee2007	0.982	0.862	1.118	-0.277	0.782
annee2008	1.068	0.941	1.213	1.017	0.309
annee2009	1.064	0.936	1.208	0.945	0.345
annee2010	1.090	0.960	1.236	1.333	0.183
annee2011	1.004	0.883	1.142	0.060	0.952
annee2012	1.056	0.930	1.199	0.835	0.404
annee2013	1.091	0.962	1.238	1.353	0.176
annee2014	1.062	0.935	1.207	0.932	0.352
annee2015	1.111	0.977	1.263	1.610	0.107
mois7	0.942	0.817	1.085	-0.830	0.406
mois8	0.902	0.776	1.050	-1.334	0.182
Joursjeudi	1.032	0.948	1.123	0.733	0.463
Jourslundi	1.025	0.942	1.116	0.582	0.560
Joursmardi	1.050	0.965	1.142	1.139	0.255
Joursmercredi	1.043	0.959	1.135	0.985	0.325
Jourssamedi	1.047	0.962	1.139	1.070	0.285
Joursvendredi	1.043	0.959	1.135	0.984	0.325
Vacances1	1.062	0.922	1.224	0.838	0.402
heat_wave1:mois7	0.670	0.481	0.935	-2.355	0.019
heat_wave1:mois8	NA	NA	NA	NA	NA
Standard errors: MLE

Dijon

summ(m.outcome[[4]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1449 (23 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	80.462
Pseudo-R² (Cragg-Uhler)	0.055
Pseudo-R² (McFadden)	0.013
AIC	6099.954
BIC	6247.755

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.628	3.191	4.125	19.669	0.000
heat_wave1	0.981	0.565	1.704	-0.068	0.946
annee2001	1.053	0.908	1.222	0.684	0.494
annee2002	1.114	0.961	1.291	1.436	0.151
annee2003	1.207	1.040	1.400	2.472	0.013
annee2004	1.034	0.890	1.201	0.438	0.661
annee2005	1.051	0.906	1.220	0.662	0.508
annee2006	1.102	0.951	1.277	1.289	0.197
annee2007	1.090	0.941	1.262	1.153	0.249
annee2008	1.117	0.965	1.293	1.483	0.138
annee2009	1.129	0.976	1.307	1.631	0.103
annee2010	1.110	0.958	1.285	1.391	0.164
annee2011	1.138	0.984	1.316	1.744	0.081
annee2012	1.202	1.041	1.388	2.510	0.012
annee2013	1.244	1.079	1.435	3.001	0.003
annee2014	1.190	1.031	1.374	2.373	0.018
annee2015	1.226	1.060	1.417	2.754	0.006
mois7	1.107	0.954	1.285	1.340	0.180
mois8	1.065	0.907	1.251	0.775	0.439
Joursjeudi	1.016	0.925	1.117	0.338	0.736
Jourslundi	1.037	0.944	1.139	0.763	0.445
Joursmardi	1.056	0.961	1.159	1.130	0.258
Joursmercredi	1.051	0.957	1.155	1.042	0.297
Jourssamedi	1.033	0.940	1.135	0.674	0.500
Joursvendredi	1.033	0.941	1.135	0.682	0.495
Vacances1	0.928	0.800	1.077	-0.985	0.325
heat_wave1:mois7	1.119	0.617	2.031	0.370	0.711
heat_wave1:mois8	2.130	1.189	3.816	2.540	0.011
Standard errors: MLE

Grenoble

summ(m.outcome[[5]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	98.713
Pseudo-R² (Cragg-Uhler)	0.065
Pseudo-R² (McFadden)	0.014
AIC	7115.077
BIC	7263.301

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	6.898	6.269	7.590	39.578	0.000
heat_wave1	0.936	0.676	1.296	-0.399	0.690
annee2001	1.064	0.954	1.187	1.111	0.267
annee2002	1.004	0.898	1.122	0.064	0.949
annee2003	1.195	1.072	1.332	3.208	0.001
annee2004	0.978	0.875	1.093	-0.395	0.693
annee2005	1.006	0.900	1.124	0.100	0.920
annee2006	0.979	0.876	1.094	-0.376	0.707
annee2007	0.923	0.825	1.034	-1.380	0.168
annee2008	1.029	0.922	1.149	0.513	0.608
annee2009	1.013	0.907	1.131	0.227	0.820
annee2010	1.042	0.933	1.163	0.730	0.465
annee2011	1.070	0.959	1.193	1.214	0.225
annee2012	1.095	0.982	1.220	1.634	0.102
annee2013	1.170	1.051	1.302	2.875	0.004
annee2014	1.117	1.003	1.245	2.013	0.044
annee2015	1.204	1.082	1.341	3.400	0.001
mois7	0.936	0.832	1.052	-1.112	0.266
mois8	0.930	0.820	1.055	-1.126	0.260
Joursjeudi	1.019	0.948	1.095	0.516	0.606
Jourslundi	1.071	0.997	1.149	1.884	0.060
Joursmardi	1.015	0.944	1.091	0.400	0.689
Joursmercredi	1.014	0.943	1.090	0.374	0.708
Jourssamedi	1.040	0.969	1.118	1.084	0.278
Joursvendredi	1.025	0.954	1.101	0.666	0.506
Vacances1	1.008	0.896	1.134	0.135	0.893
heat_wave1:mois7	1.309	0.902	1.902	1.416	0.157
heat_wave1:mois8	1.442	1.011	2.057	2.020	0.043
Standard errors: MLE

La Havre

summ(m.outcome[[6]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1467 (5 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	41.739
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	6610.519
BIC	6758.666

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.154	4.614	5.757	29.029	0.000
heat_wave1	1.457	0.722	2.938	1.051	0.293
annee2001	0.988	0.868	1.124	-0.186	0.853
annee2002	1.034	0.911	1.174	0.515	0.607
annee2003	0.992	0.872	1.129	-0.120	0.904
annee2004	1.092	0.963	1.237	1.370	0.171
annee2005	0.938	0.824	1.069	-0.955	0.340
annee2006	1.166	1.030	1.320	2.419	0.016
annee2007	1.026	0.903	1.165	0.394	0.693
annee2008	1.009	0.888	1.146	0.132	0.895
annee2009	1.021	0.899	1.160	0.322	0.748
annee2010	1.046	0.921	1.188	0.696	0.486
annee2011	1.035	0.912	1.175	0.535	0.593
annee2012	1.004	0.883	1.141	0.062	0.950
annee2013	1.066	0.940	1.209	0.996	0.319
annee2014	1.016	0.894	1.155	0.245	0.806
annee2015	1.121	0.990	1.270	1.801	0.072
mois7	1.020	0.891	1.168	0.290	0.772
mois8	1.027	0.889	1.187	0.367	0.714
Joursjeudi	1.025	0.943	1.115	0.588	0.556
Jourslundi	1.030	0.948	1.119	0.694	0.487
Joursmardi	1.043	0.960	1.134	1.004	0.315
Joursmercredi	1.050	0.967	1.142	1.161	0.246
Jourssamedi	0.988	0.908	1.075	-0.277	0.782
Joursvendredi	0.975	0.896	1.061	-0.576	0.565
Vacances1	0.934	0.816	1.069	-0.993	0.321
heat_wave1:mois7	0.819	0.395	1.698	-0.537	0.591
heat_wave1:mois8	1.056	0.492	2.263	0.139	0.890
Standard errors: MLE

Lille

summ(m.outcome[[7]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	67.392
Pseudo-R² (Cragg-Uhler)	0.045
Pseudo-R² (McFadden)	0.008
AIC	8564.230
BIC	8707.179

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	19.223	18.159	20.348	101.804	0.000
heat_wave1	1.338	1.188	1.508	4.792	0.000
annee2001	0.974	0.912	1.039	-0.808	0.419
annee2002	0.971	0.910	1.037	-0.870	0.384
annee2003	0.989	0.926	1.057	-0.316	0.752
annee2004	0.962	0.901	1.027	-1.158	0.247
annee2005	0.982	0.920	1.048	-0.547	0.584
annee2006	0.971	0.909	1.037	-0.878	0.380
annee2007	0.928	0.869	0.991	-2.214	0.027
annee2008	0.931	0.872	0.995	-2.110	0.035
annee2009	0.937	0.878	1.001	-1.931	0.054
annee2010	0.977	0.915	1.042	-0.708	0.479
annee2011	0.948	0.888	1.012	-1.605	0.109
annee2012	0.938	0.878	1.001	-1.922	0.055
annee2013	0.975	0.913	1.040	-0.766	0.444
annee2014	0.976	0.915	1.042	-0.719	0.472
annee2015	0.974	0.913	1.040	-0.791	0.429
mois7	1.028	0.958	1.102	0.759	0.448
mois8	1.021	0.947	1.102	0.544	0.586
Joursjeudi	1.062	1.016	1.110	2.666	0.008
Jourslundi	1.071	1.025	1.119	3.052	0.002
Joursmardi	1.086	1.040	1.135	3.716	0.000
Joursmercredi	1.063	1.017	1.111	2.729	0.006
Jourssamedi	1.031	0.986	1.078	1.344	0.179
Joursvendredi	1.063	1.017	1.110	2.708	0.007
Vacances1	0.953	0.888	1.023	-1.333	0.183
heat_wave1:mois7	0.851	0.726	0.997	-2.001	0.045
heat_wave1:mois8	NA	NA	NA	NA	NA
Standard errors: MLE

Lyon

summ(m.outcome[[8]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	387.612
Pseudo-R² (Cragg-Uhler)	0.232
Pseudo-R² (McFadden)	0.043
AIC	8596.027
BIC	8744.270

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	16.701	15.711	17.753	90.323	0.000
heat_wave1	0.969	0.771	1.217	-0.275	0.784
annee2001	1.042	0.972	1.116	1.150	0.250
annee2002	1.024	0.955	1.098	0.660	0.509
annee2003	1.223	1.142	1.310	5.748	0.000
annee2004	0.947	0.882	1.017	-1.505	0.132
annee2005	0.944	0.879	1.014	-1.584	0.113
annee2006	1.000	0.932	1.074	0.014	0.989
annee2007	0.942	0.877	1.011	-1.653	0.098
annee2008	1.029	0.960	1.103	0.798	0.425
annee2009	0.954	0.889	1.024	-1.295	0.195
annee2010	1.009	0.941	1.082	0.251	0.802
annee2011	0.972	0.906	1.043	-0.790	0.429
annee2012	0.994	0.927	1.066	-0.164	0.870
annee2013	0.936	0.872	1.005	-1.809	0.070
annee2014	0.964	0.898	1.034	-1.022	0.307
annee2015	1.033	0.964	1.108	0.925	0.355
mois7	1.030	0.956	1.110	0.775	0.438
mois8	1.011	0.932	1.096	0.259	0.796
Joursjeudi	1.037	0.989	1.086	1.515	0.130
Jourslundi	1.042	0.995	1.091	1.730	0.084
Joursmardi	1.038	0.991	1.088	1.587	0.112
Joursmercredi	1.044	0.997	1.094	1.824	0.068
Jourssamedi	1.008	0.962	1.056	0.327	0.744
Joursvendredi	1.088	1.039	1.139	3.592	0.000
Vacances1	0.953	0.884	1.027	-1.261	0.207
heat_wave1:mois7	1.212	0.936	1.569	1.462	0.144
heat_wave1:mois8	1.976	1.551	2.516	5.517	0.000
Standard errors: MLE

Marseille

summ(m.outcome[[9]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	168.026
Pseudo-R² (Cragg-Uhler)	0.108
Pseudo-R² (McFadden)	0.019
AIC	8708.576
BIC	8851.524

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	20.565	19.458	21.734	107.164	0.000
heat_wave1	1.227	1.108	1.358	3.944	0.000
annee2001	0.947	0.889	1.009	-1.672	0.095
annee2002	0.988	0.928	1.052	-0.387	0.699
annee2003	1.110	1.042	1.181	3.253	0.001
annee2004	0.906	0.850	0.966	-3.012	0.003
annee2005	0.873	0.818	0.932	-4.102	0.000
annee2006	0.887	0.831	0.947	-3.599	0.000
annee2007	0.912	0.855	0.972	-2.825	0.005
annee2008	0.946	0.888	1.008	-1.699	0.089
annee2009	0.948	0.890	1.011	-1.635	0.102
annee2010	0.928	0.871	0.989	-2.285	0.022
annee2011	0.901	0.845	0.960	-3.191	0.001
annee2012	0.942	0.884	1.004	-1.832	0.067
annee2013	0.950	0.891	1.012	-1.592	0.111
annee2014	0.876	0.821	0.935	-4.000	0.000
annee2015	0.963	0.904	1.026	-1.166	0.244
mois7	1.050	0.981	1.125	1.408	0.159
mois8	1.029	0.955	1.108	0.744	0.457
Joursjeudi	1.056	1.012	1.103	2.482	0.013
Jourslundi	1.050	1.006	1.096	2.210	0.027
Joursmardi	1.067	1.022	1.114	2.970	0.003
Joursmercredi	1.089	1.043	1.136	3.897	0.000
Jourssamedi	1.046	1.002	1.092	2.048	0.041
Joursvendredi	1.069	1.024	1.116	3.037	0.002
Vacances1	0.938	0.876	1.005	-1.818	0.069
heat_wave1:mois7	0.925	0.802	1.067	-1.067	0.286
heat_wave1:mois8	NA	NA	NA	NA	NA
Standard errors: MLE

Montpellier

summ(m.outcome[[10]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	40.829
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	6663.645
BIC	6811.888

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.452	4.904	6.061	31.362	0.000
heat_wave1	1.013	0.725	1.416	0.076	0.940
annee2001	1.102	0.978	1.242	1.601	0.109
annee2002	0.976	0.864	1.104	-0.380	0.704
annee2003	1.076	0.952	1.215	1.173	0.241
annee2004	0.961	0.850	1.087	-0.628	0.530
annee2005	1.021	0.905	1.153	0.342	0.733
annee2006	1.035	0.917	1.169	0.557	0.577
annee2007	0.958	0.847	1.084	-0.675	0.500
annee2008	1.015	0.899	1.147	0.244	0.807
annee2009	1.071	0.950	1.207	1.115	0.265
annee2010	1.020	0.903	1.152	0.316	0.752
annee2011	0.986	0.873	1.115	-0.218	0.827
annee2012	1.042	0.924	1.177	0.672	0.502
annee2013	1.004	0.888	1.134	0.061	0.952
annee2014	1.101	0.977	1.241	1.580	0.114
annee2015	1.185	1.052	1.334	2.805	0.005
mois7	0.959	0.840	1.094	-0.622	0.534
mois8	0.977	0.848	1.126	-0.317	0.751
Joursjeudi	1.034	0.955	1.120	0.833	0.405
Jourslundi	1.006	0.928	1.090	0.145	0.884
Joursmardi	1.016	0.938	1.101	0.396	0.692
Joursmercredi	1.070	0.989	1.158	1.679	0.093
Jourssamedi	0.982	0.905	1.065	-0.445	0.656
Joursvendredi	1.048	0.968	1.135	1.156	0.248
Vacances1	1.040	0.911	1.187	0.579	0.562
heat_wave1:mois7	1.188	0.810	1.741	0.882	0.378
heat_wave1:mois8	1.069	0.664	1.721	0.275	0.784
Standard errors: MLE

Nancy

summ(m.outcome[[11]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1470 (2 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	112.652
Pseudo-R² (Cragg-Uhler)	0.074
Pseudo-R² (McFadden)	0.016
AIC	6974.316
BIC	7122.521

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.968	5.390	6.609	34.341	0.000
heat_wave1	1.226	0.745	2.018	0.800	0.424
annee2001	1.112	0.991	1.248	1.799	0.072
annee2002	1.109	0.988	1.246	1.749	0.080
annee2003	1.197	1.066	1.344	3.044	0.002
annee2004	0.905	0.801	1.022	-1.610	0.107
annee2005	0.983	0.872	1.107	-0.286	0.775
annee2006	1.066	0.948	1.199	1.063	0.288
annee2007	1.003	0.891	1.129	0.053	0.958
annee2008	0.961	0.852	1.083	-0.655	0.513
annee2009	0.999	0.887	1.125	-0.019	0.985
annee2010	1.168	1.042	1.310	2.668	0.008
annee2011	1.100	0.980	1.235	1.610	0.107
annee2012	1.040	0.925	1.170	0.662	0.508
annee2013	1.111	0.990	1.248	1.790	0.073
annee2014	1.050	0.934	1.181	0.820	0.412
annee2015	1.184	1.055	1.328	2.883	0.004
mois7	1.054	0.933	1.192	0.845	0.398
mois8	1.059	0.928	1.209	0.851	0.395
Joursjeudi	0.979	0.907	1.056	-0.551	0.581
Jourslundi	1.020	0.946	1.100	0.523	0.601
Joursmardi	0.969	0.897	1.046	-0.817	0.414
Joursmercredi	0.993	0.920	1.071	-0.191	0.848
Jourssamedi	0.985	0.913	1.063	-0.393	0.694
Joursvendredi	1.015	0.941	1.095	0.393	0.694
Vacances1	0.946	0.837	1.069	-0.890	0.374
heat_wave1:mois7	0.859	0.501	1.472	-0.553	0.580
heat_wave1:mois8	1.363	0.806	2.307	1.155	0.248
Standard errors: MLE

Nantes

summ(m.outcome[[12]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	82.874
Pseudo-R² (Cragg-Uhler)	0.055
Pseudo-R² (McFadden)	0.011
AIC	7555.730
BIC	7703.973

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.609	7.914	9.364	50.175	0.000
heat_wave1	1.029	0.551	1.923	0.089	0.929
annee2001	1.106	1.005	1.217	2.057	0.040
annee2002	1.089	0.989	1.199	1.731	0.083
annee2003	1.124	1.020	1.237	2.373	0.018
annee2004	1.090	0.990	1.200	1.757	0.079
annee2005	1.087	0.987	1.198	1.690	0.091
annee2006	1.065	0.966	1.174	1.266	0.206
annee2007	1.093	0.992	1.203	1.804	0.071
annee2008	1.041	0.944	1.147	0.805	0.421
annee2009	1.070	0.972	1.179	1.379	0.168
annee2010	1.149	1.045	1.264	2.860	0.004
annee2011	1.095	0.994	1.205	1.844	0.065
annee2012	1.113	1.011	1.225	2.178	0.029
annee2013	1.166	1.061	1.282	3.175	0.001
annee2014	1.153	1.048	1.268	2.933	0.003
annee2015	1.209	1.100	1.328	3.947	0.000
mois7	0.929	0.838	1.030	-1.406	0.160
mois8	0.880	0.788	0.983	-2.260	0.024
Joursjeudi	1.042	0.979	1.108	1.287	0.198
Jourslundi	1.048	0.985	1.115	1.480	0.139
Joursmardi	1.039	0.976	1.105	1.194	0.232
Joursmercredi	1.019	0.957	1.084	0.581	0.561
Jourssamedi	1.000	0.939	1.064	-0.012	0.990
Joursvendredi	1.009	0.948	1.074	0.280	0.780
Vacances1	1.071	0.965	1.188	1.292	0.196
heat_wave1:mois7	1.068	0.559	2.040	0.200	0.841
heat_wave1:mois8	1.709	0.895	3.265	1.624	0.104
Standard errors: MLE

Nice

summ(m.outcome[[13]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	149.774
Pseudo-R² (Cragg-Uhler)	0.097
Pseudo-R² (McFadden)	0.019
AIC	7847.145
BIC	7990.093

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.361	10.550	12.236	64.264	0.000
heat_wave1	1.357	1.210	1.522	5.205	0.000
annee2001	1.044	0.961	1.134	1.013	0.311
annee2002	0.992	0.912	1.079	-0.183	0.854
annee2003	1.140	1.050	1.238	3.124	0.002
annee2004	0.897	0.823	0.978	-2.462	0.014
annee2005	0.871	0.799	0.951	-3.096	0.002
annee2006	0.962	0.883	1.048	-0.883	0.377
annee2007	0.876	0.803	0.956	-2.982	0.003
annee2008	0.928	0.852	1.011	-1.706	0.088
annee2009	0.949	0.871	1.033	-1.217	0.223
annee2010	0.882	0.809	0.962	-2.834	0.005
annee2011	0.884	0.811	0.964	-2.777	0.005
annee2012	0.883	0.810	0.963	-2.800	0.005
annee2013	0.893	0.819	0.974	-2.567	0.010
annee2014	0.950	0.872	1.034	-1.184	0.237
annee2015	0.957	0.879	1.042	-1.004	0.315
mois7	0.950	0.862	1.046	-1.046	0.295
mois8	0.953	0.860	1.057	-0.903	0.367
Joursjeudi	1.003	0.946	1.063	0.103	0.918
Jourslundi	1.058	0.999	1.121	1.939	0.053
Joursmardi	1.083	1.023	1.146	2.737	0.006
Joursmercredi	1.035	0.977	1.096	1.160	0.246
Jourssamedi	1.030	0.972	1.091	1.008	0.313
Joursvendredi	1.056	0.998	1.119	1.875	0.061
Vacances1	1.055	0.958	1.163	1.089	0.276
heat_wave1:mois7	0.708	0.538	0.932	-2.464	0.014
heat_wave1:mois8	NA	NA	NA	NA	NA
Standard errors: MLE

Paris

summ(m.outcome[[14]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	4095.363
Pseudo-R² (Cragg-Uhler)	0.938
Pseudo-R² (McFadden)	0.238
AIC	13154.420
BIC	13297.368

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	101.121	98.628	103.677	362.458	0.000
heat_wave1	2.825	2.723	2.930	55.704	0.000
annee2001	1.028	0.999	1.057	1.919	0.055
annee2002	0.978	0.951	1.006	-1.526	0.127
annee2003	1.096	1.066	1.128	6.407	0.000
annee2004	0.908	0.882	0.935	-6.483	0.000
annee2005	0.910	0.884	0.936	-6.402	0.000
annee2006	0.929	0.903	0.957	-4.949	0.000
annee2007	0.910	0.884	0.936	-6.398	0.000
annee2008	0.915	0.889	0.942	-6.008	0.000
annee2009	0.912	0.886	0.939	-6.245	0.000
annee2010	0.916	0.890	0.943	-5.904	0.000
annee2011	0.926	0.900	0.953	-5.220	0.000
annee2012	0.875	0.850	0.900	-9.032	0.000
annee2013	0.936	0.909	0.963	-4.495	0.000
annee2014	0.918	0.892	0.945	-5.798	0.000
annee2015	0.925	0.899	0.952	-5.275	0.000
mois7	1.037	1.005	1.070	2.267	0.023
mois8	1.011	0.978	1.046	0.644	0.519
Joursjeudi	1.064	1.044	1.085	6.288	0.000
Jourslundi	1.059	1.039	1.080	5.872	0.000
Joursmardi	1.069	1.049	1.090	6.791	0.000
Joursmercredi	1.062	1.041	1.082	6.083	0.000
Jourssamedi	1.033	1.013	1.053	3.248	0.001
Joursvendredi	1.056	1.036	1.077	5.531	0.000
Vacances1	0.948	0.919	0.978	-3.331	0.001
heat_wave1:mois7	0.432	0.406	0.460	-26.385	0.000
heat_wave1:mois8	NA	NA	NA	NA	NA
Standard errors: MLE

Rennes

summ(m.outcome[[15]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1441 (31 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	67.556
Pseudo-R² (Cragg-Uhler)	0.047
Pseudo-R² (McFadden)	0.012
AIC	5750.377
BIC	5898.024

