1. Mortality of IHD in Chinese Population from 2010 to 2015
IHD crude mortality rates was 108.74 per 100,000 in 2015(Table 1, Fig. 1). Age-standardized IHD mortality rates among people increased 5.46% (from 86.29 to 91.00 per 100,000) from 2010 to 2015 (Fig. 2). Gender stratified analyses showed the mortality was consistently increasing in male and female during the study period. Age-standardized IHD mortality rates among people aged 40 years and older increased 5.51% (from 145.95 to 153.99 per 100,000) from 2010 to 2015 (Table 2). Gender differences were also observed, with IHD mortality increasing more slowly for males (5.18 per 100,000) than females (9.51 per 100,000) .
Table 1
Crude mortality of IHD in people over 40 years old in different regions from 2010 to 2015 (1/100,000)
| | urban | rural | male | female | east | central | west | total |
2010 | 183.04 | 208.87 | 212,24 | 183.60 | 201.69 | 227.20 | 152.40 | 197.99 |
2011 | 185.65 | 212.93 | 216.76 | 187.06 | 199.25 | 234.41 | 157.91 | 201.71 |
2012 | 190.13 | 235.96 | 230.59 | 202.48 | 206.65 | 250.34 | 184.24 | 216.59 |
2013 | 202.31 | 248.77 | 244.87 | 213,87 | 220.27 | 263.63 | 195.52 | 229.39 |
2014 | 194.78 | 233.20 | 229.40 | 203.91 | 209.10 | 245.55 | 189.70 | 216.67 |
2015 | 198.73 | 236.58 | 233.56 | 208.80 | 215.34 | 254.34 | 186.47 | 221.17 |
Gender and urban/rural stratified analyses showed the mortality was apparently different between males and female, urban and rural during the study period (Table 2). Age-standardized IHD mortality rates in female and urban were significantly lower than in male and rural. During the study period, Age-standardized IHD mortality rates among people in rural increased 9.61% from 154.79/100,000 to 169.67/100,000, but the rates decreased 1.14% in urban (from 133.89/100,000 to 132.36/100,000).
In this study, superposition effect of gender and urban/rural in age-standardized mortality rates showed that the top 3 were rural males, urban males and rural females respectively, the lowest is urban females. Age-standardized IHD mortality rates of urban males and females decreased 2.00% and 0.60%( 152.97/ 100,000 and 115.12/100,000 in 2010, 149.91/ 100,000 and 114.43/ 100,000 in 2015, respectively), however, Age-standardized IHD mortality rates of rural males and females increased 6.17% and 13.8%( 183.66/ 100,000 and 128.16/100,000 in 2010, 195.00/ 100,000 and 144.92/ 100,000 in 2015, respectively).
Table 2
Standardized mortality of IHD in people over 40 years of age in different regions from 2010 to 2015 (1/100,000)
year | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 |
Urban | | | | | | |
male | 152.97 | 153.31 | 154.5 | 159.38 | 150.23 | 149.91 |
female | 115.12 | 113.28 | 115.86 | 120.07 | 116.44 | 114.43 |
total | 133.89 | 133.11 | 135.17 | 139.83 | 133.5 | 132.36 |
Rural | | | | | | |
male | 183.66 | 184.82 | 202.73 | 212.04 | 195.63 | 195 |
female | 128.16 | 131.13 | 145.43 | 152.17 | 142.18 | 144.92 |
total | 154.79 | 156.88 | 173.15 | 181.49 | 168.3 | 169.67 |
Urban + Rural | | | | | |
male | 170.59 | 171.62 | 181.87 | 189.37 | 175.38 | 175.77 |
female | 122.56 | 123.69 | 132.81 | 138.48 | 130.81 | 132.07 |
