Interpretation of "Average Patient Survival Rate Change, Center-Reported (CR) vs CR + Social Security Death Master File". The following examples refer to the kidney table.

The columns in this table refer to the completeness of follow-up provided by the transplant centers. This reports the percentage of days that are targeted for inclusion during the follow-up period relative to the number of days that were actually reported with follow-up forms submitted by the center. See the Center-Specific Reports Analytic Conventions for further information.

A facility with 80% follow-up might follow 80% of its patients for the entire time, or might follow all of its patients for 80% of the target time. In either case there is a period of time equal to 20% of the target window of follow-up time during which an event (death) might happen but would not be reported by the center. This table uses the SSDMF to ascertain the probability that this "missed time" has similar probabilities of containing deaths, compared to the "covered time".

The first two rows show that most kidney programs have at least 90% complete follow-up.

The first set of survival statistics is not weighted by facility size. That is, in these averages, a facility performing 4 transplants is treated the same as one with 150 transplants. Separately for each completion percentage (column), most groups of facilities show no difference in survival when the extra ascertainment of the SSDMF is used. The "center change" row refers to the average of the changes at all the centers, and is not the difference in the two averages. The second set of survival statistics shows these same averages, giving more weight to the larger volume transplant centers.

The largest differences are seen in facilities with the lowest percentage of follow-up completed. For these facilities, overall survival is higher when the extra ascertainment is used. This is possible because though the SSDMF adds information about deaths, it also adds years of follow-up to the survival model for patients who were otherwise excluded because they were not reported on by the center. A summary of the deaths and years of follow-up added by use of the SSDMF is shown in the last two sections of the table. For example, in the kidney tables, the SSDMF adds about 0.7 deaths (unweighted for facility size) per facility, and at the same time adds about 10.5 years of follow-up, resulting in an overall survival rate that is very similar to that based only on center-reported information.

For full details on the calculation of center-specific survival, see the Center-Specific Reports Analytic Conventions.