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.269	2.850	3.749	16.928	0.000
heat_wave1	1.468	0.604	3.564	0.847	0.397
annee2001	0.990	0.845	1.162	-0.118	0.906
annee2002	1.147	0.983	1.338	1.738	0.082
annee2003	1.050	0.896	1.231	0.608	0.543
annee2004	0.963	0.818	1.133	-0.457	0.648
annee2005	1.034	0.881	1.213	0.408	0.683
annee2006	1.153	0.986	1.348	1.778	0.075
annee2007	1.000	0.852	1.174	-0.004	0.997
annee2008	1.040	0.888	1.219	0.489	0.625
annee2009	1.066	0.910	1.249	0.789	0.430
annee2010	0.923	0.784	1.086	-0.969	0.332
annee2011	1.196	1.026	1.394	2.294	0.022
annee2012	1.173	1.006	1.367	2.043	0.041
annee2013	1.181	1.013	1.378	2.119	0.034
annee2014	1.166	0.999	1.361	1.948	0.051
annee2015	1.144	0.979	1.337	1.690	0.091
mois7	0.974	0.824	1.150	-0.313	0.754
mois8	0.898	0.751	1.075	-1.173	0.241
Joursjeudi	1.053	0.951	1.165	0.997	0.319
Jourslundi	1.061	0.960	1.173	1.157	0.247
Joursmardi	1.008	0.910	1.117	0.154	0.878
Joursmercredi	0.967	0.873	1.072	-0.632	0.527
Jourssamedi	1.009	0.911	1.118	0.179	0.858
Joursvendredi	0.963	0.867	1.068	-0.718	0.473
Vacances1	1.071	0.907	1.266	0.808	0.419
heat_wave1:mois7	0.672	0.264	1.709	-0.835	0.404
heat_wave1:mois8	1.238	0.496	3.088	0.458	0.647
Standard errors: MLE

Rouen

summ(m.outcome[[16]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	105.308
Pseudo-R² (Cragg-Uhler)	0.069
Pseudo-R² (McFadden)	0.014
AIC	7380.118
BIC	7523.066

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.969	8.245	9.757	51.086	0.000
heat_wave1	1.814	1.566	2.102	7.930	0.000
annee2001	1.014	0.921	1.117	0.285	0.776
annee2002	0.997	0.905	1.099	-0.054	0.957
annee2003	1.066	0.967	1.174	1.286	0.198
annee2004	1.022	0.928	1.126	0.441	0.659
annee2005	0.997	0.905	1.099	-0.056	0.955
annee2006	1.019	0.925	1.124	0.389	0.697
annee2007	1.044	0.949	1.150	0.887	0.375
annee2008	0.971	0.881	1.071	-0.588	0.557
annee2009	0.989	0.897	1.090	-0.222	0.825
annee2010	0.993	0.901	1.094	-0.150	0.881
annee2011	1.044	0.948	1.149	0.874	0.382
annee2012	0.989	0.897	1.090	-0.229	0.819
annee2013	1.006	0.913	1.108	0.121	0.904
annee2014	0.998	0.906	1.100	-0.044	0.965
annee2015	1.095	0.996	1.204	1.881	0.060
mois7	1.056	0.956	1.167	1.071	0.284
mois8	1.016	0.912	1.133	0.295	0.768
Joursjeudi	1.045	0.981	1.114	1.364	0.172
Jourslundi	1.029	0.966	1.098	0.888	0.374
Joursmardi	1.047	0.982	1.115	1.397	0.162
Joursmercredi	1.078	1.011	1.148	2.310	0.021
Jourssamedi	1.044	0.980	1.113	1.325	0.185
Joursvendredi	1.006	0.943	1.073	0.180	0.857
Vacances1	0.899	0.813	0.994	-2.080	0.038
heat_wave1:mois7	0.683	0.543	0.860	-3.252	0.001
heat_wave1:mois8	NA	NA	NA	NA	NA
Standard errors: MLE

Strasbourg

summ(m.outcome[[17]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	131.002
Pseudo-R² (Cragg-Uhler)	0.086
Pseudo-R² (McFadden)	0.018
AIC	7198.176
BIC	7346.419

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.219	7.525	8.977	46.790	0.000
heat_wave1	1.334	1.040	1.712	2.268	0.023
annee2001	0.982	0.887	1.086	-0.355	0.723
annee2002	0.959	0.866	1.062	-0.803	0.422
annee2003	0.997	0.900	1.104	-0.061	0.952
annee2004	0.921	0.831	1.020	-1.577	0.115
annee2005	0.869	0.783	0.964	-2.643	0.008
annee2006	0.919	0.828	1.019	-1.610	0.107
annee2007	0.978	0.884	1.082	-0.432	0.666
annee2008	0.942	0.851	1.044	-1.136	0.256
annee2009	0.959	0.866	1.062	-0.808	0.419
annee2010	0.926	0.836	1.026	-1.465	0.143
annee2011	0.888	0.801	0.985	-2.238	0.025
annee2012	0.895	0.807	0.992	-2.108	0.035
annee2013	1.002	0.906	1.108	0.038	0.969
annee2014	0.972	0.878	1.076	-0.548	0.584
annee2015	0.987	0.892	1.093	-0.249	0.803
mois7	1.077	0.967	1.199	1.343	0.179
mois8	1.002	0.892	1.126	0.035	0.972
Joursjeudi	0.981	0.916	1.051	-0.551	0.582
Jourslundi	1.058	0.989	1.131	1.635	0.102
Joursmardi	1.000	0.934	1.071	0.012	0.991
Joursmercredi	1.066	0.996	1.140	1.853	0.064
Jourssamedi	0.980	0.915	1.049	-0.582	0.561
Joursvendredi	1.008	0.941	1.079	0.222	0.824
Vacances1	0.945	0.848	1.052	-1.036	0.300
heat_wave1:mois7	1.070	0.795	1.442	0.447	0.655
heat_wave1:mois8	1.307	0.980	1.744	1.821	0.069
Standard errors: MLE

Toulouse

summ(m.outcome[[18]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(27)	92.470
Pseudo-R² (Cragg-Uhler)	0.061
Pseudo-R² (McFadden)	0.012
AIC	7799.560
BIC	7947.802

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	10.041	9.288	10.855	57.998	0.000
heat_wave1	1.096	0.633	1.898	0.329	0.742
annee2001	1.065	0.974	1.164	1.382	0.167
annee2002	1.095	1.002	1.196	2.007	0.045
annee2003	1.181	1.081	1.290	3.697	0.000
annee2004	1.042	0.953	1.139	0.905	0.366
annee2005	0.996	0.910	1.090	-0.083	0.934
annee2006	1.023	0.935	1.120	0.496	0.620
annee2007	0.989	0.903	1.083	-0.243	0.808
annee2008	1.062	0.971	1.160	1.315	0.189
annee2009	1.036	0.947	1.133	0.772	0.440
annee2010	1.083	0.991	1.183	1.768	0.077
annee2011	1.115	1.021	1.218	2.433	0.015
annee2012	1.036	0.948	1.133	0.780	0.435
annee2013	1.128	1.034	1.232	2.701	0.007
annee2014	1.038	0.950	1.136	0.825	0.409
annee2015	1.116	1.022	1.218	2.439	0.015
mois7	1.075	0.981	1.178	1.552	0.121
mois8	1.032	0.934	1.140	0.619	0.536
Joursjeudi	1.069	1.009	1.133	2.262	0.024
Jourslundi	1.018	0.960	1.079	0.587	0.557
Joursmardi	1.032	0.973	1.094	1.047	0.295
Joursmercredi	1.043	0.984	1.106	1.418	0.156
Jourssamedi	1.015	0.957	1.076	0.498	0.619
Joursvendredi	1.080	1.019	1.144	2.606	0.009
Vacances1	0.926	0.845	1.015	-1.637	0.102
heat_wave1:mois7	1.126	0.627	2.023	0.397	0.692
heat_wave1:mois8	1.237	0.708	2.162	0.747	0.455
Standard errors: MLE

In cities which had not experienced a heat wave during the month of june, the coefficient associated to the interaction between heat wave and the month of august could not be estimated (according to the output tables): does it mean that for the cities the month of july was taken as reference and the only coefficient estimated is the effect of an heat wave during the month of august compared to the month of july?

Metanalysis

We only do a Meta-analysis for the cities providing an estimation for the coefficient associated to interaction with both july and august month

## 
## Random-Effects Model (k = 11; tau^2 estimator: REML)
## 
##   logLik  deviance       AIC       BIC      AICc 
##   2.0694   -4.1388   -0.1388    0.4663    1.5755   
## 
## tau^2 (estimated amount of total heterogeneity): 0 (SE = 0.0189)
## tau (square root of estimated tau^2 value):      0
## I^2 (total heterogeneity / total variability):   0.00%
## H^2 (total variability / sampling variability):  1.00
## 
## Test for Heterogeneity:
## Q(df = 10) = 4.4174, p-val = 0.9266
## 
## Model Results:
## 
## estimate      se    zval    pval    ci.lb   ci.ub 
##   0.1235  0.0670  1.8427  0.0654  -0.0079  0.2548  . 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## 
## Random-Effects Model (k = 11; tau^2 estimator: REML)
## 
##   logLik  deviance       AIC       BIC      AICc 
##  -0.6020    1.2039    5.2039    5.8091    6.9182   
## 
## tau^2 (estimated amount of total heterogeneity): 0.0169 (SE = 0.0298)
## tau (square root of estimated tau^2 value):      0.1300
## I^2 (total heterogeneity / total variability):   24.55%
## H^2 (total variability / sampling variability):  1.33
## 
## Test for Heterogeneity:
## Q(df = 10) = 10.9506, p-val = 0.3614
## 
## Model Results:
## 
## estimate      se    zval    pval   ci.lb   ci.ub 
##   0.3977  0.0818  4.8654  <.0001  0.2375  0.5580  *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Interaction with the month of July

Interaction with the month of August

Second model Strategy

We created a new variable indicating whether we are in the first part of the summer (june and july, summer1=1) or in the second part of the summer (august, summer1=0)

for (i in 1:length(villes_s)){
  villes_s[[i]]$summer1<-ifelse(villes_s[[i]]$mois %in% c("6","7"),1,0)
}

m.outcome3<-list()
v.coeff3<-c()
v.sterr3<-c()


for (i in 1:length(villes_s)){
m.outcome3[[i]]<-glm(nocc_tot~heat_wave+annee+summer1+summer1*heat_wave+Jours+Vacances,data=villes_s[[i]],family=poisson)
v.coeff3[i]<-m.outcome3[[i]]$coefficients[26]
v.sterr3[i]<-summary(m.outcome3[[i]])$coefficients[26,2]
}

names(m.outcome3)<-names(villes_s)

#lapply(m.outcome3,function(x){summary(x)})

Results per city

summ(m.outcome3[[1]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	41.708
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	7277.419
BIC	7415.055

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	7.093	6.402	7.859	37.451	0.000
heat_wave1	1.302	1.080	1.569	2.763	0.006
annee2001	1.011	0.910	1.122	0.197	0.844
annee2002	1.043	0.940	1.157	0.798	0.425
annee2003	1.113	1.003	1.235	2.015	0.044
annee2004	0.980	0.882	1.089	-0.379	0.705
annee2005	0.977	0.879	1.085	-0.436	0.663
annee2006	1.044	0.940	1.160	0.804	0.421
annee2007	1.031	0.929	1.144	0.574	0.566
annee2008	1.035	0.932	1.148	0.641	0.521
annee2009	1.010	0.909	1.121	0.179	0.858
annee2010	1.042	0.939	1.156	0.781	0.435
annee2011	1.017	0.916	1.129	0.314	0.754
annee2012	1.106	0.999	1.225	1.935	0.053
annee2013	0.996	0.896	1.106	-0.079	0.937
annee2014	1.064	0.960	1.180	1.184	0.236
annee2015	1.055	0.951	1.170	1.008	0.314
summer1	1.044	0.997	1.094	1.832	0.067
Joursjeudi	1.033	0.964	1.107	0.918	0.359
Jourslundi	1.074	1.003	1.150	2.043	0.041
Joursmardi	1.045	0.976	1.120	1.267	0.205
Joursmercredi	1.062	0.991	1.137	1.713	0.087
Jourssamedi	0.989	0.922	1.060	-0.311	0.756
Joursvendredi	1.081	1.010	1.157	2.240	0.025
Vacances1	0.998	0.954	1.044	-0.088	0.930
heat_wave1:summer1	0.788	0.612	1.015	-1.847	0.065
Standard errors: MLE

Bordeaux

summ(m.outcome3[[2]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	136.260
Pseudo-R² (Cragg-Uhler)	0.089
Pseudo-R² (McFadden)	0.017
AIC	7933.635
BIC	8071.289

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	10.978	10.099	11.934	56.235	0.000
heat_wave1	1.740	1.525	1.984	8.254	0.000
annee2001	1.040	0.956	1.131	0.908	0.364
annee2002	1.016	0.933	1.106	0.369	0.712
annee2003	1.070	0.983	1.165	1.561	0.119
annee2004	0.945	0.867	1.031	-1.272	0.203
annee2005	0.930	0.853	1.015	-1.629	0.103
annee2006	0.990	0.909	1.079	-0.226	0.821
annee2007	0.942	0.864	1.027	-1.363	0.173
annee2008	0.986	0.905	1.074	-0.315	0.753
annee2009	0.956	0.877	1.042	-1.028	0.304
annee2010	1.030	0.946	1.121	0.676	0.499
annee2011	0.987	0.906	1.074	-0.312	0.755
annee2012	0.964	0.885	1.051	-0.831	0.406
annee2013	1.011	0.929	1.101	0.261	0.794
annee2014	1.034	0.950	1.125	0.767	0.443
annee2015	1.069	0.983	1.163	1.561	0.119
summer1	1.020	0.982	1.059	1.003	0.316
Joursjeudi	1.050	0.992	1.111	1.689	0.091
Jourslundi	1.047	0.990	1.108	1.596	0.110
Joursmardi	1.022	0.966	1.082	0.752	0.452
Joursmercredi	1.040	0.983	1.100	1.357	0.175
Jourssamedi	1.044	0.987	1.105	1.492	0.136
Joursvendredi	1.031	0.975	1.091	1.071	0.284
Vacances1	1.001	0.964	1.039	0.038	0.970
heat_wave1:summer1	0.822	0.679	0.995	-2.015	0.044
Standard errors: MLE

Clermont-Ferrand

summ(m.outcome3[[3]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1466 (6 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	69.046
Pseudo-R² (Cragg-Uhler)	0.047
Pseudo-R² (McFadden)	0.011
AIC	6529.727
BIC	6667.275

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	4.758	4.195	5.397	24.281	0.000
heat_wave1	1.750	1.455	2.105	5.934	0.000
annee2001	0.903	0.791	1.031	-1.512	0.131
annee2002	0.997	0.876	1.134	-0.046	0.964
annee2003	1.049	0.922	1.193	0.725	0.469
annee2004	0.956	0.839	1.090	-0.670	0.503
annee2005	0.956	0.839	1.090	-0.667	0.505
annee2006	0.916	0.802	1.045	-1.306	0.192
annee2007	0.980	0.861	1.116	-0.302	0.762
annee2008	1.066	0.939	1.211	0.990	0.322
annee2009	1.064	0.936	1.208	0.946	0.344
annee2010	1.089	0.960	1.236	1.325	0.185
annee2011	1.004	0.883	1.141	0.053	0.957
annee2012	1.053	0.928	1.196	0.802	0.422
annee2013	1.089	0.960	1.236	1.327	0.184
annee2014	1.060	0.933	1.204	0.897	0.370
annee2015	1.111	0.977	1.263	1.609	0.108
summer1	1.044	0.987	1.105	1.504	0.133
Joursjeudi	1.030	0.947	1.121	0.687	0.492
Jourslundi	1.025	0.942	1.115	0.567	0.571
Joursmardi	1.048	0.964	1.140	1.105	0.269
Joursmercredi	1.042	0.957	1.133	0.947	0.344
Jourssamedi	1.046	0.961	1.137	1.038	0.299
Joursvendredi	1.042	0.957	1.133	0.948	0.343
Vacances1	1.005	0.951	1.062	0.192	0.848
heat_wave1:summer1	0.658	0.473	0.916	-2.481	0.013
Standard errors: MLE

Dijon

summ(m.outcome3[[4]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1449 (23 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	78.516
Pseudo-R² (Cragg-Uhler)	0.054
Pseudo-R² (McFadden)	0.013
AIC	6097.900
BIC	6235.144

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.508	3.040	4.047	17.193	0.000
heat_wave1	2.105	1.685	2.631	6.548	0.000
annee2001	1.052	0.907	1.221	0.670	0.503
annee2002	1.112	0.959	1.289	1.409	0.159
annee2003	1.199	1.034	1.391	2.397	0.017
annee2004	1.034	0.890	1.201	0.439	0.661
annee2005	1.051	0.906	1.219	0.661	0.509
annee2006	1.106	0.955	1.282	1.342	0.180
annee2007	1.094	0.945	1.266	1.198	0.231
annee2008	1.121	0.968	1.297	1.529	0.126
annee2009	1.129	0.976	1.307	1.630	0.103
annee2010	1.111	0.959	1.287	1.406	0.160
annee2011	1.139	0.985	1.317	1.760	0.078
annee2012	1.207	1.046	1.394	2.572	0.010
annee2013	1.248	1.082	1.439	3.049	0.002
annee2014	1.195	1.035	1.380	2.436	0.015
annee2015	1.230	1.064	1.421	2.802	0.005
summer1	1.039	0.975	1.107	1.189	0.235
Joursjeudi	1.020	0.928	1.120	0.407	0.684
Jourslundi	1.038	0.945	1.140	0.782	0.434
Joursmardi	1.057	0.963	1.161	1.160	0.246
Joursmercredi	1.053	0.958	1.156	1.071	0.284
Jourssamedi	1.035	0.942	1.137	0.724	0.469
Joursvendredi	1.036	0.943	1.138	0.731	0.465
Vacances1	1.020	0.959	1.084	0.624	0.533
heat_wave1:summer1	0.518	0.384	0.699	-4.297	0.000
Standard errors: MLE

Grenoble

summ(m.outcome3[[5]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	95.485
Pseudo-R² (Cragg-Uhler)	0.063
Pseudo-R² (McFadden)	0.013
AIC	7114.305
BIC	7251.941

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	6.800	6.108	7.570	35.027	0.000
heat_wave1	1.354	1.155	1.587	3.734	0.000
annee2001	1.064	0.954	1.187	1.122	0.262
annee2002	1.005	0.899	1.123	0.087	0.931
annee2003	1.187	1.065	1.322	3.106	0.002
annee2004	0.980	0.876	1.095	-0.364	0.716
annee2005	1.003	0.898	1.120	0.049	0.961
annee2006	0.980	0.876	1.095	-0.361	0.718
annee2007	0.922	0.823	1.032	-1.413	0.158
annee2008	1.027	0.920	1.147	0.480	0.631
annee2009	1.012	0.906	1.130	0.215	0.830
annee2010	1.041	0.933	1.162	0.719	0.472
annee2011	1.069	0.959	1.192	1.202	0.229
annee2012	1.092	0.980	1.217	1.590	0.112
annee2013	1.167	1.049	1.299	2.842	0.004
annee2014	1.115	1.001	1.242	1.970	0.049
annee2015	1.211	1.088	1.347	3.512	0.000
summer1	1.009	0.961	1.059	0.366	0.714
Joursjeudi	1.017	0.946	1.092	0.452	0.651
Jourslundi	1.071	0.997	1.149	1.884	0.060
Joursmardi	1.014	0.944	1.090	0.384	0.701
Joursmercredi	1.012	0.942	1.087	0.321	0.748
Jourssamedi	1.039	0.967	1.116	1.043	0.297
Joursvendredi	1.024	0.953	1.100	0.640	0.522
Vacances1	0.953	0.909	0.999	-2.021	0.043
heat_wave1:summer1	0.841	0.675	1.049	-1.539	0.124
Standard errors: MLE

La Havre

summ(m.outcome3[[6]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1467 (5 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	41.401
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	6606.857
BIC	6744.422

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.199	4.591	5.888	25.962	0.000
heat_wave1	1.539	1.140	2.077	2.818	0.005
annee2001	0.987	0.868	1.123	-0.197	0.844
annee2002	1.034	0.910	1.174	0.509	0.610
annee2003	0.991	0.871	1.128	-0.133	0.894
annee2004	1.092	0.963	1.237	1.370	0.171
annee2005	0.939	0.824	1.069	-0.952	0.341
annee2006	1.165	1.029	1.319	2.404	0.016
annee2007	1.027	0.904	1.166	0.404	0.686
annee2008	1.009	0.888	1.147	0.141	0.888
annee2009	1.024	0.901	1.163	0.361	0.718
annee2010	1.046	0.922	1.188	0.699	0.485
annee2011	1.035	0.912	1.176	0.538	0.591
annee2012	1.004	0.884	1.141	0.067	0.946
annee2013	1.066	0.940	1.210	0.999	0.318
annee2014	1.017	0.895	1.156	0.258	0.796
annee2015	1.122	0.991	1.271	1.811	0.070
summer1	0.992	0.938	1.050	-0.269	0.788
Joursjeudi	1.026	0.944	1.115	0.604	0.546
Jourslundi	1.030	0.948	1.120	0.698	0.485
Joursmardi	1.045	0.962	1.135	1.043	0.297
Joursmercredi	1.051	0.967	1.142	1.177	0.239
Jourssamedi	0.989	0.909	1.076	-0.260	0.795
Joursvendredi	0.976	0.897	1.062	-0.564	0.572
Vacances1	0.951	0.900	1.004	-1.810	0.070
heat_wave1:summer1	0.790	0.554	1.126	-1.302	0.193
Standard errors: MLE

Lille

summ(m.outcome3[[7]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	66.819
Pseudo-R² (Cragg-Uhler)	0.045
Pseudo-R² (McFadden)	0.008
AIC	8562.804
BIC	8700.457

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	19.131	17.939	20.403	89.882	0.000
heat_wave1	1.340	1.189	1.509	4.811	0.000
annee2001	0.973	0.912	1.039	-0.817	0.414
annee2002	0.971	0.910	1.036	-0.887	0.375
annee2003	0.988	0.925	1.056	-0.343	0.731
annee2004	0.962	0.901	1.027	-1.158	0.247
annee2005	0.982	0.920	1.048	-0.539	0.590
annee2006	0.971	0.909	1.038	-0.864	0.387
annee2007	0.929	0.870	0.992	-2.190	0.029
annee2008	0.932	0.873	0.996	-2.085	0.037
annee2009	0.938	0.878	1.001	-1.922	0.055
annee2010	0.977	0.915	1.043	-0.701	0.483
annee2011	0.948	0.888	1.012	-1.597	0.110
annee2012	0.939	0.879	1.002	-1.891	0.059
annee2013	0.976	0.914	1.041	-0.741	0.459
annee2014	0.977	0.916	1.043	-0.687	0.492
annee2015	0.975	0.913	1.040	-0.766	0.444
summer1	1.006	0.977	1.036	0.397	0.691
Joursjeudi	1.063	1.017	1.110	2.708	0.007
Jourslundi	1.071	1.025	1.119	3.064	0.002
Joursmardi	1.087	1.041	1.136	3.745	0.000
Joursmercredi	1.064	1.018	1.112	2.763	0.006
Jourssamedi	1.031	0.987	1.078	1.372	0.170
Joursvendredi	1.063	1.018	1.111	2.737	0.006
Vacances1	0.977	0.950	1.006	-1.561	0.119
heat_wave1:summer1	0.852	0.728	0.999	-1.978	0.048
Standard errors: MLE

