8
C. M. García-Witulski and M. J. Rabassa
Table 1 Descriptive statistics
Panel A. Historical daily temperature data ◦ C (2004–2010, n = 25,409)
By region
Buenos aires
metropolitan
area
Pampean
Northwest
Northeast
Cuyo
(n %)
(4.5)
(44.04)
(25.69)
(13)
(12.74)
Average
Mean
17.66
16.95
19.95
21.84
17.68
S.D.
4.79
5.21
5.55
4.48
6.22
Minimum
Mean
12.59
10.72
13.08
16.01
10.27
S.D.
4.61
5.15
6.28
4.47
6.35
Maximum
Mean
22.75
23.19
26.82
27.67
25.09
S.D.
5.03
5.47
5.14
4.67
6.29
Panel B. Month mortality rate per 100,000 population by age (2004–2010, n = 25,409)
By age
0–4
5–44
45–64
>64
–
(n %)
(100)
(100)
(100)
(100)
–
Mean
11.19
7.16
50.24
347.81
–
S.D.
25.96
10.22
45.58
206.68
–
Panel C. Month mortality rate per 100,000 population by region (2004–2010, n = 25,409)
By region
Buenos aires
metropolitan
area
Pampean
Northwest
Northeast
Cuyo
(n %)
(4.5)
(44.04)
(25.69)
(13)
(12.74)
Mean
59.28
61.62
31.88
39.77
44.21
S.D.
27.81
28.64
24.57
20.66
47.69
Notes The number of observations (n = 25,409) corresponds to municipality by month, i.e., 322
municipalities by 84 month, from January 2004 to December 2010. Observations with weather
record with less than 25 days in a month were not included in the analysis (6% of total observations)
temperature above or below the observed average temperature during the 2004–2010
period.
For example, Panel A of Fig. 1 shows that on average, an additional day with a
countrywide mean temperature of 5
◦ C increases overall annual mortality rate by 3.5
pp (95% CI: 3.06–3.96) with respect to the observed mortality for a day with countrywide average temperature (i.e., 18.9
◦ C). In general, Figure 1 shows a nonlinear
relationship between temperature (daily mean, maximum, and minimum, respectively) and the overall mortality rate, i.e., all age groups, genders, and regions. A
striking regularity depicted in Fig. 1 is that the estimated impacts of colder than
average temperatures are higher than the computed estimates for hotter than average
temperatures. Again, using Panel A to illustrate, any day with a mean daily temperature below the average, 18.8
◦ C, positively and significantly affects mortality rates,
C. M. García-Witulski and M. J. Rabassa
Table 1 Descriptive statistics
Panel A. Historical daily temperature data ◦ C (2004–2010, n = 25,409)
By region
Buenos aires
metropolitan
area
Pampean
Northwest
Northeast
Cuyo
(n %)
(4.5)
(44.04)
(25.69)
(13)
(12.74)
Average
Mean
17.66
16.95
19.95
21.84
17.68
S.D.
4.79
5.21
5.55
4.48
6.22
Minimum
Mean
12.59
10.72
13.08
16.01
10.27
S.D.
4.61
5.15
6.28
4.47
6.35
Maximum
Mean
22.75
23.19
26.82
27.67
25.09
S.D.
5.03
5.47
5.14
4.67
6.29
Panel B. Month mortality rate per 100,000 population by age (2004–2010, n = 25,409)
By age
0–4
5–44
45–64
>64
–
(n %)
(100)
(100)
(100)
(100)
–
Mean
11.19
7.16
50.24
347.81
–
S.D.
25.96
10.22
45.58
206.68
–
Panel C. Month mortality rate per 100,000 population by region (2004–2010, n = 25,409)
By region
Buenos aires
metropolitan
area
Pampean
Northwest
Northeast
Cuyo
(n %)
(4.5)
(44.04)
(25.69)
(13)
(12.74)
Mean
59.28
61.62
31.88
39.77
44.21
S.D.
27.81
28.64
24.57
20.66
47.69
Notes The number of observations (n = 25,409) corresponds to municipality by month, i.e., 322
municipalities by 84 month, from January 2004 to December 2010. Observations with weather
record with less than 25 days in a month were not included in the analysis (6% of total observations)
temperature above or below the observed average temperature during the 2004–2010
period.
For example, Panel A of Fig. 1 shows that on average, an additional day with a
countrywide mean temperature of 5
◦ C increases overall annual mortality rate by 3.5
pp (95% CI: 3.06–3.96) with respect to the observed mortality for a day with countrywide average temperature (i.e., 18.9
◦ C). In general, Figure 1 shows a nonlinear
relationship between temperature (daily mean, maximum, and minimum, respectively) and the overall mortality rate, i.e., all age groups, genders, and regions. A
striking regularity depicted in Fig. 1 is that the estimated impacts of colder than
average temperatures are higher than the computed estimates for hotter than average
temperatures. Again, using Panel A to illustrate, any day with a mean daily temperature below the average, 18.8
◦ C, positively and significantly affects mortality rates,