3.1 Descriptive analysis
Table 1 summarized daily weather conditions, air pollutants and hospital admissions. A total of 23555 IHD hospital admissions were recorded, which consisted of 10477 admissions CAD and 13078 for ACS. There was an average of 13 IHD hospitalizations per day. There were seven IHD daily hospital admissions on average among patients aged < 65 years, greater than those among the elderly, which had an average daily admission of 6. More males were admitted to hospital every day on average than females due to IHD (10 vs 3). During the survey period in Lanzhou, China, the average daily mean temperature, relative humidity, pressure, and wind speed were 11.3ºC, 50.8%, 848.3 hPa, and 1.1 m/s, respectively. The minimum value, 75 th percentile and the maximum value of daily average temperatures were − 12ºC, 20ºC and 31ºC, respectively. Daily mean concentrations of SO2, NO2, CO, O3 and PM2.5 were 19.8 µg/m3, 53.5 µg/m3, 1.2 mg/m3, 94.6 µg/m3 and 47.6 µg/m3, respectively.
Table 1
Descriptive statistics of daily number of ischemic heart disease hospital admissions, meteorological factors, and air pollutants during January 1, 2015 and December 31, 2019 in Lanzhou, China
|
|
Min
|
10%
|
25%
|
50%
|
75%
|
90%
|
Max
|
Mean
|
SD
|
Daily Admission
|
|
|
|
|
|
|
|
|
|
|
|
IHD
|
0
|
3
|
6
|
11
|
18
|
25
|
59
|
13
|
9
|
|
CAD
|
0
|
0
|
1
|
5
|
8
|
13
|
35
|
6
|
5
|
|
ACS
|
0
|
2
|
3
|
6
|
10
|
14
|
34
|
7
|
5
|
Sex
|
|
|
|
|
|
|
|
|
|
|
|
male
|
0
|
2
|
4
|
8
|
13
|
19
|
47
|
10
|
7
|
|
female
|
0
|
0
|
1
|
3
|
5
|
7
|
19
|
3
|
3
|
Age
|
|
|
|
|
|
|
|
|
|
|
|
< 65
|
0
|
1
|
3
|
6
|
10
|
14
|
36
|
7
|
6
|
|
≥ 65
|
0
|
1
|
2
|
5
|
8
|
11
|
26
|
6
|
4
|
Meteorology
|
|
|
|
|
|
|
|
|
|
|
|
Average temperature (ºC)
|
-12.3
|
-2.7
|
2.1
|
12.6
|
19.9
|
23.6
|
30.8
|
11.3
|
9.9
|
|
Maximum temperature (ºC)
|
-6.7
|
3.7
|
9.0
|
19.4
|
27.1
|
31.2
|
38.6
|
18.3
|
10.4
|
|
Minimum temperature (ºC)
|
-17.1
|
-7.0
|
-2.1
|
7.2
|
14.3
|
17.9
|
25.2
|
6.1
|
9.4
|
|
Diurnal temperature range (ºC)
|
1.4
|
6.4
|
9.2
|
12.3
|
15.0
|
17.6
|
25.8
|
12.2
|
4.2
|
|
Average surface temperature (ºC)
|
-9.9
|
-5.2
|
2.2
|
15.1
|
23.3
|
32.4
|
39.9
|
13.7
|
12.3
|
|
Average pressure (hPa)
|
833.0
|
841.4
|
844.1
|
848.2
|
852.3
|
855.6
|
866.3
|
848.3
|
5.4
|
|
Maximum pressure (hPa)
|
835.8
|
843.7
|
846.5
|
851.0
|
855.1
|
858.5
|
870.0
|
851.0
|
5.7
|
|
Minimum pressure (hPa)
|
825.7
|
837.8
|
840.6
|
844.4
|
848.8
|
852.1
|
863.5
|
844.7
|
5.6
|
|
Relative humidity (%)
|
12.3
|
31.3
|
39.3
|
50.8
|
61.7
|
76.0
|
95.8
|
50.8
|
15.0
|
|
Wind speed (m/s)
|
0.0
|
0.5
|
0.8
|
1.1
|
1.4
|
1.9
|
6.0
|
1.1
|
0.6
|
Pollutants
|
|
|
|
|
|
|
|
|
|
|
|
SO2 (µm/m3)
|
3.0
|
7.0
|
10.0
|
15.0
|
26.0
|
41.0
|
84.0
|
19.8
|
13.6
|
|
NO2 (µm/m3)
|
13.0
|
29.0
|
38.0
|
50.0
|
64.0
|
84.0
|
150.0
|
53.5
|
21.4
|
|
CO (mg/m3)
|
0.3
|
0.6
|
0.7
|
1.0
|
1.4
|
2.4
|
4.8
|
1.2
|
0.7
|
|
O3 (µm/m3)
|
11.0
|
46.0
|
64.0
|
90.0
|
123.0
|
149.0
|
236.0
|
94.6
|
39.5
|
|
PM2.5 (µm/m3)
|
9.0
|
22.0
|
30.0
|
42.0
|
57.0
|
82.0
|
278.0
|
47.6
|
26.1
|
|
PM10 (µm/m3)
|
16.0
|
49.0
|
70.0
|
98.0
|
136.0
|
183.0
|
1344.0
|
115.9
|
91.1
|
Note: IHD: Ischemic heart disease, CAD: Coronary artery disease, ACS: Acute coronary syndrome.
