Organ Donation and Drug Intoxication-Related Deaths in the United States (IMAGE)

Brigham and Women's Hospital

Organ Donation and Drug Intoxication-Related Deaths in the United States

Caption

Figure 1. Temporal Changes among Heart Donors Who Died from Drug Intoxication, Heart Transplant Waiting-List Size, and Rate of Donor Heart Recovery.

Panel A shows the geographic distribution of hearts recovered from donors who died from drug intoxication in 1999, and Panel B shows the geographic distribution in 2017. Map graphics were created with the use of Microsoft Excel with software powered by Bing (DSAT for MSFT, GeoNames, and Navteq). Panel C depicts the predicted annual net changes in the size of the heart transplant waiting list from 1999 to 2017, calculated by subtracting the total number of registrations that were removed from the number of new registrations. Predictions were based on ordinary least squares estimates of their mean, taking into account the time trends (the conditional mean); these nationwide time trends were estimated with the use of state-level data (Organ Procurement and Transplantation Network, 1999-2017). The conditional mean was also estimated with the use of yearly dummy variables to isolate a fully flexible time-specific component, yielding the year-specific predicted values plotted on the vertical axis. Years with net decreases in the waiting-list size are presented in blue, and years with net increases in red. I bars indicate 95% confidence intervals. Panel D shows the percentage of heart donors among the total number of persons who died from drug intoxication in United States from 1999 through 2016; nationwide time trends were estimated with the use of state-level data. A spline function was used to model the linear trend in this percentage, capturing the difference in trends before and after 2009. The conditional mean was also estimated with the use of yearly dummy variables to isolate a fully flexible time-specific component, yielding the year-specific predicted values plotted on the vertical axis. Shaded circles represent significant differences in these coefficients from the baseline year (1999), as determined with the use of White-Huber robust standard errors at the 1% significance level. I bars indicate 95% confidence intervals.

Credit

Brigham and Women's Hospital/<em>New England Journal of Medicine</em>

Usage Restrictions

Credit Brigham and NEJM.

License

Licensed content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.