Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Oct 1.
Published in final edited form as: Spine (Phila Pa 1976). 2019 Oct 1;44(19):1371–1380. doi: 10.1097/BRS.0000000000003083

Utilization and outcomes for spine surgery in the United States and Canada

Peter Cram 1,2,3,4, Bruce E Landon 5,6, John Matelski 2, Vicki Ling 3, Anthony V Perruccio 3,7,8, J Michael Paterson 3, Raja Rampersaud 7,9,10
PMCID: PMC6746582  NIHMSID: NIHMS1526542  PMID: 31261267

Abstract

Study design:

Retrospective cohort study.

Objective:

Examine variation in spine surgery utilization between the province of Ontario and state of New York among all patients and pre-specified patient subgroups.

Summary of Background Data:

Spine surgery is common and costly. Within-country variation in utilization is well studied, but there has been little exploration of variation in spine surgery utilization between countries.

Methods:

We used population level administrative data from Ontario (years 2011–2015) and New York (2011–2014) to identify all adults who underwent inpatient spinal decompression or fusion surgery using relevant procedure codes. Patients were stratified according to age and surgical urgency (elective vs emergent). We calculated standardized utilization rates (procedures per-10,000 population per-year) for each jurisdiction. We compared Ontario and New York with respect to patient demographics and the percentage of hospitals performing spine surgery. We compared utilization rates of spinal decompression and fusion surgery in Ontario and New York among all patients and after stratifying by surgical urgency and patient age.

Results:

Patients in Ontario were older than patients in New York for both decompression (mean age 58.8 vs. 51.3 years; P<.001) and fusion (58.1 vs. 54.9; P<.001). A smaller percentage of hospitals in Ontario than New York performed decompression (26.1% vs 54.9%; P<.001) or fusion (15.2% vs 56.7%; P<.001). Overall, utilization of spine surgery (decompression plus fusion) in Ontario was 6.6 procedures per-10,000 population per-year and in New York was 16.5 per-10,000 per-year (P<.001). Ontario-New York differences in utilization were smaller for emergent cases (2.0 per-10,000 in Ontario vs. 2.5 in New York; P<.001), but larger for elective cases (4.6 vs 13.9; P<.001). The lower utilization in Ontario was particularly large among younger patients (age <60).

Conclusion:

We found significantly lower utilization of spine surgery in Ontario compared to New York. These differences should inform policy reforms in both jurisdictions.

Keywords: decompression, fusion, spine surgery, United States, Canada, variation, back pain, health services research

Introduction

Approximately 900,000 American and 30,000 Canadian adults undergo spine surgery annually1,2 with inpatient direct hospital cost per-procedure ranging from $4,500-$30,000 depending upon procedure and country.3,4 Researchers and policy makers believe that utilization of spine surgery in the US is significantly higher than in other industrialized countries but international comparisons of spine surgery are largely lacking.5,6 Meanwhile in Ontario surgeons and patients are concerned about inadequate access to spine surgery resulting in treatment delays and unwarranted suffering.7 Neither country has a clear understanding of their spine surgery utilization rates or how they directly compare. In an era when both countries are looking externally for solutions to vexing problems,8,9 international comparisons can provide new insights.

An estimated 10%−15% of the population has chronic back pain10,11 and while pain often improves with conservative management, surgery is extremely effective in some cases.12,13 Given continued uncertainty over precisely which patients will benefit from surgery, patients and surgeons have considerable discretion whether to choose surgery and which procedure is best (spinal decompression alone vs fusion).14,15 As a result most countries are struggling to balance an epidemic of chronic back pain with concerns about growth in spine surgery and cost implications.16,17 While an array of international studies have done an excellent job in characterizing the structural differences between nations’ healthcare systems at a macro-level,18,19 more granular data comparing healthcare utilization and outcomes for individual procedures and conditions remains extremely limited.20

We used contemporary administrative health data from Ontario (Canada) and New York State (US) to examine differences in utilization of in-patient decompression alone and spinal fusion procedures. We hypothesized that spine surgery utilization would be lower in Ontario than New York and that lower utilization in Ontario would primarily be manifest by a smaller percentage of hospitals performing spine surgery and lower utilization of elective spine procedures.

Methods

Data

Building on prior international comparative work,21,22 we used administrative data from the most populous Canadian province (Ontario: population 14.3 million, median age 39.8 years, 68% White, 5% Black) and a large US state (New York: population 19.8 million, median age 38.4 years, 66% White, 16% Black.23,24 Ontario and New York share a common border, are diverse, and have an extremely large city (Toronto and New York City), and significant rural areas. Our primary data source for Ontario was the 2011–2015 Discharge Abstract Database (DAD) held and analyzed at ICES(ices.on.ca). These administrative records provide information on all hospitalizations (excluding psychiatric admissions) paid for by the Ontario Health Insurance Plan (OHIP); OHIP provides health insurance to all legal residents of Ontario (~99% of the population) and pays for virtually 100% of inpatient hospitalizations. Ontario’s DAD provides information regarding demographic characteristics (age, sex), primary and secondary diagnoses coded using International Classification of Diseases Version 10 (ICD-10) codes, procedures captured using Canadian Classification of Health Interventions (CCI) codes, discharge disposition (e.g., died-in-hospital, home, transfer to another acute-care hospital), a unique patient identifier (used to track patients over time) and unique hospital identifiers. Comorbid conditions present at the time of the index hospital stay were identified using the Quan ICD-10 adaptation of the Elixhauser comorbidity coding scheme, which maps ICD-10 coded comorbidities to ICD-9.25

Our primary data for New York was the 2011–2014 State Inpatient Database (SID).22,26 The SID contains administrative data for all patients admitted to all acute care hospitals, excluding a small number of Veterans Administration and psychiatric hospitals. Data elements for each admission include patient demographics, primary and secondary diagnosis and procedures (coded using ICD9-CM codes), discharge disposition, patient identifier, and hospital identifier. Comorbid conditions present for the index hospital stay were captured using algorithms developed by Elixhauser et al.27

Estimates of the New York population were obtained from US Census Data; estimates of the Ontario population were obtained from analogous Canadian Census Data. We linked the New York data to the American Hospital Association annual survey to ascertain information regarding hospital teaching status and bed size. We linked the Ontario DAD to information from the Ontario Ministry of Health and Long-Term Care for hospital-level data.