|
3.2 Correlation analysis
Generally, the number of IHD hospital admissions was positively correlated with two air pollutants (NO2, r = 0.06, P < 0.05; O3, r = 0.01, P > 0.05, Fig. 1) and two meteorological factors (humidity, r = 0.06, P < 0.05; pressure, r = 0.03, P > 0.05). Conversely, there were negative correlation between the number of IHD hospital admissions and temperature (r=-0.04, P > 0.05), diurnal temperature range (r=-0.02, P > 0.05) and wind speed (r=-0.09, P < 0.01). In addition, there were negatively correlation between meteorological factors except three groups, temperature and diurnal temperature range (r = 0.23, P < 0.01), temperature and wind speed (r = 0.25, P < 0.01), pressure and humidity (r = 0.16, P < 0.01). However, spearman correlation coefficients between meteorological factors and air pollutants were showed that temperature, humidity, and wind speed was negatively correlated with air pollutants except O3.
The daily number of IHD and other IHD subtypes hospital admissions during January 1, 2015 and December 31, 2019 in Lanzhou, China were shown in Fig. 2. Since October 2016, the daily number of IHD hospital admissions had increased significantly and reached the peak of 59 admissions in November 2017 (Table 1). The daily number of CAD hospital admissions rose gradually during the period from 2015 to 2016, followed by an incidence peak of 35 admissions in 2017 (Table 1). The daily number of ACS hospital admissions was relatively stable during the period from 2015 to 2016, followed in 2017 by an incidence peak of 34 admissions (Table 1). The number of IHD and other IHD subtypes hospital admissions increased in spring and autumn, decreased in summer and winter, and decreased briefly and significantly in January each year.
The Fig. 3 three-dimensional (3D) displayed the exposure-lag-response association of temperatures and lag days and the RR with the reference value of 20ºC. At first, the minimum value of temperature (-12ºC) and the maximum value of temperature (30ºC) seemed to reduce hospital admissions, but their harmful effect increased with lag time. The harmful effect of the minimum value of temperature (-12ºC) on IHD was reached maximum at approximately lags 20–27 days. However, the maximum harmful effect of the maximum value of temperature (30ºC) appeared after lag 26. Besides, between 0ºC-15ºC, the harmful effect appeared at lags 0–3 days and lags 28–30 days. Figure 4 was more intuitive to show the effect of temperature on IHD hospital admissions.
Figure 5 depicted the lag patterns in the effects of different temperature on IHD hospital admissions at a lag of 0–30 days. It shows that the reduction in RR at 12 ºC at lag10-15 is statistically significant. The shaded area in the figure represents the 95% CI, and the lag10-15 segment has a 95% CI < 1. The temperature effects of 2ºC and 19ºC occurred on the lag 0 day and then attenuated gradually until lag 4 days and lag 20 days, respectively.
Figure 6 showed the cumulative effects in terms of exposure-response between different temperature and IHD hospital admissions. At the temperature of -12ºC, -5ºC, -2ºC, 23ºC, 27ºC, 30ºC, the cumulative RR of IHD hospital admissions were both negatively correlated, but it was opposite at 2ºC and 19ºC. At the temperature of -12ºC, the cumulative RR of IHD hospital admissions declined with the lowest cumulative RR of 0.39 (95% confidence interval (CI): 0.24–0.65), occurred at a lag of 17–18 days and there appeared a statistically significant negative effect over the lag of 5–30 days. At the temperature of 19ºC, the statistically significant harmful effect also appeared over the lag of 10–20 days and the cumulative RR of IHD hospital admissions increased, with the highest cumulative RR being 1.02 (95% CI: 1.00-1.03), occurring at a lag of 18 days.
