Pancreaticoduodenectomy in Australia: a national quality improvement clinical registry is long overdue 'Practice makes perfect' is an axiom of life and for those who believe in centralizing low-volume, complex, high-risk operations. During the last 40 years, multiple studies, systemic reviews and meta-analysis have attempted to determine whether centralizing pancreaticoduodenectomy (PD) into high-volume hospitals improves outcomes. The many studies over so long attest to their limitations and conflicted interpretation. 1,2 The centralists quote the many volume-outcome studies showing lower mortality in high-volume hospitals. They will in future quote two studies in this issue of the journal. In one study, Kovoor et al. 3 have extended their earlier review 2 by undertaking a meta-analysis that confirms a lower in-hospital mortality in high-volume hospitals whatever the cut off. 4,5 In the other study, Yau et al. 6 report how a 16% statewide mortality prompted the Western Australia Health Department to centralize PD to three, now two, specialist units. 7 Their 90-day mortality fell from 4.9% to 1.2% with a statistical reduction in major PD-related complications.Australian regionalists will claim that previous, predominately overseas, studies were based on observational, administrative data, rarely risk adjusted and that almost one-third found no volumeoutcome relationship. 2 High volume in some studies overlapped with low volumes in others. They believe volume may be proxy for other factors and that low-volume, specialist units can achieve good outcomes. [8][9][10] The regionalists will also emphasize the imperative to fully utilize Australia's regional hospitals. The centralists know this as their hospitals are already at capacity and facing increasing demands. They have limited ability to take on cases that could be managed in a regional hospital. The adverse consequences of the COVID-19 pandemic on PD centralization are already being reported. 11 Centralizing PD would disadvantage regional patients and their families. Almost half of all PD are still undertaken in low-volume hospitals 12,13 and this may in part reflect patient preference. Centralists believe patients cannot possibly understand data that confound the experts.Since the two WA units were established in their current hospitals, the contemporary statewide 90-day mortality is three (0.6%) in over 500 consecutive PD (pers. comm., 2021). Previous studies from other Australian states suggest an overall mortality between 3% and 4%, [8][9][10][12][13][14][15][16][17][18][19][20] similar to Europe, but greater than the USA. [21][22][23] Almost one-fifth of American patients offered a PD would accept a sixfold increased risk in exchange for locally delivered care. 24 It is not known how far Australian patients would travel for a sixfold reduction in mortality.Other states will object to such comparison noting their different geography and population distribution. They will also argue that