OPO Specific Report - Table 3: Measures of Organ Donation Rates { View Data for OPOs ]

The nation's Organ Procurement and Transplantation Network (OPTN) consists of 59 separate Donation Service Areas (DSAs). A DSA consists of an Organ Procurement Organization (OPO), at least one transplant center, and two or more hospitals. Table 3 of the OPO-specific reports (www.ustransplant.org/opo-report.html) provides measures of donation rates and the basic components that determine these rates for each of the DSAs. These new reports provide the OPO community and the public with additional insight into the steps inherent in obtaining deceased donors for organ transplantation.

As an OPO works with hospitals, families, and others in the DSA to turn eligible candidates into actual donors, two separate processes are at work. First, eligible deaths must be identified and referred by each hospital to the OPO -- this is reflected in the notification rate reported here. Second, these referrals may be converted into actual donors by OPO and hospital staff -- this is the donation rate. The separation of these two concepts helps to provide insight into the overall success of an OPO.

The characteristics of DSAs vary widely with respect to many factors that affect donation rates. Such factors include characteristics of the population served, such as age, as well as the characteristics of the hospitals in these areas, such as trauma center designation. Although these factors are not influenced by the OPO, they impact measures of donation. Table 3 presents expected rates that take into account these case-mix differences to be used for comparison to observed notification and donation rates. The SRTR continues to work to improve the methodology for these adjustments, taking advantage of newly available data sources and developing new statistical models.

A sub-table, 3A, shows donation rates by individual hospital within the DSA.

Statistics Presented

The first section of Table 3 presents observed donor recovery and referral data reported to the OPTN by each OPO from July 2002 to June 2003.

1. Deceased Donors: All deceased donors recovered in this DSA, with at least one organ recovered for the purpose of transplant.
   
   
2. Deceased Donors Meeting Eligibility Criteria: Deceased heart-beating donors aged 0-70 years old.
   
   
3. Additional Donors: Deceased donors exceeding eligibility criteria above, such as those at least 71 years old, or Donation after Cardiac Death (DCD) donors.
   
   
4. Eligible Deaths: These candidates include any heart-beating individuals meeting the definition of death by neurological criteria; aged 0-70 years old; who have not been diagnosed with exclusionary medical conditions such as leukemia, etc.
   
   
5. Observed (Crude) Donation Rate: The number of Deceased Donors Meeting Eligibility Criteria per 100 Eligible Deaths.
   
   
6. Notifiable Death Count: A subset of all in-hospital deaths age 0-70 years, with no exclusionary medical diagnoses for possible donation. This is a more broadly-defined -- and bigger -- group than only those who may be good candidates for donation (about 12 percent end up being organ donors), but excludes poor candidates as much as is possible with available population-level death records. The number of Notifiable Deaths is based on methodology to determine the relevant subset of deaths in a geographic area, described in Ojo et al¹.
   
   
7. Observed (Crude) Notification Rate: The number of Eligible Deaths per 100 Notifiable Deaths.
   

Expected Rates Differ from National Averages

The observed donation rate and observed notification rate are measures of the different aspects in the deceased donor recovery process. However, a large amount of variation in these observed rates exists among DSAs. This variation may be explained by the characteristics of the hospitals, population and geographic area served by a DSA. For example, the donation rate is expected to be higher in DSAs that have a large number of hospitals with a level 1 or 2 trauma center. Conversely, DSAs that contain many hospitals with fewer than 100 beds are not expected to have a high donation rate. The adjusted measures provided in the second section of Table 3 take these various characteristics into consideration when attempting to explain the true differences in donation and notification rates found between service areas.

1. Expected Notification Rate: The Expected Notification Rate is the rate "expected" for an OPO, based on the national experience of all OPOs serving Donation Service Areas with similar characteristics. Providing a comparison point for the Observed Notification Rate, this measure shows the national notification rate, adjusted to the characteristics of notifiable deaths (distribution of age, sex, race and cause of death) within the DSA.
   
   
2. Expected Donation Rate (Hospital Characteristics): The Expected Donation Rate is the rate "expected" for an OPO based on the national experience of other DSAs serving hospitals with similar characteristics. Providing a first comparison point for the Observed Donation Rate, the Expected Donation Rate is adjusted for the following hospital characteristics: Level 1 or 2 Trauma Center, Metropolitan Statistical Area Size, CMS Case Mix Index, Total Bed Size, Primary Service and Hospital Control/Ownership.
   
   
3. Expected Donation Rate (Hospital Characteristics, Notification Rate): This rate provides a second comparison point for the Observed Donation Rate that takes into account differences in Notification Rates. This adjusted measure is an estimate of the donation rate for an OPO derived from the hospital characteristics listed above (#2), plus an additional adjustment of the Expected Notification Rate. By incorporating the Expected Notification Rate, this comparison point also reflects the demographic mix of deaths within the DSA.
   

