Spatiotemporal variations in cardiovascular disease mortality in China from 1991 to 2009

In China, the spatiotemporal variations in cardiovascular disease (CVD) mortality are seldom characterized to understand their epidemiological features. It would be helpful to evaluate the performance of CVD-related interventions for subsequent adjustments. The 2010 Census data as well as the coronary heart disease (CHD) and stroke mortality data from the Disease Surveillance Points (DSPs) were used to calculate the age standardized death rates (ASDRs) of CVD in the DSP counties during 1991–1995, 1996–2000, 2004–2005, and 2006–2009. The ordinary kriging (OK) method was used to estimate the county-level death rates of CHD and stroke and achieved satisfactory results. The goodness-of-fit between measured and estimated values of CVD mortality was significant at the 0.01 level (0.34 < R 2 < 0.98). The counties with high CHD death rates (> 75 per 10 5 ) were located in the Northwest, North, and Northeast in 1991–2000 and then extended toward the North, Central, and South, yielding an inverted-triangle-shaped area in 2004–2009. The counties with a CHD death rate increase greater than 100% were concentrated in the Northeast and South. The Northeast-Southwest regions with a high stroke death rate gradient (> 150 per 10 5 ) narrowed in1991–2000, was followed by a slight expansion during 2004–2005, finally reducing in 2006–2009. The counties with a stroke mortality increase greater than 100% were scattered across the Northeast, Northwest, Central, and South. The epidemiological characteristics of both CHD and stroke mortality in China was spatiotemporally featured on the county level during 1991–2009.