Figure 7 showed the exposure-response relationship curves between whole temperature range and daily IHD hospital admissions in a lag of 3, 7, 14, 21 and 30 days. The curves were generally similar in both shape and reference temperature. There were consistent non-linear and inverse J-shaped associations between whole temperature range and IHD hospital admissions using different cumulative lags, with higher RRs at about temperatures of 8–9ºC in the lag 3, 7 and 30 days and there appeared a statistically significant negative effect of temperature < -5 ºC in the lag of 14, 21 and 30 days.
Table 2 presented the maximum RR of cold and hot temperatures for hospital admissions in detail. The maximum RR of extremely cold temperature (-12ºC) on IHD hospital admissions for the male group, female group, < 65 years group and ≥ 65 years group were 1.048 (95% CI: 0.963–1.140), 1.060 (95% CI: 0.923–1.218), 1.055 (95% CI: 0.946–1.175), 1.064 (95% CI: 0.972–1.165), respectively. The maximum RR of extremely hot temperature (30ºC) on IHD hospital admissions for the male group, female group, < 65 years group and ≥ 65 years group were 0.897 (95% CI: 0.767–1.048), 1.111 (95% CI: 0.906–1.362), 1.021 (95% CI: 0.957–1.091), 0.957 (95% CI: 0.900-1.018), respectively. There was no significant difference in extremely cold temperature, but the harmful effects of extremely hot temperatures seemed higher among female and the < 65 years group compared with male and the ≥ 65 years group. For the CAD hospital admissions, the maximum RR of extremely cold temperature for the male group, female group, < 65 years group and ≥ 65 years group were 1.074 (95% CI: 0.947–1.216), 1.336 (95% CI: 1.111–1.607), 1.136 (95% CI: 1.001–1.288), 1.117 (95% CI: 0.956–1.306), respectively. Remarkably, there was a statistically significant positive correlation in the maximum RR of extremely cold temperature on CAD hospital admissions for female group and < 65 years group, suggesting that the risk increased significantly in female and < 65 years of age at -12ºC. The maximum RR of extremely hot temperature on CAD hospital admissions for the male group, female group, < 65 years group and ≥ 65 years group were 0.907 (95% CI: 0.724–1.136), 1.121 (95% CI: 0.857–1.466), 1.028 (95% CI: 0.937–1.128), 0.951 (95% CI: 0.872–1.038), respectively. Similarly, there was no significant difference in extremely cold temperature, but the harmful effects of extremely hot temperatures seemed to be higher among female and the < 65 years group compared with male and the ≥ 65 years group. For the ACS hospital admissions, the maximum RR of extremely cold temperature for the male group, female group, < 65 years group and ≥ 65 years group were 1.030 (95% CI: 0.950–1.118), 0.937 (95% CI: 0.695–1.007), 0.892 (95% CI: 0.619–1.289), 0.908 (95% CI: 0.795–1.037), respectively. The maximum RR of extremely hot temperature on ACS hospital admissions for the male group, female group, < 65 years group and ≥ 65 years group were 0.889 (95% CI: 0.754–1.044), 1.113 (95% CI: 0.858–1.445), 1.016 (95% CI: 0.944–1.092), 0.938 (95% CI: 0.772–1.140), respectively. The harmful effects of extremely cold temperatures seemed to be higher among male compared with the < 65 years group, female group and the ≥ 65 years group and the harmful effects of extremely heat temperatures seemed higher among female and the < 65 years group compared with male and the ≥ 65 years group.
Table 2
Maximum relative risks (95%CIs) (%) of cold and heat effects between temperatures and daily IHD, CAD, ACS hospital admissions, with reference at 20ºC.