total | 145.95 | 147.04 | 156.93 | 163.76 | 152.99 | 153.99 |
East | | | | | | |
male | 165.17 | 160.66 | 166.15 | 171.62 | 161.97 | 164.21 |
female | 118.5 | 116.91 | 120.17 | 125.53 | 117.53 | 117.93 |
total | 140.93 | 138.08 | 142.53 | 148.34 | 139.42 | 140.83 |
Central | | | | | | |
male | 202.93 | 206.87 | 216.34 | 224.24 | 204.83 | 208.63 |
female | 144.46 | 148.01 | 158.15 | 165.01 | 155.29 | 159.37 |
total | 173.01 | 176.67 | 186.84 | 194.36 | 179.93 | 184.09 |
West | | | | | | |
male | 135.27 | 139.28 | 158.79 | 169.69 | 158.3 | 151.91 |
female | 97.41 | 99.11 | 117.05 | 121.42 | 120.61 | 118.93 |
total | 116.05 | 118.84 | 137.76 | 145.6 | 139.62 | 135.71 |
As shown in Table 2, East–central-west differences were observed on the age-standardized IHD mortality among people aged 40 and over. Mortality rates were highest in the central region and lowest in the west. Mortality rates in the eastern region first increased and then decreased in total (from 140.93/100,000 to 140.83/100,000). Nevertheless mortality rates in the western and central region increased over time (from 116.05/100,000 to 135.71/100,000, from 173.01/100,000 to 184.09/100,000, respectively). The increasing slope was highest from 2010 to 2013.
Gender and region stratified analyses showed the mortality was different. For males, overall, the mortality rates in the east region showed a downward trend (from 165.17/100,000 to 164.21/100,000). However, age-standardized IHD mortality rates increased 6.40% from 173.01/100,000 to 184.09/100,000 in the central region, and it increased 16.94% from 116.05/100,000 to 135.71/100,000 in the west region. For females, in totally, age-standardized IHD mortality rates slightly decreased from 118.50/100,000 to 117.93/100,000 in the east region. The mortality of west and central regions increased in totally (from 97.41/100,000 to 118.93/100,000, from 144.46/100,000 to 159.37/100,000, respectively) although the mortality was lower after 2013.
2. The mortality of IHD in Chinese Population from 2010 to 2015 in multilevel models
The multilevel model 1(Table 3, at the end of the text) indicates that age and gender-adjusted IHD mortality increased between 2010 and 2015 (RR = 1.016, 95% CI: 1.012, 1.020), and it increased with age (RR = 1.839, 95% CI: 1.835, 1.842). IHD mortality was lower among women (RR = 0.622, 95% CI: 0.612, 0.632). The provincial and DSP intercept variance was statistically significant and important (MRRp= 1.514 and MRRd= 1.459) indicating multilevel models need to be used. IHD mortality was explained by adjustment for region in model 2. Compared to the north region, IHD mortality was significantly lower in the east (RR = 0.435, 95% CI :0.307, 0.616), southwest (RR = 0.360, 95% CI :0.246, 0.525) and south(RR = 0.512, 95% CI: 0.329,0.795). The provincial and DSP intercept variance was significantly different (MRRp= 1.258 and MRRd= 1.430). To explore the role of socioeconomic factors on the IHD mortality, urbanization rate of each DSP was added in the multilevel model 3. As shown in Table 3, comparing with the low urbanization rate areas, the RR and 95% CI of the areas with high urbanization rate was 0.728(0.631, 0.840).
