Waiting for liver transplantation in Canada: waitlist history 2000–2004 and sensitivity analysis for the future

Abstract

This study comprises a historical review of liver transplants performed in Canada during [2000][2001][2002][2003][2004], and sensitivity analyses to assess the impact of possible changes in the operation of the waitlists. In the first part, overall statistics are reported, and the notable impact that blood type plays in affecting waiting time is discussed. Waiting times and numbers of transplants are also reported by gender, age, and geographic region (waitlist), and statistical analyses of the patient placement and cadaveric donations processes are performed. These analyses establish that the service times of an appropriate queuing model are closely approximated by an exponential distribution. Consequently, the resulting distribution for the waiting time from placement until transplant is well described by a different exponential distribution. The GI/M/1 queuing model is then used to perform a number of sensitivity analyses. The sensitivity analyses attempt to quantify the impact of no change in policy via a lottery system, and likely increases in cadaveric and/or living donor sources that would be needed to bring stability to the system. The results can be used by relevant authorities, together with information on success rates for publicity campaigns and living donor matching campaigns, to assess where further efforts should be made to reduce waiting times.

A queuing model to address waiting time inconsistency in solid-organ transplantation

“…Several prior studies assessing waiting times for solid-organ transplantation [1][2][3][4][5][6][7][8] have revealed a common and concerning pattern: recipients of blood type O wait on average substantially longer than blood type A, while those of blood type A in turn wait longer than patients with blood type AB. The waiting times for patients with blood type B are sometimes somewhat longer than type O, although this observation is inconsistent [1][2][3][4][5][6][7][8].…”

“…The waiting times for patients with blood type B are sometimes somewhat longer than type O, although this observation is inconsistent [1][2][3][4][5][6][7][8]. This waiting time pattern for solid-organ transplantation arises in many countries and/or health-systems, seemingly regardless of organ type.…”

A model for deceased-donor transplant queue waiting times

“…Given data, the model can then be calibrated by estimating the parameters using a maximum-likelihood-based approach. The suitability of assuming exponential interavailability times was discussed at length in Stanford et al [21], and that result was employed in [22] to show why it is reasonable to make that assumption in the presence of random ABO-compatible transplantation.…”

“…A sample of these studies follows which gives an indication of the variety of contexts in which this problem has been recognized: kidneys in Germany (Glander et al [7] and Liefeldt et al [18]), hearts in the UK (Hussey et al [9]), kidneys in the US (Zenios et al [29]), and livers in the US (Barone et al [2]) and in Canada (Stanford et al [21]). …”

“…Stanford et al [21] combined a statistical analysis of patient placements and deceaseddonor organ availability with a section presenting simple queueing models to assess the consequences of three strategies to try to close a well-documented gap between the demand Recently, Stanford et al [22] used a queueing model to respond to a recent development in addressing the so-called "Blood Type O Problem" (see [7]), in which too many type O organs are cross-transplanted to compatible blood groups, causing notably longer waits for blood type O recipients due to the reduced supply of O organs. Glander et al [7] noted that this frequently has led to situations in which blood type O recipients experience worse health outcomes in terms of statistically-significant higher mortality rates during their longer waiting times.…”