Disease
|
Class
|
the minimum value of temperature (-12ºC)
|
lag
|
the maximum value of temperature (30ºC)
|
lag
|
IHD
|
male
|
1.048 (0.963–1.140)
|
14
|
0.897(0.767–1.048)
|
1
|
|
female
|
1.060(0.923–1.218)
|
9
|
1.111(0.906–1.362)
|
1
|
|
< 65
|
1.055(0.946–1.175)
|
8
|
1.021(0.957–1.091)
|
6
|
|
≥ 65
|
1.064(0.972–1.165)
|
15
|
0.957(0.900-1.018)
|
4
|
CAD
|
male
|
1.074(0.947–1.216)
|
14
|
0.907(0.724–1.136)
|
0
|
|
female
|
1.336(1.111–1.607)
|
7
|
1.121(0.857–1.466)
|
0
|
|
< 65
|
1.136(1.001–1.288)
|
10
|
1.028(0.937–1.128)
|
6
|
|
≥ 65
|
1.117(0.956–1.306)
|
7
|
0.951(0.872–1.038)
|
5
|
ACS
|
male
|
1.030(0.950–1.118)
|
13
|
0.889(0.754–1.044)
|
0
|
|
female
|
0.937(0.695–1.007)
|
6
|
1.113(0.858–1.445)
|
0
|
|
< 65
|
0.892(0.619–1.289)
|
0
|
1.0159(0.944–1.092)
|
6
|
|
≥ 65
|
0.908(0.795–1.037)
|
5
|
0.938(0.772–1.140)
|
0
|
Table 3 presented the maximum cumulative RR of cold and hot temperatures for hospital admissions in detail. For the IHD hospital admissions, compared with the reference temperature (20ºC), the maximum cumulative RRs associated with extremely cold temperature were 1.559 (95% CI: 0.619–3.925) for male group, 1.303 (95% CI: 0.379–4.483) for female group, 1.049 (95% CI: 0.404–2.725) for < 65 years group and 1.972 (95% CI: 0.739 5.265) for ≥ 65 years group. The maximum cumulative RRs associated with extremely heat temperature were 0.588 (95% CI: 0.362–0.953) for male group, 1.312 (95% CI: 0.832–2.071) for female group, 0.734 (95% CI: 0.435–1.240) for < 65 years group and 0.546 (95% CI: 0.337–0.884) for ≥ 65 years group. Furthermore, for the CAD hospital admissions, the maximum cumulative RRs associated with extremely cold temperature were 2.638 (95% CI: 0.656–10.599) for male group, 3.156 (95% CI: 0.595–16.735) for female group, 2.227 (95% CI: 0.526–9.439) for < 65 years group and 3.352 (95% CI: 0.817–13.746) for ≥ 65 years group. The maximum cumulative RRs associated with extremely hot temperature were 0.498 (95% CI: 0.244–1.017) for male group, 1.513 (95% CI: 0.817-2.800) for female group, 0.711 (95% CI: 0.352–1.436) for < 65 years group and 0.475 (95% CI: 0.246–0.915) for ≥ 65 years group. For the ACS hospital admissions, the maximum cumulative RRs associated with extremely cold temperature were 1.141 (95% CI: 0.457–2.845) for male group, 0.358 (95% CI: 0.097–1.323) for female group, 0.565 (95% CI: 0.184–1.737) for < 65 years group and 1.478 (95% CI: 0.503–4.346) for ≥ 65 years group. The maximum cumulative RRs associated with extremely hot temperature were 0.633 (95% CI: 0.391–1.024) for male group, 1.222 (95% CI: 0.701–2.131) for female group, 0.745 (95% CI: 0.418–1.329) for < 65 years group and 0.592 (95% CI: 0.339–1.033) for ≥ 65 years group.
Table 3
Maximum cumulative relative risks (95%CIs) (%) of cold and heat effects between temperatures and daily ischemic heart disease (IHD), coronary artery disease (CAD), acute coronary syndrome (ACS) hospital admissions, with reference at 20ºC.
Disease
|
Class
|
Minimum value of temperature (-12ºC)
|
lag
|
Maximum value of temperature (30ºC)
|
lag
|
IHD
|
male
|
1.559(0.619–3.925)
|
0–21
|
0.588(0.362–0.953)
|
0–24
|
|
female
|
1.303(0.379–4.483)
|
0–25
|
1.312(0.832–2.071)
|
0–5
|
|
< 65
|
1.049(0.404–2.725)
|
0–16
|
0.734(0.435–1.240)
|
0–23
|
|
≥ 65
|
1.972(0.739–5.265)
|
0–22
|
0.546(0.337–0.884)
|
0–30
|
CAD
|
male
|
2.638(0.656–10.599)
|
0–23
|
0.498(0.244–1.017)
|
0–24
|
|
female
|
3.156(0.595–16.735)
|
0–26
|
1.513(0.817-2.800)
|
0–7
|
|
< 65
|
2.227(0.526–9.439)
|
0–19
|
0.711(0.352–1.436)
|
0–30
|
|
≥ 65
|
3.352(0.817–13.746)
|
0–25
|
0.475(0.246–0.915)
|
0–29
|
ACS
|
male
|
1.141(0.457–2.845)
|
0–19
|
0.633(0.391–1.024)
|
0–20
|
|
female
|
0.358(0.097–1.323)
|
0–13
|
1.222(0.701–2.131)
|
0–3
|
|
< 65
|
0.565(0.184–1.737)
|
0–20
|
0.745(0.4181.329)
|
0–23
|
|
≥ 65
|
1.478(0.503–4.346)
|
0–20
|
0.592(0.339–1.033)
|
0–30
|
IHD: ischemic heart disease; CAD: coronary artery disease; ACS: acute coronary disease. |