Patterns of geographic variability in mortality and eligible deaths between organ procurement organizations

• Published in: American journal of transplantation Vol. 19; no. 10; pp. 2756 - 2763
• Main Authors: Cannon, Robert M., Jones, Christopher M., Davis, Eric G., Franklin, Glen A., Gupta, Meera, Shah, Malay B.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Limited 01-10-2019
• Subjects: Blood & organ donations; clinical research/practice; Death; donors and donation; Epidemiology; Female; Geographical distribution; Geography; Health care industry; health services and outcomes research; Humans; Male; Mortality; organ procurement; organ procurement and allocation; organ procurement organization; Organ Transplantation - mortality; Registries - statistics & numerical data; registry/registry analysis; Tissue and Organ Procurement - organization & administration; Tissue and Organ Procurement - standards; Tissue and Organ Procurement - statistics & numerical data; Tissue Donors - supply & distribution; United States; Waiting Lists - mortality; United States; United States > US; organ procurement; registry/registry analysis; organ procurement organization; clinical research/practice; organ procurement and allocation; donors and donation; health services and outcomes research
• ISSN: 1600-6135; 1600-6143
• DOI: 10.1111/ajt.15390

Eligible deaths are currently used as the denominator of the donor conversion ratio to mitigate the effect of varying mortality patterns in the populations served by different organ procurement organizations (OPOs). Eligible death is an OPO‐reported metric rather than a product of formal epidemiological analysis, however, and may be confounded with OPO performance. Using Scientific Registry of Transplant Recipients and Centers for Disease Control and Prevention data, patterns of mortality and eligible deaths within each OPO were analyzed with the use of formal geostatistical analysis to determine whether eligible deaths truly reflect the geographic patterns they are intended to mitigate. There was a 2.1‐fold difference in mortality between the OPOs with the highest and lowest rates, with significant positive spatial autocorrelation evident in mortality rates (Moran I = .110; P < .001), meaning geographically proximate OPOs tended to have similar mortality rates. The eligible death ratio demonstrated greater variability, with a 4.5‐fold difference between the OPOs with the highest and lowest rates. Contrary to the pattern of mortality rates, the geographic distribution of eligible deaths among OPOs was random (Moran I = −.002; P = .410). This finding suggests geographic patterns do not play a significant role in eligible deaths, thus questioning its continuing use in OPO performance comparisons.
Eligible deaths do not reflect underlying geographic patterns of mortality, which calls into question their value as an organ procurement organization performance metric. See Doby et al's commentary on page 2675.