Cohort generation

Our cohorts were constructed at 2 levels: patient level, where everyone who underwent spine surgery was counted once; and procedure level, where patients were counted once for each procedure they underwent.

We identified adults aged 18–105 years who underwent decompression alone or fusion between January 1, 2011 and September 30, 2015 in Ontario and September 30, 2014 in New York using CCI codes in Ontario and ICD9-CM codes for New York that have been used in prior spine surgery research (Appendix 1).28–30 We excluded patients with missing age or sex. We used a 90-day lookback (October 1, 2010-December 31, 2010) to exclude procedures performed during the 90-days prior to study initiation to avoid counting readmissions or staged procedures as a new procedure. Patients who had both a fusion code and a decompression code during the same surgery were categorized as having had fusion surgery (Appendix 1); patients who underwent laminectomy/laminotomy (with or without a discectomy) were categorized has having decompression surgery. We decided a priori that repeat spine procedures performed ≤ 90-days after a prior procedure would be counted as either a staged procedure and/or readmission; any procedures that occurred > 90-days after the index procedure were counted as new procedures.

Statistical analyses

First, at the patient level, we compared demographics and comorbidity of patients who underwent spine surgery in Ontario and New York using bivariate measures. Analyses were conducted separately for the decompression and fusion cohorts. We then stratified patients into elective and emergent procedures (Appendix 2): elective procedures were categorized as scheduled admissions without evidence of trauma, metastatic cancer, or active infection: emergent procedures were those categorized as emergency cases or cases with evidence of trauma, metastatic cancer, or infection. Subjects who met neither our elective nor emergent criteria (0 in Ontario, 430 in New York) were judged indeterminate urgency and excluded from subgroup analysis. Second, we compared the percentage of all acute care hospitals in Ontario and New York that performed decompression and fusion and the mean annual hospital decompression and fusion volumes after excluding hospitals with implausibly low volumes (< 1 procedure per-year). We also identified the number of physicians in Ontario and New York who performed an average of 1-or-more spine procedures per-year in our dataset. We then calculated the mean, median and inter-quartile ranges for spine surgery volume for Ontario and New York surgeons.

Third we compared the overall per-capita spine surgery rates (procedures per-10,000 per-year) for adults in Ontario and New York. The numerator for Ontario was the total number of spine surgery procedures (decompressions and fusions) performed and the denominator was the number of adults age ≥ 18-years residing in Ontario in 2013. For New York utilization, we identified all spine procedures performed in New York (N=103,790), but subsequently used zip code of residence to exclude procedures performed on patients who resided outside of New York State (N=9,283) resulting in a final cohort of 94,507 procedures performed in New York on residents of New York State; our denominator for New York was the number of adults age ≥ 18-years residing in New York in 2013. We then conducted subgroup analyses focusing on elective procedures, emergency procedures, decompression alone, and fusion alone. We calculated age and sex standardized utilization rates (Ontario as the reference) using direct standardization. Fourth, we compared the percentage of patients in Ontario and New York who underwent multiple procedures.

Fifth, we compared unadjusted outcomes in Ontario and New York including hospital acute length of stay (LOS), in-hospital mortality, and hospital readmission within 30-days and 90-days of discharge among those who survived to discharge. Analyses were conducted for all patients as well as the subgroups described previously. We used generalized estimating equations to compare adjusted outcomes in Ontario and New York using 3 sequential models: Model 1 adjusted for only patient age and sex; Model 2 adjusted for Model 1 factors plus hospital decompression or fusion volume, surgical urgency (elective, emergent) and hospital LOS; Model 3 included all Model 2 factors plus comorbid conditions. This analysis was approved by the Research Ethics Board at University Health Network, Toronto. Our analyses were pre-specified prior to the initiation of any statistical analysis and are publicly available through Open Science (https://osf.io/m8jqf/). All analyses were performed using either SAS (Cary, North Carolina) or R statistical software packages.

Results

Our analyses included 15,103 patients in Ontario and 15,294 in New York who received decompression and 16,968 in Ontario and 79,716 in New York who received fusion (Table 1). Patients receiving spine surgery in Ontario were older and had fewer comorbidities (Table 1). A smaller percentage of Ontario than New York hospitals performed decompression (26.1% vs 54.9%; P<.001) and fusion (15.2% vs. 56.7%; P<.001)(Table 2). Mean hospital decompression volume was higher for Ontario than New York (79.7 vs. 34.4 procedures per-year; P=.025)(Table 2) but similar for fusion (166.8 vs. 184.6; P=.658); in both Ontario and New York 25% of hospitals had annual decompression volumes of less than 10 and fusion volumes of less than 35. In New York there were 548 surgeons who performed 4-or-more procedures during our 4-year study window (36.4 surgeons per-million population) while Ontario had 228 surgeons who performed 5-or-more procedures during the 5-year study window (21.0 surgeons per-million).

Table 1:

Characteristics of patients who underwent Decompression or Fusion in Ontario (2015–2015) or New York State (2011–2014).i Cells represent number and percentages except where specified.

