Operator volume is not associated with mortality following percutaneous coronary intervention: insights from the British Cardiovascular Intervention Society registry

Operator volume is not associated with mortality following percutaneous coronary intervention:... Abstract Aims The relationship between operator volume and outcomes for percutaneous coronary intervention (PCI) has been studied in the past, but recent analyses of national data covering the modern radial, acute coronary syndrome-dominant era are limited. Changing in case-mix, practice, and service provision mean that previously described volume–outcome relationships may no longer be relevant, and a reassessment in contemporary practice is needed. We aim to assess whether operator volume is associated with independently reported 30-day mortality in a contemporary PCI cohort. Methods and results This observational cohort study analysed procedures recorded in the British Cardiovascular Intervention Society PCI database from 2013 to 2014 in England and Wales. Mixed effects multiple logistic regression modelling was used to account for operator and centre level effects and to adjust for potential confounders. Volume is defined as the total number of procedures the operator was responsible for in the previous 12 months. A total of 133 970 procedures were analysed. Median volume across all procedures was 178 per year (interquartile range 128–239). The 30-day mortality rate was 2.6%. After adjustment for case-mix, the association between volume and mortality was negligible (odds ratio per 100 procedures 0.99, 95% confidence interval 0.93–1.05; P = 0.725). Sensitivity analyses showed similar results amongst high-risk PCI subsets and in-hospital outcomes. Conclusion There is no evidence that mortality differs by operator volume in the UK. Volume–outcome relationships in PCI should be carefully monitored in response to future changes in practice. Operator volume , PCI , Mortality Introduction The impact of operator volume on outcomes following interventional medical procedures is of great interest across many specialties,1,2 including percutaneous coronary intervention (PCI). The current body of evidence describing this relationship is discrepant, with some studies reporting that higher operator volume is related to improved outcomes via reductions in in-hospital mortality or other adverse events,3–7 and others reporting no such association.8–16 While the exact nature and extent of the volume–outcome relationship in PCI is unclear, there remains an intuitive concern that interventional cardiologists who do more cases are likely to be better operators. Therefore, there is a consensus that operators should exceed a minimum number of procedures per year (ppy) to maintain a high standard of manual skills and sound clinical judgement in order to achieve competency. Current national and international guidelines offer recommendations on minimum operator PCI volume17–19 that are summarized in Table 1. These guidelines rely on data from no later than 2010,16 and reflect the typical activity levels, patient characteristics, and operator practices at that time. In many countries, the case-mix for PCI has changed from predominantly elective to an increased proportion of acute coronary syndromes (ACSs), and this has been accompanied by a growth in PCI services to increase patient access. In addition, much of the interventional equipment, technologies, and pharmacology have evolved and therefore, existing literature regarding the association between PCI volume and outcome may no longer apply to contemporary practice. Two recent studies have examined patient outcomes in relation to volume in contemporary data,7,15 though data from a European perspective with a different healthcare delivery model are lacking. Table 1 Minimum volume recommendations in different regions Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 Table 1 Minimum volume recommendations in different regions Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 There are also several methodological limitations in many volume–outcome studies.20 The existing literature almost exclusively relies on a categorized definition of annualized volume that differs greatly from study to study, for instance the high/low volume threshold has varied from as low as 11 ppy14 or as high as 10011 and synthesizing such results is challenging. The use of fixed calendar year volume is also problematic as, for instance, outcomes of procedures performed in January are modelled based on procedural volume up to 11 months later in December. Future volume should not be used to predict previous outcomes. In this recent (2013–2014) British cohort study, we report crude and risk-adjusted short-term PCI mortality outcomes in relation to operator volume. It is the first time that this relationship has been examined from a national perspective in Europe, and the first to use a pragmatic, non-categorized definition of operator volume. The suitability of existing international volume guidelines is assessed. Methods Study design This retrospective, observational cohort study analysed procedures recorded in the BCIS PCI database in the 2-year period from 1 January 2013 to 31 December 2014 to assess whether operator volume is associated with independently reported 30-day mortality. Mixed effects multiple logistic regression modelling was used to account for operator and centre level effects and to adjust for potential confounders. The BCIS database The BCIS collects data on all PCI procedures in the UK. Data input on every case is mandated by UK Good Practice guidelines and is a specified responsibility of senior operators as part of their revalidation by the General Medical Council (GMC). The data collection is co-ordinated by the National Institute of Cardiovascular Outcomes Research (NICOR) via a centralized electronic database.21 The BCIS-NICOR registry comprises 113 variables, including clinical variables, procedural parameters, and patient outcomes. Mortality tracking is undertaken by NHS Digital linkage with each patients’ NHS number that provides a unique identifier for any person registered with the NHS in England and Wales. Because, it is a legal requirement for all deaths in the UK to be registered, these life status data are robust. Volume definition The GMC registration number of the ‘consultant responsible for the procedure’ was used to identify operators. This is a unique identifier of the consultant PCI operator and has been part of the BCIS registry since 2012. Our annualized volume metric, updated each month, is defined as the total number of procedures the consultant was responsible for in the previous 12 months across all NHS centres. For example, consultant volume in February 2013 is measured from February 2012 to January 2013. Cohort selection All PCI procedures undertaken in the NHS in England and Wales recorded in the BCIS registry from 1 January 2013 to 31 December 2014 were considered in this analysis. Mortality tracking was not available in Northern Ireland or Scotland, hence procedures performed in these countries were excluded, though any activity in these nations contributed to each consultant’s volume should an operator have crossed geographic boundaries during the study period. Similarly, activity in 2012 contributed to each consultant’s volume where necessary. Procedures were excluded where mortality, indication, sex, or age were missing, and where patients were recorded as younger than 18 or over 100 years old. Additional exclusions were made where volume could not be reliably determined: three out of 87 centres had missing data for over 10% of the ‘consultant responsible for the procedure’ field and all procedures from these centres were removed from the analysis (n = 6145 procedures, 4.1%). Missing consultant identifier rates for all remaining centres was less than 3.3% and these procedures were removed (n = 2891, 1.9%). Procedures where the consultant was in their first year in the registry were removed (7053, 4.7%) as volume could not be calculated in these cases. For further details refer to the Supplementary material online. Statistical analysis Baseline patient and procedural characteristics were reported, stratified by volume quartiles (Q1 0–128; Q2 129–178; Q3 179–239; Q4 240–714), such that each strata contained an equal number of procedures (approximate due to ties), with the possibility that the same operator was present in multiple strata if her case volume moved across strata boundaries during the study period. These strata boundaries were reused to show baseline characteristic for elective and ACS procedures, separately. Histograms and quantiles were used to describe the distribution of average volume across operators. To investigate the association between volume and 30-day mortality in the presence of confounding and clustering effects, we used multivariable, mixed effects logistic regression modelling, using multiple imputation to account for missing values. First, missing values were imputed using fully conditional specification multiple imputation with 20 imputed datasets created. Continuous variables were imputed using predictive mean matching and categorical variables using multinomial logistic regression. Imputation was not necessary for consultant and centre identifiers, volume metrics and mortality, as these were complete by design. A complete list of variables used in the imputation model is provided in the Supplementary material online. Information on completeness for each variable is provided in Supplementary material online, Table S1. Effect estimates across imputed datasets were pooled using Rubin’s rules.22 The multivariable mixed effects model adjusted for age, sex, ethnicity, PCI indication, cardiogenic shock, intra-aortic balloon pump (IABP) support, cardiopulmonary support, inotropic support, ventilation, left ventricular ejection fraction, myocardial infarction (MI) history, coronary artery bypass grafting (CABG) history, high cholesterol, hypertension, diabetes status, renal status, smoking status, and centre volume (defined as the number of procedures performed by a centre in the previous 12 months and updated monthly). Covariates were included to adjust the models, not to provide information on their associations with outcomes, which has been more robustly achieved with patient-level analyses in other work.23–25 Operator- and centre-level random effects were included, and these were not nested as operators often worked across multiple centres. The Wald-like test due to Li et al.26 was used to provide reference P-values for model odds ratios (ORs) and associated confidence intervals (CIs). The intraclass correlation coefficient due to Wu et al.27 was used to assess the variation in outcomes explained by operator- and centre-level clustering before and after the inclusion of fixed-effects. These values are reported in Supplementary material online. Smoothed curves which showed the observed (unadjusted) mortality and model-adjusted mortality against volume to explore possible non-linear relationships. The analysis was repeated in the subset of patients undergoing PCI for ACS, with ACS volume derived in an analogous way to overall volume. Both volume metrics were considered simultaneously in the same regression model, so that the model comparison test assesses the improvement in the goodness-of-fit of the model when adding both volume and ACS volume as linear effects. Subsetting was carried out after multiple imputation so that the same imputed data were used throughout. The analysis was also repeated in the subset of patients undergoing primary PCI in exactly the manner described for the ACS sub-analysis, using primary PCI volume. Due to the size of the dataset, we focus on effect estimates and their interpretations and not on P-values.28 Arbitrary significance thresholds are not used. Additional methodological details are provided in Supplementary material online. Sensitivity analyses We varied our methods in four different ways to assess the consistency of our results under alternative study designs. The first of these was to change the outcome from 30-day mortality to centre-reported in-hospital mortality (was the patient discharged alive or not) and also in-hospital major adverse cardiovascular events (MACE, defined here as MI, emergency reintervention via CABG or PCI, and mortality). Second, operator volume was modelled dichotomously rather than continuously, with a threshold of 75 ppy (reflecting the BCIS guidelines recommending a minimum of 150 procedures every 2 years) and also 50 ppy (as per ECS/EAPCI and ACCF/AHA/SCAI guidelines). Third, the interactive effect between operator volume and centre volume was examined. The additional predictive value of an interaction term between dichotomized operator volume (at 75 ppy) and categorized centre volume (0–300, 301–600, 601–1200, 1201+ ppy) was assessed using a Wald test.26 Fourth, operator-modifiable factors (stent type, access site, and adjunct pharmacotherapy) were added as covariates (these were excluded in the primary analysis due to their dependence on decisions made by operators possibly relating to operator experience). A model that did not include IABP support, cardiopulmonary support, or inotropic support was also considered (these variables were included in the primary analysis as they are strong markers of haemodynamic instability and their use may be a matter of necessity rather than choice). Software All analyses were performed using R version 3.2.2.29 The tidy verse data manipulation and visualization suite30 was used throughout. Multiple imputation and model-pooling were implemented using the mice package,31 mixed effects modelling using the lme4 package,32 and restricted cubic splines using the rms package.33 The analysis script is available on request from the first author. Results A total of 158 492 PCI procedures were recorded to the BCIS audit in England and Wales between 2013 and 2014. Following exclusions as described in Supplementary material online, Figure S1, there were 133 970 (84.5%) procedures available for analysis. In total, there were 84 centres and 540 unique consultant GMC number identifiers in this cohort. This equates to an average of 6.4 consultants per centre, 124 procedures per operator per year and 797 procedures per centre per year. The mean age of the study cohort was 65.1 years (standard deviation 12.1), and 74.3% of procedures were in male patients. Elective PCI accounted for 34.6% of procedures. There were 6141 (4.6%) procedures from operators whose volume was under 75 ppy. Figure 1 shows the distribution of the average operator volume during the study. The median operator performed 135 ppy [interquartile range (IQR) 93–188 ppy]. There were 114 operators (21.1%) who had an average volume of less than 75 ppy, and 77 operators (14.3%) who performed less than 50 ppy on average. These operators contributed 4127 (3.1%) and 944 (0.7%) procedures, respectively. Figure 1 View largeDownload slide Distribution of average operator volume. Figure 1 View largeDownload slide Distribution of average operator volume. Variables correlated with volume Table 2 reports patient and procedural factors overall and by volume, stratified by quartiles: Q1 0–128; Q2 129–178; Q3 179–239; Q4 240–714. The 30-day mortality rate was 2.6% though this differed significantly by volume, with mortality decreasing as volume increased, from 2.9% in the lowest volume stratum to 2.5% in the highest. Some factors relating to cardiovascular risk were typically more common when operator volume was higher, for instance previous MI (24.2% lowest volume to 29.6% highest volume), previous CABG (8.9−previous stroke (3.8–4.7%), hypertension (52.3–59.1%), and peripheral vascular disease (3.9–5.8%). However, shock and ventilation were proportionally lower when volume was higher. Radial access was more common in high volume operators (57.2–71.8%), as was left main stem intervention (3.6–7.2%) and multivessel PCI (10.4–17.8%).Tables 3 and 4 reports these factors in elective-only and ACS-only cohorts, respectively. Table 2 Patient and procedural variables by quartile, all procedures Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1-Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 2 Patient and procedural variables by quartile, all procedures Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1-Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 3 Patient and procedural variables by quartile, elective procedures Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 3 Patient and procedural variables by quartile, elective procedures Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 4 Patient and procedural variables by quartile, ACS procedures Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 4 Patient and procedural variables by quartile, ACS procedures Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Multivariable models The relationship between 30-day mortality and volume in 2013–2014 was explored in multivariable mixed effects logistic regression model. A scatter plot of model predicted mortality versus average operator volume is shown in Supplementary material online, Figure S2. Figure 2 presents ORs for operator volume for all, ACS-only, and primary PCI-only cohorts, with these values tabulated in Table 5. The full models are described in Supplementary material online, Tables S2a, S2b, and S2c. After adjustment for case-mix, there was no strong evidence of an independent linear association, with small effect sizes and high P-values. Potential non-linear relationships between volume and 30-day mortality were explored graphically by plotting observed and model-adjusted mortality against volume. Figures 3–5 demonstrate relatively stable model-adjusted mortality as volume varied. Table 5 Operator volume odds ratios for mixed effect logistic regression models 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 ACS, acute coronary syndromes; CI, confidence interval; MACE, major adverse cardiovascular event; OR, odds ratio. Table 5 Operator volume odds ratios for mixed effect logistic regression models 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 ACS, acute coronary syndromes; CI, confidence interval; MACE, major adverse cardiovascular event; OR, odds ratio. Figure 2 View largeDownload slide Operator volume odds ratios. Figure 2 View largeDownload slide Operator volume odds ratios. Figure 3 View largeDownload slide Operator volume vs. mortality, all procedures. Figure 3 View largeDownload slide Operator volume vs. mortality, all procedures. Figure 4 View largeDownload slide Operator volume vs. mortality, ACS procedures. Figure 4 View largeDownload slide Operator volume vs. mortality, ACS procedures. Figure 5 View largeDownload slide Operator volume vs. mortality, primary procedures. Figure 5 View largeDownload slide Operator volume vs. mortality, primary procedures. Sensitivity analyses The findings of the primary analyses were supported in all sensitivity analyses. Table 5 also lists ORs for in-hospital mortality and in-hospital MACE, and for operator volume dichotomized at 75 and 50 ppy. In these models, associations for operator volume were relative small and not statistically significant. The Wald test examining the value of an interaction term suggested no significant interactive effects between operator volume and centre volume (all procedures, P = 0.473; ACS, P = 0.740; primary, P = 0.629). The inclusion in the model of factors that depend on operator choice did not alter the direction or strength of association of volume with mortality (OR 1.02, 95% CI 0.96–1.08; P = 0.533). Similarly, excluding variables relating to cardiac support did not alter results (OR 1.00, 95% CI 0.94–1.06; P = 0.923). Further details of these sensitivity analyses are available in Supplementary material online, Tables S3a, S3b, S3c, S4a, S4b, S4c and S5. We do not adjust for multiple comparisons when reporting these sensitivity analyses. Discussion This nationally representative study is the first to investigate the relationship between operator volume and outcomes within the contemporary PCI era in Europe, where revascularization following ACS is the most common indication for intervention and transradial access is the default for the majority of countries within Europe. It is the only study, where volume is not calculated based on future operator activity and does not rely solely on a contrived dichotomization or categorization of this metric. Our analysis found no relationship between 30-day mortality following PCI and the number of cases performed by a PCI operator. Despite an inverse association between volume and mortality using crude, unadjusted data, we found that after adjusting for patient-level risk factors we can rule out any meaningful change in the odds of 30-day mortality as volume increases. This was observed both when operator volume was modelled continuously, and when volume was dichotomized at 50 or 75 ppy in line with international guidelines. Similar results were obtained in analyses of ACS only and primary PCI only procedures, and of centre-reported in-hospital mortality and in-hospital MACE. Procedures performed by lower volume operators were on average performed to treat patients at higher baseline risk. This is consistent with the finding that there are proportionally more ACS-indicated procedures when volume is low than when volume is high. Conversely, baseline cardiovascular risk typically increased with volume, a pattern that broadly persisted when looking at elective and ACS procedures, separately. The existing literature investigating the relationship of operator volume on outcomes following PCI reports discrepant findings, with some studies reporting increases in adverse events with lower operator volume after risk-adjustment,3–7 and others finding no such association.8–16 Further, significant heterogeneity exists in the statistical methods used, particularly regarding volume threshold definitions and the settings from which the data are derived. Despite this, a large meta-analysis34 pooled aggregate data from 23 studies which included 1 109 103 patients, and found no strong evidence of short-term mortality reductions against annualized dichotomized operator volume (high vs. low volume OR 0.96, 95% CI 0.86–1.08 in all studies; OR 0.90, 95% CI 0.79–1.01 in high and very-high quality studies only), although there was some evidence of a reduction in MACE for high-volume operators (high vs. low volume OR 0.62, 95% CI 0.40–0.97).This meta-analysis was undertaken prior to a study by Badheka et al.6 in 457 498 procedures identified from the National Inpatient Sample (NIS) database between 2005 and 2009 that showed lower in-hospital mortality in higher volume quartiles compared with the lowest quartile (1–15 ppy), with the highest volume quartile (>100 ppy) associated with the most significant reduction (OR 0.65, 95% CI 0.58–0.73; P < 0.001). However, as these studies stem from procedures performed no later than 2010, their relevance in informing contemporary best practice is diminished and should be interpreted with this in mind. In particular, this period has seen a rapid transition from trans-femoral to trans-radial access in the UK, with radial procedures rising from 16% in 2005 to over 75% in 2014.35 The benefits of radial access has been demonstrated in randomized controlled trials,36,37 and as the strength of this benefit is positively associated with operator volume,25 radial uptake may have altered the volume–outcome relationship. Two contemporary studies using data after 2010 can offer some insight here. An analysis of 3 747 866 procedures from July 2009 to March 2015 from the NCDR CathPCI Registry,7 which captures over 90% of PCIs in the USA, found a small increase in in-hospital mortality for each 50-case decrease in annual volume (OR 1.04, 95% CI 1.03–1.05), and this relationship persisted in stable, unstable angina/ST-elevation myocardial infarction (STEMI) and STEMI PCI subgroups. Radial access rates in this study were only 15.2%, and the inclusion of data from as early as 2009 without considering temporal trends or contemporary subgroups, limits the relevance of this study to the UK experience. An analysis of 323 322 procedures in 2014 and 2015 from the Japanese PCI Registry15 reported no significant differences in in-hospital mortality or a composite of peri-procedural complications outcome between operator volume deciles. Despite drawing data from a similar era and comparable radial access rates (61.3% in Japan vs. 67.5% in our study), geo-cultural disparities and average volume levels make translating these findings to Europe difficult. Our study adds a European perspective, reflecting a recent PCI era and showing that increasing volume is not associated with better mortality outcomes, and also that the majority of operators have caseloads exceeding previously defined volume thresholds. Median annualized volume in our study was 135 ppy (IQR 93–188). This is markedly higher than that found in the NIS database in the USA6 (between 33 and 58 ppy), the CathPCI Registry7 (59 ppy, IQR 21–106), and in the J-PCI registry in Japan15 (28 ppy, IQR 10–56). Furthermore, 44% of operators performed <50 PCI ppy in the CathPCI registry, whereas in the current analysis with only 8.9% and 16.9% of operators performing less than 50 and 75 ppy, respectively. Such differences in operator volume distribution across the two healthcare systems may contribute to the differences in the reported findings. The 150 minimum operator volume over 2 years recommended by BCIS and EAPCI17,38 was met or exceeded in 95.4% of analysed procedures, as measured using volume (i.e. exceeding 75 ppy). A sensitivity analysis found no difference in 30-day mortality above or below this threshold after risk adjustment. The ACCF/AHA/SCAI guidelines stipulating a minimum of 50 ppy were not met in just 1.4% of cases, making a well-powered and balanced analysis of this particular recommendation challenging. There are fewer studies of operator volume and outcome for primary PCI. Two studies4,12 investigating this relationship directly found post-adjustment in-hospital mortality reductions for operators performing over 11 primary procedures annually (relative risk high vs. low 0.43, 95% CI 0.21–0.83) or over 10 primary procedures annually (OR 0.66, 95% CI 0.48–0.92). A small (N = 331) single-centre, three-operator study was also conducted39 but with only one operator in each volume ‘group’, it is impossible to consider the effect of volume independent of other operator factors. Finally in the CathPCI study, the relationship between operator volume and in-hospital mortality was significant in patients presenting with STEMI with similar findings in the PPCI cohort (OR 1.03; 95% CI 1.02–1.04). The findings of these studies are not supported by our data, which suggest mortality following primary PCI is not associated with either total volume or primary specific volume, though there was considerable uncertainty around the central estimates. The need to achieve adequate geographical coverage to ensure timely access to primary PCI services, meet minimum centre activity standards, and accommodate a changing population means that future services are likely to continue to see a period of transition, affecting both centre and operator volumes. Volume–related outcome patterns in the primary PCI setting should therefore continue to be closely monitored. Centre volume was not the focus of this study, but it was included as a potential confounder in the regression models. The latest European Heart Journal STEMI guidelines recommend that STEMI patients should be transferred to 24/7 high-volume PCI centres because of reduced mortality irrespective of the chosen treatment strategy.40 A recent study using the BCIS registry from 2007 to 201341 found no change in 30-day mortality as centre volume varied. Limitations Investigations into volume–outcome relationships are hindered by the fact that volume cannot be independently randomized. This restricts the body of statistical evidence to non-randomized volume exposures, where the influence of confounding factors must always be carefully considered and controlled for where possible. Our volume metric measures the number of procedures as the responsible consultant, but it was necessary to exclude cases where consultant identifier was not known or where volume, mortality, or other key variables were unavailable, though this study was still able to retain 84.5% of all available procedures, which compares favourably with other recent registry-based volume studies, for example with 54.5%6 and 73.8%15 in the NIS database and the J-PCI registry, respectively. We note that although consultant identifier is not likely to be missing at random, missing values were present in high and low-volume centres similarly, and the use of random effects mitigates the potential bias induced by these exclusions. Where volume is very low or zero, it is unclear whether this is because the operator was inactive (for example due to maternity leave), was working outside the UK, had not yet been appointed to consultant grade, or whether an uncommon or incorrect operator alias was used. Furthermore, only three operators had volume recorded greater than 500 ppy, with just one exceeding 700 ppy, and so the volume–mortality relationship in this range cannot be adequately disassociated