Since DSAs serve such a wide variety of hospital, population and geographic characteristics, these expected rates provide a better point of comparison to observed rates than do national averages.

Statistical Comparisons

The expected measures in Table 3 are provided for comparison to the observed measures in the form of a standardized ratio. This standardized ratio is calculated as the ratio of observed to expected where 1.0 is equal to the reference. A ratio above 1.0 indicates that the observed measure for an OPO is greater than the expected, while a ratio below 1.0 indicates that the observed measure is less than what would expected given the national experience.

The 95% confidence intervals of these ratios are provided to describe the uncertainty of the estimated expected measures. The width of the confidence interval varies by DSA, depending on the amount of data available and the variability within the data. The p-values given test for statistical significance between the observed and expected measures. A p-value less than or equal to 0.05 indicates statistical significance, meaning that the difference between the ratio indicated and 1.0 (also the difference between observed and expected) is large enough that it would occur solely by random chance less than 5% of the time. A p-value greater than 0.05 indicates that the result is not statistically significant and the difference could more likely be due to random chance.

Data Sources

Counts of actual donors are taken from Deceased Donor Registration forms submitted to the OPTN during the applicable time period. Data from these forms are used to classify these donors as "eligible" or "additional" donors.

Eligible deaths, referrals, and consents are reported by each OPO, aggregated to a hospital level, to the OPTN. Collection of these data began in August 2001. Hospital-level figures, reported in table 3A, are taken directly from this referral information. See note below regarding consistency of these self-reported data and how it is addressed in the methodology for this table. Demographic information regarding eligible deaths is not currently available and, therefore, is not reflected in the development of these measures. The feasibility of expanding the data collection to include this information is being investigated.

Notifiable death count is based on methodology to determine the relevant subset of deaths in a geographic area, described in Ojo et al¹. The primary data source for these counts is the National Center for Health Statistics (NCHS) Underlying and Multiple Cause-of-Death Public-Use Data File, 1998. Data were obtained from all certificates filed for deaths occurring in each state for 1998. Continued refinement of this methodology will include updating this data source.

Characteristics of the hospitals, used for computing the expected donation rate within each OPO, are taken from the American Hospital Association Annual Survey database, 2000-2001. These characteristics include such items as Level 1 or 2 Trauma Center, Metropolitan Statistical Area Size, Total Bed Size, Primary Service and Hospital Control/Ownership. Individual OPOs have been given an opportunity, through the OPTN data collection system, to verify Trauma Level designation.

The Centers for Medicare and Medicaid Services (CMS) Case Mix Index, also used for computing the expected donation rate, was obtained from www.cms.hhs.gov. The index is used for measuring the costliness of cases treated by a hospital relative to the cost of the national average of all Medicare hospital cases. In other words, this index provides an indication of the severity of cases that are treated by a particular hospital and is found to be positively associated with donation rates.

Note on self-reported data: Eligible deaths, the denominator for the donation rate, are newly collected, self-reported data that are not currently audited. Therefore, data reporting may still be subject to differences in reporting consistency among OPOs. By presenting a donation rate that takes into account the expected notification rate, these tables address this issue by adjusting the expected donation rate accordingly. This adjustment is based on the number of eligible deaths expected within a DSA relative to how many are actually reported. For instance, a DSA with a conservative definition of eligible deaths would likely have a high observed donation rate and a lower than expected notification rate. By adjusting for this notification rate, the second expected donation rate adjusts for possible differences in referral reporting patterns among OPOs.

Future revisions: Analytical work regarding the methodology used in developing these expected measures of donation will continue by the SRTR. In the future, updated information contained in Tables 3 and 3a will be released on a semi-annual basis.

Contact Information

The Scientific Registry of Transplant Recipients (SRTR) supports the ongoing evaluation of the scientific and clinical status of solid organ transplantation including kidney, heart, liver, heart-lung, lung, intestine, and pancreas transplants. This work is done under Contract 231-00-0116 with the U.S. Department of Health and Human Services (HHS), Health Resources and Services Administration (HRSA). At present the SRTR analyzes data regarding approximately 200,000 transplant recipients and their donors. For further information about analytical methods used in developing the above measures, please visit www.ustransplant.org/opo-report.html. In addition, the SRTR help-desk can be reached at (734) 665-4108 (ext. 267), or via email at mail@ustransplant.org.

References

¹ Ojo AO, Wolfe RA, Leichtman AB, et.al. A Practical Approach to Evaluate the Potential Donor Pool and Trends in Cadaveric Kidney Donation. Transplantation 1999; 67:548-566.

View Data for OPOs