Know your PSR cohorts: transplant and event timelines made easy…?

Cohorts graphic

Know your PSR cohorts: transplant and event timelines made easy…?

By: Nino Dzebisashvili, Ph.D

The first step towards understanding PSR methodology is knowing which transplants and events (death/graft loss) are included in the current and future PSR releases. I have always used an Excel matrix with cohort start and end dates to help navigate through PSR cohort timelines. If you are a visual learner like me, then this lookup chart may simplify cohort and event assignments for you. The chart shows transplant and event accrual timelines for the next four PSR releases and can be easily referenced to understand the impact of observed (or potential) death or graft losses on upcoming reports. In addition, I will review the key inclusion/exclusion rules for adult one-year patient and graft survival and apply these rules to hypothetical scenarios.

The lookup chart displays accrual timelines for (1) transplant cohorts and (2) post-transplant events. The top part of the chart shows transplant cohort inclusion timelines for the next four PSR releases.

The upcoming December 2017 release, for example, will include transplants performed between 6/30/14 and 12/31/2016. Each PSR release uses a 2.5-year accrual window that closes one year prior to the PSR release date (June 30 or December 31) to allow for sufficient follow-up time to monitor for post-transplant events. For example, the December 2017 cohort accrual ends on 12/31/2016 and the June 2018 cohort accrual ends on 6/30/17.

To capture post-transplant events all qualified transplants (within each 2.5-year cohort) are observed for up to one year for evidence of death or graft loss. The observation (follow-up) window closes six months prior to the report release date. For example, the follow-up window for the December 2017 cohort closes on 6/30/17 and the follow-up window for June 2018 cohort closes on 12/31/17.

Scenario #1: Patient A is transplanted on 09/05/15 and Patient B is transplanted on 9/05/14. It’s the end of summer 2017, which future PSR cohorts will include these transplants?

Patient A will be included in the following three cohorts: December 2017, June 2018 and December 2018. Patient B will only be included in the December 2017 cohort.

The lower part of the lookup chart shows accrual timelines for transplant events. If an event (death or graft loss) occurs within 365 days of transplant for transplants in the (un-shaded) two-year timeline they will impact the respective cohorts.

Scenario #2: Patient C was transplanted 9/15/15 and had a graft loss on 8/15/16.     Which releases are affected?

This transplant will be included in three cohorts – December 2017, June 2018 and December 2018 and the graft loss will impact all these releases.

Scenario #3: Patient D was transplanted on 8/15/14 and lost her graft on 10/15/15. Which releases are affected?

This transplant will be included in the December 2017 cohort. But since the graft loss occurred more than 365 days after the initial transplant, none of the releases will be impacted for events. 

Transplants occurring within the orange shaded areas employ a so called “six-month follow-up rule” to determine inclusion in the upcoming or future releases. The six-month rule is only applied to patients transplanted within the final six months of the cohort end date. These patients are only followed up for six months for death or graft loss (the follow-up period stops at the end of the gray shaded area). According to the six-month follow-up rule, if a patient experiences death or graft loss within six months to the day of the month of transplant, the event(s) will be included in the upcoming release but any events(s) between six months and 365 days of transplant will be captured in subsequent release(s).

Scenario #4: Patient E was transplanted on 9/01/16 and experienced graft loss four months later on 12/01/16. Which releases are impacted?

Patient E will be included in the December 2017, June 2018, December 2018 and June 2019 cohorts and the graft loss will impact all of the listed releases. Patient E has met the six-month follow-up inclusion criteria because the graft loss was observed less than six months after transplant. Therefore, this event will first appear in the December 2017 release.

Scenario #5: Patient F was transplanted on 03/01/17 and experienced graft loss five months later on 08/01/17. Which releases will be impacted?

Patient F will be included in the following cohorts: June 2018, December 2018 and June 2019. The graft loss will impact all of the listed releases. Patient F has met the six-month follow-up inclusion criteria (graft was lost within six months of transplant) and, thus, the graft loss will be included in the June 2018 release.

Scenario #6: Patient G was transplanted on 07/01/16 and graft loss was observed on 02/01/17 (more than six months but within 365 days of transplant). Which releases are impacted?

Patient G will be included in the December 2017, June 2018, December 2018 and June 2019 cohorts and the graft loss will impact all but the December 2017 cohort. Patient G did not meet the six-month follow-up inclusion criteria because the graft loss was observed more than six months after transplant.

This list outlines a few additional inclusion/exclusion criteria used in adult one-year graft and patient survival analysis.

  • Patients under 18 are excluded.
  • Multi-organ transplants are excluded from both cohort and event calculations. However, multi-organ events now appear on special new multi-organ pages of the SRTR PSRs.  There is no risk adjustment applied to multi-organ reporting.
  • Each transplant/re-transplant is an independent event in graft failure counting, meaning that both the first transplant and the re-transplant(s) are counted in transplant cohort calculations for graft failure.
  • First transplants are followed for death, even if a re-transplant occurs. Death after re-transplant counts as a death if the first transplant occurred within one year of the death. Each patient counts once and only once in death cohorts.
  • Re-transplants are followed for graft failure, but not death. Deaths following re-transplants count as graft failures for the subsequent transplants.
  • Re-transplant counts as graft loss for the first transplant (if the first transplant happened within a prior year).

Scenario #6 – Patient H was re-transplanted on 7/5/2015 and lost his graft on 5/5/16. Which releases are affected?

The patient’s re-transplant will be counted in the December 2017, June 2018 and December 2018 cohorts and the graft loss will be counted for the same three cohorts.

Scenario #7 – Patient J is re-transplanted on 8/5/2015 and dies on 6/5/16. Which releases are affected?

Re-transplant and graft loss will be captured in the December 2017, June 2018 and December 2018 releases. However, the death will not be counted because it occurred after re-transplant, unless the first transplant occurred within one year prior to the death.

Scenario #8 – Patient L was first transplanted on 5/5/15, re-transplanted on 8/5/15 and lost his graft on 6/5/16 (greater than one year after the original transplant). Which releases are affected?

This first transplant will be counted in the December 2017 and June 2018 cohorts and the re-transplant will be counted as a graft loss in both releases.

The re-transplant will be included in the December 2017, June 2018 and December 2018 graft failure, but not death, cohorts. The graft loss (following re-transplant) will impact all three releases.

Scenario #9 – Patient M was first transplanted on 5/5/15, re-transplanted on 8/5/15 and died on 6/5/16 (greater than one year after the original transplant). Which releases are affected?

The first transplant will be counted in the December 2017 and June 2018 graft failure and death cohorts and the re-transplant will be counted as a graft loss for both releases.

The second transplant will be included in the December 2017, June 2018 and December 2018 graft failure, but not death, cohorts and the death will not impact any of the cohorts because the death occurred after the re-transplant and greater than one year after the original transplant. However, the death, which is considered a graft loss, will impact the December 2017, June 2018 and December 2018 releases.

Scenario #10 – Patient N was first transplanted on 5/5/15, re-transplanted on 10/5/2015 and died on 4/5/16 (less than one year after the first transplant). Which releases are affected?

The first transplant will be counted in the December 2017 and June 2018 death and graft failure cohorts. The graft loss at the time of re-transplant (5/5/15) will impact the same two releases. The death will be counted once as an event after the first transplant and will impact the December 2017 and June 2018 releases.

The re-transplant will be included in the December 2017, June 2018 and December 2018 graft failure, but not death, cohorts and the graft failure (as a result of death) will impact the same three releases.

 

 

What do you think?

Join the discussion below.

No Comments

Post A Comment