from the relationship of mortality with individual operators. The relationship between volume and mortality in very low volume or very high volume cases should be interpreted with caution. In particular, the potential association of mortality and volume below recommended operator minimums cannot be addressed by our analysis, although we do not observe any signals towards compromised outcomes in operators performing below national/international recommendations. Centres and operators were anonymized, and additional variables at these levels, for example operator age, years since qualification, or total career PCI volume could not be accounted for. We considered in-hospital mortality and MACE as secondary outcomes, although this information cannot be independently validated and is therefore less robust than 30-day mortality. We cannot rule out that other clinical end-points may be independently associated with operator volume. Finally, we were unable to consider whether total career volume may influence the relationships that we report, as operator identifiable information (the GMC number) was only available since 2012. Conclusion In our 2-year study in PCI procedures in England and Wales, we find no direct relationship between 30-day mortality and operator volume after adjusting for patient characteristics. This finding holds when looking at ACS and primary procedures, separately. A vast majority of operators had caseloads exceeding the yearly minimum recommended by BCIS, EAPCI, and ACCF/AHA/SCAI. Supplementary material Supplementary material is available at European Heart Journal online. Conflict of interest: N.C. has received research grants from Boston Scientific, Haemonetics and HeartFLow and speaker/consultancy from Boston, Haemonetics, Heartflow and Abbott. All other authors declared no conflict of interest. References 1 Halm EA , Lee C , Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature . Ann Intern Med 2002 ; 137 : 511 – 520 . Google Scholar CrossRef Search ADS PubMed 2 Morche J , Mathes T , Pieper D. Relationship between surgeon volume and outcomes: a systematic review of systematic reviews . Syst Rev 2016 ; 5 : 204. Google Scholar CrossRef Search ADS PubMed 3 McGrath PD , Wennberg DE , Malenka DJ , Kellett MA , Ryan TJ , O’meara JR , Bradley WA , Hearne MJ , Hettleman B , Robb JF , Shubrooks S , VerLee P , Watkins MW , Lucas FL , O’Connor GT. Operator volume and outcomes in 12,998 percutaneous coronary interventions. Northern New England Cardiovascular Disease Study Group . J Am Coll Cardiol 1998 ; 31 : 570 – 576 . Google Scholar CrossRef Search ADS PubMed 4 Vakili BA , Kaplan R , Brown DL. Volume-outcome relation for physicians and hospitals performing angioplasty for acute myocardial infarction in New York state . Circulation 2001 ; 104 : 2171 – 2176 . Google Scholar CrossRef Search ADS PubMed 5 Hannan EL , Wu C , Walford G , King SB , Holmes DR , Ambrose JA , Sharma S , Katz S , Clark LT , Jones RH. Volume-outcome relationships for percutaneous coronary interventions in the stent era . Circulation 2005 ; 112 : 1171 – 1179 . Google Scholar CrossRef Search ADS PubMed 6 Badheka AO , Patel NJ , Grover P , Singh V , Patel N , Arora S , Chothani A , Mehta K , Deshmukh A , Savani GT , Patel A , Panaich SS , Shah N , Rathod A , Brown M , Mohamad T , Tamburrino FV , Kar S , Makkar R , O’Neill WW , De Marchena E , Schreiber T , Grines CL , Rihal CS , Cohen MG. Impact of annual operator and institutional volume on percutaneous coronary intervention outcomes: a 5-year United States experience (2005-2009) . Circulation 2014 ; 130 : 1392 – 1406 . Google Scholar CrossRef Search ADS PubMed 7 Fanaroff AC , Zakroysky P , Dai D , Wojdyla D , Sherwood MW , Roe MT , Wang TY , Peterson ED , Gurm HS , Cohen MG , Messenger JC , Rao SV. Outcomes of PCI in relation to procedural characteristics and operator volumes in the United States . J Am Coll Cardiol 2017 ; 69 : 2913 – 2924 . Google Scholar CrossRef Search ADS PubMed 8 Xie Y , Rizzo JA , Brown DL. A modified method for estimating volume–outcome relationships: application to percutaneous coronary intervention . J Med Econ 2008 ; 11 : 57 – 70 . Google Scholar CrossRef Search ADS PubMed 9 Harjai KJ , Berman AD , Grines CL , Kahn J , Marsalese D , Mehta RH , Schreiber T , Boura JA , O’Neill WW. Impact of interventionalist volume, experience, and board certification on coronary angioplasty outcomes in the era of stenting . Am J Cardiol 2004 ; 94 : 421 – 426 . Google Scholar CrossRef Search ADS PubMed 10 Shook TL , Sun GW , Burstein S , Eisenhauer AC , Matthews RV. Comparison of percutaneous transluminal coronary angioplasty outcome and hospital costs for low-volume and high-volume operators . Am J Cardiol 1996 ; 77 : 331 – 336 . Google Scholar CrossRef Search ADS PubMed 11 Madan M , Nikhil J , Hellkamp AS , Pieper KS , Labinaz M , Cohen EA , Buller CE , Cantor WJ , Seidelin P , Ducas J , Carere RG , Natarajan MK , Conor O’Shea J , Tcheng JE; for the ESPRIT Investigators . Effect of operator and institutional volume on clinical outcomes after percutaneous coronary interventions performed in Canada and the United States: a brief report from the Enhanced Suppression of the Platelet glycoprotein IIb/IIIa Receptor with Integrilin Therapy (ESPRIT) study . Can J Cardiol 2009 ; 25 : e269 – e272 . Google Scholar CrossRef Search ADS PubMed 12 Srinivas VS , Hailpern SM , Koss E , Monrad ES , Alderman MH. Effect of physician volume on the relationship between hospital volume and mortality during primary angioplasty . J Am Coll Cardiol 2009 ; 53 : 574 – 579 . Google Scholar CrossRef Search ADS PubMed 13 Hannan EL , Racz M , Ryan TJ , McCallister BD , Johnson LW , Arani DT , Guerci AD , Sosa J , Topol EJ. Coronary angioplasty volume-outcome relationships for hospitals and cardiologists . JAMA 1997 ; 277 : 892 – 898 . Google Scholar CrossRef Search ADS PubMed 14 Vakili BA , Brown DL ; 1995 Coronary Angioplasty Reporting System of the New York State Department of Health . Relation of total annual coronary angioplasty volume of physicians and hospitals on outcomes of primary angioplasty for acute myocardial infarction (data from the 1995 Coronary Angioplasty Reporting System of the New York State Department of Health) . Am J Cardiol 2003 ; 91 : 726 – 728 . Google Scholar CrossRef Search ADS PubMed 15 Inohara T , Kohsaka S , Yamaji K , Amano T , Fujii K , Oda H , Uemura S , Kadota K , Miyata H , Nakamura M , Inohara T , Kohsaka S , Yamaji K , Amano T , Fujii K , Oda H , Uemura S , Kadota K , Miyata H , Nakamura M , Kadota K , Shiode N , Tanaka N , Amano T , Uemura S , Akasaka T , Morino Y , Fujii K , Hikichi H , Amano T , Fujii K , Kohsaka S , Ishii H , Tanabe K , Ozaki Y , Sumitsuji S , Iida O , Hara H , Takashima H , Shirai S , Nansato M , Inohara T , Ueda Y , Numasawa Y , Noma S ; J-PCI Registry Investigators . Impact of institutional and operator volume on short-term outcomes of percutaneous coronary intervention: a report from the Japanese Nationwide Registry . JACC Cardiovasc Interv 2017 ; 10 : 918 – 927 . Google Scholar CrossRef Search ADS PubMed 16 Jolly SS , Cairns J , Yusuf S , Niemela K , Steg PG , Worthley M , Ferrari E , Cantor WJ , Fung A , Valettas N , Rokoss M , Olivecrona GK , Widimsky P , Cheema AN , Gao P , Mehta SR. Procedural volume and outcomes with radial or femoral access for coronary angiography and intervention . J Am Coll Cardiol 2014 ; 63 : 954 – 963 . Google Scholar CrossRef Search ADS PubMed 17 Banning AP , Baumbach A , Blackman D , Curzen N , Devadathan S , Fraser D , Ludman P , Norell M , Muir D , Nolan J , Redwood S ; British Cardiovascular Intervention society . Percutaneous coronary intervention in the UK: recommendations for good practice 2015 . Heart 2015 ; 101 : 1 – 13 . Google Scholar CrossRef Search ADS PubMed 18 Harold JG , Bass TA , Bashore TM , Brindiss RG , Brush JE , Burke JA , Dehmers GJ , Deychak YA , Jneids H , Jolliss JG , Landzberg JS , Levine GN , McClurken JB , Messengers JC , Moussas ID , Muhlestein JB , Pomerantz RM , Sanborn TA , Sivaram CA , Whites CJ , Williamss ES , Halperin JL , Beckman JA , Bolger A , Byrne JG , Lester SJ , Merli GJ , Muhlestein JB , Pina IL , Wang A , Weitz H. ACCF/AHA/SCAI 2013 update of the clinical competence statement on coronary artery interventional procedures . Catheter Cardiovasc Interv 2013 ; 82 : E69 – E111 . Google Scholar CrossRef Search ADS PubMed 19 Windecker S , Kolh P , Alfonso F , Collet J-P , Cremer J , Falk V , Filippatos G , Hamm C , Head SJ , Jüni P , Kappetein AP , Kastrati A , Knuuti J , Landmesser U , Laufer G , Neumann F-J , Richter DJ , Schauerte P , Sousa Uva M , Stefanini GG , Taggart DP , Torracca L , Valgimigli M , Wijns W , Witkowski A. 2014 ESC/EACTS Guidelines on myocardial revascularization . EuroIntervention 2015 ; 10 : 1024 – 1094 . Google Scholar CrossRef Search ADS PubMed 20 Rashid M , Sperrin M , Ludman PF , Neill DO , Nicholas O , Belder MAD , Mamas MA. Impact of operator volume for percutaneous coronary intervention on clinical outcomes: what do the numbers say? Eur Hear J Qual Care Clin Outcomes 2016 ; 2 : 16 – 22 . Google Scholar CrossRef Search ADS 21 Ludman PF. British Cardiovascular Intervention Society Registry for audit and quality assessment of percutaneous coronary interventions in the United Kingdom . Heart 2011 ; 97 : 1293 – 1297 . Google Scholar CrossRef Search ADS PubMed 22 Rubin DB. Multiple Imputation for Nonresponse in Surveys . Hoboken, New Jersey : John Wiley & Sons, Inc. ; 2004 . 23 Hulme W , Sperrin M , Kontopantelis E , Ratib K , Ludman P , Sirker A , Kinnaird T , Curzen N , Kwok CS , Belder MD , Nolan J , Mamas MA. Increased radial access is not associated with worse femoral outcomes for percutaneous coronary intervention in the United Kingdom . Circ Cardiovasc Interv 2017 ; 10 : e004279. Google Scholar CrossRef Search ADS PubMed 24 Mamas MA , Nolan J , Belder MA. D , Zaman A , Kinnaird T , Curzen N , Kwok CS , Buchan I , Ludman P , Kontopantelis E ; British Cardiovascular Intervention Society (BCIS) and the National Institute for Clinical Outcomes Research (NICOR) . Changes in arterial access site and association with mortality in the United Kingdom . Circulation 2016 ; 133 : 1655 – 1667 . Google Scholar CrossRef Search ADS PubMed 25 Hulme W , Sperrin M , Rushton H , Ludman PF , Belder M , Curzen N , Kinnaird T , Kwok CS , Buchan I , Nolan J , Mamas MA. Is there a relationship of operator and center volume with access site-related outcomes? An analysis from the British Cardiovascular Intervention Society Circ Cardiovasc Interv 2016 ; 9 : e003333. Google Scholar CrossRef Search ADS PubMed 26 Li KH , Raghunathan TE , Rubin DB. Large-sample significance levels from multiply imputed data using moment-based statistics and an f reference distribution . J Am Stat Assoc 1991 ; 86 : 1065 . 27 Wu S , Crespi CM , Wong WK. Comparison of methods for estimating the intraclass correlation coefficient for binary responses in cancer prevention cluster randomized trials . Contemp Clin Trials 2012 ; 33 : 869 – 880 . Google Scholar CrossRef Search ADS PubMed 28 Lin M , Lucas HC , Shmueli G. Research commentary—too big to fail: large samples and the p-value problem . Inf Syst Res 2013 ; 24 : 906 – 917 . Google Scholar CrossRef Search ADS 29 Team RC . R: A Language and Environment for Statistical Computing . Vienna, Austria : R Foundation for Statistical Computing ; 2015 . 30 Wickham H. Easily Install and Load ‘Tidyverse’ Packages [R package tidyverse version 1.1.1]. Comprehensive R Archive Network (CRAN). R Studio. 31 Buuren S. V , Groothuis-Oudshoorn K. mice: multivariate imputation by chained equations in R . J Stat Softw 2011 ; 45 : 1 – 67 . Google Scholar CrossRef Search ADS 32 Bates D , Mächler M , Bolker B , Walker S. Fitting linear mixed-effects models using lme4 . J Stat Softw 2015 ; 67 : 48. Google Scholar CrossRef Search ADS 33 Harrell FE. Regression Modeling Strategies . Cham : Springer International Publishing ; 2015 . Google Scholar CrossRef Search ADS 34 Strom JB , Wimmer NJ , Wasfy JH , Kennedy K , Yeh RW. Association between operator procedure volume and patient outcomes in percutaneous coronary intervention: a systematic review and meta-analysis . Circ Cardiovasc Qual Outcomes 2014 ; 7 : 560 – 566 . Google Scholar CrossRef Search ADS PubMed 35 Ludman P. BCIS Audit Report for 2014 Activity. 2015 . https://www.bcis.org.uk/wp-content/uploads/2017/01/BCIS-audit-2014.pdf (5 March 2018). 36 Andò G , Capodanno D. Radial versus femoral access in invasively managed patients with acute coronary syndrome . Ann Intern Med 2015 ; 163 : 932. Google Scholar CrossRef Search ADS PubMed 37 Ruiz-Rodriguez E , Asfour A , Lolay G , Ziada KM , Abdel-Latif AK. Systematic review and meta-analysis of major cardiovascular outcomes for radial versus femoral access in patients with acute coronary syndrome . South Med J 2016 ; 109 : 61 – 76 . Google Scholar CrossRef Search ADS PubMed 38 Authors/Task Force members , Windecker S , Kolh P , Alfonso F , Collet J-P , Cremer J , Falk V , Filippatos G , Hamm C , Head SJ , Jüni P , Kappetein AP , Kastrati A , Knuuti J , Landmesser U , Laufer G , Neumann F-J , Richter DJ , Schauerte P , Sousa Uva M , Stefanini GG , Taggart DP , Torracca L , Valgimigli M , Wijns W , Witkowski A. 2014 ESC/EACTS Guidelines on myocardial revascularization . Eur Heart J 2014 ; 35 : 2541 – 2619 . Google Scholar CrossRef Search ADS PubMed 39 Politi A , Galli M , Zerboni S , Michi R , Marco FD , Llambro M , Ferrari G. Operator volume and outcomes of primary angioplasty for acute myocardial infarction in a single high-volume centre . J Cardiovasc Med 2006 ; 7 : 761 – 767 . Google Scholar CrossRef Search ADS 40 Ibanez B , James S , Agewall S , Antunes MJ , Bucciarelli-Ducci C , Bueno H , Caforio ALP , Creas F , Goudevenos JA , Halvorsen S , Hindriks G , Kastrati A , Lenzen MJ , Prescott E , Roffi M , Valgimigli M , Varenhorst C , Vranckx P , Widimsky P ; ESC Scientific Document Group . 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force for the management of acute myocardial infarction patients presenting with ST-segment elevation of the European Society of Cardiology (ESC) . Eur Heart J 2018 ; 39 : 119 – 177 . Google Scholar CrossRef Search ADS PubMed 41 O’Neill D , Nicholas O , Gale CP , Ludman P , Belder MA. D , Timmis A , Fox KA , Simpson IA , Redwood S , Ray SG. Total center percutaneous coronary intervention volume and 30-day mortality . Circ Cardiovasc Qual Outcomes 2017 ; 10 : e003186 . Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Heart Journal Oxford University Press

Operator volume is not associated with mortality following percutaneous coronary intervention: insights from the British Cardiovascular Intervention Society registry

Loading next page...