Lyon

summ(m.outcome3[[8]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	384.730
Pseudo-R² (Cragg-Uhler)	0.231
Pseudo-R² (McFadden)	0.043
AIC	8594.910
BIC	8732.563

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	16.370	15.282	17.535	79.708	0.000
heat_wave1	1.923	1.747	2.117	13.348	0.000
annee2001	1.041	0.971	1.116	1.141	0.254
annee2002	1.023	0.954	1.097	0.643	0.521
annee2003	1.216	1.135	1.302	5.603	0.000
annee2004	0.947	0.882	1.017	-1.504	0.133
annee2005	0.943	0.878	1.012	-1.622	0.105
annee2006	1.005	0.937	1.078	0.132	0.895
annee2007	0.943	0.878	1.012	-1.626	0.104
annee2008	1.030	0.961	1.104	0.828	0.408
annee2009	0.955	0.889	1.025	-1.288	0.198
annee2010	1.009	0.941	1.082	0.261	0.794
annee2011	0.972	0.906	1.043	-0.783	0.434
annee2012	0.995	0.928	1.068	-0.126	0.900
annee2013	0.938	0.873	1.007	-1.768	0.077
annee2014	0.965	0.899	1.036	-0.985	0.325
annee2015	1.038	0.968	1.113	1.046	0.296
summer1	1.021	0.989	1.053	1.276	0.202
Joursjeudi	1.037	0.990	1.087	1.546	0.122
Jourslundi	1.042	0.995	1.091	1.729	0.084
Joursmardi	1.038	0.991	1.087	1.571	0.116
Joursmercredi	1.044	0.997	1.094	1.824	0.068
Jourssamedi	1.008	0.962	1.056	0.330	0.741
Joursvendredi	1.088	1.039	1.140	3.612	0.000
Vacances1	0.982	0.953	1.012	-1.171	0.242
heat_wave1:summer1	0.584	0.507	0.673	-7.464	0.000
Standard errors: MLE

Marseille

summ(m.outcome3[[9]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	166.066
Pseudo-R² (Cragg-Uhler)	0.107
Pseudo-R² (McFadden)	0.019
AIC	8708.535
BIC	8846.189

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	20.185	18.961	21.488	94.138	0.000
heat_wave1	1.229	1.110	1.360	3.973	0.000
annee2001	0.947	0.889	1.009	-1.688	0.091
annee2002	0.987	0.927	1.051	-0.416	0.677
annee2003	1.108	1.041	1.179	3.206	0.001
annee2004	0.906	0.850	0.966	-3.012	0.003
annee2005	0.874	0.819	0.932	-4.087	0.000
annee2006	0.888	0.832	0.948	-3.555	0.000
annee2007	0.913	0.857	0.974	-2.779	0.005
annee2008	0.948	0.890	1.010	-1.651	0.099
annee2009	0.949	0.891	1.011	-1.618	0.106
annee2010	0.929	0.871	0.990	-2.269	0.023
annee2011	0.901	0.845	0.961	-3.175	0.001
annee2012	0.944	0.886	1.006	-1.773	0.076
annee2013	0.951	0.893	1.014	-1.544	0.123
annee2014	0.878	0.823	0.937	-3.941	0.000
annee2015	0.964	0.905	1.028	-1.118	0.264
summer1	1.021	0.992	1.051	1.400	0.162
Joursjeudi	1.058	1.013	1.104	2.557	0.011
Jourslundi	1.050	1.006	1.097	2.230	0.026
Joursmardi	1.069	1.024	1.115	3.024	0.002
Joursmercredi	1.090	1.045	1.138	3.963	0.000
Jourssamedi	1.047	1.003	1.094	2.099	0.036
Joursvendredi	1.070	1.025	1.117	3.089	0.002
Vacances1	0.981	0.954	1.010	-1.294	0.196
heat_wave1:summer1	0.923	0.800	1.066	-1.091	0.275
Standard errors: MLE

Montpellier

summ(m.outcome3[[10]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	39.679
Pseudo-R² (Cragg-Uhler)	0.027
Pseudo-R² (McFadden)	0.006
AIC	6660.796
BIC	6798.449

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.525	4.905	6.222	28.166	0.000
heat_wave1	1.087	0.766	1.542	0.465	0.642
annee2001	1.103	0.979	1.242	1.607	0.108
annee2002	0.977	0.864	1.105	-0.368	0.713
annee2003	1.071	0.949	1.209	1.107	0.268
annee2004	0.962	0.850	1.088	-0.620	0.535
annee2005	1.021	0.904	1.153	0.335	0.737
annee2006	1.036	0.917	1.170	0.567	0.571
annee2007	0.957	0.846	1.083	-0.693	0.488
annee2008	1.014	0.898	1.145	0.225	0.822
annee2009	1.070	0.949	1.207	1.108	0.268
annee2010	1.020	0.903	1.152	0.318	0.750
annee2011	0.986	0.872	1.115	-0.224	0.823
annee2012	1.041	0.922	1.175	0.648	0.517
annee2013	1.003	0.887	1.133	0.042	0.966
annee2014	1.098	0.974	1.237	1.533	0.125
annee2015	1.189	1.056	1.337	2.870	0.004
summer1	0.983	0.932	1.037	-0.619	0.536
Joursjeudi	1.033	0.954	1.118	0.793	0.428
Jourslundi	1.006	0.929	1.090	0.154	0.877
Joursmardi	1.016	0.938	1.101	0.399	0.690
Joursmercredi	1.070	0.989	1.158	1.673	0.094
Jourssamedi	0.980	0.904	1.062	-0.493	0.622
Joursvendredi	1.046	0.966	1.133	1.115	0.265
Vacances1	1.004	0.953	1.058	0.155	0.877
heat_wave1:summer1	1.060	0.723	1.554	0.300	0.764
Standard errors: MLE

Nancy

summ(m.outcome3[[11]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1470 (2 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	111.688
Pseudo-R² (Cragg-Uhler)	0.074
Pseudo-R² (McFadden)	0.016
AIC	6971.280
BIC	7108.899

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	6.017	5.368	6.746	30.790	0.000
heat_wave1	1.674	1.421	1.972	6.160	0.000
annee2001	1.111	0.990	1.248	1.791	0.073
annee2002	1.112	0.990	1.248	1.796	0.073
annee2003	1.195	1.064	1.342	3.018	0.003
annee2004	0.905	0.801	1.022	-1.610	0.107
annee2005	0.983	0.873	1.108	-0.277	0.781
annee2006	1.066	0.948	1.198	1.060	0.289
annee2007	1.005	0.893	1.131	0.080	0.937
annee2008	0.962	0.853	1.085	-0.628	0.530
annee2009	0.999	0.887	1.126	-0.011	0.991
annee2010	1.168	1.042	1.310	2.668	0.008
annee2011	1.100	0.980	1.236	1.619	0.105
annee2012	1.043	0.927	1.172	0.697	0.486
annee2013	1.113	0.992	1.250	1.818	0.069
annee2014	1.053	0.936	1.183	0.856	0.392
annee2015	1.183	1.055	1.327	2.879	0.004
summer1	0.994	0.945	1.047	-0.212	0.832
Joursjeudi	0.981	0.909	1.059	-0.483	0.629
Jourslundi	1.021	0.947	1.101	0.542	0.588
Joursmardi	0.970	0.899	1.047	-0.780	0.435
Joursmercredi	0.995	0.922	1.073	-0.129	0.897
Jourssamedi	0.986	0.914	1.064	-0.370	0.711
Joursvendredi	1.017	0.943	1.096	0.425	0.671
Vacances1	0.992	0.943	1.043	-0.332	0.740
heat_wave1:summer1	0.645	0.504	0.827	-3.469	0.001
Standard errors: MLE

Nantes

summ(m.outcome3[[12]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	80.828
Pseudo-R² (Cragg-Uhler)	0.054
Pseudo-R² (McFadden)	0.011
AIC	7553.777
BIC	7691.430

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.128	7.397	8.932	43.550	0.000
heat_wave1	1.753	1.485	2.070	6.632	0.000
annee2001	1.107	1.005	1.218	2.070	0.038
annee2002	1.090	0.990	1.201	1.760	0.078
annee2003	1.126	1.023	1.240	2.425	0.015
annee2004	1.090	0.990	1.200	1.757	0.079
annee2005	1.086	0.986	1.196	1.666	0.096
annee2006	1.063	0.964	1.171	1.227	0.220
annee2007	1.090	0.990	1.200	1.758	0.079
annee2008	1.039	0.942	1.145	0.760	0.447
annee2009	1.070	0.971	1.178	1.362	0.173
annee2010	1.148	1.044	1.263	2.844	0.004
annee2011	1.094	0.994	1.204	1.829	0.067
annee2012	1.109	1.008	1.221	2.120	0.034
annee2013	1.164	1.058	1.280	3.132	0.002
annee2014	1.150	1.045	1.264	2.874	0.004
annee2015	1.206	1.098	1.325	3.902	0.000
summer1	1.056	1.012	1.101	2.538	0.011
Joursjeudi	1.039	0.977	1.106	1.214	0.225
Jourslundi	1.047	0.984	1.114	1.459	0.145
Joursmardi	1.036	0.974	1.103	1.134	0.257
Joursmercredi	1.017	0.955	1.082	0.517	0.605
Jourssamedi	0.998	0.937	1.062	-0.065	0.948
Joursvendredi	1.007	0.946	1.072	0.228	0.820
Vacances1	1.000	0.960	1.042	0.018	0.985
heat_wave1:summer1	0.624	0.493	0.790	-3.916	0.000
Standard errors: MLE

Nice

summ(m.outcome3[[13]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	148.666
Pseudo-R² (Cragg-Uhler)	0.097
Pseudo-R² (McFadden)	0.019
AIC	7846.253
BIC	7983.907

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.379	10.465	12.372	56.956	0.000
heat_wave1	1.356	1.209	1.521	5.192	0.000
annee2001	1.044	0.961	1.135	1.024	0.306
annee2002	0.993	0.913	1.080	-0.160	0.873
annee2003	1.142	1.052	1.239	3.163	0.002
annee2004	0.897	0.823	0.978	-2.462	0.014
annee2005	0.871	0.798	0.950	-3.108	0.002
annee2006	0.961	0.882	1.047	-0.917	0.359
annee2007	0.875	0.802	0.954	-3.018	0.003
annee2008	0.927	0.850	1.010	-1.742	0.081
annee2009	0.948	0.871	1.032	-1.228	0.219
annee2010	0.882	0.808	0.962	-2.845	0.004
annee2011	0.884	0.810	0.964	-2.788	0.005
annee2012	0.882	0.808	0.962	-2.847	0.004
annee2013	0.891	0.817	0.972	-2.602	0.009
annee2014	0.948	0.870	1.032	-1.232	0.218
annee2015	0.956	0.878	1.041	-1.039	0.299
summer1	0.996	0.959	1.036	-0.183	0.854
Joursjeudi	1.001	0.945	1.061	0.049	0.961
Jourslundi	1.058	0.999	1.120	1.924	0.054
Joursmardi	1.082	1.022	1.145	2.699	0.007
Joursmercredi	1.033	0.975	1.095	1.114	0.265
Jourssamedi	1.029	0.971	1.090	0.969	0.333
Joursvendredi	1.055	0.996	1.117	1.837	0.066
Vacances1	1.006	0.969	1.045	0.326	0.744
heat_wave1:summer1	0.709	0.539	0.933	-2.454	0.014
Standard errors: MLE

Paris

summ(m.outcome3[[14]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	4090.262
Pseudo-R² (Cragg-Uhler)	0.938
Pseudo-R² (McFadden)	0.238
AIC	13157.520
BIC	13295.174

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	98.819	96.067	101.649	318.754	0.000
heat_wave1	2.828	2.726	2.933	55.786	0.000
annee2001	1.028	0.999	1.057	1.894	0.058
annee2002	0.977	0.950	1.006	-1.581	0.114
annee2003	1.095	1.065	1.126	6.335	0.000
annee2004	0.908	0.882	0.935	-6.482	0.000
annee2005	0.910	0.884	0.937	-6.377	0.000
annee2006	0.930	0.903	0.957	-4.913	0.000
annee2007	0.911	0.885	0.937	-6.326	0.000
annee2008	0.916	0.890	0.943	-5.934	0.000
annee2009	0.912	0.886	0.939	-6.218	0.000
annee2010	0.917	0.891	0.944	-5.878	0.000
annee2011	0.926	0.900	0.953	-5.195	0.000
annee2012	0.876	0.851	0.902	-8.947	0.000
annee2013	0.937	0.910	0.964	-4.432	0.000
annee2014	0.919	0.893	0.946	-5.707	0.000
annee2015	0.926	0.900	0.953	-5.211	0.000
summer1	1.025	1.012	1.038	3.708	0.000
Joursjeudi	1.065	1.045	1.086	6.408	0.000
Jourslundi	1.060	1.039	1.080	5.896	0.000
Joursmardi	1.070	1.049	1.090	6.870	0.000
Joursmercredi	1.063	1.042	1.083	6.180	0.000
Jourssamedi	1.034	1.014	1.054	3.326	0.001
Joursvendredi	1.057	1.037	1.078	5.609	0.000
Vacances1	0.980	0.968	0.993	-3.090	0.002
heat_wave1:summer1	0.435	0.409	0.463	-26.289	0.000
Standard errors: MLE

Rennes

summ(m.outcome3[[15]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1441 (31 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	66.808
Pseudo-R² (Cragg-Uhler)	0.046
Pseudo-R² (McFadden)	0.012
AIC	5747.126
BIC	5884.226

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.015	2.584	3.518	14.026	0.000
heat_wave1	1.816	1.465	2.251	5.450	0.000
annee2001	0.990	0.844	1.161	-0.126	0.900
annee2002	1.147	0.983	1.339	1.746	0.081
annee2003	1.051	0.897	1.232	0.615	0.538
annee2004	0.963	0.818	1.133	-0.457	0.648
annee2005	1.039	0.886	1.218	0.469	0.639
annee2006	1.149	0.983	1.343	1.740	0.082
annee2007	0.999	0.851	1.172	-0.015	0.988
annee2008	1.039	0.887	1.218	0.478	0.633
annee2009	1.066	0.909	1.249	0.786	0.432
annee2010	0.922	0.784	1.085	-0.973	0.331
annee2011	1.196	1.026	1.393	2.291	0.022
annee2012	1.172	1.005	1.366	2.030	0.042
annee2013	1.179	1.011	1.376	2.100	0.036
annee2014	1.165	0.998	1.359	1.935	0.053
annee2015	1.143	0.978	1.336	1.680	0.093
summer1	1.083	1.011	1.161	2.269	0.023
Joursjeudi	1.052	0.951	1.164	0.980	0.327
Jourslundi	1.061	0.959	1.173	1.149	0.251
Joursmardi	1.009	0.911	1.118	0.175	0.861
Joursmercredi	0.966	0.872	1.071	-0.649	0.516
Jourssamedi	1.009	0.911	1.117	0.171	0.864
Joursvendredi	0.962	0.867	1.068	-0.726	0.468
Vacances1	1.044	0.976	1.116	1.251	0.211
heat_wave1:summer1	0.559	0.394	0.792	-3.270	0.001
Standard errors: MLE

Rouen

summ(m.outcome3[[16]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	104.174
Pseudo-R² (Cragg-Uhler)	0.069
Pseudo-R² (McFadden)	0.014
AIC	7379.252
BIC	7516.905

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.654	7.869	9.518	44.486	0.000
heat_wave1	1.818	1.569	2.106	7.958	0.000
annee2001	1.013	0.920	1.116	0.272	0.786
annee2002	0.996	0.904	1.098	-0.076	0.939
annee2003	1.064	0.966	1.172	1.252	0.211
annee2004	1.022	0.928	1.126	0.441	0.659
annee2005	0.998	0.905	1.099	-0.045	0.964
annee2006	1.021	0.927	1.125	0.424	0.671
annee2007	1.046	0.951	1.152	0.923	0.356
annee2008	0.973	0.882	1.073	-0.552	0.581
annee2009	0.990	0.898	1.091	-0.209	0.834
annee2010	0.993	0.901	1.095	-0.138	0.890
annee2011	1.044	0.949	1.150	0.885	0.376
annee2012	0.991	0.899	1.092	-0.183	0.855
annee2013	1.008	0.915	1.110	0.157	0.876
annee2014	1.000	0.908	1.102	0.004	0.997
annee2015	1.097	0.998	1.206	1.919	0.055
summer1	1.039	0.995	1.085	1.718	0.086
Joursjeudi	1.047	0.983	1.116	1.420	0.156
Jourslundi	1.030	0.966	1.098	0.903	0.366
Joursmardi	1.048	0.983	1.117	1.437	0.151
Joursmercredi	1.079	1.013	1.150	2.360	0.018
Jourssamedi	1.045	0.981	1.114	1.365	0.172
Joursvendredi	1.007	0.944	1.074	0.218	0.828
Vacances1	0.945	0.906	0.985	-2.654	0.008
heat_wave1:summer1	0.682	0.542	0.858	-3.272	0.001
Standard errors: MLE

Strasbourg

summ(m.outcome3[[17]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	128.906
Pseudo-R² (Cragg-Uhler)	0.084
Pseudo-R² (McFadden)	0.018
AIC	7196.272
BIC	7333.926

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	7.671	6.943	8.474	40.073	0.000
heat_wave1	1.747	1.515	2.014	7.683	0.000
annee2001	0.981	0.887	1.086	-0.370	0.712
annee2002	0.957	0.864	1.059	-0.849	0.396
annee2003	0.995	0.898	1.102	-0.104	0.917
annee2004	0.921	0.831	1.020	-1.577	0.115
annee2005	0.869	0.783	0.964	-2.644	0.008
annee2006	0.922	0.832	1.022	-1.541	0.123
annee2007	0.980	0.886	1.085	-0.385	0.700
annee2008	0.945	0.853	1.046	-1.091	0.275
annee2009	0.960	0.867	1.063	-0.792	0.428
annee2010	0.927	0.837	1.027	-1.446	0.148
annee2011	0.889	0.802	0.986	-2.224	0.026
annee2012	0.898	0.809	0.995	-2.050	0.040
annee2013	1.004	0.908	1.110	0.069	0.945
annee2014	0.973	0.879	1.076	-0.536	0.592
annee2015	0.992	0.897	1.097	-0.163	0.870
summer1	1.075	1.026	1.126	3.061	0.002
Joursjeudi	0.983	0.917	1.052	-0.502	0.615
Jourslundi	1.058	0.989	1.132	1.648	0.099
Joursmardi	1.002	0.936	1.072	0.045	0.964
Joursmercredi	1.067	0.997	1.141	1.883	0.060
Jourssamedi	0.982	0.916	1.051	-0.533	0.594
Joursvendredi	1.009	0.943	1.080	0.259	0.795
Vacances1	1.012	0.968	1.058	0.537	0.592
heat_wave1:summer1	0.805	0.662	0.978	-2.180	0.029
Standard errors: MLE

Toulouse

summ(m.outcome3[[18]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	89.937
Pseudo-R² (Cragg-Uhler)	0.060
Pseudo-R² (McFadden)	0.011
AIC	7798.093
BIC	7935.746

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	9.660	8.853	10.540	50.964	0.000
heat_wave1	1.359	1.207	1.531	5.053	0.000
annee2001	1.064	0.973	1.163	1.365	0.172
annee2002	1.093	1.001	1.194	1.976	0.048
annee2003	1.177	1.078	1.285	3.631	0.000
annee2004	1.042	0.953	1.139	0.905	0.365
annee2005	0.997	0.911	1.091	-0.066	0.947
annee2006	1.026	0.938	1.123	0.563	0.573
annee2007	0.991	0.905	1.085	-0.193	0.847
annee2008	1.064	0.974	1.163	1.369	0.171
annee2009	1.037	0.948	1.134	0.790	0.430
annee2010	1.084	0.992	1.184	1.787	0.074
annee2011	1.116	1.022	1.218	2.450	0.014
annee2012	1.039	0.950	1.136	0.844	0.398
annee2013	1.131	1.036	1.235	2.760	0.006
annee2014	1.042	0.953	1.139	0.896	0.370
annee2015	1.119	1.024	1.221	2.499	0.012
summer1	1.042	1.002	1.084	2.055	0.040
Joursjeudi	1.072	1.012	1.136	2.363	0.018
Jourslundi	1.019	0.961	1.080	0.626	0.531
Joursmardi	1.034	0.975	1.096	1.124	0.261
Joursmercredi	1.046	0.987	1.108	1.504	0.133
Jourssamedi	1.017	0.960	1.079	0.580	0.562
Joursvendredi	1.082	1.022	1.147	2.683	0.007
Vacances1	0.990	0.953	1.029	-0.506	0.613
heat_wave1:summer1	0.894	0.712	1.122	-0.966	0.334
Standard errors: MLE

Estimations are OK but interepretation might be too simplified, we can not distiguish between a heat wave in july and a heat wave in june.

