Geographic variation in opioid mortality by race/ethnicity in the United States, 1999-2016: Identifying epidemic hotspots
We performed joinpoint regressions by state, race/ethnicity, and opioid type. Below, each line is the result of a single joinpoint regression. We arrange the results approximately in their geographical location (except AK and HI) so the geograhpical patterning is clearer. Individual states can be explored in the tab above. Solid lines indicate a statistically significant slope (i.e., increase or decrease). Solid points on the line represent a statistically significant change in the slope before and after that point. Raw data (and 95% confidence intervals) are presented in the back. We defined statistical significance as P<0.01; however, you can adjust this value below.
This is an interactive companion to our paper, which was presented at PAA 2018 and EPC 2018. To learn more about the paper and code (or report bugs), see our Github repository. Data from 1999 to 2015 come from the restricted-access multiple cause of death files provided by the National Center for Health Statistics. When possible, we supplement this data with 2016 data from CDC WONDER. A table of the average annual percent change (AAPC) is shown at the bottom. The AAPC can be interpreted as the estimated average change from the beginning of the period to the end of the period.
Geographic variation in opioid mortality by race/ethnicity in the United States, 1999-2016: Identifying epidemic hotspots
We performed joinpoint regressions by state, race/ethnicity, and opioid type. Below, each line is the result of a single joinpoint regression. Solid lines indicate a statistically significant slope (i.e., increase or decrease). Solid points on the line represent a statistically significant change in the slope before and after that point. Raw data (and 95% confidence intervals) are presented in the back. We defined statistical significance as P<0.01; however, you can adjust this value below. There is a clear geographical patterning for some opioids --- see the national results tab for more.
This is an interactive companion to our paper, which was presented at PAA 2018 and EPC 2018. To learn more about the paper and code (or report bugs), see our Github repository. Data from 1999 to 2015 come from the restricted-access multiple cause of death files provided by the National Center for Health Statistics. When possible, we supplement this data with 2016 data from CDC WONDER. A table of the average annual percent change (AAPC) is shown at the bottom. The AAPC can be interpreted as the estimated average change from the beginning of the period to the end of the period.
Geographic variation in opioid mortality by race/ethnicity in the United States, 1999-2016: Identifying epidemic hotspots
We define epidemic hotspots as areas with both high rates of mortality as well as rapid increases in their mortality rates. Below, you can change the threshold for what is considered a 'high' mortality rate and a 'rapid' increase by adjusting the boundaries of the middle category. The text on the below (on the right) will indicate what percentile your chosen boundaries represent in the data. The plot only shows statistically significant increases, which we define as P < 0.01; however, you can adjust the level of statistical significance below.
This is an interactive companion to our paper, which was presented at PAA 2018 and EPC 2018. To learn more about the paper and code (or report bugs), see our Github repository. Data from 1999 to 2015 come from the restricted-access multiple cause of death files provided by the National Center for Health Statistics. When possible, we supplement this data with 2016 data from CDC WONDER.