 
/lp/ou_press/operator-volume-is-not-associated-with-mortality-following-zq3OkSBE0R
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com.
ISSN
0195-668X
eISSN
1522-9645
D.O.I.
10.1093/eurheartj/ehy125
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims The relationship between operator volume and outcomes for percutaneous coronary intervention (PCI) has been studied in the past, but recent analyses of national data covering the modern radial, acute coronary syndrome-dominant era are limited. Changing in case-mix, practice, and service provision mean that previously described volume–outcome relationships may no longer be relevant, and a reassessment in contemporary practice is needed. We aim to assess whether operator volume is associated with independently reported 30-day mortality in a contemporary PCI cohort. Methods and results This observational cohort study analysed procedures recorded in the British Cardiovascular Intervention Society PCI database from 2013 to 2014 in England and Wales. Mixed effects multiple logistic regression modelling was used to account for operator and centre level effects and to adjust for potential confounders. Volume is defined as the total number of procedures the operator was responsible for in the previous 12 months. A total of 133 970 procedures were analysed. Median volume across all procedures was 178 per year (interquartile range 128–239). The 30-day mortality rate was 2.6%. After adjustment for case-mix, the association between volume and mortality was negligible (odds ratio per 100 procedures 0.99, 95% confidence interval 0.93–1.05; P = 0.725). Sensitivity analyses showed similar results amongst high-risk PCI subsets and in-hospital outcomes. Conclusion There is no evidence that mortality differs by operator volume in the UK. Volume–outcome relationships in PCI should be carefully monitored in response to future changes in practice. Operator volume , PCI , Mortality Introduction The impact of operator volume on outcomes following interventional medical procedures is of great interest across many specialties,1,2 including percutaneous coronary intervention (PCI). The current body of evidence describing this relationship is discrepant, with some studies reporting that higher operator volume is related to improved outcomes via reductions in in-hospital mortality or other adverse events,3–7 and others reporting no such association.8–16 While the exact nature and extent of the volume–outcome relationship in PCI is unclear, there remains an intuitive concern that interventional cardiologists who do more cases are likely to be better operators. Therefore, there is a consensus that operators should exceed a minimum number of procedures per year (ppy) to maintain a high standard of manual skills and sound clinical judgement in order to achieve competency. Current national and international guidelines offer recommendations on minimum operator PCI volume17–19 that are summarized in Table 1. These guidelines rely on data from no later than 2010,16 and reflect the typical activity levels, patient characteristics, and operator practices at that time. In many countries, the case-mix for PCI has changed from predominantly elective to an increased proportion of acute coronary syndromes (ACSs), and this has been accompanied by a growth in PCI services to increase patient access. In addition, much of the interventional equipment, technologies, and pharmacology have evolved and therefore, existing literature regarding the association between PCI volume and outcome may no longer apply to contemporary practice. Two recent studies have examined patient outcomes in relation to volume in contemporary data,7,15 though data from a European perspective with a different healthcare delivery model are lacking. Table 1 Minimum volume recommendations in different regions Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 Table 1 Minimum volume recommendations in different regions Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 Guidelines Coverage Recommended minimum operator volume, annualized Publication year The European Society of Cardiology (ESC)/European Association of PCI (EAPCI) Europe 75 2014 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) USA 50 2013 British Cardiovascular Intervention Society (BCIS) UK 75, averaged across 2 years 2015 There are also several methodological limitations in many volume–outcome studies.20 The existing literature almost exclusively relies on a categorized definition of annualized volume that differs greatly from study to study, for instance the high/low volume threshold has varied from as low as 11 ppy14 or as high as 10011 and synthesizing such results is challenging. The use of fixed calendar year volume is also problematic as, for instance, outcomes of procedures performed in January are modelled based on procedural volume up to 11 months later in December. Future volume should not be used to predict previous outcomes. In this recent (2013–2014) British cohort study, we report crude and risk-adjusted short-term PCI mortality outcomes in relation to operator volume. It is the first time that this relationship has been examined from a national perspective in Europe, and the first to use a pragmatic, non-categorized definition of operator volume. The suitability of existing international volume guidelines is assessed. Methods Study design This retrospective, observational cohort study analysed procedures recorded in the BCIS PCI database in the 2-year period from 1 January 2013 to 31 December 2014 to assess whether operator volume is associated with independently reported 30-day mortality. Mixed effects multiple logistic regression modelling was used to account for operator and centre level effects and to adjust for potential confounders. The BCIS database The BCIS collects data on all PCI procedures in the UK. Data input on every case is mandated by UK Good Practice guidelines and is a specified responsibility of senior operators as part of their revalidation by the General Medical Council (GMC). The data collection is co-ordinated by the National Institute of Cardiovascular Outcomes Research (NICOR) via a centralized electronic database.21 The BCIS-NICOR registry comprises 113 variables, including clinical variables, procedural parameters, and patient outcomes. Mortality tracking is undertaken by NHS Digital linkage with each patients’ NHS number that provides a unique identifier for any person registered with the NHS in England and Wales. Because, it is a legal requirement for all deaths in the UK to be registered, these life status data are robust. Volume definition The GMC registration number of the ‘consultant responsible for the procedure’ was used to identify operators. This is a unique identifier of the consultant PCI operator and has been part of the BCIS registry since 2012. Our annualized volume metric, updated each month, is defined as the total number of procedures the consultant was responsible for in the previous 12 months across all NHS centres. For example, consultant volume in February 2013 is measured from February 2012 to January 2013. Cohort selection All PCI procedures undertaken in the NHS in England and Wales recorded in the BCIS registry from 1 January 2013 to 31 December 2014 were considered in this analysis. Mortality tracking was not available in Northern Ireland or Scotland, hence procedures performed in these countries were excluded, though any activity in these nations contributed to each consultant’s volume should an operator have crossed geographic boundaries during the study period. Similarly, activity in 2012 contributed to each consultant’s volume where necessary. Procedures were excluded where mortality, indication, sex, or age were missing, and where patients were recorded as younger than 18 or over 100 years old. Additional exclusions were made where volume could not be reliably determined: three out of 87 centres had missing data for over 10% of the ‘consultant responsible for the procedure’ field and all procedures from these centres were removed from the analysis (n = 6145 procedures, 4.1%). Missing consultant identifier rates for all remaining centres was less than 3.3% and these procedures were removed (n = 2891, 1.9%). Procedures where the consultant was in their first year in the registry were removed (7053, 4.7%) as volume could not be calculated in these cases. For further details refer to the Supplementary material online. Statistical analysis Baseline patient and procedural characteristics were reported, stratified by volume quartiles (Q1 0–128; Q2 129–178; Q3 179–239; Q4 240–714), such that each strata contained an equal number of procedures (approximate due to ties), with the possibility that the same operator was present in multiple strata if her case volume moved across strata boundaries during the study period. These strata boundaries were reused to show baseline characteristic for elective and ACS procedures, separately. Histograms and quantiles were used to describe the distribution of average volume across operators. To investigate the association between volume and 30-day mortality in the presence of confounding and clustering effects, we used multivariable, mixed effects logistic regression modelling, using multiple imputation to account for missing values. First, missing values were imputed using fully conditional specification multiple imputation with 20 imputed datasets created. Continuous variables were imputed using predictive mean matching and categorical variables using multinomial logistic regression. Imputation was not necessary for consultant and centre identifiers, volume metrics and mortality, as these were complete by design. A complete list of variables used in the imputation model is provided in the Supplementary material online. Information on completeness for each variable is provided in Supplementary material online, Table S1. Effect estimates across imputed datasets were pooled using Rubin’s rules.22 The multivariable mixed effects model adjusted for age, sex, ethnicity, PCI indication, cardiogenic shock, intra-aortic balloon pump (IABP) support, cardiopulmonary support, inotropic support, ventilation, left ventricular ejection fraction, myocardial infarction (MI) history, coronary artery bypass grafting (CABG) history, high cholesterol, hypertension, diabetes status, renal status, smoking status, and centre volume (defined as the number of procedures performed by a centre in the previous 12 months and updated monthly). Covariates were included to adjust the models, not to provide information on their associations with outcomes, which has been more robustly achieved with patient-level analyses in other work.23–25 Operator- and centre-level random effects were included, and these were not nested as operators often worked across multiple centres. The Wald-like test due to Li et al.26 was used to provide reference P-values for model odds ratios (ORs) and associated confidence intervals (CIs). The intraclass correlation coefficient due to Wu et al.27 was used to assess the variation in outcomes explained by operator- and centre-level clustering before and after the inclusion of fixed-effects. These values are reported in Supplementary material online. Smoothed curves which showed the observed (unadjusted) mortality and model-adjusted mortality against volume to explore possible non-linear relationships. The analysis was repeated in the subset of patients undergoing PCI for ACS, with ACS volume derived in an analogous way to overall volume. Both volume metrics were considered simultaneously in the same regression model, so that the model comparison test assesses the improvement in the goodness-of-fit of the model when adding both volume and ACS volume as linear effects. Subsetting was carried out after multiple imputation so that the same imputed data were used throughout. The analysis was also repeated in the subset of patients undergoing primary PCI in exactly the manner described for the ACS sub-analysis, using primary PCI volume. Due to the size of the dataset, we focus on effect estimates and their interpretations and not on P-values.28 Arbitrary significance thresholds are not used. Additional methodological details are provided in Supplementary material online. Sensitivity analyses We varied our methods in four different ways to assess the consistency of our results under alternative study designs. The first of these was to change the outcome from 30-day mortality to centre-reported in-hospital mortality (was the patient discharged alive or not) and also in-hospital major adverse cardiovascular events (MACE, defined here as MI, emergency reintervention via CABG or PCI, and mortality). Second, operator volume was modelled dichotomously rather than continuously, with a threshold of 75 ppy (reflecting the BCIS guidelines recommending a minimum of 150 procedures every 2 years) and also 50 ppy (as per ECS/EAPCI and ACCF/AHA/SCAI guidelines). Third, the interactive effect between operator volume and centre volume was examined. The additional predictive value of an interaction term between dichotomized operator volume (at 75 ppy) and categorized centre volume (0–300, 301–600, 601–1200, 1201+ ppy) was assessed using a Wald test.26 Fourth, operator-modifiable factors (stent type, access site, and adjunct pharmacotherapy) were added as covariates (these were excluded in the primary analysis due to their dependence on decisions made by operators possibly relating to operator experience). A model that did not include IABP support, cardiopulmonary support, or inotropic support was also considered (these variables were included in the primary analysis as they are strong markers of haemodynamic instability and their use may be a matter of necessity rather than choice). Software All analyses were performed using R version 3.2.2.29 The tidy verse data manipulation and visualization suite30 was used throughout. Multiple imputation and model-pooling were implemented using the mice package,31 mixed effects modelling using the lme4 package,32 and restricted cubic splines using the rms package.33 The analysis script is available on request from the first author. Results A total of 158 492 PCI procedures were recorded to the BCIS audit in England and Wales between 2013 and 2014. Following exclusions as described in Supplementary material online, Figure S1, there were 133 970 (84.5%) procedures available for analysis. In total, there were 84 centres and 540 unique consultant GMC number identifiers in this cohort. This equates to an average of 6.4 consultants per centre, 124 procedures per operator per year and 797 procedures per centre per year. The mean age of the study cohort was 65.1 years (standard deviation 12.1), and 74.3% of procedures were in male patients. Elective PCI accounted for 34.6% of procedures. There were 6141 (4.6%) procedures from operators whose volume was under 75 ppy. Figure 1 shows the distribution of the average operator volume during the study. The median operator performed 135 ppy [interquartile range (IQR) 93–188 ppy]. There were 114 operators (21.1%) who had an average volume of less than 75 ppy, and 77 operators (14.3%) who performed less than 50 ppy on average. These operators contributed 4127 (3.1%) and 944 (0.7%) procedures, respectively. Figure 1 View largeDownload slide Distribution of average operator volume. Figure 1 View largeDownload slide Distribution of average operator volume. Variables correlated with volume Table 2 reports patient and procedural factors overall and by volume, stratified by quartiles: Q1 0–128; Q2 129–178; Q3 179–239; Q4 240–714. The 30-day mortality rate was 2.6% though this differed significantly by volume, with mortality decreasing as volume increased, from 2.9% in the lowest volume stratum to 2.5% in the highest. Some factors relating to cardiovascular risk were typically more common when operator volume was higher, for instance previous MI (24.2% lowest volume to 29.6% highest volume), previous CABG (8.9−previous stroke (3.8–4.7%), hypertension (52.3–59.1%), and peripheral vascular disease (3.9–5.8%). However, shock and ventilation were proportionally lower when volume was higher. Radial access was more common in high volume operators (57.2–71.8%), as was left main stem intervention (3.6–7.2%) and multivessel PCI (10.4–17.8%).Tables 3 and 4 reports these factors in elective-only and ACS-only cohorts, respectively. Table 2 Patient and procedural variables by quartile, all procedures Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1-Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 2 Patient and procedural variables by quartile, all procedures Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) Variables % non missing All procedures N = 133 970 Operator volume quartile: N (%) Q1: 0–128 (N = 33 715) Q2: 129–178 (N = 33 712) Q3: 179-239 (N = 33 097) Q4: 240–714 (N = 33 446) Age  <50 100 14 735 (11.0) 3923 (11.6) 3853 (11.4) 3566 (10.8) 3393 (10.1)  50–59 29 580 (22.1) 7660 (22.7) 7563 (22.4) 7252 (21.9) 7105 (21.2)  60–69 39 092 (29.2) 9783 (29.0) 9836 (29.2) 9772 (29.5) 9701 (29.0)  70–79 33 418 (24.9) 8173 (24.2) 8252 (24.5) 8219 (24.8) 8774 (26.2)  80+ 17 145 (12.8) 4176 (12.4) 4208 (12.5) 4288 (13.0) 4473 (13.4) Sex  Male 100 99 548 (74.3) 24 967 (74.1) 25 159 (74.6) 24 513 (74.1) 24 909 (74.5)  Female 34 422 (25.7) 8748 (25.9) 8553 (25.4) 8584 (25.9) 8537 (25.5) Ethnicity  Asian 79.9 7587 (7.1) 2417 (9.1) 1971 (7.5) 1750 (6.3) 1449 (5.5)  Black 835 (0.8) 253 (1.0) 234 (0.9) 181 (0.7) 167 (0.6)  White 90 476 (84.5) 20 795 (78.6) 21 687 (82.0) 24 460 (87.9) 23 534 (89.6)  Other 8113 (7.6) 2998 (11.3) 2549 (9.6) 1438 (5.2) 1128 (4.3) Smoking  Never 91.1 45 631 (37.4) 12 053 (39.5) 11 421 (37.3) 11 136 (37.0) 11 021 (35.7)  Former 47 446 (38.9) 11 065 (36.3) 11 686 (38.1) 11 702 (38.9) 12 993 (42.1)  Current 29 009 (23.8) 7403 (24.3) 7547 (24.6) 7224 (24.0) 6835 (22.2) Dialysis 93.5 1275 (1.0) 336 (1.1) 348 (1.1) 269 (0.9) 322 (1.0) Diabetes 96.1 27 333 (21.2) 6815 (20.9) 6874 (21.1) 6714 (21.2) 6930 (21.7) PCI 98.0 32 454 (24.7) 7395 (22.5) 7738 (23.5) 7969 (24.6) 9352 (28.3) CABG 75.8 9500 (9.4) 2451 (8.9) 2146 (8.8) 2209 (9.6) 2694 (10.0) MI 94.3 33 548 (26.6) 7635 (24.2) 8108 (25.9) 8362 (26.5) 9443 (29.6) CVA 95.1 5386 (4.2) 1211 (3.8) 1320 (4.1) 1367 (4.3) 1488 (4.7) HC 95.1 67 412 (52.9) 15 436 (47.9) 17 316 (53.7) 17 208 (54.5) 17 452 (55.6) Hypertension 95.1 71 072 (55.8) 16 844 (52.3) 17 997 (55.8) 17 661 (56.0) 18 570 (59.1) PVD 95.1 6340 (5.0) 1268 (3.9) 1596 (4.9) 1644 (5.2) 1832 (5.8) VHD 95.1 2216 (1.7) 426 (1.3) 546 (1.7) 635 (2.0) 609 (1.9) LVEF  Good (>50%) 48.6 45 639 (70.1) 10 456 (70.6) 11 063 (69.5) 11 545 (71.0) 12 575 (69.3)  Fair (30–50%) 15 342 (23.6) 3439 (23.2) 3923 (24.7) 3578 (22.0) 4402 (24.3)  Poor (<30%) 4128 (6.3) 905 (6.1) 924 (5.8) 1131 (7.0) 1168 (6.4) Indication  Elective 100 46 373 (34.6) 11 206 (33.2) 11 414 (33.9) 11 328 (34.2) 12 425 (37.1)  UA/NSTEMI 50 562 (37.7) 12 641 (37.5) 12 836 (38.1) 12 484 (37.7) 12 601 (37.7)  STEMI 37 035 (27.6) 9868 (29.3) 9462 (28.1) 9285 (28.1) 8420 (25.2) Shock 99.3 3804 (2.9) 1074 (3.2) 1005 (3.0) 922 (2.8) 803 (2.4) Ventilated 89.4 2368 (2.0) 703 (2.3) 601 (2.0) 570 (1.9) 494 (1.7) IABP 97.0 1848 (1.4) 473 (1.4) 456 (1.4) 475 (1.5) 444 (1.3) CP support 97.0 139 (0.1) 41 (0.1) 39 (0.1) 30 (0.1) 29 (0.1) Inotropes 97.0 1311 (1.0) 353 (1.1) 308 (1.0) 329 (1.0) 321 (1.0) Access  Femoral 98.9 38 302 (28.9) 13 019 (39.5) 9852 (29.6) 7125 (21.7) 8306 (24.9)  Radial 89 449 (67.5) 18 876 (57.2) 22 309 (67.0) 24 322 (74.0) 23 942 (71.8)  Multiple 4379 (3.3) 1020 (3.1) 1070 (3.2) 1307 (4.0) 982 (2.9)  Other 346 (0.3) 69 (0.2) 61 (0.2) 123 (0.4) 93 (0.3) Left-main 100 6419 (4.8) 1214 (3.6) 1269 (3.8) 1533 (4.6) 2403 (7.2) Multivessel 100 17 899 (13.4) 3520 (10.4) 4168 (12.4) 4257 (12.9) 5954 (17.8) CTO 92.4 9091 (7.3) 1941 (6.5) 1977 (6.6) 2252 (7.2) 2921 (8.9) Invasive imaging 94.8 9143 (7.2) 1422 (4.4) 2027 (6.4) 2274 (7.3) 3420 (10.8) Pressure wire 94.8 10 102 (8.0) 2290 (7.0) 2508 (8.0) 2397 (7.7) 2907 (9.2) Stent type  No stents 96.9 10 069 (7.8) 2142 (6.6) 2314 (7.2) 2569 (8.0) 3044 (9.2)  BMS only 10 299 (7.9) 2984 (9.2) 2946 (9.2) 2574 (8.0) 1795 (5.4)  DES only 106 667 (82.2) 26 670 (82.3) 26 024 (80.9) 26 300 (81.8) 27 673 (83.8)  BMS and DES 2729 (2.1) 621 (1.9) 883 (2.7) 710 (2.2) 515 (1.6) Anti-platelets  Clopidogrel 80.9 79 157 (73.1) 19 821 (74.9) 19 697 (72.4) 19 856 (69.4) 19 783 (75.9)  Prasugrel 7236 (6.7) 1684 (6.4) 2133 (7.8) 2111 (7.4) 1308 (5.0)  Ticagrelor 21 646 (20.0) 4887 (18.5) 5281 (19.4) 6568 (23.0) 4910 (18.8)  Ticlopidine 294 (0.3) 82 (0.3) 78 (0.3) 74 (0.3) 60 (0.2) Warfarin 94.8 1554 (1.2) 313 (1.0) 351 (1.1) 436 (1.4) 454 (1.4) Bivalirudin 94.8 6878 (5.4) 1572 (4.9) 1768 (5.6) 2161 (6.8) 1377 (4.4) GP IIb/IIIa 96.3 22 345 (17.3) 6743 (20.6) 5921 (18.4) 5140 (16.3) 4541 (13.9) 30-day death 100 3538 (2.6) 978 (2.9) 854 (2.5) 869 (2.6) 837 (2.5) In-hospital death 98.9 2086 (1.6) 553 (1.7) 494 (1.5) 493 (1.5) 546 (1.6) In-hospital MACE 98.9 2984 (2.3) 816 (2.4) 694 (2.1) 713 (2.2) 761 (2.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1-Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 3 Patient and procedural variables by quartile, elective procedures Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 3 Patient and procedural variables by quartile, elective procedures Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) Variables % non missing Elective procedures (N = 46 373) Operator volume quartile: N (%) Q1: 0–128 (N = 11 206) Q2: 129–178 (N = 11 414) Q3: 179–239 (N = 11 328) Q4: 240–714 (N = 12 425) Age  <50 100 3581 (7.7) 936 (8.4) 932 (8.2) 817 (7.2) 896 (7.2)  50–59 9740 (21.0) 2445 (21.8) 2438 (21.4) 2347 (20.7) 2510 (20.2)  60–69 15 202 (32.8) 3678 (32.8) 3745 (32.8) 3760 (33.2) 4019 (32.3)  70–79 13 029 (28.1) 3108 (27.7) 3107 (27.2) 3182 (28.1) 3632 (29.2)  80+ 4821 (10.4) 1039 (9.3) 1192 (10.4) 1222 (10.8) 1368 (11.0) Sex  Male 100 35 546 (76.7) 8472 (75.6) 8748 (76.6) 8706 (76.9) 9620 (77.4)  Female 10 827 (23.3) 2734 (24.4) 2666 (23.4) 2622 (23.1) 2805 (22.6) Ethnicity  Asian 78.5 2676 (7.3) 780 (9.0) 688 (7.7) 605 (6.4) 603 (6.4)  Black 251 (0.7) 61 (0.7) 77 (0.9) 59 (0.6) 54 (0.6)  White 29 978 (82.3) 6478 (74.9) 6971 (78.2) 8179 (86.9) 8350 (88.4)  Other 3515 (9.7) 1330 (15.4) 1177 (13.2) 573 (6.1) 435 (4.6) Smoking  Never 90.3 17 026 (40.7) 4324 (42.7) 4136 (40.3) 4140 (40.7) 4426 (39.2)  Former 19 864 (47.4) 4474 (44.2) 4790 (46.6) 4906 (48.2) 5694 (50.4)  Current 4990 (11.9) 1318 (13.0) 1349 (13.1) 1138 (11.2) 1185 (10.5) Dialysis 95.3 469 (1.1) 116 (1.1) 133 (1.2) 104 (1.0) 116 (1.0) Diabetes 96.2 10 518 (23.6) 2501 (22.9) 2580 (23.3) 2578 (23.9) 2859 (24.1) PCI 98.2 18 028 (39.6) 3788 (34.5) 4187 (37.6) 4468 (40.2) 5585 (45.4) CABG 77.5 4496 (12.5) 1039 (11.2) 976 (11.4) 1134 (13.9) 1347 (13.4) MI 93.9 15 353 (34.9) 3217 (30.1) 3538 (33.1) 3892 (35.9) 4706 (39.7) CVA 95.0 1632 (3.7) 346 (3.2) 388 (3.6) 422 (3.9) 476 (4.1) HC 95.0 28 088 (63.7) 6430 (59.7) 7373 (67.5) 7115 (66.0) 7170 (61.8) Hypertension 95.0 28 170 (63.9) 6509 (60.4) 7064 (64.7) 6985 (64.8) 7612 (65.6) PVD 95.0 2210 (5.0) 405 (3.8) 582 (5.3) 640 (5.9) 583 (5.0) VHD 95.0 1032 (2.3) 179 (1.7) 271 (2.5) 305 (2.8) 277 (2.4) LVEF  Good (>50%) 60.4 22 664 (80.9) 4814 (82.8) 5273 (79.6) 6053 (81.8) 6524 (79.7)  Fair (30–50%) 4402 (15.7) 805 (13.8) 1134 (17.1) 1085 (14.7) 1378 (16.8)  Poor (<30%) 957 (3.4) 194 (3.3) 215 (3.2) 264 (3.6) 284 (3.5) Ventilated 86.4 145 (0.4) 45 (0.5) 39 (0.4) 28 (0.3) 33 (0.3) IABP 97.3 106 (0.2) 21 (0.2) 29 (0.3) 19 (0.2) 37 (0.3) CP support 97.3 5 (0.0) 1 (0.0) 2 (0.0) 1 (0.0) 1 (0.0) Inotropes 97.3 39 (0.1) 11 (0.1) 10 (0.1) 8 (0.1) 10 (0.1) Access  Femoral 99.1 14 816 (32.3) 4501 (40.9) 3772 (33.4) 2900 (25.7) 3643 (29.4)  Radial 29 149 (63.5) 6139 (55.8) 7101 (62.9) 7691 (68.3) 8218 (66.3)  Multiple 1802 (3.9) 327 (3.0) 392 (3.5) 598 (5.3) 485 (3.9)  Other 167 (0.4) 26 (0.2) 26 (0.2) 75 (0.7) 40 (0.3) Left-main 100 2470 (5.3) 398 (3.6) 461 (4.0) 621 (5.5) 990 (8.0) Multivessel 100 7342 (15.8) 1350 (12.0) 1701 (14.9) 1699 (15.0) 2592 (20.9) CTO 92.7 5851 (13.6) 1069 (10.8) 1229 (12.1) 1559 (14.4) 1994 (16.4) Invasive imaging 96.1 4151 (9.3) 566 (5.2) 903 (8.3) 1007 (9.3) 1675 (14.1) Pressure wire 96.1 6931 (15.5) 1521 (13.9) 1724 (15.9) 1646 (15.1) 2040 (17.1) Stent type  No stents 96.5 4036 (9.0) 776 (7.3) 890 (8.2) 974 (8.9) 1396 (11.4)  BMS only 2084 (4.7) 582 (5.4) 595 (5.5) 514 (4.7) 393 (3.2)  DES only 37 669 (84.2) 9136 (85.4) 9069 (83.4) 9220 (84.2) 10 244 (83.7)  BMS and DES 973 (2.2) 209 (2.0) 317 (2.9) 246 (2.2) 201 (1.6) Anti-platelets  Clopidogrel 83.1 35 382 (91.8) 8363 (93.3) 8541 (91.7) 8971 (90.6) 9507 (91.7)  Prasugrel 954 (2.5) 167 (1.9) 232 (2.5) 291 (2.9) 264 (2.5)  Ticagrelor 2146 (5.6) 417 (4.7) 517 (5.6) 623 (6.3) 589 (5.7)  Ticlopidine 67 (0.2) 18 (0.2) 19 (0.2) 17 (0.2) 13 (0.1) Warfarin 95.4 785 (1.8) 160 (1.5) 175 (1.6) 229 (2.1) 221 (1.9) Bivalirudin 95.4 230 (0.5) 50 (0.5) 47 (0.4) 61 (0.6) 72 (0.6) GP IIb/IIIa 96.6 1854 (4.1) 565 (5.2) 478 (4.4) 428 (3.9) 383 (3.2) 30-day death 100 165 (0.4) 31 (0.3) 34 (0.3) 48 (0.4) 52 (0.4) In-hospital death 99.1 51 (0.1) 11 (0.1) 7 (0.1) 10 (0.1) 23 (0.2) In-hospital MACE 99.1 272 (0.6) 66 (0.6) 68 (0.6) 69 (0.6) 69 (0.6) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 4 Patient and procedural variables by quartile, ACS procedures Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Table 4 Patient and procedural variables by quartile, ACS procedures Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) Variables % non missing ACS procedures N = 87 597 Operator volume quartile: n (%) Q1: 0–128 (N = 22 509) Q2: 129–178 (N = 22 298) Q3: 179–239 (N = 21 769) Q4: 240–714 (N = 21 021) Age  <50 100 11 154 (12.7) 2987 (13.3) 2921 (13.1) 2749 (12.6) 2497 (11.9)  50–59 19 840 (22.6) 5215 (23.2) 5125 (23.0) 4905 (22.5) 4595 (21.9)  60–69 23 890 (27.3) 6105 (27.1) 6091 (27.3) 6012 (27.6) 5682 (27.0)  70–79 20 389 (23.3) 5065 (22.5) 5145 (23.1) 5037 (23.1) 5142 (24.5)  80+ 12 324 (14.1) 3137 (13.9) 3016 (13.5) 3066 (14.1) 3105 (14.8) Sex  Male 100 64 002 (73.1) 16 495 (73.3) 16 411 (73.6) 15 807 (72.6) 15 289 (72.7)  Female 23 595 (26.9) 6014 (26.7) 5887 (26.4) 5962 (27.4) 5732 (27.3) Ethnicity  Asian 80.6 4911 (7.0) 1637 (9.2) 1283 (7.3) 1145 (6.2) 846 (5.0)  Black 584 (0.8) 192 (1.1) 157 (0.9) 122 (0.7) 113 (0.7)  White 60 498 (85.7) 14 317 (80.4) 14 716 (84.0) 16 281 (88.4) 15 184 (90.2)  Other 4598 (6.5) 1668 (9.4) 1372 (7.8) 865 (4.7) 693 (4.1) Smoking  Never 91.6 28 605 (35.7) 7729 (37.9) 7285 (35.7) 6996 (35.2) 6595 (33.7)  Former 27 582 (34.4) 6591 (32.3) 6896 (33.8) 6796 (34.2) 7299 (37.3)  Current 24 019 (29.9) 6085 (29.8) 6198 (30.4) 6086 (30.6) 5650 (28.9) Dialysis 92.6 806 (1.0) 220 (1.0) 215 (1.0) 165 (0.8) 206 (1.1) Diabetes 96.0 16 815 (20.0) 4314 (19.9) 4294 (19.9) 4136 (19.8) 4071 (20.3) PCI 97.9 14 426 (16.8) 3607 (16.5) 3551 (16.3) 3501 (16.4) 3767 (18.1) CABG 74.9 5004 (7.6) 1412 (7.8) 1170 (7.4) 1075 (7.3) 1347 (8.0) MI 93.9 18 195 (22.1) 4418 (21.2) 4570 (22.1) 4470 (21.6) 4737 (23.7) CVA 95.2 3754 (4.5) 865 (4.0) 932 (4.4) 945 (4.6) 1012 (5.1) HC 95.2 39 324 (47.2) 9006 (42.0) 9943 (46.6) 10 093 (48.6) 10 282 (51.9) Hypertension 95.2 42 902 (51.5) 10 335 (48.2) 10 933 (51.2) 10 676 (51.4) 10 958 (55.3) PVD 95.2 4130 (5.0) 863 (4.0) 1014 (4.8) 1004 (4.8) 1249 (6.3) VHD 95.2 1184 (1.4) 247 (1.2) 275 (1.3) 330 (1.6) 332 (1.7) LVEF  Good (>50%) 42.3 22 975 (62.0) 5642 (62.8) 5790 (62.3) 5492 (62.0) 6051 (60.8)  Fair (30–50%) 10 940 (29.5) 2634 (29.3) 2789 (30.0) 2493 (28.2) 3024 (30.4)  Poor (<30%) 3171 (8.6) 711 (7.9) 709 (7.6) 867 (9.8) 884 (8.9) Indication  UA/NSTEMI 100 50 562 (57.7) 12 641 (56.2) 12 836 (57.6) 12 484 (57.3) 12 601 (59.9)  STEMI 37 035 (42.3) 9868 (43.8) 9462 (42.4) 9285 (42.7) 8420 (40.1) Shock 99.0 3804 (4.4) 1074 (4.8) 1005 (4.6) 922 (4.3) 803 (3.8) Ventilated 91.1 2223 (2.8) 658 (3.2) 562 (2.8) 542 (2.7) 461 (2.4) IABP 96.8 1742 (2.1) 452 (2.1) 427 (2.0) 456 (2.2) 407 (2.0) CP support 96.8 134 (0.2) 40 (0.2) 37 (0.2) 29 (0.1) 28 (0.1) Inotropes 96.8 1272 (1.5) 342 (1.6) 298 (1.4) 321 (1.5) 311 (1.5) Access  Femoral 98.8 23 486 (27.1) 8518 (38.7) 6080 (27.6) 4225 (19.5) 4663 (22.3)  Radial 60 300 (69.7) 12 737 (57.9) 15 208 (69.1) 16 631 (76.9) 15 724 (75.1)  Multiple 2577 (3.0) 693 (3.2) 678 (3.1) 709 (3.3) 497 (2.4)  Other 179 (0.2) 43 (0.2) 35 (0.2) 48 (0.2) 53 (0.3) Left-main 100 3949 (4.5) 816 (3.6) 808 (3.6) 912 (4.2) 1413 (6.7) Multivessel 100 10 557 (12.1) 2170 (9.6) 2467 (11.1) 2558 (11.8) 3362 (16.0) CTO 92.3 3240 (4.0) 872 (4.4) 748 (3.8) 693 (3.4) 927 (4.5) Invasive imaging 94.0 4992 (6.1) 856 (4.0) 1124 (5.4) 1267 (6.2) 1745 (8.8) Pressure wire 94.0 3171 (3.8) 769 (3.6) 784 (3.8) 751 (3.7) 867 (4.4) Stent type  No stents 97.0 6033 (7.1) 1366 (6.3) 1424 (6.7) 1595 (7.5) 1648 (7.9)  BMS only 8215 (9.7) 2402 (11.1) 2351 (11.0) 2060 (9.7) 1402 (6.7)  DES only 68 998 (81.2) 17 534 (80.7) 16 955 (79.6) 17 080 (80.6) 17 429 (83.8)  BMS and DES 1756 (2.1) 412 (1.9) 566 (2.7) 464 (2.2) 314 (1.5) Anti-platelets  Clopidogrel 79.7 43 775 (62.7) 11 458 (65.4) 11 156 (62.4) 10 885 (58.2) 10 276 (65.5)  Prasugrel 6282 (9.0) 1517 (8.7) 1901 (10.6) 1820 (9.7) 1044 (6.7)  Ticagrelor 19 500 (27.9) 4470 (25.5) 4764 (26.6) 5945 (31.8) 4321 (27.5)  Ticlopidine 227 (0.3) 64 (0.4) 59 (0.3) 57 (0.3) 47 (0.3) Warfarin 94.5 769 (0.9) 153 (0.7) 176 (0.8) 207 (1.0) 233 (1.2) Bivalirudin 94.5 6648 (8.0) 1522 (7.1) 1721 (8.2) 2100 (10.2) 1305 (6.6) GP IIb/IIIa 96.2 20 491 (24.3) 6178 (28.3) 5443 (25.7) 4712 (22.7) 4158 (20.3) 30-day death 100 3373 (3.9) 947 (4.2) 820 (3.7) 821 (3.8) 785 (3.7) In-hospital death 98.8 2035 (2.4) 542 (2.4) 487 (2.2) 483 (2.3) 523 (2.5) In-hospital MACE 98.8 2712 (3.1) 750 (3.4) 626 (2.8) 644 (3.0) 692 (3.3) BMS, bare metal stents; CABG, coronary artery bypass grafting; CP, cardiopulmonary; CTO, chronic total occlusion; CVA, cardiovascular accident; DES, drug-eluting Stents; GP, glycoprotein; HC, hypercholesterolaemia; IABP, intra-aortic balloon pump; LVEF, left-ventricular ejection fraction; MACE, major adverse cardiovascular event; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease; Q1–Q4, 1st to 4th quartile; STEMI, ST-elevation myocardial infarction; UA, unstable angina; VHD, valvular heart disease. Multivariable models The relationship between 30-day mortality and volume in 2013–2014 was explored in multivariable mixed effects logistic regression model. A scatter plot of model predicted mortality versus average operator volume is shown in Supplementary material online, Figure S2. Figure 2 presents ORs for operator volume for all, ACS-only, and primary PCI-only cohorts, with these values tabulated in Table 5. The full models are described in Supplementary material online, Tables S2a, S2b, and S2c. After adjustment for case-mix, there was no strong evidence of an independent linear association, with small effect sizes and high P-values. Potential non-linear relationships between volume and 30-day mortality were explored graphically by plotting observed and model-adjusted mortality against volume. Figures 3–5 demonstrate relatively stable model-adjusted mortality as volume varied. Table 5 Operator volume odds ratios for mixed effect logistic regression models 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 ACS, acute coronary syndromes; CI, confidence interval; MACE, major adverse cardiovascular event; OR, odds ratio. Table 5 Operator volume odds ratios for mixed effect logistic regression models 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 30-day mortality In-hospital mortality In-hospital MACE OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-value All PCIs N = 133 970 N = 138 627 N = 138 627  per 100 ppy 0.99 (0.93–1.05) 0.725 1.02 (0.93–1.12) 0.643 0.98 (0.92–1.05) 0.630  Less than 75 ppy 1.04 (0.85–1.27) 0.688 1.00 (0.76–1.32) 0.990 1.13 (0.92–1.38) 0.244  Less than 50 ppy 1.01 (0.71–1.45) 0.938 0.85 (0.52–1.39) 0.526 1.03 (0.72–1.46) 0.890 ACS only N = 87 597 N = 91 178 N = 91 178  per 100 ppy 0.96 (0.83–1.11) 0.545 0.92 (0.74–1.15) 0.462 1.03 (0.86–1.22) 0.774  Less than 75 ppy 1.09 (0.88–1.35) 0.417 1.01 (0.76–1.34) 0.947 1.15 (0.92–1.43) 0.225  Less than 50 ppy 1.10 (0.76–1.59) 0.609 0.87 (0.53–1.43) 0.576 1.01 (0.69–1.50) 0.946 Primary only N = 36 048 N = 37 732 N = 37 732  per 100 ppy 0.93 (0.84–1.03) 0.149 1.01 (0.88–1.16) 0.909 0.98 (0.88–1.10) 0.788  Less than 75 ppy 1.07 (0.82–1.39) 0.606 1.13 (0.82–1.55) 0.448 1.24 (0.95–1.61) 0.110  Less than 50 ppy 1.27 (0.81–2.01) 0.295 1.11 (0.65–1.89) 0.711 1.15 (0.73–1.82) 0.550 ACS, acute coronary syndromes; CI, confidence interval; MACE, major adverse cardiovascular event; OR, odds ratio. Figure 2 View largeDownload slide Operator volume odds ratios. Figure 2 View largeDownload slide Operator volume odds ratios. Figure 3 View largeDownload slide Operator volume vs. mortality, all procedures. Figure 3 View largeDownload slide Operator volume vs. mortality, all procedures. Figure 4 View largeDownload slide Operator volume vs. mortality, ACS procedures. Figure 4 View largeDownload slide Operator volume vs. mortality, ACS procedures. Figure 5 View largeDownload slide Operator volume vs. mortality, primary procedures. Figure 5 View largeDownload slide Operator volume vs. mortality, primary procedures. Sensitivity analyses The findings of the primary analyses were supported in all sensitivity analyses. Table 5 also lists ORs for in-hospital mortality and in-hospital MACE, and for operator volume dichotomized at 75 and 50 ppy. In these models, associations for operator volume were relative small and not statistically significant. The Wald test examining the value of an interaction term suggested no significant interactive effects between operator volume and centre volume (all procedures, P = 0.473; ACS, P = 0.740; primary, P = 0.629). The inclusion in the model of factors that depend on operator choice did not alter the direction or strength of association of volume with mortality (OR 1.02, 95% CI 0.96–1.08; P = 0.533). Similarly, excluding variables relating to cardiac support did not alter results (OR 1.00, 95% CI 0.94–1.06; P = 0.923). Further details of these sensitivity analyses are available in Supplementary material online, Tables S3a, S3b, S3c, S4a, S4b, S4c and S5. We do not adjust for multiple comparisons when reporting these sensitivity analyses. Discussion This nationally representative study is the first to investigate the relationship between operator volume and outcomes within the contemporary PCI era in Europe, where revascularization following ACS is the most common indication for intervention and transradial access is the default for the majority of countries within Europe. It is the only study, where volume is not calculated based on future operator activity and does not rely solely on a contrived dichotomization or categorization of this metric. Our analysis found no relationship between 30-day mortality following PCI and the number of cases performed by a PCI operator. Despite an inverse association between volume and mortality using crude, unadjusted data, we found that after adjusting for patient-level risk factors we can rule out any meaningful change in the odds of 30-day mortality as volume increases. This was observed both when operator volume was modelled continuously, and when volume was dichotomized at 50 or 75 ppy in line with international guidelines. Similar results were obtained in analyses of ACS only and primary PCI only procedures, and of centre-reported in-hospital mortality and in-hospital MACE. Procedures performed by lower volume operators were on average performed to treat patients at higher baseline risk. This is consistent with the finding that there are proportionally more ACS-indicated procedures when volume is low than when volume is high. Conversely, baseline cardiovascular risk typically increased with volume, a pattern that broadly persisted when looking at elective and ACS procedures, separately. The existing literature investigating the relationship of operator volume on outcomes following PCI reports discrepant findings, with some studies reporting increases in adverse events with lower operator volume after risk-adjustment,3–7 and others finding no such association.8–16 Further, significant heterogeneity exists in the statistical methods used, particularly regarding volume threshold definitions and the settings from which the data are derived. Despite this, a large meta-analysis34 pooled aggregate data from 23 studies which included 1 109 103 patients, and found no strong evidence of short-term mortality reductions against annualized dichotomized operator volume (high vs. low volume OR 0.96, 95% CI 0.86–1.08 in all studies; OR 0.90, 95% CI 0.79–1.01 in high and very-high quality studies only), although there was some evidence of a reduction in MACE for high-volume operators (high vs. low volume OR 0.62, 95% CI 0.40–0.97).This meta-analysis was undertaken prior to a study by Badheka et al.6 in 457 498 procedures identified from the National Inpatient Sample (NIS) database between 2005 and 2009 that showed lower in-hospital mortality in higher volume quartiles compared with the lowest quartile (1–15 ppy), with the highest volume quartile (>100 ppy) associated with the most significant reduction (OR 0.65, 95% CI 0.58–0.73; P < 0.001). However, as these studies stem from procedures performed no later than 2010, their relevance in informing contemporary best practice is diminished and should be interpreted with this in mind. In particular, this period has seen a rapid transition from trans-femoral to trans-radial access in the UK, with radial procedures rising from 16% in 2005 to over 75% in 2014.35 The benefits of radial access has been demonstrated in randomized controlled trials,36,37 and as the strength of this benefit is positively associated with operator volume,25 radial uptake may have altered the volume–outcome relationship. Two contemporary studies using data after 2010 can offer some insight here. An analysis of 3 747 866 procedures from July 2009 to March 2015 from the NCDR CathPCI Registry,7 which captures over 90% of PCIs in the USA, found a small increase in in-hospital mortality for each 50-case decrease in annual volume (OR 1.04, 95% CI 1.03–1.05), and this relationship persisted in stable, unstable angina/ST-elevation myocardial infarction (STEMI) and STEMI PCI subgroups. Radial access rates in this study were only 15.2%, and the inclusion of data from as early as 2009 without considering temporal trends or contemporary subgroups, limits the relevance of this study to the UK experience. An analysis of 323 322 procedures in 2014 and 2015 from the Japanese PCI Registry15 reported no significant differences in in-hospital mortality or a composite of peri-procedural complications outcome between operator volume deciles. Despite drawing data from a similar era and comparable radial access rates (61.3% in Japan vs. 67.5% in our study), geo-cultural disparities and average volume levels make translating these findings to Europe difficult. Our study adds a European perspective, reflecting a recent PCI era and showing that increasing volume is not associated with better mortality outcomes, and also that the majority of operators have caseloads exceeding previously defined volume thresholds. Median annualized volume in our study was 135 ppy (IQR 93–188). This is markedly higher than that found in the NIS database in the USA6 (between 33 and 58 ppy), the CathPCI Registry7 (59 ppy, IQR 21–106), and in the J-PCI registry in Japan15 (28 ppy, IQR 10–56). Furthermore, 44% of operators performed <50 PCI ppy in the CathPCI registry, whereas in the current analysis with only 8.9% and 16.9% of operators performing less than 50 and 75 ppy, respectively. Such differences in operator volume distribution across the two healthcare systems may contribute to the differences in the reported findings. The 150 minimum operator volume over 2 years recommended by BCIS and EAPCI17,38 was met or exceeded in 95.4% of analysed procedures, as measured using volume (i.e. exceeding 75 ppy). A sensitivity analysis found no difference in 30-day mortality above or below this threshold after risk adjustment. The ACCF/AHA/SCAI guidelines stipulating a minimum of 50 ppy were not met in just 1.4% of cases, making a well-powered and balanced analysis of this particular recommendation challenging. There are fewer studies of operator volume and outcome for primary PCI. Two studies4,12 investigating this relationship directly found post-adjustment in-hospital mortality reductions for operators performing over 11 primary procedures annually (relative risk high vs. low 0.43, 95% CI 0.21–0.83) or over 10 primary procedures annually (OR 0.66, 95% CI 0.48–0.92). A small (N = 331) single-centre, three-operator study was also conducted39 but with only one operator in each volume ‘group’, it is impossible to consider the effect of volume independent of other operator factors. Finally in the CathPCI study, the relationship between operator volume and in-hospital mortality was significant in patients presenting with STEMI with similar findings in the PPCI cohort (OR 1.03; 95% CI 1.02–1.04). The findings of these studies are not supported by our data, which suggest mortality following primary PCI is not associated with either total volume or primary specific volume, though there was considerable uncertainty around the central estimates. The need to achieve adequate geographical coverage to ensure timely access to primary PCI services, meet minimum centre activity standards, and accommodate a changing population means that future services are likely to continue to see a period of transition, affecting both centre and operator volumes. Volume–related outcome patterns in the primary PCI setting should therefore continue to be closely monitored. Centre volume was not the focus of this study, but it was included as a potential confounder in the regression models. The latest European Heart Journal STEMI guidelines recommend that STEMI patients should be transferred to 24/7 high-volume PCI centres because of reduced mortality irrespective of the chosen treatment strategy.40 A recent study using the BCIS registry from 2007 to 201341 found no change in 30-day mortality as centre volume varied. Limitations Investigations into volume–outcome relationships are hindered by the fact that volume cannot be independently randomized. This restricts the body of statistical evidence to non-randomized volume exposures, where the influence of confounding factors must always be carefully considered and controlled for where possible. Our volume metric measures the number of procedures as the responsible consultant, but it was necessary to exclude cases where consultant identifier was not known or where volume, mortality, or other key variables were unavailable, though this study was still able to retain 84.5% of all available procedures, which compares favourably with other recent registry-based volume studies, for example with 54.5%6 and 73.8%15 in the NIS database and the J-PCI registry, respectively. We note that although consultant identifier is not likely to be missing at random, missing values were present in high and low-volume centres similarly, and the use of random effects mitigates the potential bias induced by these exclusions. Where volume is very low or zero, it is unclear whether this is because the operator was inactive (for example due to maternity leave), was working outside the UK, had not yet been appointed to consultant grade, or whether an uncommon or incorrect operator alias was used. Furthermore, only three operators had volume recorded greater than 500 ppy, with just one exceeding 700 ppy, and so the volume–mortality relationship in this range cannot be adequately disassociated from the relationship of mortality with individual operators. The relationship between volume and mortality in very low volume or very high volume cases should be interpreted with caution. In particular, the potential association of mortality and volume below recommended operator minimums cannot be addressed by our analysis, although we do not observe any signals towards compromised outcomes in operators performing below national/international recommendations. Centres and operators were anonymized, and additional variables at these levels, for example operator age, years since qualification, or total career PCI volume could not be accounted for. We considered in-hospital mortality and MACE as secondary outcomes, although this information cannot be independently validated and is therefore less robust than 30-day mortality. We cannot rule out that other clinical end-points may be independently associated with operator volume. Finally, we were unable to consider whether total career volume may influence the relationships that we report, as operator identifiable information (the GMC number) was only available since 2012. Conclusion In our 2-year study in PCI procedures in England and Wales, we find no direct relationship between 30-day mortality and operator volume after adjusting for patient characteristics. This finding holds when looking at ACS and primary procedures, separately. A vast majority of operators had caseloads exceeding the yearly minimum recommended by BCIS, EAPCI, and ACCF/AHA/SCAI. Supplementary material Supplementary material is available at European Heart Journal online. Conflict of interest: N.C. has received research grants from Boston Scientific, Haemonetics and HeartFLow and speaker/consultancy from Boston, Haemonetics, Heartflow and Abbott. All other authors declared no conflict of interest. References 1 Halm EA , Lee C , Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature . Ann Intern Med 2002 ; 137 : 511 – 520 . Google Scholar CrossRef Search ADS PubMed 2 Morche J , Mathes T , Pieper D. Relationship between surgeon volume and outcomes: a systematic review of systematic reviews . Syst Rev 2016 ; 5 : 204. Google Scholar CrossRef Search ADS PubMed 3 McGrath PD , Wennberg DE , Malenka DJ , Kellett MA , Ryan TJ , O’meara JR , Bradley WA , Hearne MJ , Hettleman B , Robb JF , Shubrooks S , VerLee P , Watkins MW , Lucas FL , O’Connor GT. Operator volume and outcomes in 12,998 percutaneous coronary interventions. Northern New England Cardiovascular Disease Study Group . J Am Coll Cardiol 1998 ; 31 : 570 – 576 . Google Scholar CrossRef Search ADS PubMed 4 Vakili BA , Kaplan R , Brown DL. Volume-outcome relation for physicians and hospitals performing angioplasty for acute myocardial infarction in New York state . Circulation 2001 ; 104 : 2171 – 2176 . Google Scholar CrossRef Search ADS PubMed 5 Hannan EL , Wu C , Walford G , King SB , Holmes DR , Ambrose JA , Sharma S , Katz S , Clark LT , Jones RH. Volume-outcome relationships for percutaneous coronary interventions in the stent era . Circulation 2005 ; 112 : 1171 – 1179 . Google Scholar CrossRef Search ADS PubMed 6 Badheka AO , Patel NJ , Grover P , Singh V , Patel N , Arora S , Chothani A , Mehta K , Deshmukh A , Savani GT , Patel A , Panaich SS , Shah N , Rathod A , Brown M , Mohamad T , Tamburrino FV , Kar S , Makkar R , O’Neill WW , De Marchena E , Schreiber T , Grines CL , Rihal CS , Cohen MG. Impact of annual operator and institutional volume on percutaneous coronary intervention outcomes: a 5-year United States experience (2005-2009) . Circulation 2014 ; 130 : 1392 – 1406 . Google Scholar CrossRef Search ADS PubMed 7 Fanaroff AC , Zakroysky P , Dai D , Wojdyla D , Sherwood MW , Roe MT , Wang TY , Peterson ED , Gurm HS , Cohen MG , Messenger JC , Rao SV. Outcomes of PCI in relation to procedural characteristics and operator volumes in the United States . J Am Coll Cardiol 2017 ; 69 : 2913 – 2924 . Google Scholar CrossRef Search ADS PubMed 8 Xie Y , Rizzo JA , Brown DL. A modified method for estimating volume–outcome relationships: application to percutaneous coronary intervention . J Med Econ 2008 ; 11 : 57 – 70 . Google Scholar CrossRef Search ADS PubMed 9 Harjai KJ , Berman AD , Grines CL , Kahn J , Marsalese D , Mehta RH , Schreiber T , Boura JA , O’Neill WW. Impact of interventionalist volume, experience, and board certification on coronary angioplasty outcomes in the era of stenting . Am J Cardiol 2004 ; 94 : 421 – 426 . Google Scholar CrossRef Search ADS PubMed 10 Shook TL , Sun GW , Burstein S , Eisenhauer AC , Matthews RV. Comparison of percutaneous transluminal coronary angioplasty outcome and hospital costs for low-volume and high-volume operators . Am J Cardiol 1996 ; 77 : 331 – 336 . Google Scholar CrossRef Search ADS PubMed 11 Madan M , Nikhil J , Hellkamp AS , Pieper KS , Labinaz M , Cohen EA , Buller CE , Cantor WJ , Seidelin P , Ducas J , Carere RG , Natarajan MK , Conor O’Shea J , Tcheng JE; for the ESPRIT Investigators . Effect of operator and institutional volume on clinical outcomes after percutaneous coronary interventions performed in Canada and the United States: a brief report from the Enhanced Suppression of the Platelet glycoprotein IIb/IIIa Receptor with Integrilin Therapy (ESPRIT) study . Can J Cardiol 2009 ; 25 : e269 – e272 . Google Scholar CrossRef Search ADS PubMed 12 Srinivas VS , Hailpern SM , Koss E , Monrad ES , Alderman MH. Effect of physician volume on the relationship between hospital volume and mortality during primary angioplasty . J Am Coll Cardiol 2009 ; 53 : 574 – 579 . Google Scholar CrossRef Search ADS PubMed 13 Hannan EL , Racz M , Ryan TJ , McCallister BD , Johnson LW , Arani DT , Guerci AD , Sosa J , Topol EJ. Coronary angioplasty volume-outcome relationships for hospitals and cardiologists . JAMA 1997 ; 277 : 892 – 898 . Google Scholar CrossRef Search ADS PubMed 14 Vakili BA , Brown DL ; 1995 Coronary Angioplasty Reporting System of the New York State Department of Health . Relation of total annual coronary angioplasty volume of physicians and hospitals on outcomes of primary angioplasty for acute myocardial infarction (data from the 1995 Coronary Angioplasty Reporting System of the New York State Department of Health) . Am J Cardiol 2003 ; 91 : 726 – 728 . Google Scholar CrossRef Search ADS PubMed 15 Inohara T , Kohsaka S , Yamaji K , Amano T , Fujii K , Oda H , Uemura S , Kadota K , Miyata H , Nakamura M , Inohara T , Kohsaka S , Yamaji K , Amano T , Fujii K , Oda H , Uemura S , Kadota K , Miyata H , Nakamura M , Kadota K , Shiode N , Tanaka N , Amano T , Uemura S , Akasaka T , Morino Y , Fujii K , Hikichi H , Amano T , Fujii K , Kohsaka S , Ishii H , Tanabe K , Ozaki Y , Sumitsuji S , Iida O , Hara H , Takashima H , Shirai S , Nansato M , Inohara T , Ueda Y , Numasawa Y , Noma S ; J-PCI Registry Investigators . Impact of institutional and operator volume on short-term outcomes of percutaneous coronary intervention: a report from the Japanese Nationwide Registry . JACC Cardiovasc Interv 2017 ; 10 : 918 – 927 . Google Scholar CrossRef Search ADS PubMed 16 Jolly SS , Cairns J , Yusuf S , Niemela K , Steg PG , Worthley M , Ferrari E , Cantor WJ , Fung A , Valettas N , Rokoss M , Olivecrona GK , Widimsky P , Cheema AN , Gao P , Mehta SR. Procedural volume and outcomes with radial or femoral access for coronary angiography and intervention . J Am Coll Cardiol 2014 ; 63 : 954 – 963 . Google Scholar CrossRef Search ADS PubMed 17 Banning AP , Baumbach A , Blackman D , Curzen N , Devadathan S , Fraser D , Ludman P , Norell M , Muir D , Nolan J , Redwood S ; British Cardiovascular Intervention society . Percutaneous coronary intervention in the UK: recommendations for good practice 2015 . Heart 2015 ; 101 : 1 – 13 . Google Scholar CrossRef Search ADS PubMed 18 Harold JG , Bass TA , Bashore TM , Brindiss RG , Brush JE , Burke JA , Dehmers GJ , Deychak YA , Jneids H , Jolliss JG , Landzberg JS , Levine GN , McClurken JB , Messengers JC , Moussas ID , Muhlestein JB , Pomerantz RM , Sanborn TA , Sivaram CA , Whites CJ , Williamss ES , Halperin JL , Beckman JA , Bolger A , Byrne JG , Lester SJ , Merli GJ , Muhlestein JB , Pina IL , Wang A , Weitz H. ACCF/AHA/SCAI 2013 update of the clinical competence statement on coronary artery interventional procedures . Catheter Cardiovasc Interv 2013 ; 82 : E69 – E111 . Google Scholar CrossRef Search ADS PubMed 19 Windecker S , Kolh P , Alfonso F , Collet J-P , Cremer J , Falk V , Filippatos G , Hamm C , Head SJ , Jüni P , Kappetein AP , Kastrati A , Knuuti J , Landmesser U , Laufer G , Neumann F-J , Richter DJ , Schauerte P , Sousa Uva M , Stefanini GG , Taggart DP , Torracca L , Valgimigli M , Wijns W , Witkowski A. 2014 ESC/EACTS Guidelines on myocardial revascularization . EuroIntervention 2015 ; 10 : 1024 – 1094 . Google Scholar CrossRef Search ADS PubMed 20 Rashid M , Sperrin M , Ludman PF , Neill DO , Nicholas O , Belder MAD , Mamas MA. Impact of operator volume for percutaneous coronary intervention on clinical outcomes: what do the numbers say? Eur Hear J Qual Care Clin Outcomes 2016 ; 2 : 16 – 22 . Google Scholar CrossRef Search ADS 21 Ludman PF. British Cardiovascular Intervention Society Registry for audit and quality assessment of percutaneous coronary interventions in the United Kingdom . Heart 2011 ; 97 : 1293 – 1297 . Google Scholar CrossRef Search ADS PubMed 22 Rubin DB. Multiple Imputation for Nonresponse in Surveys . Hoboken, New Jersey : John Wiley & Sons, Inc. ; 2004 . 23 Hulme W , Sperrin M , Kontopantelis E , Ratib K , Ludman P , Sirker A , Kinnaird T , Curzen N , Kwok CS , Belder MD , Nolan J , Mamas MA. Increased radial access is not associated with worse femoral outcomes for percutaneous coronary intervention in the United Kingdom . Circ Cardiovasc Interv 2017 ; 10 : e004279. Google Scholar CrossRef Search ADS PubMed 24 Mamas MA , Nolan J , Belder MA. D , Zaman A , Kinnaird T , Curzen N , Kwok CS , Buchan I , Ludman P , Kontopantelis E ; British Cardiovascular Intervention Society (BCIS) and the National Institute for Clinical Outcomes Research (NICOR) . Changes in arterial access site and association with mortality in the United Kingdom . Circulation 2016 ; 133 : 1655 – 1667 . Google Scholar CrossRef Search ADS PubMed 25 Hulme W , Sperrin M , Rushton H , Ludman PF , Belder M , Curzen N , Kinnaird T , Kwok CS , Buchan I , Nolan J , Mamas MA. Is there a relationship of operator and center volume with access site-related outcomes? An analysis from the British Cardiovascular Intervention Society Circ Cardiovasc Interv 2016 ; 9 : e003333. Google Scholar CrossRef Search ADS PubMed 26 Li KH , Raghunathan TE , Rubin DB. Large-sample significance levels from multiply imputed data using moment-based statistics and an f reference distribution . J Am Stat Assoc 1991 ; 86 : 1065 . 27 Wu S , Crespi CM , Wong WK. Comparison of methods for estimating the intraclass correlation coefficient for binary responses in cancer prevention cluster randomized trials . Contemp Clin Trials 2012 ; 33 : 869 – 880 . Google Scholar CrossRef Search ADS PubMed 28 Lin M , Lucas HC , Shmueli G. Research commentary—too big to fail: large samples and the p-value problem . Inf Syst Res 2013 ; 24 : 906 – 917 . Google Scholar CrossRef Search ADS 29 Team RC . R: A Language and Environment for Statistical Computing . Vienna, Austria : R Foundation for Statistical Computing ; 2015 . 30 Wickham H. Easily Install and Load ‘Tidyverse’ Packages [R package tidyverse version 1.1.1]. Comprehensive R Archive Network (CRAN). R Studio. 31 Buuren S. V , Groothuis-Oudshoorn K. mice: multivariate imputation by chained equations in R . J Stat Softw 2011 ; 45 : 1 – 67 . Google Scholar CrossRef Search ADS 32 Bates D , Mächler M , Bolker B , Walker S. Fitting linear mixed-effects models using lme4 . J Stat Softw 2015 ; 67 : 48. Google Scholar CrossRef Search ADS 33 Harrell FE. Regression Modeling Strategies . Cham : Springer International Publishing ; 2015 . Google Scholar CrossRef Search ADS 34 Strom JB , Wimmer NJ , Wasfy JH , Kennedy K , Yeh RW. Association between operator procedure volume and patient outcomes in percutaneous coronary intervention: a systematic review and meta-analysis . Circ Cardiovasc Qual Outcomes 2014 ; 7 : 560 – 566 . Google Scholar CrossRef Search ADS PubMed 35 Ludman P. BCIS Audit Report for 2014 Activity. 2015 . https://www.bcis.org.uk/wp-content/uploads/2017/01/BCIS-audit-2014.pdf (5 March 2018). 36 Andò G , Capodanno D. Radial versus femoral access in invasively managed patients with acute coronary syndrome . Ann Intern Med 2015 ; 163 : 932. Google Scholar CrossRef Search ADS PubMed 37 Ruiz-Rodriguez E , Asfour A , Lolay G , Ziada KM , Abdel-Latif AK. Systematic review and meta-analysis of major cardiovascular outcomes for radial versus femoral access in patients with acute coronary syndrome . South Med J 2016 ; 109 : 61 – 76 . Google Scholar CrossRef Search ADS PubMed 38 Authors/Task Force members , Windecker S , Kolh P , Alfonso F , Collet J-P , Cremer J , Falk V , Filippatos G , Hamm C , Head SJ , Jüni P , Kappetein AP , Kastrati A , Knuuti J , Landmesser U , Laufer G , Neumann F-J , Richter DJ , Schauerte P , Sousa Uva M , Stefanini GG , Taggart DP , Torracca L , Valgimigli M , Wijns W , Witkowski A. 2014 ESC/EACTS Guidelines on myocardial revascularization . Eur Heart J 2014 ; 35 : 2541 – 2619 . Google Scholar CrossRef Search ADS PubMed 39 Politi A , Galli M , Zerboni S , Michi R , Marco FD , Llambro M , Ferrari G. Operator volume and outcomes of primary angioplasty for acute myocardial infarction in a single high-volume centre . J Cardiovasc Med 2006 ; 7 : 761 – 767 . Google Scholar CrossRef Search ADS 40 Ibanez B , James S , Agewall S , Antunes MJ , Bucciarelli-Ducci C , Bueno H , Caforio ALP , Creas F , Goudevenos JA , Halvorsen S , Hindriks G , Kastrati A , Lenzen MJ , Prescott E , Roffi M , Valgimigli M , Varenhorst C , Vranckx P , Widimsky P ; ESC Scientific Document Group . 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force for the management of acute myocardial infarction patients presenting with ST-segment elevation of the European Society of Cardiology (ESC) . Eur Heart J 2018 ; 39 : 119 – 177 . Google Scholar CrossRef Search ADS PubMed 41 O’Neill D , Nicholas O , Gale CP , Ludman P , Belder MA. D , Timmis A , Fox KA , Simpson IA , Redwood S , Ray SG. Total center percutaneous coronary intervention volume and 30-day mortality . Circ Cardiovasc Qual Outcomes 2017 ; 10 : e003186 . Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Heart JournalOxford University Press

Published: Mar 22, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off