Metanalysis of the results for the interaction term

meta_summer1<-rma(v.coeff3,sei=v.sterr3)

summary(meta_summer1)

## 
## Random-Effects Model (k = 18; tau^2 estimator: REML)
## 
##   logLik  deviance       AIC       BIC      AICc 
##   0.6469   -1.2937    2.7063    4.3727    3.5634   
## 
## tau^2 (estimated amount of total heterogeneity): 0.0433 (SE = 0.0197)
## tau (square root of estimated tau^2 value):      0.2080
## I^2 (total heterogeneity / total variability):   82.77%
## H^2 (total variability / sampling variability):  5.80
## 
## Test for Heterogeneity:
## Q(df = 17) = 216.6913, p-val < .0001
## 
## Model Results:
## 
## estimate      se     zval    pval    ci.lb    ci.ub 
##  -0.3401  0.0569  -5.9818  <.0001  -0.4515  -0.2287  *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

forest(meta_summer1,transf = exp, slab = names(villes_s), refline=1,xlab="RR Interaction First part of the summer", header="City")

Second model strategy - BIS

I define the two periods for summer season by taking the first period as one going from the 1st of June until the 15th of July and the second until the end of August

## Creating variable for day of the season, going from 1 at the 1st of june of every year and until 92 for 30th of september (except of bisextiles years, 2000, 2004, 2008, 2012)

for (i in 1:18){
villes_s[[i]]$day<-ifelse(villes_s[[i]]$annee %in% c("2000","2004","2008","2012"),as.numeric(strftime(villes_s[[i]]$Dates,format="%j"))-152,as.numeric(strftime(villes_s[[i]]$Dates, format = "%j"))-151)   
 }

for (i in 1:length(villes_s)){
  villes_s[[i]]$summer1<-ifelse(villes_s[[i]]$day < 45,1,0)
}

m.outcome3<-list()

for (i in 1:length(villes_s)){
m.outcome3[[i]]<-glm(nocc_tot~heat_wave+annee+summer1+summer1*heat_wave+Jours+Vacances,data=villes_s[[i]],family=poisson)
}

names(m.outcome3)<-names(villes_s)
#lapply(m.outcome3,function(x){summary(x)})

There is a problem with the city of rouen because it only experienced heat waves during the second part of the summer. I withdraw it from the analysis:

villes_s2<-villes_s[-16]

m.outcome3<-list()
v.coeff3<-c()
v.sterr3<-c()

for (i in 1:length(villes_s2)){
m.outcome3[[i]]<-glm(nocc_tot~heat_wave+annee+summer1+summer1*heat_wave+Jours+Vacances,data=villes_s2[[i]],family=poisson)
v.coeff3[i]<-m.outcome3[[i]]$coefficients[26]
v.sterr3[i]<-summary(m.outcome3[[i]])$coefficients[26, 2]
}

names(m.outcome3)<-names(villes_s2)

#lapply(m.outcome3,function(x){summary(x)})

Results per city

summ(m.outcome3[[1]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	42.132
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	7276.995
BIC	7414.631

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	7.068	6.344	7.874	35.463	0.000
heat_wave1	1.203	1.053	1.375	2.721	0.007
annee2001	1.012	0.912	1.124	0.230	0.818
annee2002	1.041	0.938	1.155	0.757	0.449
annee2003	1.120	1.011	1.242	2.161	0.031
annee2004	0.980	0.882	1.089	-0.379	0.705
annee2005	0.977	0.879	1.086	-0.433	0.665
annee2006	1.025	0.923	1.139	0.469	0.639
annee2007	1.031	0.929	1.145	0.584	0.559
annee2008	1.035	0.933	1.149	0.652	0.515
annee2009	1.009	0.909	1.121	0.176	0.860
annee2010	1.047	0.944	1.162	0.869	0.385
annee2011	1.017	0.916	1.129	0.317	0.751
annee2012	1.108	1.000	1.228	1.967	0.049
annee2013	0.994	0.895	1.105	-0.103	0.918
annee2014	1.063	0.959	1.179	1.163	0.245
annee2015	1.060	0.956	1.176	1.105	0.269
summer1	1.051	0.992	1.112	1.700	0.089
Joursjeudi	1.031	0.962	1.104	0.859	0.390
Jourslundi	1.073	1.002	1.149	2.017	0.044
Joursmardi	1.042	0.973	1.116	1.173	0.241
Joursmercredi	1.059	0.989	1.134	1.636	0.102
Jourssamedi	0.989	0.922	1.060	-0.311	0.756
Joursvendredi	1.079	1.008	1.155	2.193	0.028
Vacances1	1.013	0.955	1.075	0.443	0.658
heat_wave1:summer1	0.676	0.448	1.021	-1.861	0.063
Standard errors: MLE

Bordeaux

summ(m.outcome3[[2]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	137.440
Pseudo-R² (Cragg-Uhler)	0.090
Pseudo-R² (McFadden)	0.017
AIC	7932.456
BIC	8070.109

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.211	10.259	12.250	53.417	0.000
heat_wave1	1.654	1.493	1.833	9.606	0.000
annee2001	1.040	0.956	1.132	0.911	0.362
annee2002	1.016	0.934	1.106	0.375	0.708
annee2003	1.084	0.997	1.179	1.883	0.060
annee2004	0.945	0.867	1.031	-1.271	0.204
annee2005	0.933	0.855	1.017	-1.570	0.116
annee2006	0.974	0.894	1.061	-0.601	0.548
annee2007	0.941	0.863	1.026	-1.370	0.171
annee2008	0.986	0.905	1.074	-0.323	0.747
annee2009	0.958	0.879	1.044	-0.970	0.332
annee2010	1.030	0.946	1.120	0.674	0.500
annee2011	0.987	0.906	1.075	-0.295	0.768
annee2012	0.965	0.886	1.052	-0.803	0.422
annee2013	1.006	0.924	1.095	0.145	0.885
annee2014	1.033	0.949	1.124	0.757	0.449
annee2015	1.069	0.983	1.162	1.553	0.120
summer1	1.002	0.955	1.051	0.081	0.936
Joursjeudi	1.048	0.990	1.109	1.623	0.105
Jourslundi	1.044	0.987	1.105	1.499	0.134
Joursmardi	1.018	0.962	1.078	0.629	0.529
Joursmercredi	1.038	0.981	1.099	1.303	0.193
Jourssamedi	1.041	0.984	1.101	1.387	0.166
Joursvendredi	1.028	0.972	1.088	0.964	0.335
Vacances1	0.991	0.943	1.041	-0.374	0.708
heat_wave1:summer1	0.718	0.540	0.955	-2.276	0.023
Standard errors: MLE

Clermont-Ferrand

summ(m.outcome3[[3]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1466 (6 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	63.101
Pseudo-R² (Cragg-Uhler)	0.043
Pseudo-R² (McFadden)	0.010
AIC	6535.672
BIC	6673.219

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	4.947	4.327	5.655	23.417	0.000
heat_wave1	1.585	1.336	1.880	5.289	0.000
annee2001	0.903	0.791	1.031	-1.509	0.131
annee2002	0.997	0.877	1.135	-0.039	0.969
annee2003	1.065	0.937	1.211	0.969	0.333
annee2004	0.956	0.839	1.090	-0.670	0.503
annee2005	0.956	0.839	1.090	-0.671	0.502
annee2006	0.911	0.798	1.040	-1.381	0.167
annee2007	0.980	0.860	1.115	-0.312	0.755
annee2008	1.066	0.939	1.210	0.981	0.327
annee2009	1.063	0.936	1.208	0.942	0.346
annee2010	1.089	0.960	1.235	1.321	0.186
annee2011	1.008	0.887	1.146	0.121	0.904
annee2012	1.056	0.930	1.199	0.839	0.402
annee2013	1.089	0.959	1.235	1.318	0.188
annee2014	1.059	0.933	1.203	0.884	0.377
annee2015	1.093	0.962	1.242	1.363	0.173
summer1	1.003	0.934	1.077	0.088	0.930
Joursjeudi	1.030	0.946	1.121	0.685	0.493
Jourslundi	1.028	0.944	1.119	0.636	0.525
Joursmardi	1.052	0.967	1.144	1.186	0.236
Joursmercredi	1.044	0.959	1.136	0.999	0.318
Jourssamedi	1.044	0.960	1.136	1.010	0.312
Joursvendredi	1.043	0.958	1.134	0.968	0.333
Vacances1	0.987	0.916	1.062	-0.357	0.721
heat_wave1:summer1	0.793	0.554	1.136	-1.263	0.206
Standard errors: MLE

Dijon

summ(m.outcome3[[4]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1449 (23 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	62.281
Pseudo-R² (Cragg-Uhler)	0.043
Pseudo-R² (McFadden)	0.010
AIC	6114.135
BIC	6251.380

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.543	3.045	4.122	16.375	0.000
heat_wave1	1.529	1.284	1.822	4.758	0.000
annee2001	1.053	0.908	1.222	0.680	0.496
annee2002	1.113	0.960	1.290	1.423	0.155
annee2003	1.266	1.096	1.462	3.203	0.001
annee2004	1.034	0.891	1.202	0.444	0.657
annee2005	1.051	0.906	1.219	0.653	0.513
annee2006	1.071	0.924	1.242	0.915	0.360
annee2007	1.094	0.945	1.267	1.206	0.228
annee2008	1.122	0.969	1.298	1.540	0.124
annee2009	1.130	0.976	1.308	1.638	0.101
annee2010	1.111	0.960	1.287	1.412	0.158
annee2011	1.140	0.986	1.318	1.767	0.077
annee2012	1.208	1.047	1.395	2.583	0.010
annee2013	1.249	1.083	1.441	3.060	0.002
annee2014	1.196	1.036	1.381	2.444	0.015
annee2015	1.205	1.043	1.393	2.531	0.011
summer1	1.024	0.947	1.108	0.606	0.545
Joursjeudi	1.026	0.934	1.127	0.530	0.596
Jourslundi	1.043	0.950	1.145	0.877	0.380
Joursmardi	1.059	0.965	1.163	1.206	0.228
Joursmercredi	1.058	0.963	1.162	1.174	0.240
Jourssamedi	1.035	0.942	1.137	0.720	0.471
Joursvendredi	1.036	0.943	1.138	0.734	0.463
Vacances1	1.021	0.941	1.107	0.495	0.620
heat_wave1:summer1	0.786	0.566	1.093	-1.430	0.153
Standard errors: MLE

Grenoble

summ(m.outcome3[[5]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	96.487
Pseudo-R² (Cragg-Uhler)	0.064
Pseudo-R² (McFadden)	0.013
AIC	7113.303
BIC	7250.939

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	6.809	6.077	7.630	33.042	0.000
heat_wave1	1.313	1.154	1.493	4.142	0.000
annee2001	1.064	0.954	1.187	1.122	0.262
annee2002	1.005	0.899	1.123	0.087	0.931
annee2003	1.195	1.073	1.331	3.243	0.001
annee2004	0.976	0.873	1.091	-0.428	0.669
annee2005	1.004	0.899	1.122	0.070	0.944
annee2006	0.974	0.872	1.089	-0.455	0.649
annee2007	0.922	0.823	1.032	-1.412	0.158
annee2008	1.027	0.920	1.147	0.481	0.631
annee2009	1.013	0.907	1.131	0.230	0.818
annee2010	1.041	0.933	1.162	0.719	0.472
annee2011	1.070	0.959	1.193	1.210	0.226
annee2012	1.093	0.981	1.218	1.606	0.108
annee2013	1.167	1.049	1.299	2.842	0.004
annee2014	1.115	1.001	1.242	1.970	0.049
annee2015	1.208	1.086	1.344	3.483	0.000
summer1	1.009	0.950	1.073	0.305	0.760
Joursjeudi	1.016	0.946	1.092	0.440	0.660
Jourslundi	1.071	0.998	1.150	1.896	0.058
Joursmardi	1.013	0.943	1.089	0.356	0.722
Joursmercredi	1.011	0.941	1.087	0.303	0.762
Jourssamedi	1.039	0.967	1.116	1.044	0.297
Joursvendredi	1.023	0.952	1.099	0.620	0.535
Vacances1	0.954	0.896	1.015	-1.487	0.137
heat_wave1:summer1	0.788	0.608	1.022	-1.799	0.072
Standard errors: MLE

La Havre

summ(m.outcome3[[6]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1467 (5 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	45.620
Pseudo-R² (Cragg-Uhler)	0.031
Pseudo-R² (McFadden)	0.007
AIC	6602.638
BIC	6740.203

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	4.728	4.145	5.392	23.155	0.000
heat_wave1	1.268	1.046	1.538	2.417	0.016
annee2001	0.992	0.872	1.128	-0.125	0.901
annee2002	1.032	0.909	1.172	0.484	0.628
annee2003	1.000	0.880	1.137	0.001	0.999
annee2004	1.092	0.963	1.237	1.370	0.171
annee2005	0.939	0.824	1.070	-0.942	0.346
annee2006	1.157	1.022	1.310	2.308	0.021
annee2007	1.029	0.906	1.169	0.443	0.658
annee2008	1.012	0.890	1.150	0.179	0.858
annee2009	1.018	0.896	1.157	0.270	0.787
annee2010	1.048	0.923	1.190	0.724	0.469
annee2011	1.036	0.913	1.177	0.550	0.582
annee2012	1.007	0.886	1.144	0.103	0.918
annee2013	1.068	0.941	1.211	1.022	0.307
annee2014	1.020	0.897	1.160	0.305	0.761
annee2015	1.125	0.993	1.274	1.850	0.064
summer1	1.087	1.015	1.165	2.373	0.018
Joursjeudi	1.026	0.944	1.115	0.603	0.546
Jourslundi	1.029	0.947	1.119	0.680	0.497
Joursmardi	1.046	0.962	1.136	1.056	0.291
Joursmercredi	1.051	0.967	1.142	1.172	0.241
Jourssamedi	0.991	0.911	1.078	-0.212	0.832
Joursvendredi	0.976	0.897	1.062	-0.563	0.574
Vacances1	1.024	0.953	1.100	0.646	0.518
heat_wave1:summer1	1.116	0.790	1.576	0.621	0.534
Standard errors: MLE

Lille

summ(m.outcome3[[7]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	73.453
Pseudo-R² (Cragg-Uhler)	0.049
Pseudo-R² (McFadden)	0.009
AIC	8556.169
BIC	8693.823

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	19.780	18.473	21.180	85.551	0.000
heat_wave1	1.271	1.170	1.381	5.650	0.000
annee2001	0.975	0.913	1.040	-0.769	0.442
annee2002	0.970	0.909	1.035	-0.916	0.360
annee2003	0.994	0.931	1.061	-0.178	0.859
annee2004	0.962	0.901	1.027	-1.158	0.247
annee2005	0.982	0.920	1.048	-0.546	0.585
annee2006	0.957	0.896	1.022	-1.299	0.194
annee2007	0.928	0.869	0.992	-2.213	0.027
annee2008	0.932	0.872	0.995	-2.109	0.035
annee2009	0.937	0.878	1.001	-1.930	0.054
annee2010	0.982	0.920	1.048	-0.552	0.581
annee2011	0.948	0.888	1.012	-1.604	0.109
annee2012	0.938	0.879	1.002	-1.902	0.057
annee2013	0.973	0.912	1.039	-0.809	0.419
annee2014	0.975	0.914	1.041	-0.760	0.447
annee2015	0.979	0.918	1.046	-0.623	0.533
summer1	0.976	0.941	1.013	-1.262	0.207
Joursjeudi	1.059	1.014	1.107	2.570	0.010
Jourslundi	1.070	1.024	1.118	3.002	0.003
Joursmardi	1.083	1.037	1.131	3.572	0.000
Joursmercredi	1.060	1.015	1.108	2.613	0.009
Jourssamedi	1.031	0.987	1.078	1.362	0.173
Joursvendredi	1.061	1.016	1.109	2.660	0.008
Vacances1	0.954	0.918	0.992	-2.391	0.017
heat_wave1:summer1	0.701	0.543	0.905	-2.725	0.006
Standard errors: MLE

Lyon

summ(m.outcome3[[8]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	353.052
Pseudo-R² (Cragg-Uhler)	0.214
Pseudo-R² (McFadden)	0.040
AIC	8626.587
BIC	8764.241

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	16.282	15.139	17.512	75.112	0.000
heat_wave1	1.621	1.497	1.755	11.901	0.000
annee2001	1.041	0.971	1.116	1.139	0.255
annee2002	1.023	0.954	1.097	0.639	0.523
annee2003	1.252	1.170	1.339	6.528	0.000
annee2004	0.947	0.882	1.017	-1.502	0.133
annee2005	0.944	0.879	1.013	-1.599	0.110
annee2006	0.974	0.908	1.045	-0.725	0.468
annee2007	0.943	0.878	1.012	-1.618	0.106
annee2008	1.030	0.961	1.104	0.837	0.402
annee2009	0.958	0.892	1.028	-1.198	0.231
annee2010	1.010	0.941	1.083	0.266	0.790
annee2011	0.975	0.909	1.047	-0.690	0.490
annee2012	0.996	0.928	1.068	-0.115	0.908
annee2013	0.933	0.869	1.002	-1.899	0.058
annee2014	0.965	0.900	1.036	-0.973	0.331
annee2015	1.034	0.964	1.108	0.928	0.354
summer1	1.026	0.987	1.067	1.297	0.195
Joursjeudi	1.041	0.993	1.090	1.681	0.093
Jourslundi	1.046	0.999	1.096	1.902	0.057
Joursmardi	1.037	0.990	1.086	1.521	0.128
Joursmercredi	1.044	0.996	1.093	1.811	0.070
Jourssamedi	1.006	0.960	1.055	0.270	0.787
Joursvendredi	1.090	1.041	1.141	3.673	0.000
Vacances1	0.990	0.951	1.031	-0.465	0.642
heat_wave1:summer1	0.654	0.549	0.778	-4.770	0.000
Standard errors: MLE

Marseille

summ(m.outcome3[[9]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	164.302
Pseudo-R² (Cragg-Uhler)	0.106
Pseudo-R² (McFadden)	0.019
AIC	8710.299
BIC	8847.953

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	20.908	19.562	22.346	89.578	0.000
heat_wave1	1.171	1.088	1.261	4.211	0.000
annee2001	0.947	0.889	1.009	-1.674	0.094
annee2002	0.987	0.927	1.051	-0.401	0.688
annee2003	1.116	1.049	1.187	3.476	0.001
annee2004	0.906	0.850	0.966	-3.011	0.003
annee2005	0.873	0.819	0.932	-4.094	0.000
annee2006	0.885	0.829	0.944	-3.702	0.000
annee2007	0.913	0.856	0.973	-2.801	0.005
annee2008	0.947	0.889	1.009	-1.673	0.094
annee2009	0.948	0.890	1.010	-1.638	0.102
annee2010	0.929	0.871	0.990	-2.276	0.023
annee2011	0.901	0.845	0.961	-3.182	0.001
annee2012	0.945	0.887	1.007	-1.749	0.080
annee2013	0.950	0.892	1.012	-1.578	0.115
annee2014	0.877	0.822	0.936	-3.969	0.000
annee2015	0.960	0.901	1.023	-1.271	0.204
summer1	0.986	0.950	1.022	-0.785	0.432
Joursjeudi	1.059	1.014	1.105	2.585	0.010
Jourslundi	1.051	1.007	1.098	2.270	0.023
Joursmardi	1.069	1.024	1.116	3.031	0.002
Joursmercredi	1.091	1.045	1.139	3.998	0.000
Jourssamedi	1.048	1.004	1.094	2.123	0.034
Joursvendredi	1.070	1.025	1.117	3.101	0.002
Vacances1	0.960	0.924	0.997	-2.139	0.032
heat_wave1:summer1	1.087	0.812	1.456	0.562	0.574
Standard errors: MLE

Montpellier

summ(m.outcome3[[10]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	45.322
Pseudo-R² (Cragg-Uhler)	0.031
Pseudo-R² (McFadden)	0.007
AIC	6655.152
BIC	6792.806

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.260	4.638	5.966	25.849	0.000
heat_wave1	1.302	1.086	1.560	2.849	0.004
annee2001	1.102	0.978	1.242	1.601	0.109
annee2002	0.976	0.864	1.104	-0.380	0.704
annee2003	1.074	0.951	1.212	1.152	0.249
annee2004	0.960	0.849	1.086	-0.645	0.519
annee2005	1.021	0.905	1.153	0.342	0.733
annee2006	1.036	0.917	1.170	0.569	0.569
annee2007	0.958	0.847	1.084	-0.677	0.499
annee2008	1.015	0.899	1.147	0.243	0.808
annee2009	1.071	0.949	1.207	1.114	0.265
annee2010	1.023	0.906	1.155	0.366	0.714
annee2011	0.986	0.873	1.115	-0.218	0.827
annee2012	1.042	0.923	1.176	0.670	0.503
annee2013	1.004	0.888	1.134	0.059	0.953
annee2014	1.102	0.978	1.242	1.597	0.110
annee2015	1.184	1.053	1.333	2.812	0.005
summer1	1.035	0.968	1.106	1.011	0.312
Joursjeudi	1.033	0.954	1.119	0.804	0.422
Jourslundi	1.007	0.929	1.090	0.160	0.873
Joursmardi	1.017	0.939	1.102	0.415	0.678
Joursmercredi	1.067	0.986	1.155	1.612	0.107
Jourssamedi	0.984	0.908	1.067	-0.392	0.695
Joursvendredi	1.049	0.969	1.135	1.177	0.239
Vacances1	1.036	0.967	1.110	1.004	0.316
heat_wave1:summer1	0.699	0.517	0.946	-2.320	0.020
Standard errors: MLE

Nancy

summ(m.outcome3[[11]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1470 (2 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	99.989
Pseudo-R² (Cragg-Uhler)	0.066
Pseudo-R² (McFadden)	0.014
AIC	6982.979
BIC	7120.597

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	6.036	5.347	6.815	29.044	0.000
heat_wave1	1.412	1.226	1.626	4.782	0.000
annee2001	1.111	0.990	1.248	1.792	0.073
annee2002	1.109	0.987	1.245	1.746	0.081
annee2003	1.231	1.098	1.380	3.571	0.000
annee2004	0.905	0.801	1.022	-1.610	0.107
annee2005	0.983	0.873	1.108	-0.278	0.781
annee2006	1.043	0.927	1.173	0.702	0.482
annee2007	1.005	0.892	1.131	0.078	0.938
annee2008	0.962	0.853	1.085	-0.630	0.529
annee2009	0.999	0.887	1.126	-0.012	0.991
annee2010	1.165	1.039	1.306	2.613	0.009
annee2011	1.100	0.980	1.236	1.618	0.106
annee2012	1.048	0.932	1.178	0.788	0.431
annee2013	1.113	0.992	1.249	1.816	0.069
annee2014	1.052	0.936	1.183	0.853	0.394
annee2015	1.173	1.046	1.315	2.720	0.007
summer1	0.989	0.927	1.054	-0.350	0.726
Joursjeudi	0.984	0.911	1.062	-0.424	0.672
Jourslundi	1.028	0.953	1.108	0.712	0.476
Joursmardi	0.975	0.903	1.052	-0.653	0.514
Joursmercredi	0.997	0.924	1.076	-0.071	0.943
Jourssamedi	0.984	0.912	1.062	-0.411	0.681
Joursvendredi	1.016	0.942	1.095	0.409	0.682
Vacances1	0.987	0.923	1.055	-0.394	0.693
heat_wave1:summer1	0.865	0.658	1.136	-1.045	0.296
Standard errors: MLE

Nantes

summ(m.outcome3[[12]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	62.822
Pseudo-R² (Cragg-Uhler)	0.042
Pseudo-R² (McFadden)	0.008
AIC	7571.782
BIC	7709.436

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.515	7.706	9.408	42.060	0.000
heat_wave1	1.383	1.224	1.562	5.204	0.000
annee2001	1.103	1.002	1.214	1.999	0.046
annee2002	1.087	0.987	1.197	1.694	0.090
annee2003	1.152	1.048	1.268	2.922	0.003
annee2004	1.090	0.990	1.200	1.759	0.079
annee2005	1.070	0.971	1.179	1.375	0.169
annee2006	1.042	0.946	1.148	0.835	0.404
annee2007	1.086	0.986	1.195	1.670	0.095
annee2008	1.034	0.938	1.140	0.673	0.501
annee2009	1.065	0.967	1.174	1.283	0.199
annee2010	1.143	1.040	1.258	2.763	0.006
annee2011	1.090	0.990	1.200	1.749	0.080
annee2012	1.101	1.000	1.212	1.966	0.049
annee2013	1.152	1.047	1.266	2.919	0.004
annee2014	1.144	1.041	1.259	2.781	0.005
annee2015	1.201	1.093	1.319	3.812	0.000
summer1	1.010	0.959	1.064	0.378	0.706
Joursjeudi	1.041	0.979	1.108	1.274	0.203
Jourslundi	1.050	0.987	1.117	1.554	0.120
Joursmardi	1.039	0.977	1.106	1.213	0.225
Joursmercredi	1.019	0.957	1.084	0.590	0.555
Jourssamedi	1.000	0.939	1.065	0.003	0.997
Joursvendredi	1.010	0.949	1.074	0.298	0.766
Vacances1	0.980	0.929	1.035	-0.710	0.478
heat_wave1:summer1	0.839	0.540	1.304	-0.780	0.435
Standard errors: MLE

Nice

summ(m.outcome3[[13]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	152.918
Pseudo-R² (Cragg-Uhler)	0.099
Pseudo-R² (McFadden)	0.019
AIC	7842.001
BIC	7979.655