Table 3
Multilevel analysis of spatiotemporal changes in mortality of ischemic heart disease in China from 2010 to 2015
| model 1 | model 2 | model 3 | model 4 | model 5 | model 6 |
Fixed effect | relative risk(95%CI) |
Year | 1.016(1.012,1.020) | 1.020(1.016,1.024) | 1.019(1.015,1.023) | 1.019(1.015,1.023) | 1.019(1.015,1.023) | 1.019(1.015,1.023) |
Gender(ref:male) | 1 | 1 | 1 | 1 | 1 | 1 |
female | 0.622(0.612,0.632) | 0.618(0.609,0.628) | 0.618(0.608,0.627) | 0.618(0.608,0.627) | 0.619(0.609,0.629) | 0.619(0.610,0.629) |
Age(5 years old as an age group) | 1.839(1.835,1.842) | 1.848(1.844,1.851) | 1.850(1.846,1.853) | 1.850(1.846,1.853) | 1.852(1.848,1.855) | 1.852(1.848,1.855) |
Area(ref: north) | | 1 | 1 | 1 | 1 | 1 |
east | | 0.435(0.307,0.616) | 0.442(0.317,0.615) | 0.484(0.346,0.677) | 0.598(0.446,0.800) | 0.641(0.487,0.843) |
central | | 0.720(0.475,1.094) | 0.705(0.475,1.048) | 0.710(0.484,1.043) | 0.837(0.604,1.159) | 0.845(0.629,1.137) |
north | | 0.512(0.329,0.795) | 0.531(0.350,0.805) | 0.572(0.376,0.870) | 0.739(0.509,1.073) | 0.743(0.527,1.047) |
weatsouth | | 0.360(0.246,0.525) | 0.334(0.233,0.478) | 0.361(0.250,0.520) | 0.472(0.340,0.656) | 0.509(0.375,0.691) |
westnorth | | 0.865(0.589,1.270) | 0.824(0.574,1.185) | 0.757(0.526,1.090) | 0.770(0.564,1.052) | 0.795(0.596,1.060) |
eastnorth | | 1.051(0.688,1.605) | 1.051(0.705,1.568) | 0.994(0.668,1.480) | 1.023(0.754,1.389) | 0.948(0.717,1.255) |
Urbanization rate(ref: low) | | | 1 | 1 | 1 | 1 |
middle | | | 0.949(0.816,1.104) | 0.987(0.842,1.157) | 1.043(0.897,1.213) | 1.002(0.863,1.163) |
high | | | 0.728(0.631,0.840) | 0.771(0.646,0.920) | 0.820(0.684,0.984) | 0.819(0.672,0.998) |
Current smoking rate(ref: low) | | | | 1 | 1 | 1 |
middle | | | | 0.971(0.826,1.143) | 0.952(0.820,1.105) | 0.952(0.824,1.101) |
high | | | | 1.134(0.947,1.358) | 1.133(0.954,1.346) | 1.082(0.914,1.281) |
Excessive red meat intake(ref: low) | | | | 1 | 1 | 1 |
middle | | | | 0.890(0.757,1.048) | 0.893(0.765,1.043) | 0.890(0.766,1.032) |
high | | | | 0.891(0.733,1.084) | 0.918(0.763,1.103) | 0.900(0.756,1.072) |
Insufficient Vegetable and fruit intake(ref: low) | | | | 1 | 1 | 1 |
middle | | | | 0.943(0.811,1.096) | 0.941(0.811,1.092) | 0.927(0.802,1.071) |
high | | | | 0.901(0.767,1.058) | 0.898(0.769,1.048) | 0.884(0.762,1.026) |
Average meditation time(ref: low) | | | | 1 | 1 | 1 |
middle | | | | 0.895(0.767,1.045) | 0.883(0.763,1.023) | 0.851(0.739,0.980) |
high | | | | 0.933(0.789,1.105) | 0.869(0.737,1.025) | 0.868(0.741,1.018) |
Average BMI (ref:low) | | | | | 1 | 1 |
middle | | | | | 1.135(0.939,1.373) | 1.050(0.870,1.268) |
high | | | | | 1.495(1.172,1.906) | 1.436(1.135,1.817) |
Average SBP(ref:low) | | | | | 1 | 1 |
middle | | | | | 1.150(0.983,1.346) | 1.149(0.986,1.339) |
high | | | | | 1.294(1.007,1.663) | 1.310(1.019,1.684) |
Average FBG (ref:low) | | | | | 1 | 1 |
middle | | | | | 1.069(0.907,1.261) | 1.061(0.905,1.243) |
high | | | | | 1.093(0.909,1.314) | 1.079(0.903,1.290) |