All patients
Decompression Fusion
Ontario
(N=15103)		 NY
(N=15294)		 P-value Ontario
(N=16986)		 NY
(N=79716)		 P-value
Demographics
Age, mean (sd) 58.8 (16.4) 51.3 (15.4) <.001 58.1 (15.1) 54.9 (14.0) <.001
Age, <50 4514 (29.9) 7242 (47.4) <.001 4646 (27.4) 28124 (35.3) <.001
Age, 50–59 2824 (18.7) 3314 (21.7) <.001 3907 (23.0) 21606 (27.1) <.001
Age, 60–69 3294 (21.8) 2598 (17) <.001 4184 (24.6) 17152 (21.5) <.001
Age, 70–79 2960 (19.6) 1608 (10.5) <.001 3228 (19.0) 9914 (12.4) <.001
Age, 80–89 1391 (9.2) 505 (3.3) <.001 975 (5.7) 2791 (3.5) <.001
Age, ≥ 90 120 (0.8) 27 (0.2) <.001 46 (0.3) 129 (0.2) 0.003
Female 6675 (44.2) 6632 (43.4) 0.146 7869 (46.3) 41126 (51.6) <.001
Comorbid conditionsii
Congestive heart failure 145 (1.0) 166 (1.1) 0.304 148 (0.9) 1607 (2.0) <.001
Depression 128 (0.8) 1519 (9.9) <.001 213 (1.3) 12196 (15.3) <.001
Hypertension 12 (0.1) 5403 (35.3) <.001 20 (0.1) 34006 (42.7) <.001
Diabetes 1953 (12.9) 2185 (14.3) 0.001 2216 (13.0) 12835 (16.1) <.001
COPD 501 (3.3) 1830 (12) <.001 659 (3.9) 13489 (16.9) <.001
Renal Failure 124 (0.8) 285 (1.9) <.001 129 (0.8) 2182 (2.7) <.001
Scheduled/electiveiii
Decompression Fusion
Ontario
(N=10612)		 NY
(N=10027)		 P-value Ontario
(N=11779)		 NY
(N=67735)		 P-value
Age, mean (sd) 58.6 (15.7) 52.4 (15.5) <.001 58.5 (13.9) 54.7 (13.7) <.001
Age, <50 3059 (28.8) 4435 (44.3) <.001 3081 (26.2) 24073 (35.6) <.001
Age, 50–59 2085 (19.6) 2182 (21.8) <.001 2840 (24.1) 18677 (27.6) <.001
Age, 60–69 2446 (23.0) 1857 (18.5) <.001 3028 (25.7) 14599 (21.6) <.001
Age, 70–79 2178 (20.5) 1173 (11.7) <.001 2255 (19.1) 8247 (12.2) <.001
Age, 80–89 819 (7.7) 353 (3.5) <.001 561 (4.8) 2012 (3.0) <.001
Age, ≥ 90 25 (0.2) 20 (0.2) 0.684 14 (0.1) 60 (0.1) 0.406
Female 4464 (42.1) 4230 (42.2) 0.872 5860 (49.7) 35798 (52.9) <.001
Comorbid conditions2
Congestive heart failure 36 (0.3) 95 (0.9) <.001 69 (0.6) 1017 (1.5) <.001
Depression 38 (0.4) 1026 (10.2) <.001 107 (0.9) 10816 (16.0) <.001
Hypertension **iv 3673 (36.7) <.001 ** 29217 (43.2) <.001
Diabetes 1280 (12.1) 1418 (14.2) <.001 1450 (12.3) 10492 (15.5) <.001
COPD 319 (3.0) 1277 (12.7) <.001 444 (3.8) 11758 (17.4) <.001
Renal Failure 45 (0.4) 169 (1.7) <.001 69 (0.6) 1522 (2.2) <.001
Emergent
Decompression Fusion
Ontario
(N=4491)		 NY
(N=5270)		 P-value Ontario
(N=5207)		 NY
(N=11715)		 P-value
Age, mean (sd) 59.1 (18.0) 49.3 (15.2) <.001 57.4 (17.3) 56.1 (15.8) <.001
Age, <50 1455 (32.4) 2802 (53.2) <.001 1565 (30.1) 3853 (33.1) <.001
Age, 50–59 739 (16.5) 1129 (21.5) <.001 1067 (20.5) 2854 (24.5) <.001
Age, 60–69 848 (18.9) 740 (14.1) <.001 1156 (22.2) 2499 (21.5) 0.212
Age, 70–79 782 (17.4) 434 (8.2) <.001 973 (18.7) 1620 (13.9) <.001
Age, 80–89 572 (12.7) 151 (2.9) <.001 414 (8.0) 757 (6.5) <.001
Age, ≥ 90 95 (2.1) 7 (0.1) <.001 32 (0.6) 65 (0.6) 0.715
Female 2211 (49.2) 2399 (45.6) <.001 2009 (38.6) 5203 (44.7) <.001
Comorbid conditions2
Congestive heart failure 109 (2.4) 71 (1.3) <.001 79 (1.5) 579 (5.0) <.001
Depression 90 (2.0) 493 (9.4) <.001 106 (2.0) 1339 (11.5) <.001
Hypertension ** 1729 (32.9) <.001 ** 4683 (40.2) <.001
Diabetes 673 (15.0) 766 (14.6) 0.551 766 (14.7) 2297 (19.7) <.001
COPD 182 (4.1) 553 (10.5) <.001 215 (4.1) 1680 (14.4) <.001
Renal Failure 79 (1.8) 116 (2.2) 0.138 60 (1.2) 654 (5.6) <.001
Open in a new tab
i

Based upon index (first) procedure for each patient

ii

Based upon index hospitalization only

iii

Some procedures will be of indeterminate urgency (neither elective nor emergent and thus present in neither subgroup)

iv

Cell suppressed because cell size less than 5

Table 2:

Hospital characteristics

Decompression Fusion
Ontario
(N=165)		 NY
(N=224)		 P-value Ontario
(N=165)		 NY
(N=224)		 P-value
Hospitals performing procedure,i number (%) 43 (26.1) 123 (54.9) <.001 25 (15.2) 127 (56.7) <.001
Annual procedural volume [all], mean (SD) 79.7 (124.4) 34.4 (50.1) 0.025 166.8 (169.1) 184.6 (240.0) 0.658
Annual procedural volume [all], median (Inter-quartile range) 14 (2–109) 17.9 (7.7 – 38.9) N/A 120 (22–287) 95.2 (32.7 – 224) N/A
Annual procedural volume [elective], mean (SD) 55.6 (94.4) 23.0 (40.7) 0.033 116.3 (119.9) 160.2 (224.5) 0.164
Annual procedural volume [emergency], mean (SD) 24.1 (34.6) 12.0 (14.8) 0.031 50.3 (60.4) 26.3 (31.7) 0.064
Bed number, mean (SD)ii 275.5 (171.0) 475.2 (418.9) <.001 325.2 (185.5) 479.2 (416.2) 0.004
Major teaching, number (%)iii 12 (27.9) 32 (26.7) 0.967 10 (40.0) 35 (28.2) 0.314
Open in a new tab
i

Defined as performing ≥ 1 procedure per-year over the study period

ii

Defined for calendar year 2013

iii

Defined for calendar year 2013

Annual spine surgery utilization was significantly higher in New York (16.5 procedures per-10,000 per-year) than Ontario (6.6 per-10,000 per-year; P<.001)(Table 3). In analyses stratified by surgical urgency (Table 3), utilization of elective procedures in New York was approximately 300% higher than in Ontario (13.9 vs 4.6 procedures per-10,000 per-year; P<.001), while utilization of emergent spine procedures was only 25% higher in New York (2.5 vs 2.0; P<.001). In analyses stratified by surgical subtype, utilization of decompression was lower in New York than Ontario (2.1 vs 3.1; P<.001)(Table 3), while utilization of fusion was approximately 400% higher in New York than Ontario (14.3 vs 3.5; P<.001). Further analysis demonstrated that the New York-Ontario difference in utilization was substantially larger among younger patients and smaller for older patients (Figure 1 and Supplementary Table S1). For example, utilization of spine procedures in New York was 315% greater than Ontario for patients less-than 50 years of age (10.7 vs 3.4), but only 14% greater in patients age 80 and above (11.4 vs 10.0). The percentage of patients who had multiple hospitalizations for spine surgery was higher in New York (8.3%) than Ontario (5.2%)(P<.001) (Supplementary Table S2).

Table 3:

Per-capita back surgery utilizationi (procedures per-10,000 per-year)

All procedures
Ontario New York Utilizationii
Procedures Population Procedures Population Ontario NY P-value
Total 33957 10854795 94507 15053173 6.6 16.5 <.001
Scheduled/elective only
Total 23734 10854795 77831 15053173 4.6 13.9 <.001
Emergent only
Total 10223 10854795 16286 15053173 2.0 2.5 <.001
Decompression only
Total 15744 10854795 14222 15053173 3.1 2.1 <.001
Elective 11110 10854795 9237 15053173 2.2 1.4 <.001
Emergent 4634 10854795 4972 15053173 0.9 0.8 .327
Fusion only
Total 18213 10854795 80285 15053173 3.5 14.3 <.001
Elective 12624 10854795 68594 15053173 2.4 12.5 <.001
Emergent 5589 10854795 11314 15053173 1.1 1.8 <.001
Open in a new tab
i

Age and sex standardized using the 2013 Ontario population as the reference

ii

Age and sex standardized using the 2013 Ontario population as the reference

Figure 1:

Figure 1:

Open in a new tab

Per-capita utilization of spine surgery in Ontario and New York stratified by patient age and procedure urgency

In unadjusted analyses, hospital acute length of stay (LOS) was approximately 2 days longer for decompression and 3.5 days longer for fusion in Ontario compared to New York (P<.001)(Table 4); however stratified analyses demonstrate that Ontario-New York differences in LOS were quite small in the elective surgery population and much larger in the emergent group (Table 4). Unadjusted in-hospital mortality was significantly higher in Ontario for both emergent decompression and fusion cohorts (Table 4). After adjusting for patient demographics, hospital LOS and surgical urgency (Model 2), differences in mortality in Ontario and New York were no longer significant for elective or emergent decompression or fusion (Supplementary Tables S3–S7). In adjusted analyses differences in hospital LOS were reduced and readmission rates in Ontario were significantly lower than in New York (Supplementary Tables S3–S7).

Table 4:

Unadjusted outcomes

All procedures
Decompression Fusion
Ontario
(N=15744)		 NY
(N=15861)		 P-value Ontario
(N=18213)		 NY
(N=87927)		 P-value
Length-of-stay, mean (SD) 4.9 (8.7) 2.9 (4.3) <.001 7.9 (14.3) 4.4 (6.4) <.001
Discharge disposition, number (%)
Died in-hospital 102 (0.6) 19 (0.12) <.001 185 (1.0) 263 (0.30) <.001
Died in-hospital within 7-days of admission 12 (0.1) 6 (0.04) 0.161 42 (0.2) 58 (0.07) <.001
Home 13366 (84.9) 14560 (91.8) <.001 13676 (75.1) 72787 (82.8) <.001
Transfer to another acute-care hospital 684 (4.3) 59 (0.4) <.001 1230 (6.8) 515 (0.6) <.001
Post-acute-care 1560 (9.9) 1200 (7.6) <.001 3088 (17.0) 14262 (16.2) 0.015
Other 32 (0.2) 23 (0.1) 0.216 34 (0.2) 100 (0.1) 0.013
Readmission, number (%)
30-day hospital readmission 1491 (9.5) 1394 (8.8) 0.036 2174 (12.1) 11323 (12.9) 0.001
90-day hospital readmission 2022 (12.9) 1937 (12.2) 0.090 2773 (15.4) 14306 (16.3) 0.001
Scheduled/elective
Decompression Fusion
Ontario
(N=11110)		 NY
(N=10421)		 P-value Ontario
(N=12624)		 NY
(N=75074)		 P-value
Length-of-stay, mean (SD) 2.5 (2.9) 2.0 (2.3) <.001 4.6 (4.0)
 3.4 (3.2) <.001
Discharge disposition, number (%)
Died in-hospital **x **x N/A 19 (0.2) 49 (0.07) 0.002
Died in-hospital within 7-days of admission **x **x N/A 11 (0.1) 30 (0.04) 0.027
Home 10495 (94.5) 9802 (94.1) 0.202 11022 (87.3) 65295 (87.0) 0.299
Transfer to another acute-care hospital 141 (1.3) 30 (0.3) <.001 238 (1.9) 307 (0.4) <.001
Post-acute-care 448 (4.0) 577 (5.5) <.001 1328 (10.5) 9350 (12.5) <.001
Other 22 (0.2) 8 (0.1) 0.022 17 (0.1) 73 (0.1) 0.226
Readmission, number (%)
30-day hospital readmission 577 (5.2) 699 (6.7) <.001 788 (6.3) 7605 (10.1) <.001
90-day hospital readmission 838 (7.5) 1007 (9.7) <.001 1087 (8.6) 9829 (13.1) <.001
Emergent
Decompression Fusion
Ontario
(N=4634)		 NY
(N=5426)		 P-value Ontario
(N=5589)		 NY
(N=12437)		 P-value
Length-of-stay, mean (SD) 10.6 (13.9) 4.5 (6.3) <.001 15.6 (23.4) 10.5 (13.3) <.001
Discharge disposition, number (%)
Died in-hospital ** 15 (0.28) <.001 166 (3.0) 207 (1.66) <.001
Died in-hospital within 7-days of admission ** **x 0.059 31 (0.6) 28 (0.23) 0.001
Home 2871 (62.0) 4747 (87.5) <.001 2654 (47.5) 7299 (58.7) <.001
Transfer to another acute-care hospital 543 (11.7) 29 (0.5) <.001 992 (17.7) 196 (1.6) <.001
Post-acute-care 1112 (24.0) 620 (11.4) <.001 1760 (31.5) 4708 (37.9) <.001
Other 10 (0.2) 15 (0.3) 0.544 17 (0.3) 27 (0.2) 0.275
Readmission, number (%)
30-day hospital readmission 914 (20.1) 691 (12.7) <.001 1386 (25.6) 3574 (28.7) <.001
90-day hospital readmission 1184 (26.1) 926 (17.1) <.001 1686 (31.1) 4315 (35.3) <.001
Open in a new tab
x

Cell suppressed because cell size less than 5

Discussion

In an analysis of data from the US and Canada, we found significantly higher per-capita utilization of spine surgery in New York compared to Ontario. The higher utilization in New York was attributable to higher rates of elective fusion procedures particularly in younger patients. In aggregate these results provide important insight into differences in healthcare delivery in the US and Canada.

Several aspects of our study require discussion. It is important to consider our results in the context of nearly 2-decades of concern about excessive utilization of spine surgery in the US.16,31 We found that utilization of emergent spine surgery in New York and Ontario was similar, while utilization of elective surgery was markedly higher in New York. From an epidemiologic standpoint we had no reason to expect that the need for emergent spine surgery would be different between New York and Ontario populations. Alternatively, our finding that elective spine surgery utilization was 300% higher in New York than Ontario updates and expands on an extremely small body of international population-based spine surgery literature from more than two-decades ago.5,32,33

Higher utilization of spine surgery in New York was driven almost entirely by higher rates of elective fusion. Elective spine surgery, particularly fusion, has long been characterized as a prototype for preference sensitive care.34 Identified drivers of utilization have included patient and surgeon preference, overuse of imaging, surgeon-industry financial relationships, and supply of spine surgeons.35–37 Utilization of spine surgery has also been driven by an epidemic of chronic back pain and evidence of improvements in symptoms with spinal fusion in a subset of patients who have failed conservative management.38,39 However there remains significant uncertainty over how to identify patients who are most likely to benefit from surgery and which patients benefit from fusion above and beyond decompression alone.39,40

While it is tempting to assume that our findings represent excess utilization in New York, it is also important to consider potential underuse in Ontario. Data from the Canadian Institute for Health Information suggests that utilization of spine surgery in Ontario is 20%−30% below the Canadian average.41 Meanwhile, data from the Dartmouth Atlas suggests that spine surgery utilization (for spinal stenosis) in New York is well below the US average.42

It is important to consider the role of health system financing and incentives. While the recent introduction of bundled payments and accountable care organizations may change the economic incentives that have driven exceedingly high utilization of many expensive procedures in the US,18,43 fee-for-service reimbursement remains the dominant model in the US. In New York (and most of the US) fee-for-service reimbursement incentivizes both spine surgeons and hospitals to perform surgery.44 In Ontario (and most of Canada) surgeons are reimbursed on a fee-for-service basis and hospitals operate under global budgets determined largely by historical services and volumes. Any increases in spine surgery volume increase hospital “costs.” As a result, Canadian hospitals have incentive to minimize spine surgery volumes, typically done through restriction of operating room time for elective cases. Our finding that approximately 55% of New York hospitals but only 25% of Ontario hospitals provide spine surgery reflect fundamental differences in hospital financing and regulation in the US and Canada. Our finding that New York had approximately 75% more spine-surgeons per-capita than Ontario is also interesting. Due to constraints in the number of spine surgeons and hospital surgical capacity, Ontario residents often wait 6-months or more to see a surgeon and an additional 6-months or more (particularly for fusion) to receive surgery when recommended.45 Given these pressures Canadian spine surgeons commonly screen incoming referrals carefully, seeing only those patients with pathology on imaging that is most likely to be amenable to intervention.46 We suspect that differences in ability to supply spine surgery is a major factor underlying cross-border differences in utilization rates per capita.

Cognizant of intra- and cross-state variation in utilization and cost of spinal fusion across the US,4 it is important to consider the economic and clinical implications of the different utilization rates that we found. We do not know what the correct or appropriate fusion rate is. However, if we were to assume a desired fusion rate of 8.0 procedures per-10,000 per-year (i.e. halfway between that of Ontario (2.4/10,000) and New York (13.7/10,000) (Table 3), patients in New York may be undergoing an excess of 6 fusion procedures per-10,000 per-year. This would amount to 9,031 additional spine fusions annually, which at a mean national cost of $27,600 (US) per-procedure,1 translates to $249,255,600 annually. Alternatively, if we were to assume that utilization of fusion in Ontario should be increased by 6 procedures per-10,000 per-year this would translate into an additional 6,513 fusions annually, which at a mean average cost of $13,000 CDN per procedure, translates to $42,419,169 (CDN).3

A number of other findings warrant brief mention. First, our finding that the higher utilization in New York relative to Ontario was more pronounced in younger patients and less pronounced in older patients is interesting. This likely reflects several issues including patient and provider preference as well as differences in reimbursement between private insurance (more generous) common in younger patients (more generous) relative to Medicare (less generous) in the US; Ontario provincial insurance provides identical reimbursement for patients of any age. From a clinical perspective older patients and/or surgeons, regardless of geography, may be less likely to choose spinal surgery than younger patients due to perceived or actual increased risk of adverse events.47,48. Second, our finding of significantly lower rates of most comorbid conditions in Ontario likely reflects the intense pressure on US hospitals to completely code all comorbidities- a pressure that does not exist to the same extent in Canada; risk-adjustment including comorbidities in international comparative research using administrative data should consider this carefully. Third, our finding that outcomes in Ontario and New York were similar for elective procedures, but worse in Ontario for emergent procedures warrants further study. Much of the difference in outcomes (e.g., higher mortality in Ontario) appears to have been driven by the lower elective-to-emergent ratio in Ontario compared to New York. Another component of the higher mortality in Ontario can be attributed to the longer hospital length of stay in Ontario which, in turn, provided a longer observation period during which in-hospital mortality could occur. The longer hospital length of stay in Ontario compared to New York is consistent with previous Canada-US comparative studies.22 Longer hospital length of stay in Canada is related to an array of factors, most importantly a chronic shortage of post-acute-care and rehabilitation beds that delay discharge.49 Fourth, it is important to comment on differences in hospital volume and percentage of hospitals offering spine surgery: Ontario has a small percentage of hospitals offering spine surgery with higher volumes for decompression; New York has a large percentage of hospitals offering spine surgery and lower volume for decompression. At the same time, both Ontario and New York had a larger number of low-volume hospitals than might be optimal. To the extent that lower volumes have been associated with inferior surgical outcomes,50 our results highlight a potential problem in both jurisdictions. Finally, it is important to consider our work in the context of growing interest in international comparative population-based health services research.22,51 There is growing appreciation that these studies can provide important insights into the tradeoffs inherent to different healthcare delivery systems.20

Our study has several limitations. First, our analysis is limited to administrative data from 1 Canadian province and 1 US state and our results should be interpreted with this perspective. Our analyses should be replicated by comparing different Canadian provinces and US states. Second, our analysis was limited to inpatient spine procedures and did not include outpatient procedures; available data suggest that ambulatory spine procedures are much more common in the US (13%−20% of all procedures versus <5% in Canada) and the result of inclusion of these procedures would likely have been to magnify the already markedly higher US spine surgery utilization rates even further.52,53,54 Third, our study did not capture New York residents who may have received surgery outside of New York. Fourth, we did not address differences in utilization by diagnosis or surgical indication. Finally, we lacked detailed data on surgical indications including symptoms and imaging in our administrative data.

In conclusion, we found significantly lower utilization of spine surgery in Ontario when compared to New York. The difference in utilization was attributable to less elective fusion surgery, primarily in younger (i.e. non-Medicare) patients. These differences should inform broader spine surgery policy reforms in both jurisdictions.

Supplementary Material

Supplemental Data File (doc., pdf., xls., etc.)_1
Supplemental Data File (doc., pdf., xls., etc.)_2
Supplemental Data File (doc., pdf., xls., etc.)_3

Acknowledgement

This study was supported by the Institute for Clinical Evaluative Sciences (ICES), which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care (MOHLTC). The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsement by ICES or the Ontario MOHLTC is intended or should be inferred. Parts of this material are based on data and/or information compiled and provided by the Canadian Institute for Health Information (CIHI). However, the analyses, conclusions, opinions and statements expressed in the material are those of the authors and not necessarily those of CIHI.

A K24 AR062133 award from NIAMS at the NIH funds were received in support of this work.

Relevant financial activities outside the submitted work: grants.

Footnotes

Publisher's Disclaimer: The manuscript submitted does not contain information about medical device(s)/drug(s).

References

  • 1.Weiss AJ, Elixhauser A, Andrews RM. Characteristics of Operating Room Procedures in U.S. Hospitals, 2011: Statistical Brief #170 Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville (MD): Agency for Healthcare Research and Quality (US); 2006. [PubMed] [Google Scholar]
  • 2.McIntosh G, Hall H, Melles T. The incidence of spinal surgery in Canada. Can J Surg. 1998;41(1):59–66. [PMC free article] [PubMed] [Google Scholar]
  • 3.Kim S, Mortaz Hedjri S, Coyte PC, Rampersaud YR. Cost-utility of lumbar decompression with or without fusion for patients with symptomatic degenerative lumbar spondylolisthesis. Spine J. 2012;12(1):44–54. [DOI] [PubMed] [Google Scholar]
  • 4.Goz V, Rane A, Abtahi AM, Lawrence BD, Brodke DS, Spiker WR. Geographic variations in the cost of spine surgery. Spine (Phila Pa 1976). 2015;40(17):1380–1389. [DOI] [PubMed] [Google Scholar]
  • 5.Cherkin DC, Deyo RA, Loeser JD, Bush T, Waddell G. An international comparison of back surgery rates. Spine (Phila Pa 1976). 1994;19(11):1201–1206. [DOI] [PubMed] [Google Scholar]
  • 6.Deyo RA, Mirza SK. The case for restraint in spinal surgery: does quality management have a role to play? Eur Spine J. 2009;18 Suppl 3:331–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Busse JW, Riva JJ, Nash JV, et al. Surgeon attitudes toward nonphysician screening of low back or low back-related leg pain patients referred for surgical assessment: a survey of Canadian spine surgeons. Spine (Phila Pa 1976). 2013;38(7):E402–408. [DOI] [PubMed] [Google Scholar]
  • 8.Naylor CD. Canada as Single-Payer Exemplar for Universal Health Care in the United States: A Borderline Option. Jama. 2018;319(1):17–18. [DOI] [PubMed] [Google Scholar]
  • 9.Simpson J Chronic Condition. Toronto: Allan Lane; 2012. [Google Scholar]
  • 10.Newton JN, Briggs AD, Murray CJ, et al. Changes in health in England, with analysis by English regions and areas of deprivation, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015;386(10010):2257–2274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Shmagel A, Foley R, Ibrahim H. Epidemiology of Chronic Low Back Pain in US Adults: Data From the 2009–2010 National Health and Nutrition Examination Survey. Arthritis Care Res (Hoboken). 2016;68(11):1688–1694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med. 2008;358(8):794–810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Zaina F, Tomkins-Lane C, Carragee E, Negrini S. Surgical versus non-surgical treatment for lumbar spinal stenosis. Cochrane Database Syst Rev. 2016(1):Cd010264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Bernstein IA, Malik Q, Carville S, Ward S. Low back pain and sciatica: summary of NICE guidance. Bmj. 2017;356:i6748. [DOI] [PubMed] [Google Scholar]
  • 15.Ghogawala Z, Resnick DK, Glassman SD, Dziura J, Shaffrey CI, Mummaneni PV. Randomized controlled trials for degenerative lumbar spondylolisthesis: which patients benefit from lumbar fusion? Journal of neurosurgery Spine. 2017;26(2):260–266. [DOI] [PubMed] [Google Scholar]
  • 16.Deyo RA, Mirza SK, Martin BI, Kreuter W, Goodman DC, Jarvik JG. Trends, major medical complications, and charges associated with surgery for lumbar spinal stenosis in older adults. Jama. 2010;303(13):1259–1265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kobayashi K, Ando K, Nishida Y, Ishiguro N, Imagama S. Epidemiological trends in spine surgery over 10 years in a multicenter database. Eur Spine J. 2018;27(8):1698–1703. [DOI] [PubMed] [Google Scholar]
  • 18.Papanicolas I, Woskie LR, Jha AK. Health Care Spending in the United States and Other High-Income Countries. Jama. 2018;319(10):1024–1039. [DOI] [PubMed] [Google Scholar]
  • 19.Squires D, Anderson C. U.S. health care from a global perspective: spending, use of services, prices, and health in 13 countries. Issue Brief (Commonw Fund). 2015;15:1–15. [PubMed] [Google Scholar]
  • 20.Baicker K, Chandra A. Challenges in Understanding Differences in Health Care Spending Between the United States and Other High-Income Countries. Jama. 2018;319(10):986–987. [DOI] [PubMed] [Google Scholar]
  • 21.Ko DT, Krumholz HM, Wang Y, et al. Regional differences in process of care and outcomes for older acute myocardial infarction patients in the United States and Ontario, Canada. Circulation. 2007;115(2):196–203. [DOI] [PubMed] [Google Scholar]
  • 22.Cram P, Landon BE, Matelski J, et al. Hip and knee arthroplasty utilization and outcomes in the United States and Canada: an analysis of New York and Ontario administrative data. Arthritis & rheumatology (Hoboken, NJ). 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Population count, Canada, provinces, and territories 2016. 2016. Available at: https://www12.statcan.gc.ca/census-recensement/2016/dp-pd/hlt-fst/pd-pl/Table.cfm?Lang=Eng&T=101&S=50&O=A [Accessed November 26, 2018]. [Google Scholar]
  • 24.United States Census Bureau. Quick facts, New York. [Google Scholar]
  • 25.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–1139. [DOI] [PubMed] [Google Scholar]
  • 26.White C Cutting medicare hospital prices leads to a spillover reduction in hospital discharges for the nonelderly. Health Serv Res. 2014;49(5):1578–1595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Medical Care. 1998;36(1):8–27. [DOI] [PubMed] [Google Scholar]
  • 28.Bae HW, Rajaee SS, Kanim LE. Nationwide trends in the surgical management of lumbar spinal stenosis. Spine (Phila Pa 1976). 2013;38(11):916–926. [DOI] [PubMed] [Google Scholar]
  • 29.Deyo RA, Gray DT, Kreuter W, Mirza S, Martin BI. United States trends in lumbar fusion surgery for degenerative conditions. Spine (Phila Pa 1976). 2005;30(12):1441–1445; discussion 1446–1447. [DOI] [PubMed] [Google Scholar]
  • 30.Ontario ministry of health and long-term care. Quality-Based Pathway Clinical Handbook for Non-Emergent Integrated Spine Care. 2016. [Google Scholar]
  • 31.Weinstein JN, Lurie JD, Olson PR, Bronner KK, Fisher ES. United States’ trends and regional variations in lumbar spine surgery: 1992–2003. Spine (Phila Pa 1976). 2006;31(23):2707–2714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Taylor VM, Anderson GM, McNeney B, et al. Hospitalizations for back and neck problems: a comparison between the Province of Ontario and Washington State. Health Serv Res. 1998;33(4 Pt 1):929–945. [PMC free article] [PubMed] [Google Scholar]
  • 33.Lavis JN, Malter A, Anderson GM, et al. Trends in hospital use for mechanical neck and back problems in Ontario and the United States: discretionary care in different health care systems. Cmaj. 1998;158(1):29–36. [PMC free article] [PubMed] [Google Scholar]
  • 34.Lurie JD, Bell JE, Weinstein J. What rate of utilization is appropriate in musculoskeletal care? Clin Orthop Relat Res. 2009;467(10):2506–2511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Bederman SS, Mahomed NN, Kreder HJ, McIsaac WJ, Coyte PC, Wright JG. In the eye of the beholder: preferences of patients, family physicians, and surgeons for lumbar spinal surgery. Spine (Phila Pa 1976). 2010;35(1):108–115. [DOI] [PubMed] [Google Scholar]
  • 36.Bederman SS, Coyte PC, Kreder HJ, Mahomed NN, McIsaac WJ, Wright JG. Who’s in the driver’s seat? The influence of patient and physician enthusiasm on regional variation in degenerative lumbar spinal surgery: a population-based study. Spine (Phila Pa 1976). 2011;36(6):481–489. [DOI] [PubMed] [Google Scholar]
  • 37.Martin BI, Franklin GM, Deyo RA, Wickizer TM, Lurie JD, Mirza SK. How do coverage policies influence practice patterns, safety, and cost of initial lumbar fusion surgery? A population-based comparison of workers’ compensation systems. Spine J. 2014;14(7):1237–1246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus nonoperative treatment for lumbar spinal stenosis four-year results of the Spine Patient Outcomes Research Trial. Spine (Phila Pa 1976). 2010;35(14):1329–1338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Yavin D, Casha S, Wiebe S, et al. Lumbar Fusion for Degenerative Disease: A Systematic Review and Meta-Analysis. Neurosurgery. 2017;80(5):701–715. [DOI] [PubMed] [Google Scholar]
  • 40.Ghogawala Z, Dziura J, Butler WE, et al. Laminectomy plus Fusion versus Laminectomy Alone for Lumbar Spondylolisthesis. N Engl J Med. 2016;374(15):1424–1434. [DOI] [PubMed] [Google Scholar]
  • 41.Patient cost estimator. 2018. Available at: https://www.cihi.ca/en/patient-cost-estimator [Accessed August 3, 2018].
  • 42.Martin B, Tosteson A, Lurie JD, Mirza S. Variation in the care of surgical conditions: spinal stenosis. The Dartmouth Institute;2014. [PubMed] [Google Scholar]
  • 43.Barnett ML, Wilcock A, McWilliams JM, et al. Two-Year Evaluation of Mandatory Bundled Payments for Joint Replacement. N Engl J Med. 2019;380(3):252–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.The Quinn K. 8 Basic Payment Methods in Health Care. Ann Intern Med. 2015;163(4):300–306. [DOI] [PubMed] [Google Scholar]
  • 45.Zarrabian M, Bidos A, Fanti C, et al. Improving spine surgical access, appropriateness and efficiency in metropolitan, urban and rural settings. Can J Surg. 2017;60(5):342–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Busse JW, Riva JJ, Rampersaud R, et al. Spine surgeons’ requirements for imaging at the time of referral: a survey of Canadian spine surgeons. Can J Surg. 2014;57(2):E25–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Bernstein DN, Thirukumaran C, Saleh A, Molinari RW, Mesfin A. Complications and Readmission After Cervical Spine Surgery in Elderly Patients: An Analysis of 1786 Patients. World neurosurgery. 2017;103:859–868.e858. [DOI] [PubMed] [Google Scholar]
  • 48.Saleh A, Thirukumaran C, Mesfin A, Molinari RW. Complications and readmission after lumbar spine surgery in elderly patients: an analysis of 2,320 patients. Spine J. 2017;17(8):1106–1112. [DOI] [PubMed] [Google Scholar]
  • 49.Cram P, Lix LM, Bohm E, et al. Hip fracture care in Manitoba, Canada and New York State, United States: an analysis of administrative data. CMAJ open. 2019;7(1):E55–E62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Paul JC, Lonner BS, Goz V, et al. Complication rates are reduced for revision adult spine deformity surgery among high-volume hospitals and surgeons. Spine J. 2015;15(9):1963–1972. [DOI] [PubMed] [Google Scholar]
  • 51.Papanicolas I, Jha AK. Challenges in International Comparison of Health Care Systems. Jama. 2017;318(6):515–516. [DOI] [PubMed] [Google Scholar]
  • 52.Best MJ, Buller LT, Eismont FJ. National Trends in Ambulatory Surgery for Intervertebral Disc Disorders and Spinal Stenosis: A 12-Year Analysis of the National Surveys of Ambulatory Surgery. Spine (Phila Pa 1976). 2015;40(21):1703–1711. [DOI] [PubMed] [Google Scholar]
  • 53.Baird EO, Egorova NN, McAnany SJ, Qureshi SA, Hecht AC, Cho SK. National trends in outpatient surgical treatment of degenerative cervical spine disease. Global spine journal. 2014;4(3):143–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Martin CT, D’Oro A, Buser Z, et al. Trends and Costs of Anterior Cervical Discectomy and Fusion: a Comparison of Inpatient And Outpatient Procedures. The Iowa orthopaedic journal. 2018;38:167–176. [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data File (doc., pdf., xls., etc.)_1
NIHMS1526542-supplement-Supplemental_Data_File__doc___pdf___xls___etc___1.docx (33.8KB, docx)
Supplemental Data File (doc., pdf., xls., etc.)_2
NIHMS1526542-supplement-Supplemental_Data_File__doc___pdf___xls___etc___2.docx (18.1KB, docx)
Supplemental Data File (doc., pdf., xls., etc.)_3
NIHMS1526542-supplement-Supplemental_Data_File__doc___pdf___xls___etc___3.docx (54.6KB, docx)

RESOURCES