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	12.010	10.982	13.134	54.444	0.000
heat_wave1	1.296	1.165	1.442	4.774	0.000
annee2001	1.045	0.962	1.136	1.035	0.301
annee2002	0.994	0.914	1.081	-0.138	0.890
annee2003	1.150	1.059	1.248	3.335	0.001
annee2004	0.897	0.823	0.978	-2.462	0.014
annee2005	0.871	0.798	0.950	-3.118	0.002
annee2006	0.954	0.876	1.039	-1.080	0.280
annee2007	0.873	0.801	0.953	-3.050	0.002
annee2008	0.925	0.849	1.008	-1.775	0.076
annee2009	0.949	0.872	1.034	-1.197	0.231
annee2010	0.881	0.808	0.961	-2.856	0.004
annee2011	0.883	0.810	0.964	-2.799	0.005
annee2012	0.882	0.808	0.961	-2.848	0.004
annee2013	0.890	0.816	0.971	-2.634	0.008
annee2014	0.946	0.869	1.030	-1.275	0.202
annee2015	0.952	0.875	1.037	-1.126	0.260
summer1	0.945	0.899	0.992	-2.284	0.022
Joursjeudi	1.002	0.946	1.062	0.073	0.942
Jourslundi	1.058	0.999	1.120	1.914	0.056
Joursmardi	1.083	1.023	1.146	2.729	0.006
Joursmercredi	1.034	0.976	1.095	1.126	0.260
Jourssamedi	1.026	0.969	1.087	0.880	0.379
Joursvendredi	1.056	0.997	1.119	1.873	0.061
Vacances1	0.963	0.915	1.013	-1.452	0.147
heat_wave1:summer1	0.611	0.374	1.000	-1.958	0.050
Standard errors: MLE

Paris

summ(m.outcome3[[14]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	3411.810
Pseudo-R² (Cragg-Uhler)	0.902
Pseudo-R² (McFadden)	0.198
AIC	13835.972
BIC	13973.626

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	102.034	99.022	105.138	302.564	0.000
heat_wave1	2.066	2.005	2.129	47.702	0.000
annee2001	1.028	0.999	1.057	1.908	0.056
annee2002	0.969	0.942	0.997	-2.178	0.029
annee2003	1.174	1.142	1.206	11.524	0.000
annee2004	0.908	0.882	0.935	-6.484	0.000
annee2005	0.910	0.884	0.937	-6.387	0.000
annee2006	0.882	0.857	0.908	-8.455	0.000
annee2007	0.910	0.884	0.937	-6.352	0.000
annee2008	0.916	0.889	0.942	-5.967	0.000
annee2009	0.912	0.886	0.939	-6.230	0.000
annee2010	0.917	0.890	0.944	-5.890	0.000
annee2011	0.926	0.900	0.953	-5.205	0.000
annee2012	0.895	0.870	0.922	-7.487	0.000
annee2013	0.920	0.894	0.947	-5.681	0.000
annee2014	0.911	0.885	0.937	-6.338	0.000
annee2015	0.913	0.887	0.940	-6.109	0.000
summer1	0.999	0.983	1.015	-0.140	0.889
Joursjeudi	1.058	1.037	1.078	5.683	0.000
Jourslundi	1.056	1.036	1.076	5.508	0.000
Joursmardi	1.066	1.046	1.087	6.501	0.000
Joursmercredi	1.056	1.035	1.076	5.504	0.000
Jourssamedi	1.022	1.003	1.042	2.228	0.026
Joursvendredi	1.045	1.025	1.066	4.478	0.000
Vacances1	0.967	0.951	0.984	-3.855	0.000
heat_wave1:summer1	0.666	0.599	0.740	-7.583	0.000
Standard errors: MLE

Rennes

summ(m.outcome3[[15]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1441 (31 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	54.605
Pseudo-R² (Cragg-Uhler)	0.038
Pseudo-R² (McFadden)	0.009
AIC	5759.328
BIC	5896.428

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.242	2.753	3.818	14.100	0.000
heat_wave1	1.459	1.223	1.741	4.196	0.000
annee2001	0.987	0.841	1.157	-0.165	0.869
annee2002	1.144	0.980	1.335	1.705	0.088
annee2003	1.083	0.925	1.267	0.991	0.322
annee2004	0.961	0.816	1.131	-0.484	0.628
annee2005	1.034	0.882	1.213	0.414	0.679
annee2006	1.118	0.957	1.306	1.400	0.161
annee2007	0.994	0.847	1.167	-0.069	0.945
annee2008	1.035	0.883	1.213	0.421	0.674
annee2009	1.060	0.905	1.242	0.720	0.471
annee2010	0.919	0.781	1.081	-1.019	0.308
annee2011	1.191	1.022	1.388	2.242	0.025
annee2012	1.170	1.004	1.364	2.012	0.044
annee2013	1.162	0.997	1.356	1.916	0.055
annee2014	1.158	0.992	1.351	1.861	0.063
annee2015	1.138	0.974	1.330	1.628	0.104
summer1	1.013	0.929	1.104	0.287	0.774
Joursjeudi	1.049	0.948	1.161	0.924	0.355
Jourslundi	1.061	0.959	1.173	1.152	0.249
Joursmardi	1.005	0.908	1.113	0.099	0.921
Joursmercredi	0.965	0.871	1.070	-0.673	0.501
Jourssamedi	1.010	0.912	1.119	0.201	0.841
Joursvendredi	0.964	0.869	1.070	-0.688	0.491
Vacances1	1.012	0.924	1.107	0.250	0.802
heat_wave1:summer1	0.686	0.368	1.279	-1.186	0.236
Standard errors: MLE

Strasbourg

summ(m.outcome3[[16]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	118.453
Pseudo-R² (Cragg-Uhler)	0.078
Pseudo-R² (McFadden)	0.016
AIC	7206.725
BIC	7344.379

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	7.981	7.180	8.872	38.483	0.000
heat_wave1	1.584	1.407	1.783	7.630	0.000
annee2001	0.981	0.887	1.086	-0.367	0.714
annee2002	0.956	0.864	1.059	-0.858	0.391
annee2003	1.010	0.913	1.118	0.201	0.841
annee2004	0.921	0.831	1.020	-1.577	0.115
annee2005	0.868	0.782	0.963	-2.661	0.008
annee2006	0.911	0.822	1.010	-1.770	0.077
annee2007	0.980	0.886	1.084	-0.391	0.696
annee2008	0.944	0.853	1.046	-1.097	0.273
annee2009	0.960	0.867	1.062	-0.794	0.427
annee2010	0.926	0.836	1.026	-1.462	0.144
annee2011	0.890	0.803	0.988	-2.198	0.028
annee2012	0.900	0.812	0.998	-2.002	0.045
annee2013	1.002	0.907	1.109	0.048	0.962
annee2014	0.970	0.876	1.074	-0.587	0.557
annee2015	0.989	0.894	1.094	-0.215	0.830
summer1	1.031	0.974	1.092	1.059	0.290
Joursjeudi	0.983	0.918	1.053	-0.480	0.631
Jourslundi	1.061	0.992	1.135	1.721	0.085
Joursmardi	1.003	0.937	1.074	0.089	0.929
Joursmercredi	1.067	0.998	1.142	1.904	0.057
Jourssamedi	0.981	0.916	1.051	-0.537	0.592
Joursvendredi	1.009	0.942	1.080	0.245	0.806
Vacances1	1.002	0.944	1.063	0.054	0.957
heat_wave1:summer1	0.931	0.758	1.144	-0.679	0.497
Standard errors: MLE

Toulouse

summ(m.outcome3[[17]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	86.187
Pseudo-R² (Cragg-Uhler)	0.057
Pseudo-R² (McFadden)	0.011
AIC	7801.843
BIC	7939.496

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	9.823	8.956	10.774	48.458	0.000
heat_wave1	1.310	1.179	1.455	5.030	0.000
annee2001	1.064	0.974	1.163	1.376	0.169
annee2002	1.093	1.001	1.194	1.978	0.048
annee2003	1.181	1.081	1.289	3.700	0.000
annee2004	1.043	0.954	1.140	0.917	0.359
annee2005	0.997	0.911	1.091	-0.066	0.947
annee2006	1.022	0.935	1.118	0.484	0.628
annee2007	0.991	0.905	1.085	-0.195	0.846
annee2008	1.064	0.974	1.163	1.369	0.171
annee2009	1.037	0.949	1.134	0.801	0.423
annee2010	1.084	0.992	1.184	1.788	0.074
annee2011	1.117	1.023	1.220	2.471	0.013
annee2012	1.042	0.953	1.139	0.894	0.371
annee2013	1.130	1.035	1.234	2.739	0.006
annee2014	1.042	0.953	1.139	0.895	0.371
annee2015	1.117	1.024	1.220	2.479	0.013
summer1	1.023	0.975	1.074	0.931	0.352
Joursjeudi	1.074	1.014	1.138	2.414	0.016
Jourslundi	1.020	0.962	1.082	0.677	0.499
Joursmardi	1.035	0.976	1.098	1.163	0.245
Joursmercredi	1.047	0.988	1.110	1.558	0.119
Jourssamedi	1.017	0.959	1.079	0.565	0.572
Joursvendredi	1.083	1.022	1.148	2.705	0.007
Vacances1	0.988	0.939	1.039	-0.463	0.644
heat_wave1:summer1	1.003	0.692	1.454	0.017	0.987
Standard errors: MLE

Metanalysis of the results for the interaction term

meta3bis<-rma(v.coeff3,sei=v.sterr3)

summary(meta3bis)

## 
## Random-Effects Model (k = 17; tau^2 estimator: REML)
## 
##   logLik  deviance       AIC       BIC      AICc 
##   3.9484   -7.8968   -3.8968   -2.3516   -2.9737   
## 
## tau^2 (estimated amount of total heterogeneity): 0.0141 (SE = 0.0122)
## tau (square root of estimated tau^2 value):      0.1188
## I^2 (total heterogeneity / total variability):   44.37%
## H^2 (total variability / sampling variability):  1.80
## 
## Test for Heterogeneity:
## Q(df = 16) = 28.3659, p-val = 0.0286
## 
## Model Results:
## 
## estimate      se     zval    pval    ci.lb    ci.ub 
##  -0.2451  0.0466  -5.2614  <.0001  -0.3364  -0.1538  *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

forest(meta3bis,transf = exp, slab = names(villes_s2),refline=1,xlab="RR Interaction First part of the summer", header="City")

Third model strategy - considering order of heat wave

After a check of heat wave occurences, this is the list of cities and years when heat waves occured more than once per summer:

dijon: 2003 2015
grenoble: 2015
le havre: 2001 2003 2006
lyon: 2003 2015
marseille: 2003 2015
montpellier: 2003 2006 (2015)
nancy: 2015
nantes: 2003 2005
nice: 2006 2015
paris: 2006? 2015
rennes: 2001 2003 2006
rouen: 2013
strasbourg: 2015
toulouse: 2003

we have to be carefull as we saw that different extreme-heat episodes are separated by only one day: they should be then considered as part of the same heat wave.

Define new variable “order of heat wave”

I first put together heat wave separated by only one day:

for (i in 1:length(villes_s)){
villes_s[[i]]$orderhw<-NA
villes_s[[i]]$heat_wave<-as.numeric(as.character(villes_s[[i]]$heat_wave))
}

## Put together heat wave episodes separated only by one days 

for (i in 1:length(villes_s)){
  for (r in 2:nrow(villes_s[[i]])){
  villes_s[[i]]$heatwave[1]<-0  
villes_s[[i]]$heatwave[r]<-ifelse(villes_s[[i]]$heat_wave[r-1]==1 & villes_s[[i]]$heat_wave[r+1]==1,1,villes_s[[i]]$heat_wave[r])
 }
}

lapply(villes_s,function(x){table(x$heatwave,x$heat_wave)})

## $BM
##    
##        0    1
##   0 1441    0
##   1    1   30
## 
## $bordeaux
##    
##        0    1
##   0 1446    0
##   1    0   26
## 
## $clermont
##    
##        0    1
##   0 1448    0
##   1    0   24
## 
## $dijon
##    
##        0    1
##   0 1441    0
##   1    0   31
## 
## $grenoble
##    
##        0    1
##   0 1435    0
##   1    1   36
## 
## $lehavre
##    
##        0    1
##   0 1449    0
##   1    0   23
## 
## $lille
##    
##        0    1
##   0 1441    0
##   1    1   30
## 
## $lyon
##    
##        0    1
##   0 1439    0
##   1    1   32
## 
## $marseille
##    
##        0    1
##   0 1437    0
##   1    0   35
## 
## $montpellier
##    
##        0    1
##   0 1444    0
##   1    0   28
## 
## $nancy
##    
##        0    1
##   0 1440    0
##   1    0   32
## 
## $nantes
##    
##        0    1
##   0 1448    0
##   1    0   24
## 
## $nice
##    
##        0    1
##   0 1446    0
##   1    0   26
## 
## $paris
##    
##        0    1
##   0 1445    0
##   1    0   27
## 
## $rennes
##    
##        0    1
##   0 1440    0
##   1    0   32
## 
## $rouen
##    
##        0    1
##   0 1447    0
##   1    0   25
## 
## $strasbourg
##    
##        0    1
##   0 1434    0
##   1    0   38
## 
## $toulouse
##    
##        0    1
##   0 1443    0
##   1    0   29

Here we can see that only a few days that were not considered heat wave according to the first definition have been now included in a heat wave episode in order to put together to heat waves separated only by one day.

With this new heat wave variable, I can now create the variable of order of heat waves:

year<-levels(villes_s[[1]]$annee)

#define first day for each heat wave
for (i in 1:length(villes_s)){
  for (r in 2:nrow(villes_s[[i]])){
    villes_s[[i]]$firstd[1]<-0
    villes_s[[i]]$firstd[r]<-if (villes_s[[i]]$heatwave[r]==1 & villes_s[[i]]$heatwave[r-1]==0) {1} else {0}
 }
}

for (i in 1:length(villes_s)){ # pour exécuter la boucle pour chaque villes
  for (y in 1:16){ # pour exécuter la boucle pour chaque année
    for (r in 1:92){ # pour exécuter la boucle pour chaque jour 
    villes_s[[i]]$orderhw[which (villes_s[[i]]$annee == year[y] & villes_s[[i]]$day==r)]<-sum(villes_s[[i]]$firstd[which (villes_s[[i]]$annee == year[y])][1:r])
   }
  }
# pour éviter que la somme des épisodes de HW se repete même les jours hors HW
villes_s[[i]]$orderhw[which(villes_s[[i]]$heatwave==0)]<-0
}


lapply(villes_s,function(x){table(x$orderhw,x$annee)})

## $BM
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   90   91   82   92   92   81   92   92   92   90   92   91   91   91
##   1    0    2    1    1    0    0   11    0    0    0    2    0    1    1    1
##   2    0    0    0    9    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   90
##   1    2
##   2    0
## 
## $bordeaux
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   92   80   92   91   85   92   92   91   92   91   90   90   92
##   1    0    0    0    1    0    1    7    0    0    1    0    1    2    2    0
##   2    0    0    0    1    0    0    0    0    0    0    0    0    0    0    0
##   3    0    0    0   10    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   92
##   1    0
##   2    0
##   3    0
## 
## $clermont
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   92   82   92   92   91   92   92   92   92   89   90   92   92
##   1    0    0    0   10    0    0    1    0    0    0    0    3    2    0    0
##   2    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   84
##   1    6
##   2    2
## 
## $dijon
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   92   77   92   91   85   92   92   92   92   92   92   92   92
##   1    0    0    0    2    0    1    7    0    0    0    0    0    0    0    0
##   2    0    0    0    2    0    0    0    0    0    0    0    0    0    0    0
##   3    0    0    0   11    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   84
##   1    5
##   2    3
##   3    0
## 
## $grenoble
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   92   77   90   91   89   92   92   90   92   91   90   92   92
##   1    0    0    0    2    2    1    3    0    0    2    0    1    2    0    0
##   2    0    0    0    2    0    0    0    0    0    0    0    0    0    0    0
##   3    0    0    0    1    0    0    0    0    0    0    0    0    0    0    0
##   4    0    0    0   10    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   81
##   1    4
##   2    5
##   3    2
##   4    0
## 
## $lehavre
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   88   92   87   92   92   85   92   92   89   92   92   90   90   92
##   1    0    2    0    1    0    0    3    0    0    3    0    0    2    2    0
##   2    0    2    0    1    0    0    4    0    0    0    0    0    0    0    0
##   3    0    0    0    3    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   92
##   1    0
##   2    0
##   3    0
## 
## $lille
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   90   91   82   92   92   81   92   92   92   90   92   91   91   91
##   1    0    2    1    1    0    0   11    0    0    0    2    0    1    1    1
##   2    0    0    0    9    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   90
##   1    2
##   2    0
## 
## $lyon
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   92   77   92   91   85   92   92   91   92   91   92   91   92
##   1    0    0    0    1    0    1    7    0    0    1    0    1    0    1    0
##   2    0    0    0    2    0    0    0    0    0    0    0    0    0    0    0
##   3    0    0    0    1    0    0    0    0    0    0    0    0    0    0    0
##   4    0    0    0   11    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   85
##   1    4
##   2    2
##   3    1
##   4    0
## 
## $marseille
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   90   92   79   92   92   83   92   92   91   92   92   89   91   92
##   1    0    2    0    1    0    0    1    0    0    1    0    0    3    1    0
##   2    0    0    0   12    0    0    8    0    0    0    0    0    0    0    0
##   3    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   86
##   1    1
##   2    1
##   3    4
## 
## $montpellier
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   92   83   91   92   84   92   92   92   91   92   92   92   91
##   1    0    0    0    3    1    0    2    0    0    0    1    0    0    0    1
##   2    0    0    0    1    0    0    2    0    0    0    0    0    0    0    0
##   3    0    0    0    1    0    0    3    0    0    0    0    0    0    0    0
##   4    0    0    0    4    0    0    1    0    0    0    0    0    0    0    0
##    
##     2015
##   0   84
##   1    3
##   2    5
##   3    0
##   4    0
## 
## $nancy
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   90   81   92   92   86   92   92   92   90   92   90   92   92
##   1    0    0    2   11    0    0    2    0    0    0    2    0    2    0    0
##   2    0    0    0    0    0    0    4    0    0    0    0    0    0    0    0
##   3    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0
##   4    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   83
##   1    5
##   2    2
##   3    1
##   4    1
## 
## $nantes
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   91   92   92   83   91   87   87   92   92   92   92   92   91   90   92
##   1    1    0    0    1    1    1    5    0    0    0    0    0    1    2    0
##   2    0    0    0    8    0    4    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   92
##   1    0
##   2    0
## 
## $nice
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   92   83   92   92   84   92   92   89   92   92   89   92   92
##   1    0    0    0    9    0    0    5    0    0    3    0    0    3    0    0
##   2    0    0    0    0    0    0    3    0    0    0    0    0    0    0    0
##    
##     2015
##   0   89
##   1    1
##   2    2
## 
## $paris
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   91   82   92   92   86   92   92   92   92   92   89   90   91
##   1    0    0    1   10    0    0    5    0    0    0    0    0    3    2    1
##   2    0    0    0    0    0    0    1    0    0    0    0    0    0    0    0
##    
##     2015
##   0   88
##   1    3
##   2    1
## 
## $rennes
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   91   85   92   80   91   91   85   92   92   92   92   92   91   90   92
##   1    1    4    0    1    1    1    2    0    0    0    0    0    1    2    0
##   2    0    3    0   11    0    0    3    0    0    0    0    0    0    0    0
##   3    0    0    0    0    0    0    2    0    0    0    0    0    0    0    0
##    
##     2015
##   0   92
##   1    0
##   2    0
##   3    0
## 
## $rouen
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   90   91   82   92   92   86   92   92   92   92   92   91   88   91
##   1    0    2    1    1    0    0    5    0    0    0    0    0    1    3    1
##   2    0    0    0    9    0    0    1    0    0    0    0    0    0    1    0
##    
##     2015
##   0   92
##   1    0
##   2    0
## 
## $strasbourg
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   92   91   81   92   91   85   92   92   92   91   91   90   91   89
##   1    0    0    1   11    0    1    7    0    0    0    1    1    2    1    3
##   2    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0
##   3    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   82
##   1    5
##   2    2
##   3    3
## 
## $toulouse
##    
##     2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
##   0   92   91   92   79   91   92   88   92   92   91   92   90   87   91   92
##   1    0    1    0    1    1    0    2    0    0    1    0    2    5    1    0
##   2    0    0    0    1    0    0    2    0    0    0    0    0    0    0    0
##   3    0    0    0   11    0    0    0    0    0    0    0    0    0    0    0
##    
##     2015
##   0   91
##   1    1
##   2    0
##   3    0

We still have 4 heat wave episodes in a summer in some cities: Grenoble, Lyon, Montpellier in 2003, Nancy in 2015. 3 Heat waves episodes in Marseille and Lyon in 2015

Model

First version with order variable as numeric

I only put the interaction term, not the variable order alone (?). With this method I suppose that the gravity of the heat wave lineary changes according to the order.

m.outcome4<-list()
v.coeff4<-c()
v.sterr4<-c()

for (i in 1:length(villes_s)){
m.outcome4[[i]]<-glm(nocc_tot~heatwave+annee+orderhw:heatwave+Jours+Vacances,data=villes_s[[i]],family=poisson)
v.coeff4[i]<-m.outcome4[[i]]$coefficients[25]
v.sterr4[i]<-summary(m.outcome4[[i]])$coefficients[25, 2]
}

names(m.outcome4)<-names(villes_s)

#lapply(m.outcome4,function(x){summary(x)})

Results per city

summ(m.outcome4[[1]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	40.934
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	7276.193
BIC	7408.536

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	7.430	6.784	8.136	43.248	0.000
heatwave	0.735	0.508	1.064	-1.633	0.103
annee2001	1.017	0.916	1.129	0.311	0.755
annee2002	1.044	0.941	1.158	0.811	0.418
annee2003	1.104	0.994	1.227	1.850	0.064
annee2004	0.980	0.882	1.089	-0.379	0.705
annee2005	0.977	0.879	1.085	-0.441	0.659
annee2006	1.045	0.941	1.161	0.823	0.411
annee2007	1.030	0.928	1.143	0.560	0.576
annee2008	1.034	0.932	1.147	0.627	0.530
annee2009	1.009	0.909	1.121	0.170	0.865
annee2010	1.042	0.939	1.156	0.782	0.434
annee2011	1.017	0.916	1.129	0.309	0.757
annee2012	1.108	1.001	1.228	1.973	0.048
annee2013	0.995	0.896	1.105	-0.089	0.929
annee2014	1.064	0.959	1.179	1.169	0.243
annee2015	1.054	0.951	1.169	1.001	0.317
Joursjeudi	1.029	0.961	1.103	0.819	0.413
Jourslundi	1.071	1.000	1.146	1.956	0.051
Joursmardi	1.042	0.973	1.116	1.171	0.242
Joursmercredi	1.058	0.988	1.133	1.605	0.109
Jourssamedi	0.987	0.921	1.058	-0.366	0.714
Joursvendredi	1.077	1.006	1.153	2.138	0.032
Vacances1	0.976	0.939	1.014	-1.234	0.217
heatwave:orderhw	1.374	1.063	1.775	2.427	0.015
Standard errors: MLE

Bordeaux

summ(m.outcome4[[2]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	143.488
Pseudo-R² (Cragg-Uhler)	0.093
Pseudo-R² (McFadden)	0.018
AIC	7924.407
BIC	8056.766