Average HDL (ref:low) | | | | | 1 | 1 |
middle | | | | | 0.830(0.706,0.977) | 0.816(0.692,0.962) |
high | | | | | 0.746(0.619,0.899) | 0.741(0.616,0.891) |
Average total cholesterol(ref:low) | | | | | 1 | 1 |
middle | | | | | 0.932(0.772,1.125) | 0.899(0.747,1.080) |
high | | | | | 1.062(0.838,1.346) | 1.046(0.828,1.321) |
Triglyceride level (ref:low) | | | | | 1 | 1 |
middle | | | | | 1.012(0.857,1.196) | 1.080(0.913,1.278) |
high | | | | | 1.081(0.897,1.302) | 1.079(0.899,1.295) |
No health insurance ratio(ref:low) | | | | | | 1 |
middle | | | | | | 1.137(0.964,1.340) |
high | | | | | | 1.218(1.007,1.473) |
Household income level(ref:low) | | | | | | 1 |
middle | | | | | | 1.146(0.989,1.327) |
high | | | | | | 0.886(0.741,1.059) |
Random effect | | | | | | |
Variance (SE) Provincial level | 0.189(0.057) | 0.058(0.023) | 0.054(0.020) | 0.049(0.019) | 0.022(0.011) | 0.015(0.009) |
MRR | 1.514 | 1.258 | 1.248 | 1.235 | 1.152 | 1.124 |
PCV | - | 70.1% | 72.2% | 74.9% | 89.8% | 94.7% |
Variance (SE) Monitoring point level | 0.157(0.020) | 0.141(0.018) | 0.114(0.015) | 0.112(0.015) | 0.095(0.012) | 0.089(0.011) |
MRR | 1.459 | 1.430 | 1.380 | 1.376 | 1.342 | 1.329 |
PCV | - | 10.4% | 29.4% | 30.1% | 35.6% | 38.1% |
Lifestyle factors (smoking, red meat intake, vegetable and fruit deficiency, sedentary time) were included in the multilevel model 4. The results shown that smoking, red meat intake, vegetable and fruit deficiency, sedentary time did not influenced on the IHD mortality. Compared to model 3, model 4 increased approximately 0.7% of the DSP-level PCV in IHD mortality. In models incorporating the six metabolic risk factors (average BMI, systolic pressure, mean FBG, HDL cholesterol, cholesterol level and triglyceride level), only average BMI and systolic pressure were positively associated with IHD mortality (RR = 1.495, 95% CI: 1.172, 1.906, RR = 1.294, 95% CI: 1.007, 1.663)(in model 5). 1.495(95%CI: 1.172, 1.906), 1.294(95༅CI:1.007,1.663). Moderate and high HDL cholesterol level were negatively correlated to IHD mortality (RR = 0.830, 95༅CI:0.706,0.977, RR = 0.746, 95༅CI: 0.619,0.899, respectively). In order to know the role of personal economic situation on the IHD mortality, health insurance rate and family income level were added in the multilevel model 6. The low health insurance rate was positively associated with the LHD mortality(RR = 1.218, 95% CI༚1.007,1.473), indicating that the necessity and importance of continuing to expand coverage of health insurance. The relationship between family income level and IHD mortality is not obvious(RR = 0.886, 95% CI:0.741,1.059). Moderate sedentary time was negatively correlated to IHD mortality(RR = 0.851, 95% CI: 0.739,0.980) in model 6༌but it showed no correlation with IHD mortality in model4,5. In model 6, approximately 94.7% and 38.1% of the provincial-level the DSP-level variation in IHD mortality was explained by adding behavioral risk factors and socioeconomic factors in turn.