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.208	10.407	12.070	63.908	0.000
heatwave	1.155	0.937	1.424	1.348	0.178
annee2001	1.040	0.956	1.132	0.911	0.362
annee2002	1.016	0.934	1.106	0.374	0.708
annee2003	1.050	0.963	1.145	1.098	0.272
annee2004	0.945	0.867	1.031	-1.272	0.204
annee2005	0.931	0.854	1.015	-1.615	0.106
annee2006	0.995	0.913	1.084	-0.123	0.902
annee2007	0.941	0.863	1.026	-1.370	0.171
annee2008	0.986	0.905	1.074	-0.323	0.747
annee2009	0.956	0.877	1.042	-1.016	0.310
annee2010	1.030	0.946	1.120	0.674	0.501
annee2011	0.990	0.909	1.079	-0.223	0.823
annee2012	0.971	0.891	1.058	-0.665	0.506
annee2013	1.012	0.930	1.102	0.286	0.775
annee2014	1.033	0.949	1.124	0.757	0.449
annee2015	1.069	0.983	1.162	1.552	0.121
Joursjeudi	1.048	0.991	1.109	1.645	0.100
Jourslundi	1.049	0.991	1.109	1.651	0.099
Joursmardi	1.020	0.964	1.080	0.686	0.492
Joursmercredi	1.038	0.981	1.098	1.285	0.199
Jourssamedi	1.042	0.985	1.103	1.439	0.150
Joursvendredi	1.030	0.973	1.090	1.016	0.310
Vacances1	0.990	0.960	1.022	-0.594	0.552
heatwave:orderhw	1.184	1.075	1.305	3.424	0.001
Standard errors: MLE

Clermont-Ferrand

summ(m.outcome4[[3]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1466 (6 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	62.748
Pseudo-R² (Cragg-Uhler)	0.042
Pseudo-R² (McFadden)	0.010
AIC	6534.025
BIC	6666.282

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	4.961	4.435	5.550	27.991	0.000
heatwave	2.134	1.120	4.066	2.303	0.021
annee2001	0.903	0.791	1.031	-1.508	0.131
annee2002	0.997	0.877	1.135	-0.039	0.969
annee2003	1.071	0.943	1.216	1.051	0.293
annee2004	0.956	0.840	1.090	-0.669	0.503
annee2005	0.956	0.839	1.090	-0.670	0.503
annee2006	0.912	0.799	1.040	-1.372	0.170
annee2007	0.980	0.861	1.115	-0.311	0.756
annee2008	1.066	0.939	1.210	0.982	0.326
annee2009	1.063	0.936	1.208	0.943	0.346
annee2010	1.089	0.960	1.236	1.322	0.186
annee2011	1.010	0.888	1.148	0.145	0.885
annee2012	1.057	0.931	1.200	0.856	0.392
annee2013	1.089	0.959	1.235	1.318	0.187
annee2014	1.059	0.933	1.203	0.885	0.376
annee2015	1.085	0.956	1.232	1.258	0.209
Joursjeudi	1.028	0.944	1.118	0.633	0.526
Jourslundi	1.026	0.942	1.116	0.586	0.558
Joursmardi	1.050	0.965	1.142	1.140	0.254
Joursmercredi	1.043	0.958	1.135	0.975	0.330
Jourssamedi	1.044	0.960	1.136	1.006	0.314
Joursvendredi	1.040	0.956	1.131	0.907	0.365
Vacances1	0.987	0.941	1.034	-0.561	0.575
heatwave:orderhw	0.720	0.399	1.299	-1.091	0.275
Standard errors: MLE

Dijon

summ(m.outcome4[[4]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1449 (23 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	74.605
Pseudo-R² (Cragg-Uhler)	0.051
Pseudo-R² (McFadden)	0.012
AIC	6099.810
BIC	6231.776

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.645	3.206	4.143	19.772	0.000
heatwave	0.770	0.532	1.115	-1.384	0.166
annee2001	1.053	0.908	1.222	0.683	0.495
annee2002	1.114	0.961	1.291	1.430	0.153
annee2003	1.207	1.041	1.400	2.488	0.013
annee2004	1.035	0.891	1.202	0.445	0.657
annee2005	1.052	0.907	1.220	0.672	0.502
annee2006	1.109	0.957	1.286	1.380	0.168
annee2007	1.093	0.944	1.266	1.195	0.232
annee2008	1.121	0.968	1.297	1.529	0.126
annee2009	1.129	0.976	1.307	1.631	0.103
annee2010	1.111	0.960	1.287	1.409	0.159
annee2011	1.139	0.986	1.317	1.763	0.078
annee2012	1.207	1.046	1.393	2.569	0.010
annee2013	1.248	1.082	1.439	3.049	0.002
annee2014	1.195	1.035	1.379	2.430	0.015
annee2015	1.218	1.054	1.407	2.674	0.008
Joursjeudi	1.021	0.929	1.122	0.434	0.664
Jourslundi	1.038	0.945	1.140	0.786	0.432
Joursmardi	1.057	0.963	1.161	1.170	0.242
Joursmercredi	1.054	0.960	1.158	1.103	0.270
Jourssamedi	1.034	0.941	1.136	0.702	0.483
Joursvendredi	1.035	0.942	1.137	0.716	0.474
Vacances1	0.999	0.948	1.053	-0.034	0.973
heatwave:orderhw	1.369	1.164	1.610	3.788	0.000
Standard errors: MLE

Grenoble

summ(m.outcome4[[5]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	95.661
Pseudo-R² (Cragg-Uhler)	0.063
Pseudo-R² (McFadden)	0.013
AIC	7112.129
BIC	7244.471

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	6.859	6.235	7.547	39.537	0.000
heatwave	1.010	0.807	1.263	0.085	0.932
annee2001	1.065	0.954	1.187	1.123	0.262
annee2002	1.005	0.899	1.123	0.088	0.930
annee2003	1.176	1.054	1.312	2.892	0.004
annee2004	0.980	0.877	1.096	-0.349	0.727
annee2005	1.003	0.898	1.121	0.056	0.956
annee2006	0.981	0.877	1.096	-0.342	0.733
annee2007	0.922	0.823	1.032	-1.415	0.157
annee2008	1.027	0.920	1.147	0.478	0.633
annee2009	1.017	0.911	1.136	0.308	0.758
annee2010	1.041	0.933	1.162	0.718	0.473
annee2011	1.072	0.961	1.195	1.249	0.212
annee2012	1.097	0.985	1.223	1.682	0.093
annee2013	1.167	1.049	1.299	2.839	0.005
annee2014	1.114	1.000	1.241	1.967	0.049
annee2015	1.208	1.086	1.344	3.470	0.001
Joursjeudi	1.017	0.947	1.093	0.472	0.637
Jourslundi	1.070	0.996	1.148	1.857	0.063
Joursmardi	1.015	0.945	1.091	0.417	0.677
Joursmercredi	1.013	0.943	1.089	0.354	0.723
Jourssamedi	1.039	0.967	1.116	1.038	0.299
Joursvendredi	1.024	0.953	1.100	0.643	0.520
Vacances1	0.949	0.912	0.987	-2.599	0.009
heatwave:orderhw	1.090	1.001	1.187	1.987	0.047
Standard errors: MLE

La Havre

summ(m.outcome4[[6]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1467 (5 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	42.555
Pseudo-R² (Cragg-Uhler)	0.029
Pseudo-R² (McFadden)	0.006
AIC	6603.703
BIC	6735.978

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.167	4.627	5.771	29.122	0.000
heatwave	0.952	0.647	1.401	-0.248	0.804
annee2001	0.991	0.872	1.128	-0.131	0.896
annee2002	1.034	0.910	1.174	0.508	0.611
annee2003	0.989	0.869	1.125	-0.174	0.862
annee2004	1.092	0.963	1.237	1.370	0.171
annee2005	0.939	0.824	1.069	-0.951	0.342
annee2006	1.158	1.023	1.311	2.321	0.020
annee2007	1.027	0.904	1.166	0.405	0.685
annee2008	1.009	0.888	1.147	0.142	0.887
annee2009	1.026	0.903	1.165	0.394	0.693
annee2010	1.046	0.922	1.188	0.700	0.484
annee2011	1.035	0.912	1.176	0.538	0.590
annee2012	1.006	0.885	1.143	0.089	0.929
annee2013	1.068	0.941	1.211	1.020	0.308
annee2014	1.017	0.895	1.156	0.260	0.795
annee2015	1.122	0.991	1.271	1.812	0.070
Joursjeudi	1.025	0.943	1.114	0.576	0.564
Jourslundi	1.029	0.947	1.119	0.682	0.495
Joursmardi	1.044	0.961	1.134	1.021	0.307
Joursmercredi	1.050	0.966	1.140	1.142	0.253
Jourssamedi	0.989	0.908	1.076	-0.264	0.791
Joursvendredi	0.974	0.896	1.060	-0.600	0.548
Vacances1	0.953	0.910	0.999	-2.021	0.043
heatwave:orderhw	1.211	0.978	1.501	1.756	0.079
Standard errors: MLE

Lille

summ(m.outcome4[[7]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	73.904
Pseudo-R² (Cragg-Uhler)	0.049
Pseudo-R² (McFadden)	0.009
AIC	8553.719
BIC	8686.078

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	19.287	18.222	20.414	102.101	0.000
heatwave	0.826	0.657	1.038	-1.638	0.101
annee2001	0.978	0.916	1.044	-0.671	0.502
annee2002	0.971	0.910	1.037	-0.873	0.383
annee2003	0.976	0.913	1.044	-0.712	0.477
annee2004	0.962	0.901	1.027	-1.158	0.247
annee2005	0.982	0.920	1.048	-0.539	0.590
annee2006	0.974	0.912	1.040	-0.785	0.432
annee2007	0.929	0.870	0.992	-2.193	0.028
annee2008	0.932	0.873	0.996	-2.089	0.037
annee2009	0.938	0.878	1.001	-1.923	0.055
annee2010	0.978	0.916	1.043	-0.679	0.497
annee2011	0.948	0.888	1.012	-1.597	0.110
annee2012	0.941	0.881	1.005	-1.824	0.068
annee2013	0.976	0.914	1.042	-0.732	0.464
annee2014	0.978	0.916	1.043	-0.680	0.497
annee2015	0.976	0.914	1.041	-0.745	0.456
Joursjeudi	1.060	1.015	1.108	2.604	0.009
Jourslundi	1.068	1.022	1.116	2.927	0.003
Joursmardi	1.084	1.038	1.133	3.630	0.000
Joursmercredi	1.061	1.015	1.109	2.640	0.008
Jourssamedi	1.031	0.986	1.077	1.335	0.182
Joursvendredi	1.061	1.015	1.109	2.641	0.008
Vacances1	0.974	0.951	0.998	-2.099	0.036
heatwave:orderhw	1.339	1.139	1.573	3.547	0.000
Standard errors: MLE

Lyon

summ(m.outcome4[[8]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	380.940
Pseudo-R² (Cragg-Uhler)	0.229
Pseudo-R² (McFadden)	0.043
AIC	8596.699
BIC	8729.059

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	16.699	15.711	17.749	90.470	0.000
heatwave	0.905	0.779	1.052	-1.297	0.195
annee2001	1.041	0.972	1.116	1.144	0.253
annee2002	1.023	0.954	1.097	0.649	0.516
annee2003	1.197	1.117	1.283	5.090	0.000
annee2004	0.947	0.882	1.017	-1.503	0.133
annee2005	0.943	0.878	1.013	-1.616	0.106
annee2006	1.005	0.937	1.078	0.136	0.892
annee2007	0.942	0.878	1.012	-1.634	0.102
annee2008	1.030	0.960	1.104	0.821	0.412
annee2009	0.963	0.897	1.033	-1.049	0.294
annee2010	1.009	0.941	1.082	0.260	0.795
annee2011	0.981	0.914	1.052	-0.541	0.588
annee2012	0.995	0.928	1.067	-0.137	0.891
annee2013	0.938	0.873	1.007	-1.768	0.077
annee2014	0.965	0.899	1.035	-0.995	0.320
annee2015	1.037	0.968	1.112	1.029	0.303
Joursjeudi	1.040	0.993	1.090	1.666	0.096
Jourslundi	1.041	0.994	1.090	1.687	0.092
Joursmardi	1.036	0.989	1.086	1.499	0.134
Joursmercredi	1.041	0.994	1.091	1.697	0.090
Jourssamedi	1.008	0.962	1.056	0.325	0.745
Joursvendredi	1.089	1.040	1.141	3.650	0.000
Vacances1	0.973	0.948	0.999	-2.066	0.039
heatwave:orderhw	1.216	1.155	1.279	7.510	0.000
Standard errors: MLE

Marseille

summ(m.outcome4[[9]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	163.849
Pseudo-R² (Cragg-Uhler)	0.106
Pseudo-R² (McFadden)	0.019
AIC	8708.753
BIC	8841.112

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	20.600	19.494	21.770	107.401	0.000
heatwave	1.258	1.013	1.561	2.081	0.037
annee2001	0.946	0.888	1.008	-1.705	0.088
annee2002	0.987	0.927	1.051	-0.409	0.683
annee2003	1.115	1.048	1.186	3.452	0.001
annee2004	0.906	0.850	0.966	-3.011	0.003
annee2005	0.874	0.819	0.932	-4.090	0.000
annee2006	0.885	0.829	0.944	-3.707	0.000
annee2007	0.913	0.856	0.973	-2.790	0.005
annee2008	0.948	0.889	1.010	-1.661	0.097
annee2009	0.948	0.890	1.010	-1.647	0.099
annee2010	0.929	0.871	0.990	-2.272	0.023
annee2011	0.901	0.845	0.961	-3.178	0.001
annee2012	0.944	0.886	1.006	-1.774	0.076
annee2013	0.950	0.892	1.012	-1.579	0.114
annee2014	0.878	0.823	0.936	-3.954	0.000
annee2015	0.963	0.904	1.026	-1.158	0.247
Joursjeudi	1.059	1.014	1.105	2.601	0.009
Jourslundi	1.051	1.007	1.098	2.271	0.023
Joursmardi	1.069	1.024	1.116	3.036	0.002
Joursmercredi	1.091	1.045	1.139	3.995	0.000
Jourssamedi	1.048	1.004	1.095	2.140	0.032
Joursvendredi	1.071	1.026	1.117	3.112	0.002
Vacances1	0.971	0.948	0.994	-2.412	0.016
heatwave:orderhw	0.965	0.863	1.081	-0.612	0.540
Standard errors: MLE

Montpellier

summ(m.outcome4[[10]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	39.546
Pseudo-R² (Cragg-Uhler)	0.027
Pseudo-R² (McFadden)	0.006
AIC	6658.928
BIC	6791.288

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.434	4.889	6.041	31.362	0.000
heatwave	1.053	0.773	1.435	0.328	0.743
annee2001	1.103	0.979	1.242	1.605	0.108
annee2002	0.977	0.864	1.105	-0.371	0.710
annee2003	1.067	0.945	1.204	1.045	0.296
annee2004	0.962	0.851	1.089	-0.610	0.542
annee2005	1.021	0.904	1.153	0.337	0.736
annee2006	1.036	0.917	1.170	0.564	0.573
annee2007	0.958	0.846	1.083	-0.689	0.491
annee2008	1.014	0.898	1.146	0.229	0.819
annee2009	1.070	0.949	1.207	1.109	0.267
annee2010	1.021	0.904	1.153	0.330	0.742
annee2011	0.986	0.872	1.115	-0.223	0.824
annee2012	1.041	0.923	1.175	0.654	0.513
annee2013	1.003	0.888	1.133	0.046	0.963
annee2014	1.099	0.975	1.238	1.549	0.121
annee2015	1.192	1.060	1.342	2.925	0.003
Joursjeudi	1.033	0.954	1.118	0.793	0.428
Jourslundi	1.007	0.929	1.091	0.166	0.868
Joursmardi	1.016	0.938	1.101	0.400	0.689
Joursmercredi	1.070	0.988	1.158	1.671	0.095
Jourssamedi	0.980	0.904	1.063	-0.477	0.633
Joursvendredi	1.047	0.967	1.133	1.125	0.261
Vacances1	1.012	0.968	1.058	0.525	0.600
heatwave:orderhw	1.037	0.912	1.178	0.551	0.581
Standard errors: MLE

Nancy

summ(m.outcome4[[11]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1470 (2 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	103.994
Pseudo-R² (Cragg-Uhler)	0.069
Pseudo-R² (McFadden)	0.015
AIC	6976.974
BIC	7109.299

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.945	5.370	6.582	34.326	0.000
heatwave	1.805	1.370	2.378	4.197	0.000
annee2001	1.111	0.990	1.248	1.790	0.073
annee2002	1.103	0.982	1.238	1.660	0.097
annee2003	1.226	1.094	1.373	3.507	0.000
annee2004	0.905	0.801	1.022	-1.610	0.107
annee2005	0.983	0.873	1.108	-0.276	0.782
annee2006	1.054	0.937	1.185	0.878	0.380
annee2007	1.005	0.893	1.132	0.084	0.933
annee2008	0.963	0.854	1.085	-0.624	0.533
annee2009	0.999	0.887	1.126	-0.010	0.992
annee2010	1.159	1.034	1.300	2.534	0.011
annee2011	1.101	0.980	1.236	1.621	0.105
annee2012	1.048	0.932	1.178	0.785	0.432
annee2013	1.113	0.992	1.250	1.823	0.068
annee2014	1.053	0.937	1.184	0.862	0.388
annee2015	1.180	1.052	1.323	2.828	0.005
Joursjeudi	0.985	0.913	1.063	-0.391	0.696
Jourslundi	1.031	0.956	1.111	0.793	0.428
Joursmardi	0.975	0.903	1.052	-0.658	0.511
Joursmercredi	0.999	0.926	1.077	-0.038	0.970
Jourssamedi	0.988	0.916	1.066	-0.313	0.755
Joursvendredi	1.017	0.943	1.097	0.439	0.660
Vacances1	0.999	0.957	1.043	-0.048	0.962
heatwave:orderhw	0.801	0.657	0.977	-2.191	0.028
Standard errors: MLE

Nantes

summ(m.outcome4[[12]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	75.338
Pseudo-R² (Cragg-Uhler)	0.050
Pseudo-R² (McFadden)	0.010
AIC	7557.266
BIC	7689.626

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.636	7.940	9.392	50.313	0.000
heatwave	0.700	0.473	1.035	-1.787	0.074
annee2001	1.099	0.999	1.210	1.935	0.053
annee2002	1.083	0.984	1.193	1.632	0.103
annee2003	1.127	1.023	1.241	2.424	0.015
annee2004	1.090	0.990	1.200	1.761	0.078
annee2005	1.053	0.956	1.160	1.046	0.296
annee2006	1.056	0.958	1.163	1.092	0.275
annee2007	1.082	0.982	1.191	1.598	0.110
annee2008	1.030	0.935	1.136	0.602	0.547
annee2009	1.062	0.964	1.170	1.215	0.225
annee2010	1.140	1.036	1.253	2.694	0.007
annee2011	1.086	0.986	1.196	1.681	0.093
annee2012	1.100	1.000	1.211	1.958	0.050
annee2013	1.155	1.051	1.270	2.982	0.003
annee2014	1.140	1.037	1.254	2.707	0.007
annee2015	1.196	1.089	1.314	3.738	0.000
Joursjeudi	1.038	0.976	1.104	1.182	0.237
Jourslundi	1.053	0.990	1.120	1.632	0.103
Joursmardi	1.039	0.977	1.106	1.215	0.225
Joursmercredi	1.018	0.957	1.084	0.574	0.566
Jourssamedi	0.997	0.936	1.061	-0.095	0.924
Joursvendredi	1.009	0.948	1.074	0.285	0.776
Vacances1	0.973	0.939	1.007	-1.570	0.116
heatwave:orderhw	1.540	1.217	1.948	3.601	0.000
Standard errors: MLE

Nice

summ(m.outcome4[[13]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	143.895
Pseudo-R² (Cragg-Uhler)	0.094
Pseudo-R² (McFadden)	0.018
AIC	7849.025
BIC	7981.384

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.333	10.525	12.204	64.279	0.000
heatwave	1.024	0.746	1.405	0.147	0.884
annee2001	1.044	0.961	1.135	1.024	0.306
annee2002	0.993	0.913	1.080	-0.161	0.872
annee2003	1.156	1.065	1.254	3.464	0.001
annee2004	0.897	0.823	0.978	-2.462	0.014
annee2005	0.871	0.798	0.950	-3.108	0.002
annee2006	0.944	0.866	1.028	-1.327	0.184
annee2007	0.875	0.802	0.954	-3.017	0.003
annee2008	0.927	0.851	1.010	-1.741	0.082
annee2009	0.952	0.874	1.037	-1.134	0.257
annee2010	0.882	0.808	0.962	-2.845	0.004
annee2011	0.884	0.810	0.964	-2.788	0.005
annee2012	0.885	0.812	0.966	-2.751	0.006
annee2013	0.891	0.817	0.972	-2.601	0.009
annee2014	0.948	0.870	1.032	-1.230	0.219
annee2015	0.951	0.873	1.035	-1.158	0.247
Joursjeudi	1.001	0.945	1.061	0.036	0.971
Jourslundi	1.059	1.000	1.122	1.968	0.049
Joursmardi	1.084	1.024	1.147	2.770	0.006
Joursmercredi	1.035	0.977	1.096	1.160	0.246
Jourssamedi	1.028	0.970	1.089	0.929	0.353
Joursvendredi	1.056	0.997	1.118	1.859	0.063
Vacances1	1.008	0.976	1.041	0.477	0.634
heatwave:orderhw	1.200	0.937	1.536	1.443	0.149
Standard errors: MLE

Paris

summ(m.outcome4[[14]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	3440.957
Pseudo-R² (Cragg-Uhler)	0.903
Pseudo-R² (McFadden)	0.200
AIC	13804.825
BIC	13937.185

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	101.519	99.021	104.081	363.455	0.000
heatwave	3.844	3.314	4.459	17.775	0.000
annee2001	1.028	0.999	1.057	1.899	0.058
annee2002	0.969	0.942	0.997	-2.181	0.029
annee2003	1.175	1.143	1.207	11.597	0.000
annee2004	0.908	0.882	0.935	-6.486	0.000
annee2005	0.910	0.884	0.937	-6.388	0.000
annee2006	0.892	0.866	0.918	-7.744	0.000
annee2007	0.910	0.884	0.937	-6.341	0.000
annee2008	0.916	0.890	0.943	-5.954	0.000
annee2009	0.912	0.886	0.939	-6.227	0.000
annee2010	0.917	0.891	0.944	-5.887	0.000
annee2011	0.926	0.900	0.953	-5.206	0.000
annee2012	0.896	0.870	0.922	-7.450	0.000
annee2013	0.920	0.894	0.947	-5.652	0.000
annee2014	0.911	0.885	0.938	-6.312	0.000
annee2015	0.903	0.877	0.929	-6.931	0.000
Joursjeudi	1.055	1.034	1.075	5.386	0.000
Jourslundi	1.056	1.036	1.077	5.538	0.000
Joursmardi	1.066	1.046	1.087	6.527	0.000
Joursmercredi	1.060	1.040	1.081	5.949	0.000
Jourssamedi	1.019	1.000	1.039	1.939	0.052
Joursvendredi	1.047	1.027	1.067	4.620	0.000
Vacances1	0.974	0.963	0.984	-4.826	0.000
heatwave:orderhw	0.534	0.464	0.615	-8.753	0.000
Standard errors: MLE

Rennes

summ(m.outcome4[[15]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1441 (31 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	52.993
Pseudo-R² (Cragg-Uhler)	0.037
Pseudo-R² (McFadden)	0.009
AIC	5758.940
BIC	5890.767

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.282	2.861	3.764	16.993	0.000
heatwave	1.385	0.856	2.242	1.327	0.185
annee2001	0.990	0.844	1.161	-0.121	0.904
annee2002	1.143	0.980	1.334	1.700	0.089
annee2003	1.089	0.930	1.274	1.057	0.291
annee2004	0.960	0.816	1.130	-0.490	0.624
annee2005	1.029	0.878	1.206	0.351	0.725
annee2006	1.110	0.949	1.297	1.306	0.191
annee2007	0.994	0.846	1.166	-0.080	0.937
annee2008	1.034	0.882	1.212	0.412	0.680
annee2009	1.059	0.904	1.241	0.708	0.479
annee2010	0.918	0.781	1.080	-1.027	0.305
annee2011	1.190	1.022	1.387	2.233	0.026
annee2012	1.170	1.004	1.363	2.010	0.044
annee2013	1.163	0.997	1.356	1.923	0.055
annee2014	1.157	0.991	1.350	1.850	0.064
annee2015	1.137	0.973	1.329	1.619	0.105
Joursjeudi	1.048	0.947	1.160	0.911	0.362
Jourslundi	1.059	0.957	1.171	1.112	0.266
Joursmardi	1.001	0.904	1.109	0.023	0.981
Joursmercredi	0.965	0.870	1.070	-0.678	0.498
Jourssamedi	1.009	0.911	1.118	0.180	0.857
Joursvendredi	0.963	0.868	1.069	-0.706	0.480
Vacances1	1.005	0.949	1.065	0.170	0.865
heatwave:orderhw	1.009	0.770	1.323	0.066	0.947
Standard errors: MLE

Rouen

summ(m.outcome4[[16]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	104.408
Pseudo-R² (Cragg-Uhler)	0.069
Pseudo-R² (McFadden)	0.014
AIC	7377.018
BIC	7509.377

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	9.038	8.310	9.830	51.359	0.000
heatwave	0.856	0.601	1.220	-0.858	0.391
annee2001	1.025	0.931	1.129	0.501	0.617
annee2002	0.996	0.904	1.098	-0.073	0.942
annee2003	1.057	0.959	1.165	1.116	0.264
annee2004	1.022	0.928	1.126	0.442	0.659
annee2005	0.998	0.905	1.099	-0.048	0.962
annee2006	1.012	0.919	1.115	0.250	0.802
annee2007	1.046	0.950	1.151	0.911	0.362
annee2008	0.972	0.882	1.072	-0.565	0.572
annee2009	0.989	0.898	1.091	-0.213	0.831
annee2010	0.993	0.901	1.094	-0.142	0.887
annee2011	1.044	0.949	1.149	0.882	0.378
annee2012	0.996	0.903	1.097	-0.090	0.928
annee2013	1.006	0.913	1.108	0.113	0.910
annee2014	0.999	0.907	1.101	-0.018	0.985
annee2015	1.097	0.997	1.206	1.907	0.057
Joursjeudi	1.042	0.977	1.110	1.254	0.210
Jourslundi	1.025	0.961	1.093	0.750	0.453
Joursmardi	1.042	0.978	1.110	1.263	0.206
Joursmercredi	1.070	1.004	1.140	2.091	0.037
Jourssamedi	1.042	0.978	1.111	1.271	0.204
Joursvendredi	1.001	0.939	1.068	0.039	0.969
Vacances1	0.926	0.894	0.960	-4.228	0.000
heatwave:orderhw	1.481	1.188	1.846	3.490	0.000
Standard errors: MLE

Strasbourg

summ(m.outcome4[[17]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	117.929
Pseudo-R² (Cragg-Uhler)	0.078
Pseudo-R² (McFadden)	0.016
AIC	7205.249
BIC	7337.609

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.235	7.541	8.993	46.929	0.000
heatwave	1.701	1.357	2.132	4.610	0.000
annee2001	0.981	0.887	1.086	-0.362	0.717
annee2002	0.956	0.863	1.058	-0.870	0.384
annee2003	1.012	0.916	1.119	0.240	0.810
annee2004	0.921	0.831	1.020	-1.577	0.115
annee2005	0.867	0.781	0.962	-2.684	0.007
annee2006	0.911	0.822	1.010	-1.772	0.076
annee2007	0.979	0.885	1.084	-0.406	0.685
annee2008	0.944	0.852	1.045	-1.112	0.266
annee2009	0.959	0.866	1.062	-0.800	0.424
annee2010	0.925	0.835	1.025	-1.483	0.138
annee2011	0.890	0.803	0.987	-2.200	0.028
annee2012	0.899	0.811	0.997	-2.017	0.044
annee2013	1.001	0.905	1.107	0.012	0.990
annee2014	0.966	0.873	1.069	-0.666	0.506
annee2015	0.994	0.898	1.100	-0.117	0.907
Joursjeudi	0.981	0.916	1.051	-0.543	0.587
Jourslundi	1.060	0.991	1.134	1.710	0.087
Joursmardi	1.001	0.935	1.072	0.022	0.983
Joursmercredi	1.064	0.995	1.138	1.816	0.069
Jourssamedi	0.981	0.916	1.051	-0.542	0.588
Joursvendredi	1.006	0.940	1.077	0.175	0.861
Vacances1	0.979	0.943	1.017	-1.071	0.284
heatwave:orderhw	0.922	0.776	1.096	-0.920	0.358
Standard errors: MLE

Toulouse

summ(m.outcome4[[18]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	93.811
Pseudo-R² (Cragg-Uhler)	0.062
Pseudo-R² (McFadden)	0.012
AIC	7792.219
BIC	7924.578

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	10.079	9.325	10.894	58.258	0.000
heatwave	0.968	0.768	1.219	-0.278	0.781
annee2001	1.067	0.976	1.166	1.423	0.155
annee2002	1.094	1.001	1.195	1.986	0.047
annee2003	1.152	1.053	1.260	3.080	0.002
annee2004	1.045	0.955	1.142	0.957	0.339
annee2005	0.997	0.911	1.091	-0.071	0.944
annee2006	1.025	0.937	1.121	0.545	0.586
annee2007	0.990	0.905	1.084	-0.208	0.835
annee2008	1.063	0.973	1.162	1.354	0.176
annee2009	1.039	0.950	1.136	0.839	0.402
annee2010	1.084	0.992	1.184	1.783	0.075
annee2011	1.121	1.027	1.224	2.548	0.011
annee2012	1.050	0.961	1.149	1.079	0.281
annee2013	1.132	1.037	1.235	2.766	0.006
annee2014	1.041	0.952	1.138	0.876	0.381
annee2015	1.119	1.025	1.222	2.505	0.012
Joursjeudi	1.071	1.011	1.135	2.322	0.020
Jourslundi	1.020	0.962	1.081	0.652	0.514
Joursmardi	1.032	0.974	1.094	1.073	0.283
Joursmercredi	1.044	0.985	1.107	1.467	0.142
Jourssamedi	1.016	0.958	1.077	0.516	0.606
Joursvendredi	1.080	1.019	1.144	2.602	0.009
Vacances1	0.970	0.939	1.002	-1.865	0.062
heatwave:orderhw	1.169	1.053	1.299	2.912	0.004
Standard errors: MLE

Meta-analysis:

meta4<-rma(v.coeff4,sei=v.sterr4)

summary(meta4)

## 
## Random-Effects Model (k = 18; tau^2 estimator: REML)
## 
##   logLik  deviance       AIC       BIC      AICc 
##  -1.5364    3.0729    7.0729    8.7393    7.9300   
## 
## tau^2 (estimated amount of total heterogeneity): 0.0586 (SE = 0.0233)
## tau (square root of estimated tau^2 value):      0.2422
## I^2 (total heterogeneity / total variability):   92.82%
## H^2 (total variability / sampling variability):  13.94
## 
## Test for Heterogeneity:
## Q(df = 17) = 172.6965, p-val < .0001
## 
## Model Results:
## 
## estimate      se    zval    pval    ci.lb   ci.ub 
##   0.0905  0.0618  1.4644  0.1431  -0.0306  0.2116    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

forest(meta4,transf = exp, slab = names(villes_s),refline=1,xlab="RR Order Heat Wave", header="City")

There is not a consistent pattern and the pooled estimate is not significanlty different from 1.

Second version with order variable as factor without interaction

for (i in 1:length(villes_s)){
  villes_s[[i]]$orderhw<-as.factor(as.character(villes_s[[i]]$orderhw))
}

lapply(villes_s,function(x){levels(x$orderhw)})

## $BM
## [1] "0" "1" "2"
## 
## $bordeaux
## [1] "0" "1" "2" "3"
## 
## $clermont
## [1] "0" "1" "2"
## 
## $dijon
## [1] "0" "1" "2" "3"
## 
## $grenoble
## [1] "0" "1" "2" "3" "4"
## 
## $lehavre
## [1] "0" "1" "2" "3"
## 
## $lille
## [1] "0" "1" "2"
## 
## $lyon
## [1] "0" "1" "2" "3" "4"
## 
## $marseille
## [1] "0" "1" "2" "3"
## 
## $montpellier
## [1] "0" "1" "2" "3" "4"
## 
## $nancy
## [1] "0" "1" "2" "3" "4"
## 
## $nantes
## [1] "0" "1" "2"
## 
## $nice
## [1] "0" "1" "2"
## 
## $paris
## [1] "0" "1" "2"
## 
## $rennes
## [1] "0" "1" "2" "3"
## 
## $rouen
## [1] "0" "1" "2"
## 
## $strasbourg
## [1] "0" "1" "2" "3"
## 
## $toulouse
## [1] "0" "1" "2" "3"

The category “0” will be taken as the reference category in the model.

I just put in the model the “order heat wave” variable, which allow to consider mortality risk for non heat-wave days compared to mortality risk during heat-wave days, and allow also the risk of mortality to differ by heat wave timing (inspired by Anderson and Bell, 2011).

m.outcome4bis<-list()
v.coeffHW1<-c()
v.sterrHW1<-c()
v.coeffHW2<-c()
v.sterrHW2<-c()
resultsHW1<-matrix(NA,nrow = 18,ncol=4)
colnames(resultsHW1)<-c("city","HW1","HW1_2.5","HW1_97.5")
resultsHW1[,1]<-names(villes_s)

resultsHW2<-matrix(NA,nrow = 18,ncol=4)
colnames(resultsHW2)<-c("city","HW2","HW2_2.5","HW2_97.5")
resultsHW2[,1]<-names(villes_s)


for (i in 1:length(villes_s)){
m.outcome4bis[[i]]<-glm(nocc_tot~orderhw+annee+Jours+Vacances,data=villes_s[[i]],family=poisson)
v.coeffHW1[i]<-m.outcome4bis[[i]]$coefficients[2]
v.sterrHW1[i]<-summary(m.outcome4bis[[i]])$coefficients[2, 2]
v.coeffHW2[i]<-m.outcome4bis[[i]]$coefficients[3]
v.sterrHW2[i]<-summary(m.outcome4bis[[i]])$coefficients[3, 2]
resultsHW1[i,2]<-m.outcome4bis[[i]]$coefficients[2]
resultsHW1[i,3:4]<-confint(m.outcome4bis[[i]])[2,]
resultsHW2[i,2]<-m.outcome4bis[[i]]$coefficients[3]
resultsHW2[i,3:4]<-confint(m.outcome4bis[[i]])[3,]
}

names(m.outcome4bis)<-names(villes_s)

Results per city

summ(m.outcome4bis[[1]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	40.934
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	7276.193
BIC	7408.536

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	7.430	6.784	8.136	43.248	0.000
orderhw1	1.010	0.865	1.179	0.123	0.902
orderhw2	1.387	1.128	1.705	3.103	0.002
annee2001	1.017	0.916	1.129	0.311	0.755
annee2002	1.044	0.941	1.158	0.811	0.418
annee2003	1.104	0.994	1.227	1.850	0.064
annee2004	0.980	0.882	1.089	-0.379	0.705
annee2005	0.977	0.879	1.085	-0.441	0.659
annee2006	1.045	0.941	1.161	0.823	0.411
annee2007	1.030	0.928	1.143	0.560	0.576
annee2008	1.034	0.932	1.147	0.627	0.530
annee2009	1.009	0.909	1.121	0.170	0.865
annee2010	1.042	0.939	1.156	0.782	0.434
annee2011	1.017	0.916	1.129	0.309	0.757
annee2012	1.108	1.001	1.228	1.973	0.048
annee2013	0.995	0.896	1.105	-0.089	0.929
annee2014	1.064	0.959	1.179	1.169	0.243
annee2015	1.054	0.951	1.169	1.001	0.317
Joursjeudi	1.029	0.961	1.103	0.819	0.413
Jourslundi	1.071	1.000	1.146	1.956	0.051
Joursmardi	1.042	0.973	1.116	1.171	0.242
Joursmercredi	1.058	0.988	1.133	1.605	0.109
Jourssamedi	0.987	0.921	1.058	-0.366	0.714
Joursvendredi	1.077	1.006	1.153	2.138	0.032
Vacances1	0.976	0.939	1.014	-1.234	0.217
Standard errors: MLE

Bordeaux

summ(m.outcome4bis[[2]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	147.537
Pseudo-R² (Cragg-Uhler)	0.096
Pseudo-R² (McFadden)	0.018
AIC	7922.358
BIC	8060.012

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.239	10.436	12.104	63.963	0.000
orderhw1	1.392	1.220	1.588	4.918	0.000
orderhw2	0.938	0.517	1.701	-0.212	0.832
orderhw3	1.945	1.686	2.243	9.124	0.000
annee2001	1.040	0.956	1.132	0.911	0.362
annee2002	1.016	0.934	1.106	0.374	0.708
annee2003	1.054	0.966	1.149	1.181	0.237
annee2004	0.945	0.867	1.031	-1.271	0.204
annee2005	0.931	0.853	1.015	-1.621	0.105
annee2006	0.993	0.911	1.082	-0.163	0.870
annee2007	0.941	0.863	1.026	-1.370	0.171
annee2008	0.986	0.905	1.074	-0.323	0.746
annee2009	0.956	0.877	1.042	-1.022	0.307
annee2010	1.030	0.946	1.120	0.674	0.501
annee2011	0.990	0.909	1.078	-0.229	0.819
annee2012	0.971	0.891	1.058	-0.677	0.498
annee2013	1.012	0.930	1.102	0.274	0.784
annee2014	1.033	0.949	1.124	0.756	0.450
annee2015	1.069	0.983	1.162	1.553	0.121
Joursjeudi	1.045	0.988	1.106	1.533	0.125
Jourslundi	1.045	0.988	1.105	1.524	0.128
Joursmardi	1.017	0.961	1.076	0.570	0.569
Joursmercredi	1.034	0.977	1.094	1.167	0.243
Jourssamedi	1.039	0.982	1.100	1.331	0.183
Joursvendredi	1.027	0.970	1.086	0.908	0.364
Vacances1	0.991	0.960	1.022	-0.589	0.556
Standard errors: MLE

Clermont-Ferrand

summ(m.outcome4bis[[3]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1466 (6 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	62.748
Pseudo-R² (Cragg-Uhler)	0.042
Pseudo-R² (McFadden)	0.010
AIC	6534.025
BIC	6666.282

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	4.961	4.435	5.550	27.991	0.000
orderhw1	1.536	1.315	1.793	5.424	0.000
orderhw2	1.105	0.623	1.963	0.342	0.732
annee2001	0.903	0.791	1.031	-1.508	0.131
annee2002	0.997	0.877	1.135	-0.039	0.969
annee2003	1.071	0.943	1.216	1.051	0.293
annee2004	0.956	0.840	1.090	-0.669	0.503
annee2005	0.956	0.839	1.090	-0.670	0.503
annee2006	0.912	0.799	1.040	-1.372	0.170
annee2007	0.980	0.861	1.115	-0.311	0.756
annee2008	1.066	0.939	1.210	0.982	0.326
annee2009	1.063	0.936	1.208	0.943	0.346
annee2010	1.089	0.960	1.236	1.322	0.186
annee2011	1.010	0.888	1.148	0.145	0.885
annee2012	1.057	0.931	1.200	0.856	0.392
annee2013	1.089	0.959	1.235	1.318	0.187
annee2014	1.059	0.933	1.203	0.885	0.376
annee2015	1.085	0.956	1.232	1.258	0.209
Joursjeudi	1.028	0.944	1.118	0.633	0.526
Jourslundi	1.026	0.942	1.116	0.586	0.558
Joursmardi	1.050	0.965	1.142	1.140	0.254
Joursmercredi	1.043	0.958	1.135	0.975	0.330
Jourssamedi	1.044	0.960	1.136	1.006	0.314
Joursvendredi	1.040	0.956	1.131	0.907	0.365
Vacances1	0.987	0.941	1.034	-0.561	0.575
Standard errors: MLE

Dijon

summ(m.outcome4bis[[4]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1449 (23 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	77.112
Pseudo-R² (Cragg-Uhler)	0.053
Pseudo-R² (McFadden)	0.013
AIC	6099.304
BIC	6236.548

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.645	3.207	4.144	19.774	0.000
orderhw1	1.114	0.879	1.413	0.892	0.372
orderhw2	1.091	0.732	1.626	0.427	0.670
orderhw3	2.061	1.654	2.568	6.437	0.000
annee2001	1.053	0.908	1.222	0.682	0.495
annee2002	1.114	0.961	1.291	1.430	0.153
annee2003	1.203	1.037	1.395	2.436	0.015
annee2004	1.035	0.891	1.202	0.444	0.657
annee2005	1.051	0.907	1.219	0.663	0.507
annee2006	1.104	0.953	1.280	1.317	0.188
annee2007	1.093	0.944	1.266	1.195	0.232
annee2008	1.121	0.968	1.297	1.529	0.126
annee2009	1.129	0.976	1.307	1.631	0.103
annee2010	1.111	0.960	1.287	1.409	0.159
annee2011	1.139	0.985	1.317	1.763	0.078
annee2012	1.207	1.046	1.394	2.569	0.010
annee2013	1.248	1.082	1.439	3.049	0.002
annee2014	1.195	1.035	1.379	2.430	0.015
annee2015	1.228	1.062	1.419	2.780	0.005
Joursjeudi	1.019	0.928	1.120	0.395	0.693
Jourslundi	1.038	0.945	1.140	0.776	0.438
Joursmardi	1.056	0.962	1.160	1.148	0.251
Joursmercredi	1.052	0.958	1.155	1.057	0.290
Jourssamedi	1.034	0.941	1.136	0.702	0.482
Joursvendredi	1.035	0.942	1.137	0.716	0.474
Vacances1	1.000	0.949	1.054	0.004	0.997
Standard errors: MLE

Grenoble

summ(m.outcome4bis[[5]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1471 (1 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	97.702
Pseudo-R² (Cragg-Uhler)	0.065
Pseudo-R² (McFadden)	0.014
AIC	7114.088
BIC	7257.018

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	6.858	6.233	7.545	39.529	0.000
orderhw1	1.146	0.969	1.355	1.595	0.111
orderhw2	1.087	0.842	1.404	0.641	0.521
orderhw3	1.087	0.736	1.604	0.419	0.675
orderhw4	1.502	1.233	1.830	4.034	0.000
annee2001	1.065	0.954	1.187	1.123	0.262
annee2002	1.005	0.899	1.123	0.088	0.930
annee2003	1.171	1.049	1.307	2.812	0.005
annee2004	0.979	0.876	1.095	-0.366	0.714
annee2005	1.003	0.898	1.120	0.047	0.963
annee2006	0.979	0.876	1.095	-0.367	0.714
annee2007	0.922	0.823	1.032	-1.415	0.157
annee2008	1.027	0.920	1.147	0.478	0.633
annee2009	1.017	0.910	1.135	0.291	0.771
annee2010	1.041	0.933	1.162	0.718	0.473
annee2011	1.071	0.961	1.195	1.240	0.215
annee2012	1.096	0.984	1.222	1.665	0.096
annee2013	1.167	1.049	1.299	2.840	0.005
annee2014	1.114	1.000	1.241	1.967	0.049
annee2015	1.218	1.094	1.356	3.609	0.000
Joursjeudi	1.018	0.948	1.094	0.490	0.624
Jourslundi	1.070	0.996	1.149	1.862	0.063
Joursmardi	1.016	0.945	1.091	0.421	0.674
Joursmercredi	1.013	0.943	1.089	0.357	0.721
Jourssamedi	1.039	0.967	1.116	1.037	0.300
Joursvendredi	1.025	0.954	1.101	0.666	0.506
Vacances1	0.949	0.912	0.987	-2.600	0.009
Standard errors: MLE

La Havre

summ(m.outcome4bis[[6]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1467 (5 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	42.917
Pseudo-R² (Cragg-Uhler)	0.029
Pseudo-R² (McFadden)	0.007
AIC	6605.341
BIC	6742.906

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.163	4.623	5.767	29.098	0.000
orderhw1	1.177	0.945	1.466	1.458	0.145
orderhw2	1.308	0.984	1.738	1.852	0.064
orderhw3	1.810	1.226	2.672	2.984	0.003
annee2001	0.993	0.873	1.129	-0.110	0.912
annee2002	1.033	0.910	1.174	0.508	0.611
annee2003	0.986	0.866	1.122	-0.218	0.827
annee2004	1.092	0.963	1.237	1.370	0.171
annee2005	0.939	0.824	1.069	-0.951	0.342
annee2006	1.162	1.026	1.315	2.362	0.018
annee2007	1.027	0.904	1.166	0.406	0.685
annee2008	1.009	0.888	1.147	0.143	0.887
annee2009	1.025	0.902	1.164	0.382	0.702
annee2010	1.046	0.922	1.188	0.700	0.484
annee2011	1.035	0.912	1.176	0.538	0.590
annee2012	1.005	0.885	1.143	0.082	0.935
annee2013	1.067	0.941	1.211	1.013	0.311
annee2014	1.017	0.895	1.156	0.260	0.795
annee2015	1.122	0.991	1.271	1.813	0.070
Joursjeudi	1.026	0.943	1.115	0.592	0.554
Jourslundi	1.029	0.947	1.118	0.672	0.501
Joursmardi	1.045	0.961	1.135	1.032	0.302
Joursmercredi	1.050	0.967	1.141	1.161	0.246
Jourssamedi	0.990	0.909	1.077	-0.244	0.807
Joursvendredi	0.975	0.896	1.061	-0.582	0.561
Vacances1	0.954	0.910	0.999	-1.998	0.046
Standard errors: MLE

Lille

summ(m.outcome4bis[[7]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	73.904
Pseudo-R² (Cragg-Uhler)	0.049
Pseudo-R² (McFadden)	0.009
AIC	8553.719
BIC	8686.078

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	19.287	18.222	20.414	102.101	0.000
orderhw1	1.105	1.005	1.215	2.072	0.038
orderhw2	1.479	1.297	1.688	5.826	0.000
annee2001	0.978	0.916	1.044	-0.671	0.502
annee2002	0.971	0.910	1.037	-0.873	0.383
annee2003	0.976	0.913	1.044	-0.712	0.477
annee2004	0.962	0.901	1.027	-1.158	0.247
annee2005	0.982	0.920	1.048	-0.539	0.590
annee2006	0.974	0.912	1.040	-0.785	0.432
annee2007	0.929	0.870	0.992	-2.193	0.028
annee2008	0.932	0.873	0.996	-2.089	0.037
annee2009	0.938	0.878	1.001	-1.923	0.055
annee2010	0.978	0.916	1.043	-0.679	0.497
annee2011	0.948	0.888	1.012	-1.597	0.110
annee2012	0.941	0.881	1.005	-1.824	0.068
annee2013	0.976	0.914	1.042	-0.732	0.464
annee2014	0.978	0.916	1.043	-0.680	0.497
annee2015	0.976	0.914	1.041	-0.745	0.456
Joursjeudi	1.060	1.015	1.108	2.604	0.009
Jourslundi	1.068	1.022	1.116	2.927	0.003
Joursmardi	1.084	1.038	1.133	3.630	0.000
Joursmercredi	1.061	1.015	1.109	2.640	0.008
Jourssamedi	1.031	0.986	1.077	1.335	0.182
Joursvendredi	1.061	1.015	1.109	2.641	0.008
Vacances1	0.974	0.951	0.998	-2.099	0.036
Standard errors: MLE

Lyon

summ(m.outcome4bis[[8]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	406.684
Pseudo-R² (Cragg-Uhler)	0.242
Pseudo-R² (McFadden)	0.046
AIC	8574.956
BIC	8717.904

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	16.693	15.706	17.743	90.455	0.000
orderhw1	1.209	1.082	1.352	3.338	0.001
orderhw2	0.906	0.714	1.149	-0.818	0.413
orderhw3	1.030	0.753	1.408	0.183	0.855
orderhw4	2.092	1.886	2.321	13.958	0.000
annee2001	1.041	0.972	1.116	1.144	0.253
annee2002	1.023	0.954	1.097	0.650	0.515
annee2003	1.199	1.118	1.285	5.112	0.000
annee2004	0.947	0.882	1.017	-1.505	0.132
annee2005	0.942	0.877	1.011	-1.652	0.099
annee2006	0.997	0.929	1.070	-0.090	0.928
annee2007	0.942	0.878	1.012	-1.637	0.102
annee2008	1.029	0.960	1.104	0.818	0.413
annee2009	0.962	0.896	1.032	-1.081	0.280
annee2010	1.009	0.941	1.082	0.257	0.797
annee2011	0.980	0.913	1.051	-0.576	0.564
annee2012	0.995	0.928	1.067	-0.140	0.888
annee2013	0.937	0.872	1.006	-1.801	0.072
annee2014	0.965	0.899	1.035	-0.998	0.318
annee2015	1.049	0.978	1.125	1.345	0.179
Joursjeudi	1.040	0.992	1.089	1.640	0.101
Jourslundi	1.040	0.993	1.090	1.657	0.097
Joursmardi	1.043	0.995	1.092	1.760	0.078
Joursmercredi	1.046	0.998	1.095	1.881	0.060
Jourssamedi	1.007	0.961	1.055	0.277	0.782
Joursvendredi	1.093	1.044	1.144	3.791	0.000
Vacances1	0.971	0.946	0.997	-2.216	0.027
Standard errors: MLE

Marseille

summ(m.outcome4bis[[9]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	168.681
Pseudo-R² (Cragg-Uhler)	0.109
Pseudo-R² (McFadden)	0.019
AIC	8705.921
BIC	8843.575

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	20.595	19.489	21.764	107.392	0.000
orderhw1	1.144	1.004	1.304	2.025	0.043
orderhw2	1.242	1.136	1.359	4.752	0.000
orderhw3	0.953	0.760	1.194	-0.418	0.676
annee2001	0.948	0.889	1.010	-1.657	0.097
annee2002	0.987	0.927	1.051	-0.409	0.683
annee2003	1.106	1.039	1.177	3.169	0.002
annee2004	0.906	0.850	0.966	-3.011	0.003
annee2005	0.874	0.819	0.932	-4.090	0.000
annee2006	0.880	0.825	0.939	-3.850	0.000
annee2007	0.913	0.856	0.973	-2.790	0.005
annee2008	0.948	0.889	1.010	-1.661	0.097
annee2009	0.949	0.891	1.011	-1.624	0.104
annee2010	0.929	0.871	0.990	-2.272	0.023
annee2011	0.901	0.845	0.961	-3.179	0.001
annee2012	0.946	0.888	1.008	-1.703	0.089
annee2013	0.951	0.893	1.013	-1.557	0.119
annee2014	0.878	0.823	0.936	-3.954	0.000
annee2015	0.971	0.911	1.035	-0.917	0.359
Joursjeudi	1.059	1.015	1.106	2.621	0.009
Jourslundi	1.051	1.006	1.097	2.251	0.024
Joursmardi	1.069	1.024	1.116	3.029	0.002
Joursmercredi	1.092	1.047	1.140	4.045	0.000
Jourssamedi	1.048	1.004	1.095	2.144	0.032
Joursvendredi	1.071	1.026	1.118	3.129	0.002
Vacances1	0.971	0.948	0.995	-2.408	0.016
Standard errors: MLE

Montpellier

summ(m.outcome4bis[[10]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	41.027
Pseudo-R² (Cragg-Uhler)	0.028
Pseudo-R² (McFadden)	0.006
AIC	6661.447
BIC	6804.395

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.436	4.890	6.043	31.366	0.000
orderhw1	1.070	0.845	1.356	0.562	0.574
orderhw2	1.118	0.856	1.460	0.819	0.413
orderhw3	1.418	1.000	2.011	1.962	0.050
orderhw4	1.098	0.774	1.556	0.523	0.601
annee2001	1.103	0.979	1.242	1.605	0.108
annee2002	0.977	0.864	1.105	-0.371	0.711
annee2003	1.070	0.948	1.209	1.099	0.272
annee2004	0.963	0.851	1.089	-0.606	0.544
annee2005	1.021	0.904	1.153	0.337	0.736
annee2006	1.029	0.911	1.163	0.465	0.642
annee2007	0.957	0.846	1.083	-0.689	0.491
annee2008	1.014	0.898	1.146	0.229	0.819
annee2009	1.070	0.949	1.207	1.109	0.267
annee2010	1.021	0.904	1.153	0.334	0.739
annee2011	0.986	0.872	1.115	-0.223	0.824
annee2012	1.041	0.923	1.175	0.653	0.514
annee2013	1.003	0.888	1.133	0.046	0.963
annee2014	1.099	0.975	1.239	1.552	0.121
annee2015	1.194	1.061	1.344	2.939	0.003
Joursjeudi	1.034	0.954	1.119	0.813	0.416
Jourslundi	1.006	0.928	1.090	0.143	0.886
Joursmardi	1.016	0.938	1.101	0.393	0.694
Joursmercredi	1.069	0.988	1.157	1.661	0.097
Jourssamedi	0.980	0.904	1.063	-0.480	0.631
Joursvendredi	1.048	0.968	1.134	1.148	0.251
Vacances1	1.012	0.967	1.058	0.503	0.615
Standard errors: MLE

Nancy

summ(m.outcome4bis[[11]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1470 (2 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(26)	106.457
Pseudo-R² (Cragg-Uhler)	0.070
Pseudo-R² (McFadden)	0.015
AIC	6978.511
BIC	7121.423

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	5.946	5.371	6.583	34.329	0.000
orderhw1	1.477	1.294	1.686	5.770	0.000
orderhw2	0.972	0.703	1.344	-0.172	0.864
orderhw3	0.866	0.387	1.939	-0.349	0.727
orderhw4	1.111	0.552	2.236	0.296	0.767
annee2001	1.111	0.990	1.248	1.790	0.073
annee2002	1.102	0.982	1.237	1.649	0.099
annee2003	1.221	1.090	1.369	3.442	0.001
annee2004	0.905	0.801	1.022	-1.610	0.107
annee2005	0.983	0.873	1.108	-0.276	0.783
annee2006	1.062	0.944	1.194	0.998	0.318
annee2007	1.005	0.893	1.132	0.085	0.933
annee2008	0.963	0.854	1.085	-0.623	0.533
annee2009	0.999	0.887	1.126	-0.010	0.992
annee2010	1.158	1.033	1.299	2.523	0.012
annee2011	1.101	0.980	1.236	1.621	0.105
annee2012	1.047	0.932	1.177	0.774	0.439
annee2013	1.113	0.992	1.250	1.823	0.068
annee2014	1.053	0.937	1.184	0.863	0.388
annee2015	1.179	1.051	1.322	2.807	0.005
Joursjeudi	0.985	0.913	1.063	-0.394	0.693
Jourslundi	1.028	0.953	1.108	0.712	0.476
Joursmardi	0.974	0.902	1.051	-0.686	0.492
Joursmercredi	0.998	0.925	1.077	-0.040	0.968
Jourssamedi	0.989	0.916	1.067	-0.286	0.775
Joursvendredi	1.017	0.943	1.097	0.436	0.663
Vacances1	0.999	0.958	1.043	-0.025	0.980
Standard errors: MLE

Nantes

summ(m.outcome4bis[[12]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	75.338
Pseudo-R² (Cragg-Uhler)	0.050
Pseudo-R² (McFadden)	0.010
AIC	7557.266
BIC	7689.626

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.636	7.940	9.392	50.313	0.000
orderhw1	1.078	0.899	1.292	0.807	0.420
orderhw2	1.659	1.425	1.933	6.510	0.000
annee2001	1.099	0.999	1.210	1.935	0.053
annee2002	1.083	0.984	1.193	1.632	0.103
annee2003	1.127	1.023	1.241	2.424	0.015
annee2004	1.090	0.990	1.200	1.761	0.078
annee2005	1.053	0.956	1.160	1.046	0.296
annee2006	1.056	0.958	1.163	1.092	0.275
annee2007	1.082	0.982	1.191	1.598	0.110
annee2008	1.030	0.935	1.136	0.602	0.547
annee2009	1.062	0.964	1.170	1.215	0.225
annee2010	1.140	1.036	1.253	2.694	0.007
annee2011	1.086	0.986	1.196	1.681	0.093
annee2012	1.100	1.000	1.211	1.958	0.050
annee2013	1.155	1.051	1.270	2.982	0.003
annee2014	1.140	1.037	1.254	2.707	0.007
annee2015	1.196	1.089	1.314	3.738	0.000
Joursjeudi	1.038	0.976	1.104	1.182	0.237
Jourslundi	1.053	0.990	1.120	1.632	0.103
Joursmardi	1.039	0.977	1.106	1.215	0.225
Joursmercredi	1.018	0.957	1.084	0.574	0.566
Jourssamedi	0.997	0.936	1.061	-0.095	0.924
Joursvendredi	1.009	0.948	1.074	0.285	0.776
Vacances1	0.973	0.939	1.007	-1.570	0.116
Standard errors: MLE

Nice

summ(m.outcome4bis[[13]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	143.895
Pseudo-R² (Cragg-Uhler)	0.094
Pseudo-R² (McFadden)	0.018
AIC	7849.025
BIC	7981.384

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	11.333	10.525	12.204	64.279	0.000
orderhw1	1.228	1.094	1.379	3.475	0.001
orderhw2	1.473	1.181	1.839	3.428	0.001
annee2001	1.044	0.961	1.135	1.024	0.306
annee2002	0.993	0.913	1.080	-0.161	0.872
annee2003	1.156	1.065	1.254	3.464	0.001
annee2004	0.897	0.823	0.978	-2.462	0.014
annee2005	0.871	0.798	0.950	-3.108	0.002
annee2006	0.944	0.866	1.028	-1.327	0.184
annee2007	0.875	0.802	0.954	-3.017	0.003
annee2008	0.927	0.851	1.010	-1.741	0.082
annee2009	0.952	0.874	1.037	-1.134	0.257
annee2010	0.882	0.808	0.962	-2.845	0.004
annee2011	0.884	0.810	0.964	-2.788	0.005
annee2012	0.885	0.812	0.966	-2.751	0.006
annee2013	0.891	0.817	0.972	-2.601	0.009
annee2014	0.948	0.870	1.032	-1.230	0.219
annee2015	0.951	0.873	1.035	-1.158	0.247
Joursjeudi	1.001	0.945	1.061	0.036	0.971
Jourslundi	1.059	1.000	1.122	1.968	0.049
Joursmardi	1.084	1.024	1.147	2.770	0.006
Joursmercredi	1.035	0.977	1.096	1.160	0.246
Jourssamedi	1.028	0.970	1.089	0.929	0.353
Joursvendredi	1.056	0.997	1.118	1.859	0.063
Vacances1	1.008	0.976	1.041	0.477	0.634
Standard errors: MLE

Paris

summ(m.outcome4bis[[14]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	3440.957
Pseudo-R² (Cragg-Uhler)	0.903
Pseudo-R² (McFadden)	0.200
AIC	13804.825
BIC	13937.185

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	101.519	99.021	104.081	363.455	0.000
orderhw1	2.053	1.995	2.113	49.004	0.000
orderhw2	1.097	0.955	1.259	1.312	0.189
annee2001	1.028	0.999	1.057	1.899	0.058
annee2002	0.969	0.942	0.997	-2.181	0.029
annee2003	1.175	1.143	1.207	11.597	0.000
annee2004	0.908	0.882	0.935	-6.486	0.000
annee2005	0.910	0.884	0.937	-6.388	0.000
annee2006	0.892	0.866	0.918	-7.744	0.000
annee2007	0.910	0.884	0.937	-6.341	0.000
annee2008	0.916	0.890	0.943	-5.954	0.000
annee2009	0.912	0.886	0.939	-6.227	0.000
annee2010	0.917	0.891	0.944	-5.887	0.000
annee2011	0.926	0.900	0.953	-5.206	0.000
annee2012	0.896	0.870	0.922	-7.450	0.000
annee2013	0.920	0.894	0.947	-5.652	0.000
annee2014	0.911	0.885	0.938	-6.312	0.000
annee2015	0.903	0.877	0.929	-6.931	0.000
Joursjeudi	1.055	1.034	1.075	5.386	0.000
Jourslundi	1.056	1.036	1.077	5.538	0.000
Joursmardi	1.066	1.046	1.087	6.527	0.000
Joursmercredi	1.060	1.040	1.081	5.949	0.000
Jourssamedi	1.019	1.000	1.039	1.939	0.052
Joursvendredi	1.047	1.027	1.067	4.620	0.000
Vacances1	0.974	0.963	0.984	-4.826	0.000
Standard errors: MLE

Rennes

summ(m.outcome4bis[[15]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1441 (31 missing obs. deleted)
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	56.665
Pseudo-R² (Cragg-Uhler)	0.039
Pseudo-R² (McFadden)	0.010
AIC	5757.268
BIC	5894.369

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	3.288	2.867	3.771	17.023	0.000
orderhw1	1.292	0.986	1.692	1.857	0.063
orderhw2	1.573	1.266	1.956	4.086	0.000
orderhw3	0.805	0.359	1.808	-0.525	0.600
annee2001	0.988	0.843	1.159	-0.144	0.886
annee2002	1.142	0.979	1.333	1.686	0.092
annee2003	1.069	0.911	1.253	0.817	0.414
annee2004	0.959	0.815	1.129	-0.503	0.615
annee2005	1.029	0.878	1.206	0.352	0.725
annee2006	1.120	0.958	1.309	1.425	0.154
annee2007	0.992	0.845	1.165	-0.093	0.926
annee2008	1.033	0.881	1.210	0.398	0.690
annee2009	1.058	0.903	1.239	0.694	0.488
annee2010	0.917	0.780	1.079	-1.040	0.298
annee2011	1.189	1.020	1.386	2.219	0.026
annee2012	1.170	1.004	1.363	2.010	0.044
annee2013	1.164	0.998	1.358	1.939	0.052
annee2014	1.156	0.990	1.348	1.835	0.066
annee2015	1.136	0.972	1.328	1.605	0.109
Joursjeudi	1.049	0.948	1.161	0.919	0.358
Jourslundi	1.058	0.957	1.170	1.095	0.273
Joursmardi	1.001	0.904	1.108	0.013	0.990
Joursmercredi	0.966	0.871	1.071	-0.659	0.510
Jourssamedi	1.008	0.910	1.116	0.148	0.883
Joursvendredi	0.963	0.868	1.068	-0.714	0.475
Vacances1	1.004	0.948	1.064	0.142	0.887
Standard errors: MLE

Rouen

summ(m.outcome4bis[[16]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(24)	104.408
Pseudo-R² (Cragg-Uhler)	0.069
Pseudo-R² (McFadden)	0.014
AIC	7377.018
BIC	7509.377

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	9.038	8.310	9.830	51.359	0.000
orderhw1	1.268	1.079	1.491	2.885	0.004
orderhw2	1.878	1.609	2.193	7.969	0.000
annee2001	1.025	0.931	1.129	0.501	0.617
annee2002	0.996	0.904	1.098	-0.073	0.942
annee2003	1.057	0.959	1.165	1.116	0.264
annee2004	1.022	0.928	1.126	0.442	0.659
annee2005	0.998	0.905	1.099	-0.048	0.962
annee2006	1.012	0.919	1.115	0.250	0.802
annee2007	1.046	0.950	1.151	0.911	0.362
annee2008	0.972	0.882	1.072	-0.565	0.572
annee2009	0.989	0.898	1.091	-0.213	0.831
annee2010	0.993	0.901	1.094	-0.142	0.887
annee2011	1.044	0.949	1.149	0.882	0.378
annee2012	0.996	0.903	1.097	-0.090	0.928
annee2013	1.006	0.913	1.108	0.113	0.910
annee2014	0.999	0.907	1.101	-0.018	0.985
annee2015	1.097	0.997	1.206	1.907	0.057
Joursjeudi	1.042	0.977	1.110	1.254	0.210
Jourslundi	1.025	0.961	1.093	0.750	0.453
Joursmardi	1.042	0.978	1.110	1.263	0.206
Joursmercredi	1.070	1.004	1.140	2.091	0.037
Jourssamedi	1.042	0.978	1.111	1.271	0.204
Joursvendredi	1.001	0.939	1.068	0.039	0.969
Vacances1	0.926	0.894	0.960	-4.228	0.000
Standard errors: MLE

Strasbourg

summ(m.outcome4bis[[17]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	120.710
Pseudo-R² (Cragg-Uhler)	0.079
Pseudo-R² (McFadden)	0.017
AIC	7204.469
BIC	7342.122

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	8.243	7.548	9.001	46.952	0.000
orderhw1	1.582	1.429	1.752	8.823	0.000
orderhw2	1.002	0.610	1.646	0.008	0.994
orderhw3	1.470	1.051	2.056	2.248	0.025
annee2001	0.981	0.887	1.086	-0.362	0.717
annee2002	0.956	0.863	1.058	-0.874	0.382
annee2003	1.011	0.914	1.118	0.210	0.834
annee2004	0.921	0.831	1.020	-1.577	0.115
annee2005	0.867	0.781	0.962	-2.687	0.007
annee2006	0.910	0.822	1.009	-1.790	0.073
annee2007	0.979	0.885	1.084	-0.406	0.685
annee2008	0.944	0.852	1.045	-1.112	0.266
annee2009	0.959	0.866	1.062	-0.800	0.424
annee2010	0.925	0.835	1.025	-1.486	0.137
annee2011	0.890	0.802	0.987	-2.202	0.028
annee2012	0.899	0.811	0.997	-2.022	0.043
annee2013	1.000	0.905	1.106	0.009	0.992
annee2014	0.966	0.873	1.069	-0.674	0.500
annee2015	0.998	0.901	1.105	-0.039	0.969
Joursjeudi	0.980	0.915	1.049	-0.584	0.559
Jourslundi	1.058	0.989	1.132	1.644	0.100
Joursmardi	0.999	0.933	1.070	-0.025	0.980
Joursmercredi	1.063	0.993	1.137	1.768	0.077
Jourssamedi	0.981	0.916	1.051	-0.540	0.589
Joursvendredi	1.005	0.938	1.076	0.132	0.895
Vacances1	0.980	0.943	1.018	-1.051	0.293
Standard errors: MLE

Toulouse

summ(m.outcome4bis[[18]], exp = TRUE, confint = TRUE, digits = 3)

Observations	1472
Dependent variable	nocc_tot
Type	Generalized linear model
Family	poisson
Link	log

χ²(25)	94.871
Pseudo-R² (Cragg-Uhler)	0.063
Pseudo-R² (McFadden)	0.012
AIC	7793.159
BIC	7930.813

	exp(Est.)	2.5%	97.5%	z val.	p
(Intercept)	10.064	9.310	10.878	58.170	0.000
orderhw1	1.112	0.958	1.291	1.399	0.162
orderhw2	1.531	1.148	2.043	2.897	0.004
orderhw3	1.521	1.304	1.775	5.332	0.000
annee2001	1.067	0.976	1.166	1.428	0.153
annee2002	1.094	1.001	1.195	1.986	0.047
annee2003	1.153	1.054	1.262	3.107	0.002
annee2004	1.045	0.956	1.142	0.962	0.336
annee2005	0.997	0.911	1.091	-0.071	0.944
annee2006	1.021	0.933	1.117	0.459	0.647
annee2007	0.990	0.905	1.084	-0.208	0.835
annee2008	1.063	0.973	1.162	1.354	0.176
annee2009	1.039	0.950	1.136	0.843	0.399
annee2010	1.084	0.992	1.184	1.783	0.075
annee2011	1.121	1.027	1.224	2.557	0.011
annee2012	1.052	0.962	1.150	1.102	0.271
annee2013	1.132	1.037	1.235	2.771	0.006
annee2014	1.041	0.952	1.138	0.876	0.381
annee2015	1.119	1.025	1.222	2.510	0.012
Joursjeudi	1.073	1.013	1.137	2.385	0.017
Jourslundi	1.022	0.964	1.084	0.723	0.470
Joursmardi	1.035	0.976	1.097	1.138	0.255
Joursmercredi	1.047	0.987	1.109	1.530	0.126
Jourssamedi	1.017	0.959	1.078	0.548	0.584
Joursvendredi	1.081	1.021	1.146	2.654	0.008
Vacances1	0.970	0.939	1.002	-1.869	0.062
Standard errors: MLE

Meta-analysis:

For First Heat Weave Category:

## 
## Random-Effects Model (k = 18; tau^2 estimator: REML)
## 
##   logLik  deviance       AIC       BIC      AICc 
##   4.3895   -8.7790   -4.7790   -3.1126   -3.9219   
## 
## tau^2 (estimated amount of total heterogeneity): 0.0315 (SE = 0.0130)
## tau (square root of estimated tau^2 value):      0.1775
## I^2 (total heterogeneity / total variability):   90.29%
## H^2 (total variability / sampling variability):  10.30
## 
## Test for Heterogeneity:
## Q(df = 17) = 535.3299, p-val < .0001
## 
## Model Results:
## 
## estimate      se    zval    pval   ci.lb   ci.ub 
##   0.2382  0.0459  5.1840  <.0001  0.1481  0.3283  *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

For Second Heat Wave:

## 
## Random-Effects Model (k = 18; tau^2 estimator: REML)
## 
##   logLik  deviance       AIC       BIC      AICc 
##   1.4442   -2.8883    1.1117    2.7781    1.9688   
## 
## tau^2 (estimated amount of total heterogeneity): 0.0319 (SE = 0.0164)
## tau (square root of estimated tau^2 value):      0.1785
## I^2 (total heterogeneity / total variability):   75.31%
## H^2 (total variability / sampling variability):  4.05
## 
## Test for Heterogeneity:
## Q(df = 17) = 65.1263, p-val < .0001
## 
## Model Results:
## 
## estimate      se    zval    pval   ci.lb   ci.ub 
##   0.2543  0.0526  4.8369  <.0001  0.1513  0.3573  *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Effect of different Heat wave timing on non-accidental mortality in France

Anna Alari and Noemie Letellier

29 avril, 2021

First Model strategy

Results per city

Metanalysis

Second model Strategy

Results per city

Metanalysis of the results for the interaction term

Second model strategy - BIS

Results per city

Metanalysis of the results for the interaction term

Third model strategy - considering order of heat wave

Define new variable “order of heat wave”

Model

First version with order variable as numeric

Results per city

Meta-analysis:

Second version with order variable as factor without interaction

Results per city

Meta-analysis:

For First Heat Weave Category:

For Second Heat Wave:

Comparing First and Second Heat Wave effect: