Analysis of National Rates, Cost, and Sources of Cost Variation in Adult Spinal Deformity

Analysis of National Rates, Cost, and Sources of Cost Variation in Adult Spinal Deformity Abstract BACKGROUND Several studies suggest significant variation in cost for spine surgery, but there has been little research in this area for spinal deformity. OBJECTIVE To determine the utilization, cost, and factors contributing to cost for spinal deformity surgery. METHODS The cohort comprised 55 599 adults who underwent spinal deformity fusion in the 2001 to 2013 National Inpatient Sample database. Patient variables included age, gender, insurance, median income of zip code, county population, severity of illness, mortality risk, number of comorbidities, length of stay, elective vs nonelective case. Hospital variables included bed size, wage index, hospital type (rural, urban nonteaching, urban teaching), and geographical region. The outcome was total hospital cost for deformity surgery. Statistics included univariate and multivariate regression analyses. RESULTS The number of spinal deformity cases increased from 1803 in 2001 (rate: 4.16 per 100 000 adults) to 6728 in 2013 (rate: 13.9 per 100 000). Utilization of interbody fusion devices increased steadily during this time period, while bone morphogenic protein usage peaked in 2010 and declined thereafter. The mean inflation-adjusted case cost rose from $32 671 to $43 433 over the same time period. Multivariate analyses showed the following patient factors were associated with cost: age, race, insurance, severity of illness, length of stay, and elective admission (P < .01). Hospitals in the western United States and those with higher wage indices or smaller bed sizes were significantly more expensive (P < .05). CONCLUSION The rate of adult spinal deformity surgery and the mean case cost increased from 2001 to 2013, exceeding the rate of inflation. Both patient and hospital factors are important contributors to cost variation for spinal deformity surgery. Cost of surgery, Spinal fusion, Spinal deformity, Scoliosis, National inpatient sample (NIS) database, BMP, Interbody fusion devices ABBREVIATIONS ABBREVIATIONS BMP bone morphogenic protein NIS National Inpatient Sample SOI severity of illness The United States spent $86 billion on the treatment of back and neck conditions in 2005.1 National spending on spine disease increased at an average rate of 7.0% per year from 1997 to 2006 (adjusted for inflation) and has continued to climb.2 Several studies demonstrate significant variation in cost for spinal surgery.3,4 There are many factors that affect the cost of spinal procedures, including patient factors such as comorbidities and severity of illness (SOI),5 hospital factors like teaching status and bed size,5 and, of course, surgery-specific factors such as the number of levels fused. A handful of studies also suggest that geography impacts the cost and utilization of spinal surgery.6,7 Goz et al found that posterolateral fusion procedures were significantly more expensive in the northeastern United States than the Midwest based on 2012 Medicare data.6 Another study reported lumbar fusion rates to be highest in Idaho, Montana, Iowa, Florida, and Wyoming and lowest in Maine, Kentucky, Indiana, North Dakota, and New Jersey.8 Despite this analysis of cost variation for spine surgery in general,3,4,6,7 there has been relatively little research in the area of spinal deformity surgery. Recent work shows that spinal deformity is more prevalent than previously thought: approximately 32% of the adult population and more than 60% of the elderly US population suffer from scoliosis.9,10 Anecdotally, the number of spinal deformity surgeries in the United States appears to be rising, as more surgeons are trained in this subspecialty, and research reveals additional indications for deformity surgery. In addition, spinal deformity surgeries represent some of the most costly spinal surgeries,11-13 with single-center studies showing average primary surgery costs over $100 00011 and revision surgery costs of ∼$55 000 per case.14 Work from the International Spine Study Group suggests that physician preferences may significantly impact operating room costs for spinal deformity surgery, potentially by driving implant and biological choices.11 To the best of our knowledge, however, there has not been a national analysis of the rates of spinal deformity surgery or the factors affecting its cost. The goal of our study was therefore to determine the utilization rates, total hospital costs, and factors contributing to cost of spinal fusion for adult spinal deformity using the 2001 to 2013 National Inpatient Sample (NIS) database, the largest all-payer inpatient care database in the United States. METHODS Study Population All patients ≥18 yr old who underwent spinal fusions for spinal deformity from 2001 to 2013 in the Healthcare Cost and Utilization Project NIS database were analyzed in this study. IRB approval was not required for this study. Patient informed consent was not obtained, as the NIS is a national database, and patients cannot be identified or contacted. The NIS approximates a 20% stratified sample of discharges from US community hospitals and represents 95% of the US population.15 A total of 55 599 patients met the inclusion and exclusion criteria specified below. Inclusion Criteria Spinal fusions for scoliosis/spinal deformity were defined using one of the following procedure codes and one of the following diagnostic codes (associated with the hospitalization for the given surgery). Inclusion ICD-9 procedure codes: 81.00 to 81.08 (spinal fusion), 81.62 (fusion or refusion of 2-3 vertebrae), 81.63 (fusion or refusion of 4-8 vertebrae), 81.64 (fusion or refusion of 9 or more vertebrae) Inclusion ICD-9 diagnosis codes: 737.0 to 737.9 (scoliosis and curvature of the spine), 738.2 (acquired deformity of neck), 738.5 (other acquired deformity of back or spine) Exclusion Criteria To ensure a homogeneous patient population undergoing primary fusion surgery for adult spinal deformity only, we excluded patients with the following diagnosis codes. Exclusion ICD-9 procedure codes: 81.30 to 81.39 (refusions) Exclusion ICD-9 diagnosis codes: 140 to 239.9 (all neoplasms), 324.1 (intraspinal abscess), 630 to 676 (pregnancy-related diagnoses), 720.0 to 720.9 (inflammatory spondyloarthropathies), 730 to 730.99 (osteomyelitis), 733.1, 733.10, 733.13 (pathological fractures), 733.8, 733.81 to 733.82 (nonunion/mal-union of fracture), 805 to 806.9 (fractures of spinal column), 839 to 839.59 (all vertebral dislocations), E800 to E849.9 (vehicular accidents), pseudarthrosis (733.82), 741.00 to 741.03 (spina bifida with hydrocephalus), 741.90 to 741.93 (spina bifida without mention of hydrocephalus), and 742.59 (other specified congenital anomalies of spinal cord). Collected Variables Patient Variables Patient variables included age, gender, race, insurance status, mean annual income of patient's zip code (<$39 000; $39 000-47 999; $48 000-62 999; >$63 000), population of patient's county of residence (central metropolitan county with population >1 million, fringe county of metro area with >1 million population, county in metro area with 250 000 to 999 000 population, county in metro area with 50 000 to 249 000 population, micropolitan county with <50 000 population, not metropolitan or micropolitan county), elective vs nonelective case, length of stay, number of comorbidities, All Patient Refined Disease Related Group (APR DRG) SOI (minor, moderate, major, extreme), and APR DRG risk of mortality (minor, moderate, major, extreme). APR DRG comorbidities include AIDS, alcohol abuse, chronic pulmonary disease, liver disease, obesity, valvular disease, etc. APR DRG SOI and risk of mortality scores are calculated using a proprietary formula from the principal and secondary diagnoses, procedure codes, age, sex, and discharge disposition of the patient. For example, a minor SOI would be uncomplicated diabetes, moderate is diabetes with renal manifestations, major is diabetes with ketoacidosis, and extreme SOI is diabetes with hyperosmolar coma. Hospital Variables Hospital variables included geographic region (Northeast, Midwest, South, West), location (rural/urban nonteaching, urban teaching), bed size (small, medium, large), and year-specific wage index. Surgical Variables Surgical variables included use of interbody fusion device (ICD-9 84.51 or 84.59) and use of recombinant bone morphogenic protein (BMP; ICD-9 84.52). Primary Outcome The primary outcome was total hospital cost (in US dollars) for each case, calculated by multiplying the total hospital charges by each hospital's year-specific unique all-payer inpatient cost-to-charge ratio. Costs were then adjusted for inflation (ie, all costs were expressed in 2013 dollars) using the US Consumer Price Index rates.16 Statistical Analysis We collected and sorted data in Stata 13 (StataCorp, College Station, Texas) and performed statistical analyses in R (v. 3.2.3). We performed univariate and then multivariate least-squares regressions (general linear model) to evaluate the effect of each patient and hospital variable on total cost for spinal deformity surgery. To analytically handle the complex design of the NIS, we applied the survey package in R (v. 3.30-3).17 For the general linear model, we used the gamma distribution because cost is a positive continuous variable that is right skewed. This distribution assumes that the standard deviation is proportional to the mean. A logistic link function improved the stability of the model and avoided the problem of negative predicted cost. Due to the multiplicative nature of our model, coefficients were interpreted as percent changes in cost. For example, a coefficient of 1.25 indicates the category is 25% more expensive than the reference category; a coefficient of 0.75 indicates this category is 25% less expensive than the reference. RESULTS From 2001 to 2013, 55 599 fusions were performed for adult spinal deformity in the NIS database. The average patient age was 59.9 (±15.4) yr old, and the majority of patients were white (72.6%) and female (67.2%; Table 1). Of the patients, 73.9% had a minor mortality risk, although the mean number of chronic medical conditions was 5.56 (±2.8; Table 1). Of spinal deformity patients, 53.9% had a moderate SOI, and the mean length of stay was 5.07 (±4.9) d (Table 1). The majority of fusions for spinal deformity were performed at hospitals with large bed sizes (64.7%) and at urban teaching hospitals (56.5%; Table 1). The greatest number of fusions for spinal deformity was performed in the southern United States (n = 22 012; Table 1). TABLE 1. Patient and Hospital Characteristics for 55 599 Fusions for Spinal Deformity in NIS Database from 2001 to 2013   Spinal deformity  Variables  (n = # of cases, % of total)  Patient and clinical characteristics    Age (Mean ± SD)  59.9 (±15.4)  Gender     Male  18 172 (32.7%)   Female  37 383 (67.2%)  Race     White  40 379 (72.6%)   Black  2240 (4.0%)   Hispanic  1742 (3.1%)   Other  1678 (3%)  Median income in patient zipcode     1st quartile: <$39k  8657 (15.6%)   2nd quartile: $39k-47 999  10 837 (19.5%)   3rd quartile: $48k-62 999  11 563 (20.8%)   4th quartile: >$63k  11 302 (20.3%)  Insurance     Private including HMO  23 880 (43.0%)   Medicare  26 103 (46.9%)   Medicaid  2220 (4.0%)   Other (including self-pay)  3313 (6.0%)  Mortality risk     Minor  41 080 (73.9%)   Moderate  8687 (15.6%)   Major/Extreme  3798 (6.8%)  Severity of illness     Minor  10 852 (19.5%)   Moderate  29 990 (53.9%)   Major/Extreme  12 723 (22.9%)  Patient transferred in     No  35 883 (64.5%)   Yes  294 (0.5%)  Length of Stay, days (± SD)  5.07 (±4.9)  # Chronic conditions, mean (± SD)  5.56 (±2.8)  Admission type     Nonelective  3502 (6.3%)   Elective  50 221 (90.3%)  Hospital characteristics    Bedsize     Small  7594 (13.7%)   Medium  11 804 (21.2%)   Large  35 949 (64.7%)  Wage index, mean (± SD)  1.02 (±0.15)  Hospital type     Urban nonteaching/Rural  23 907 (43.0%)   Urban teaching  31 440 (56.5%)  Hospital region     Northeast  7466 (13.4%)   Midwest  12 779 (23.0%)   South  22 012 (39.6%)   West  13 342 (24.0%)    Spinal deformity  Variables  (n = # of cases, % of total)  Patient and clinical characteristics    Age (Mean ± SD)  59.9 (±15.4)  Gender     Male  18 172 (32.7%)   Female  37 383 (67.2%)  Race     White  40 379 (72.6%)   Black  2240 (4.0%)   Hispanic  1742 (3.1%)   Other  1678 (3%)  Median income in patient zipcode     1st quartile: <$39k  8657 (15.6%)   2nd quartile: $39k-47 999  10 837 (19.5%)   3rd quartile: $48k-62 999  11 563 (20.8%)   4th quartile: >$63k  11 302 (20.3%)  Insurance     Private including HMO  23 880 (43.0%)   Medicare  26 103 (46.9%)   Medicaid  2220 (4.0%)   Other (including self-pay)  3313 (6.0%)  Mortality risk     Minor  41 080 (73.9%)   Moderate  8687 (15.6%)   Major/Extreme  3798 (6.8%)  Severity of illness     Minor  10 852 (19.5%)   Moderate  29 990 (53.9%)   Major/Extreme  12 723 (22.9%)  Patient transferred in     No  35 883 (64.5%)   Yes  294 (0.5%)  Length of Stay, days (± SD)  5.07 (±4.9)  # Chronic conditions, mean (± SD)  5.56 (±2.8)  Admission type     Nonelective  3502 (6.3%)   Elective  50 221 (90.3%)  Hospital characteristics    Bedsize     Small  7594 (13.7%)   Medium  11 804 (21.2%)   Large  35 949 (64.7%)  Wage index, mean (± SD)  1.02 (±0.15)  Hospital type     Urban nonteaching/Rural  23 907 (43.0%)   Urban teaching  31 440 (56.5%)  Hospital region     Northeast  7466 (13.4%)   Midwest  12 779 (23.0%)   South  22 012 (39.6%)   West  13 342 (24.0%)  Note that percentages do not always equal 100% due to missing data. SD = standard deviation. View Large The number of cases of fusions for spinal deformity increased 273%, from 1803 in 2001 to 6728 in 2013 (Table 2). Similarly, the rate of fusions for spinal deformity rose 234%, from 4.16 per 100 000 adults in 2001 to 13.9 per 100 000 adults in 2013 (Figure 1; Table 3). This significant increase in the number and rate of spinal deformity surgery cases mirrors the increase in lumbar fusions for degenerative conditions also seen in the NIS data and reported in our prior work18 (from 53.15 in 2001 to 77.47 per 100 000 adults in 2013; Figure 1; Table 3). During the same time period, the rate of lumbar laminectomy/discectomy procedures for degenerative conditions decreased from 34.06 to 19.74 per 100 000 adults (Figure 1; Table 3). FIGURE 1. View largeDownload slide Temporal trends in spinal surgery rates in National Inpatient Sample database from 2001 to 2013: lumbar fusion for degenerative conditions (green), lumbar laminectomy for degenerative conditions (blue), and spinal fusion for spinal deformity (red). FIGURE 1. View largeDownload slide Temporal trends in spinal surgery rates in National Inpatient Sample database from 2001 to 2013: lumbar fusion for degenerative conditions (green), lumbar laminectomy for degenerative conditions (blue), and spinal fusion for spinal deformity (red). TABLE 2. Mean Cost (standard error) in US Dollars for Spinal Fusion for Deformity in NIS Database from 2001 to 2013 Year  # Cases  US cost, mean  Northeast cost, mean  Midwest cost, mean  South cost, mean  West cost, mean  Spinal fusion    2001  1803  $32 671 ($1283)  $33 746 ($3004)  $30 660 ($2529)  $28 203 ($1681)  $39 402 ($2523)  2002  2102  $33 594 ($1330)  $32 482 ($2538)  $29 334 ($2785)  $31 316 ($1830)  $41 280 ($2481)  2003  2471  $33 652 ($1736)  $26 401 ($4387)  $34 343 ($4852)  $30 694 ($1688)  $41 378 ($2556)  2004  2820  $37 052 ($1470)  $37 196 ($3608)  $35 806 ($3916)  $34 269 ($1917)  $42 367 ($2542)  2005  3258  $41 588 ($1790)  $38 839 ($2533)  $45 000 ($3970)  $36 521 ($2646)  $45 851 ($3365)  2006  3321  $39 627 ($1370)  $43 265 ($2396)  $35 095 ($2433)  $36 061 ($1379)  $49 753 ($3337)  2007  3543  $41 272 ($1578)  $38 881 ($2721)  $43 691 ($4455)  $36 544 ($1455)  $47 523 ($2714)  2008  4893  $44 299 ($1617)  $48 395 ($5359)  $44 673 ($3750)  $38 800 ($2160)  $50 490 ($2459)  2009  5109  $45 666 ($2352)  $42 029 ($3191)  $41 391 ($2467)  $44 585 ($5223)  $54 855 ($3438)  2010  6281  $46 272 ($1795)  $45 768 ($4465)  $39 513 ($1915)  $48 068 ($2981)  $50 857 ($4292)  2011  6980  $44 758 ($1654)  $42 666 ($4230)  $42 499 ($4209)  $40 406 ($1828)  $54 538 ($3571)  2012  6290  $43 300 ($911)  $39 841 ($2028)  $39 896 ($1722)  $39 795 ($1240)  $53 871 ($2222)  2013  6728  $43 433 ($882)  $38 937 ($2240)  $40 573 ($1850)  $41 625 ($1338)  $51 490 ($1802)  Year  # Cases  US cost, mean  Northeast cost, mean  Midwest cost, mean  South cost, mean  West cost, mean  Spinal fusion    2001  1803  $32 671 ($1283)  $33 746 ($3004)  $30 660 ($2529)  $28 203 ($1681)  $39 402 ($2523)  2002  2102  $33 594 ($1330)  $32 482 ($2538)  $29 334 ($2785)  $31 316 ($1830)  $41 280 ($2481)  2003  2471  $33 652 ($1736)  $26 401 ($4387)  $34 343 ($4852)  $30 694 ($1688)  $41 378 ($2556)  2004  2820  $37 052 ($1470)  $37 196 ($3608)  $35 806 ($3916)  $34 269 ($1917)  $42 367 ($2542)  2005  3258  $41 588 ($1790)  $38 839 ($2533)  $45 000 ($3970)  $36 521 ($2646)  $45 851 ($3365)  2006  3321  $39 627 ($1370)  $43 265 ($2396)  $35 095 ($2433)  $36 061 ($1379)  $49 753 ($3337)  2007  3543  $41 272 ($1578)  $38 881 ($2721)  $43 691 ($4455)  $36 544 ($1455)  $47 523 ($2714)  2008  4893  $44 299 ($1617)  $48 395 ($5359)  $44 673 ($3750)  $38 800 ($2160)  $50 490 ($2459)  2009  5109  $45 666 ($2352)  $42 029 ($3191)  $41 391 ($2467)  $44 585 ($5223)  $54 855 ($3438)  2010  6281  $46 272 ($1795)  $45 768 ($4465)  $39 513 ($1915)  $48 068 ($2981)  $50 857 ($4292)  2011  6980  $44 758 ($1654)  $42 666 ($4230)  $42 499 ($4209)  $40 406 ($1828)  $54 538 ($3571)  2012  6290  $43 300 ($911)  $39 841 ($2028)  $39 896 ($1722)  $39 795 ($1240)  $53 871 ($2222)  2013  6728  $43 433 ($882)  $38 937 ($2240)  $40 573 ($1850)  $41 625 ($1338)  $51 490 ($1802)  View Large TABLE 3. Rates of spinal fusion for deformity, lumbar laminectomy for degenerative disease, and lumbar fusion for degenerative disease in the NIS database from 2001 to 2013   Rate of surgical cases per 100 000 adult population  Year  Spinal deformity  Lumbar laminectomy  Lumbar fusion  2001  4.16  34.06  53.15  2002  4.49  33.03  55.17  2003  5.25  35.27  58.61  2004  6.02  32.88  58.66  2005  6.94  33.00  63.58  2006  6.95  29.45  65.81  2007  7.27  29.88  65.97  2008  10.02  29.70  78.47  2009  10.68  28.830  80.82  2010  12.82  28.35  85.43  2011  13.69  24.93  82.25  2012  13.08  21.78  77.39  2013  13.85  19.74  77.47    Rate of surgical cases per 100 000 adult population  Year  Spinal deformity  Lumbar laminectomy  Lumbar fusion  2001  4.16  34.06  53.15  2002  4.49  33.03  55.17  2003  5.25  35.27  58.61  2004  6.02  32.88  58.66  2005  6.94  33.00  63.58  2006  6.95  29.45  65.81  2007  7.27  29.88  65.97  2008  10.02  29.70  78.47  2009  10.68  28.830  80.82  2010  12.82  28.35  85.43  2011  13.69  24.93  82.25  2012  13.08  21.78  77.39  2013  13.85  19.74  77.47  Note that lumbar laminectomy and lumbar fusion data are from Zygourakis et al, Neurosurgery 2016 (in press).18 View Large The mean inflation-adjusted total cost per spinal deformity surgery increased 25%, from $32 671 in 2001 to $43 433 in 2013 (Table 2). The western United States was the most expensive region across all years (Table 2; Figure 2). In 2001, the mean cost per case was $39 402 in the West compared to $28 203 in the least expensive region (the South). By 2013, the mean cost per spinal deformity case was $51 490 in the western United States vs $38 937 in the least expensive region, this time the Northeast (Table 2; Figure 2). FIGURE 2. View largeDownload slide Regional trends in inflation-adjusted mean cost for spinal fusion for deformity (n = 55 599) in National Inpatient Sample database from 2001 to 2013. FIGURE 2. View largeDownload slide Regional trends in inflation-adjusted mean cost for spinal fusion for deformity (n = 55 599) in National Inpatient Sample database from 2001 to 2013. In our univariate analyses, we found the following patient variables were associated with higher cost for spinal deformity surgery: Hispanic and “other” race (as compared to white), patient county of residence (central county metro area >1 million population vs all others), higher median income in patient zip code, Medicaid insurance (vs private insurance), higher patient mortality risk, higher patient SOI, greater number of chronic conditions, and longer length of stay (Table 4; all P < .01). In addition to these patient and clinical factors, the following hospital factors were also associated with higher cost for spinal deformity fusion: higher wage index, urban teaching (vs urban nonteaching/rural hospital), and the western region (vs the Northeast; Table 4; P < .01). TABLE 4. Univariate Analysis of Predictors of Total Cost for Spinal Fusion   Spinal deformity (n = 55 559)  Variables  Mean cost, $ (SE)  P-value  Patient and clinical factors    Age (yr, Quartiles)     Q1: <52  41 456 (853)     Q2: 52–62  44 371 (863)  <.01**   Q3: 62-71  44 504 (707)  <.01**   Q4: >71  39 013 (560)  <.01**  Gender     Male  41 708 (587)     Female  42 546 (732)  .08  Race     White  42 380 (728)     Black  40 438 (897)  .02*   Hispanic  46 297 (1839)  .02*   Other  47 870 (1333)  <.01**  Patient county of residence     Central county metro area >1 million pop  47 907 (1497)     Fringe county of metro area >1 million  42 277 (1310)  <.01**   County in metro area 250-999k  41 185 (1318)  <.01**   County in metro area 50-250k  45 098 (1539)  .16   Micropolitan county (<50k)  40 184 (1394)  <.01**   Not metropolitan or micropolitan  42 801 (1618)  .01*  Median income in patient zipcode     Q1: <$39k  41 089 (834)     Q2: $39k-47 999  42 393 (735)  .02*   Q3: $48k-62 999  44 681 (838)  <.01**   Q4: >$63k  46 765 (986)  <.01**  Insurance     Private  42 621 (900)     Medicare  41 657 (605)  .16   Medicaid  48 333 (1218)  <.01**   Other  40 604 (766)  .04*  Mortality risk     Minor  38 032 (513)     Moderate  49 455 (937)  <.01**   Major/Extreme  76 270 (1786)  <.01**  Severity of illness     Minor  31 399 (428)     Moderate  36 871 (449)  <.01**   Major/Extreme  65 547 (1395)  <.01**  Patient transferred in     No  44 402 (707)     Yes  66 846 (15 243)  .07  Length of stay (d, quartiles)     Q1: <3  26 751 (304)     Q2: 4  37 539 (536)  <.01**   Q3: 5–6  46 041 (849)  <.01**   Q4: >6  69 961 (1216)  <.01**  Elective admission     No  41 717 (1207)     Yes  42 679 (684)  .45  Chronic conditions (number, quartiles)     Q1: ≤4  40 986 (810)     Q2: 5  43 187 (847)  <.01**   Q3: 6–7  45 270 (846)  <.01**   Q4: >7  51 591 (1062)  <.01**  Hospital factors    Bedsize     Small  41 332 (1301)     Medium  43 012 (1723)  .43   Large  42 220 (771)  .56  Wage index (quartiles)     Q1: <0.9138  36 658 (1144)     Q2: 0.9151-0.9801  41 597 (1018)  <.01**   Q3: 0.9801-1.0945  42 560 (1371)  <.01**   Q4: >1.0945  48 648 (1175)  <.01**  Hospital type     Urban nonteaching/rural  40 083 (1027)     Urban teaching  43 968 (830)  <.01**  Hospital region     Northeast  39 962 (1235)     Midwest  39 861 (1390)  .96   South  39 803 (1108)  .92   West  49 978 (1224)  <.01**    Spinal deformity (n = 55 559)  Variables  Mean cost, $ (SE)  P-value  Patient and clinical factors    Age (yr, Quartiles)     Q1: <52  41 456 (853)     Q2: 52–62  44 371 (863)  <.01**   Q3: 62-71  44 504 (707)  <.01**   Q4: >71  39 013 (560)  <.01**  Gender     Male  41 708 (587)     Female  42 546 (732)  .08  Race     White  42 380 (728)     Black  40 438 (897)  .02*   Hispanic  46 297 (1839)  .02*   Other  47 870 (1333)  <.01**  Patient county of residence     Central county metro area >1 million pop  47 907 (1497)     Fringe county of metro area >1 million  42 277 (1310)  <.01**   County in metro area 250-999k  41 185 (1318)  <.01**   County in metro area 50-250k  45 098 (1539)  .16   Micropolitan county (<50k)  40 184 (1394)  <.01**   Not metropolitan or micropolitan  42 801 (1618)  .01*  Median income in patient zipcode     Q1: <$39k  41 089 (834)     Q2: $39k-47 999  42 393 (735)  .02*   Q3: $48k-62 999  44 681 (838)  <.01**   Q4: >$63k  46 765 (986)  <.01**  Insurance     Private  42 621 (900)     Medicare  41 657 (605)  .16   Medicaid  48 333 (1218)  <.01**   Other  40 604 (766)  .04*  Mortality risk     Minor  38 032 (513)     Moderate  49 455 (937)  <.01**   Major/Extreme  76 270 (1786)  <.01**  Severity of illness     Minor  31 399 (428)     Moderate  36 871 (449)  <.01**   Major/Extreme  65 547 (1395)  <.01**  Patient transferred in     No  44 402 (707)     Yes  66 846 (15 243)  .07  Length of stay (d, quartiles)     Q1: <3  26 751 (304)     Q2: 4  37 539 (536)  <.01**   Q3: 5–6  46 041 (849)  <.01**   Q4: >6  69 961 (1216)  <.01**  Elective admission     No  41 717 (1207)     Yes  42 679 (684)  .45  Chronic conditions (number, quartiles)     Q1: ≤4  40 986 (810)     Q2: 5  43 187 (847)  <.01**   Q3: 6–7  45 270 (846)  <.01**   Q4: >7  51 591 (1062)  <.01**  Hospital factors    Bedsize     Small  41 332 (1301)     Medium  43 012 (1723)  .43   Large  42 220 (771)  .56  Wage index (quartiles)     Q1: <0.9138  36 658 (1144)     Q2: 0.9151-0.9801  41 597 (1018)  <.01**   Q3: 0.9801-1.0945  42 560 (1371)  <.01**   Q4: >1.0945  48 648 (1175)  <.01**  Hospital type     Urban nonteaching/rural  40 083 (1027)     Urban teaching  43 968 (830)  <.01**  Hospital region     Northeast  39 962 (1235)     Midwest  39 861 (1390)  .96   South  39 803 (1108)  .92   West  49 978 (1224)  <.01**  *Indicates statistically significant with P-value < .05; **indicates statistically significant with P-value < .01. SE = standard error; Q = quartile. View Large After adjusting for covariates in regression analyses, the western region was 29% more expensive than the Northeast for spinal deformity surgery (Table 5). Higher hospital wage index and smaller bed size were also correlated with higher cost for spinal deformity surgery (Table 5; P < .05). In addition to these hospital factors, the following patient variables were significantly associated with cost for spinal deformity surgery: age, race, insurance, SOI, length of stay, and elective admission status (Table 5). Of these, SOI was the most significant driver of cost, where the cost for patients with major/extreme SOI was 52% greater than that for patients with minor SOI (P < .01; Table 4). TABLE 5. Multivariate Analysis of Predictors of Total Cost for Spinal Fusion for Deformity   Estimate  P-Value  Significance  Patient/Clinical factors        Age (10 yr increments)  0.97  <.01  **  Gender (female vs male)  1.00  .77    Race (vs white)     Black  0.95  <.01  **   Hispanic  0.99  .46     Other  1.01  .63    Insurance (vs private)     Medicare  0.96  <.01  **   Medicaid  0.93  <.01  **   Other  0.93  <.01  **  Mortality risk (vs minor)     Moderate  0.98  .04  *   Major/Extreme  0.97  .09    Severity of Illness (vs minor)     Moderate  1.15  <.01  **   Major/Extreme  1.52  <.01  **  Length of stay (d)  1.08  <.01  **  Elective admission (vs nonelective)  1.15  <.01  **  Hospital factors    Bedsize (vs small)     Medium  1.01  .72     Large  0.95  .04  *  Wage index (0.1 increments)  1.04  <.01  **  Hospital type (vs urban nonteaching/rural)     Urban teaching  1.00  .8    Hospital region (vs northeast)     Midwest region  1.17  <.01  **   South region  1.16  <.01  **   West region  1.29  <.01  **    Estimate  P-Value  Significance  Patient/Clinical factors        Age (10 yr increments)  0.97  <.01  **  Gender (female vs male)  1.00  .77    Race (vs white)     Black  0.95  <.01  **   Hispanic  0.99  .46     Other  1.01  .63    Insurance (vs private)     Medicare  0.96  <.01  **   Medicaid  0.93  <.01  **   Other  0.93  <.01  **  Mortality risk (vs minor)     Moderate  0.98  .04  *   Major/Extreme  0.97  .09    Severity of Illness (vs minor)     Moderate  1.15  <.01  **   Major/Extreme  1.52  <.01  **  Length of stay (d)  1.08  <.01  **  Elective admission (vs nonelective)  1.15  <.01  **  Hospital factors    Bedsize (vs small)     Medium  1.01  .72     Large  0.95  .04  *  Wage index (0.1 increments)  1.04  <.01  **  Hospital type (vs urban nonteaching/rural)     Urban teaching  1.00  .8    Hospital region (vs northeast)     Midwest region  1.17  <.01  **   South region  1.16  <.01  **   West region  1.29  <.01  **  *Indicates statistically significant with P-value < .05; **P-value < .01. View Large Finally, because implants represent a considerable portion of spinal deformity costs, we analyzed the utilization of interbody fusion devices and BMP over this period of time. Use of interbody fusion devices increased significantly, from 6.42% in 2002 to 61.58% of spinal deformity cases in 2013 (Table 6; Figure 3). BMP utilization, however, rose from 2002 to 2010 (0.9-40.58%), then declined to 22.76% of spinal deformity cases in 2013 (Table 6; Figure 3). FIGURE 3. View largeDownload slide Temporal trends of utilization of recombinant bone morphogenic protein (BMP) and interbody fusion devices. FIGURE 3. View largeDownload slide Temporal trends of utilization of recombinant bone morphogenic protein (BMP) and interbody fusion devices. TABLE 6. Utilization Rates of BMP and Interbody Fusion Devices in Spinal Deformity Cases in NIS Database, 2002 to 2013   BMP Utilization  Interbody Fusion Utilization  Year  # Cases  % Cases  # Cases  % Cases  2002  19  0.9  135  6.42  2003  241  9.75  676  27.36  2004  604  21.42  963  34.15  2005  905  27.78  1290  39.59  2006  1120  33.72  1459  43.93  2007  1280  36.13  1651  46.6  2008  1863  38.07  2507  51.24  2009  2139  41.87  2851  55.8  2010  2549  40.58  3513  55.93  2011  2181  31.25  4271  61.19  2012  1608  25.56  3734  59.36  2013  1531  22.76  4143  61.58    BMP Utilization  Interbody Fusion Utilization  Year  # Cases  % Cases  # Cases  % Cases  2002  19  0.9  135  6.42  2003  241  9.75  676  27.36  2004  604  21.42  963  34.15  2005  905  27.78  1290  39.59  2006  1120  33.72  1459  43.93  2007  1280  36.13  1651  46.6  2008  1863  38.07  2507  51.24  2009  2139  41.87  2851  55.8  2010  2549  40.58  3513  55.93  2011  2181  31.25  4271  61.19  2012  1608  25.56  3734  59.36  2013  1531  22.76  4143  61.58  View Large DISCUSSION Our analyses show the number of spinal fusions for deformity increased 273%, and the rate of spinal deformity surgery increased 234% from 2001 to 2013. The increasing rates of spinal deformity surgeries may be due to several factors, including the training of more surgeons in this subspecialty, a heightened focus on spinal alignment parameters in neurosurgical training,19 and the evolving body of research showing improvement in health-related quality of life with spinal deformity surgery.20-22 Due to the significant attention paid to adult spinal deformity in the scientific literature and at recent meetings and conferences, it is also possible that the increased rate of deformity surgery reflects the growing use of scoliosis/deformity diagnosis codes, which would previously have been coded differently. We also find that the mean cost per spinal deformity case increased 25% over this time period, similar to the cumulative US inflation rate of 31.6%.16 In addition, there may be factors specific to spinal deformity surgery that help explain the rise in costs. Spinal deformity cases have some of the highest revision rates in all of spine surgery,23 and the main causes include proximal junctional kyphosis, pseudarthrosis, and infection. To prevent such complications, deformity surgeons have recently added costly items like increased interbody fusions, BMP, vertebroplasty, and ligament augmentation to their armamentarium. In particular, BMP utilization has recently increased in most types of spine surgery,24 with the exception of anterior cervical fusion, in which BMP utilization decreased after the 2008 US Food and Drug Administration safety warning.25 The frequent use of BMP in spinal deformity may be largely driven by its association with lower rates of pseudarthrosis necessitating reoperation.26 However, while we see an increase in BMP utilization from 2002 to 2010, its use in spinal deformity cases decreased from 2010 to 2013, which may reflect backlash from the aforementioned safety concerns. In contrast, interbody fusion utilization has increased steadily from 2002 to 2013 and may contribute partially to increased surgical costs for these cases. Another significant finding is that several patient factors (namely, patient age, race, insurance, SOI, length of stay, and elective admission status) are significantly associated with cost for spinal deformity surgery. Hospitals with smaller bed sizes are more expensive for spinal deformity fusions, supporting the principle of economy of scale. This also suggests that a system that aims to deliver the highest value care may consider focusing its spinal deformity surgeries in large, urban hospitals with high volumes. Our analysis also reveals that the western United States is the most expensive region across all studied years, even after controlling for covariates. This confirms prior studies showing large geographic variation in cost for spine surgery in general.3,5-7 Although one group describes the highest costs for two spinal fusion procedures in the northeastern United States,6 several papers report the highest costs in the West, as we have found.7,27 We hypothesize that systems delivery issues, rather than implant price or surgeon practice patterns alone, may contribute to the higher costs for spinal deformity surgery in the western United States. For example, the wage index in western states like California may not adequately account for the significantly higher cost of labor, and regional policies like the minimum nurse-to-patient ratios mandated in California may further increase costs in these western locations.28 We used the NIS database for this study because it represents the most comprehensive national database of patients from all payers, with data from more than 7 million hospital stays a year, representing 95% of the US population.15 It is specifically designed to enable researchers and policymakers to make national estimates of health care utilization (with complex statistical design to avoid sampling error from different states) and to analyze trends over time, which is the goal of this paper. Because the government and insurers are making payment and coverage decisions based on these databases, it is essential that we as front-line neurosurgical providers understand this data, even if it is imperfect. Limitations The NIS database has several important limitations. As a national administrative database, it relies on ICD-9 coding data, which may be unreliable29,30 and at times heterogeneous. In particular, three of the codes utilized in this study (81.63, 81.63, and 81.64) include both primary fusions and refusions, which may reflect the fact that many deformity operations are performed after initial smaller fusion operations and are sometimes required to treat iatrogenic deformity. In addition, as with many databases, there is some missing information, particularly with regards to race and income, but the amount of missing data is fairly low. Although we convert charges into costs using cost-to-charge ratios, NIS costs still inherently rely on charge data, which may not be an accurate reflection of true cost. Finally, we cannot extract data on the number of levels of spinal surgery, which directly impacts the cost of spinal deformity surgery. However, we do not have any reason to believe that surgeons in the western United States are routinely operating on more levels than those in other parts of the country, although studies suggest different rates of utilization of spinal fusions in various parts of the country.8 Despite these limitations, our study is unique in that it represents, to the best of our knowledge, the first national analysis of the utilization rates and the cost drivers of spinal deformity surgery. Despite its limitations, the NIS database represents the most comprehensive national database of patients from all payers and is specifically designed to enable researchers and policymakers to analyze national health care utilization rates. This type of research is an essential step towards providing the most cost-effective and highest value neurosurgical care. CONCLUSION The number and rate of fusions for spinal deformity rose substantially from 2001 to 2013 in the NIS database. Utilization of interbody fusion rates increased steadily during this time period, while BMP usage peaked in 2010 and declined thereafter. The increased rate of spinal deformity surgery is likely influenced by several factors, including the expanding body of research demonstrating improvement in health-related quality of life measures with spinal deformity surgery. In addition, the mean cost per spinal deformity case increased at a pace similar to inflation. Both patient and hospital factors are important contributors to the variation in cost for spinal deformity surgery. After adjusting for covariates, hospitals in the western United States and those with higher wage indices or smaller bed sizes were significantly more expensive for adult spinal deformity surgery. Disclosures Dr Zygourakis has received travel grants from Nuvasive and Globus to attend resident education courses. Ms Keefe is a consultant for DePuy Spine. Dr Ratliff is a consultant and receives royalties and travel funding from Stryker. Dr Mummaneni has received grants and honoraria from AOSpine, honoraria from Globus and DePuy, royalties from Thieme, Springer, and Taylor and Francis publishers, and is a stock holder in Spinicity/ISD. Dr Ames has received grants and research funding from DePuy Synthes Spine, is a consultant for DePuy, Medtronic, and Stryker, is a stock/shareholder with Baxano and Doctor's Research Group, and receives royalties from Aesculap and Biomet Spine. Dr Zygourakis is supported by a research fellowship from the UCSF Center for Healthcare Value. The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Martin BI, Deyo RA, Mirza SK et al.   Expenditures and health status among adults with back and neck problems. JAMA . 2008; 299( 6): 656- 664. Google Scholar CrossRef Search ADS PubMed  2. Martin BI, Turner JA, Mirza SK, Lee MJ, Comstock BA, Deyo RA. Trends in health care expenditures, utilization, and health status among US adults with spine problems, 1997-2006. Spine (Phila Pa 1976) . 2009; 34( 19): 2077- 2084. Google Scholar CrossRef Search ADS PubMed  3. Ugiliweneza B, Kong M, Nosova K et al.   Spinal surgery: variations in health care costs and implications for episode-based bundled payments. Spine (Phila Pa 1976) . 2014; 39( 15): 1235- 1242. Google Scholar CrossRef Search ADS PubMed  4. Schoenfeld AJ, Harris MB, Liu H, Birkmeyer JD. Variations in Medicare payments for episodes of spine surgery. Spine J . 2014; 14( 12): 2793- 2798. Google Scholar CrossRef Search ADS PubMed  5. Missios S, Bekelis K. Hospitalization cost after spine surgery in the United States of America. J Clin Neurosci . 2015; 22( 10): 1632- 1637. Google Scholar CrossRef Search ADS PubMed  6. 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. Google Scholar CrossRef Search ADS PubMed  7. Cook C, Santos GC, Lima R, Pietrobon R, Jacobs DO, Richardson W. Geographic variation in lumbar fusion for degenerative disorders: 1990 to 2000. Spine J . 2007; 7( 5): 552- 557. Google Scholar CrossRef Search ADS PubMed  8. 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. Google Scholar CrossRef Search ADS PubMed  9. Robin GC, Span Y, Steinberg R, Makin M, Menczel J. Scoliosis in the elderly: a follow-up study. Spine (Phila Pa 1976) . 1982; 7( 4): 355- 359. Google Scholar CrossRef Search ADS PubMed  10. Schwab F, Dubey A, Gamez L et al.   Adult scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population. Spine (Phila Pa 1976) . 2005; 30( 9): 1082- 1085. Google Scholar CrossRef Search ADS PubMed  11. McCarthy IM, Hostin RA, Ames CP et al.   Total hospital costs of surgical treatment for adult spinal deformity: an extended follow-up study. Spine J . 2014; 14( 10): 2326- 2333. Google Scholar CrossRef Search ADS PubMed  12. McCarthy IM, Hostin RA, O’Brien MF et al.   Analysis of the direct cost of surgery for four diagnostic categories of adult spinal deformity. Spine J . 2013; 13( 12): 1843- 1848. Google Scholar CrossRef Search ADS PubMed  13. McCarthy I, Hostin R, O’Brien M, Saigal R, Ames CP. Health economic analysis of adult deformity surgery. Neurosurg Clin N Am . 2013; 24( 2): 293- 304. Google Scholar CrossRef Search ADS PubMed  14. Theologis AA, Miller L, Callahan M et al.   The economic impact of revision surgery for proximal junctional failure after adult spinal deformity surgery: a cost analysis of 57 operations in a 10-year experience at a major deformity center. Spine (Phila Pa 1976) . 2016; 41( 16): E964- E972. Google Scholar CrossRef Search ADS PubMed  15. Houchens RL, Ross DN, Elixhauser A, Jiang J. Nationwide Inpatient Sample (NIS) Redesign Final Report . 2014. HCUP NIS Related Reports ONLINE. U.S. Agency for Healthcare Research and Quality. Available at: http://www.hcupus.ahrq.gov/db/nation/nis/nisrelatedreports.jsp. Published April 4, 2014. Accessed March 27, 2017. 16. US Inflation Calculator. Consumer price index data from 1913 to 2016. Available at: http://www.usinflationcalculator.com/inflation/consumer-price-index-and-annual-percent-changes-from-1913-to-2008/. Accessed March 27, 2017. 17. Houchens RL, Ross D, Elixhauser A. Using the HCUP National Inpatient Sample to Estimate Trends . 2015. HCUP Methods Series Report # 2006-05 ONLINE. Available at: http://www.hcup-us.ahrq.gov/reports/methods/methods.jsp. Published January 4, 2016. Accessed March 27, 2017. 18. Zygourakis CC, Liu CY, Wakam G et al.   Geographic and hospital variation in cost of Lumbra laminectomy and lumbar fusion for degenerative conditions [published online ahead of print March 10, 2017]. Neurosurgery . 2017. doi: 10.1093/neuros/nyx047. 19. Ames CP, Smith JS, Scheer JK et al.   Impact of spinopelvic alignment on decision making in deformity surgery in adults: a review. J Neurosurg Spine . 2012; 16( 6): 547- 564. Google Scholar CrossRef Search ADS PubMed  20. Smith JS, Shaffrey CI, Glassman SD et al.   Risk-benefit assessment of surgery for adult scoliosis: an analysis based on patient age. Spine (Phila Pa 1976) . 2011; 36( 10): 817- 824. Google Scholar CrossRef Search ADS PubMed  21. Schwab F, Lafage V, Patel A, Farcy JP. Sagittal plane considerations and the pelvis in the adult patient. Spine (Phila Pa 1976) . 2009; 34( 17): 1828- 1833. Google Scholar CrossRef Search ADS PubMed  22. Schwab F, Patel A, Ungar B, Farcy JP, Lafage V. Adult spinal deformity-postoperative standing imbalance: how much can you tolerate? An overview of key parameters in assessing alignment and planning corrective surgery. Spine (Phila Pa 1976) . 2010; 35( 25): 2224- 2231. Google Scholar CrossRef Search ADS PubMed  23. Smith JS, Klineberg E, Lafage V et al.   Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery. J Neurosurg Spine . 2016; 25( 1): 1- 14. Google Scholar CrossRef Search ADS PubMed  24. Louie PK, Hassanzadeh H, Singh K. Epidemiologic trends in the utilization, demographics, and cost of bone morphogenetic protein in spinal fusions. Curr Rev Musculoskelet Med . 2014; 7( 3): 177- 181. Google Scholar CrossRef Search ADS PubMed  25. Singh K, Nandyala SV, Marquez-Lara A, Fineberg SJ. Epidemiological trends in the utilization of bone morphogenetic protein in spinal fusions from 2002 to 2011. Spine (Phila Pa 1976) . 2014; 39( 6): 491- 496. Google Scholar CrossRef Search ADS PubMed  26. Paul JC, Lonner BS, Vira S, Kaye ID, Errico TJ. Use of recombinant bone morphogenetic protein is associated with reduced risk of reoperation after spine fusion for adult spinal deformity. Spine (Phila Pa 1976) . 2016; 41( 1): E15- E21. Google Scholar CrossRef Search ADS PubMed  27. Thakore RV, Greenberg SE, Bulka CM, Ehrenfeld JM, Obremskey WT, Sethi MK. Geographic variations in hospital charges and Medicare payments for major joint arthroplasty. J Arthroplasty . 2015; 30( 5): 728- 732. Google Scholar CrossRef Search ADS PubMed  28. Coffman JM, Seago JA, Spetz J. Minimum nurse-to-patient ratios in acute care hospitals in California. Health Aff (Millwood) . 2002; 21( 5): 53- 64. Google Scholar CrossRef Search ADS PubMed  29. Faciszewski T, Johnson L, Noren C, Smith MD. Administrative databases' complication coding in anterior spinal fusion procedures. What does it mean? Spine (Phila Pa 1976) . 1995; 20( 16): 1783- 1788. Google Scholar CrossRef Search ADS PubMed  30. Gologorsky Y, Knightly JJ, Chi JH, Groff MW. The Nationwide Inpatient Sample database does not accurately reflect surgical indications for fusion. J Neurosurg Spine . 2014; 21( 6): 984- 993. Google Scholar CrossRef Search ADS PubMed  COMMENTS The authors propose an interesting study on the variation in cost for spinal deformity surgery in the United States. For this purpose, they utilized the National Inpatient Sample (NIS) database from 2001–2013 to determine the factors that contribute to the cost of spinal deformity surgery. The authors found that there was a significant increase in the number and rate of spinal deformity surgery cases and a parallel increase in the cost per case over the reported years. The western region was found to be 29% more expensive than the northeastern region for spinal deformity surgery in the United States. The authors subdivided the factors that affected cost into hospital factors and patient factors. After adjusting for covariates in regression analyses, hospitals of higher wage indices and smaller bed sizes were mostly associated with an increase in cost. Patients with higher severity of illness were also identified as a significant driver of cost. This article is well-written and novel in terms of using a national registry, the NIS database, to report the regional variation of cost per case of spinal deformity surgery. The advantage of using such database is that it represents a 20% stratified sample of all US community hospitals. The authors should be commended for this important addition to our understanding of the most significant factors that affect the cost of spinal deformity procedures. Moving forward, it will be interesting to utilize the findings of this study in future cost-effectiveness reports. Seba Ramhmdani Ali Bydon Baltimore, Maryland The authors of this analysis present further evidence on the increasing social burden that the United States has experienced in the first decade of this century. While there are deep inherent limitations to subject selection and the analysis of cost in these patients, it must be acknowledged that the major trends reported here are consistent with previous smaller scale evaluations and that an enhanced focus on the trends of this clinical problem is required. Peter G. Passias New York, New York Copyright © 2017 by the Congress of Neurological Surgeons http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Neurosurgery Oxford University Press

Loading next page...
 
/lp/ou_press/analysis-of-national-rates-cost-and-sources-of-cost-variation-in-adult-vRjW8ogwis
Publisher
Oxford University Press
Copyright
Copyright © 2017 by the Congress of Neurological Surgeons
ISSN
0148-396X
eISSN
1524-4040
D.O.I.
10.1093/neuros/nyx218
Publisher site
See Article on Publisher Site

Abstract

Abstract BACKGROUND Several studies suggest significant variation in cost for spine surgery, but there has been little research in this area for spinal deformity. OBJECTIVE To determine the utilization, cost, and factors contributing to cost for spinal deformity surgery. METHODS The cohort comprised 55 599 adults who underwent spinal deformity fusion in the 2001 to 2013 National Inpatient Sample database. Patient variables included age, gender, insurance, median income of zip code, county population, severity of illness, mortality risk, number of comorbidities, length of stay, elective vs nonelective case. Hospital variables included bed size, wage index, hospital type (rural, urban nonteaching, urban teaching), and geographical region. The outcome was total hospital cost for deformity surgery. Statistics included univariate and multivariate regression analyses. RESULTS The number of spinal deformity cases increased from 1803 in 2001 (rate: 4.16 per 100 000 adults) to 6728 in 2013 (rate: 13.9 per 100 000). Utilization of interbody fusion devices increased steadily during this time period, while bone morphogenic protein usage peaked in 2010 and declined thereafter. The mean inflation-adjusted case cost rose from $32 671 to $43 433 over the same time period. Multivariate analyses showed the following patient factors were associated with cost: age, race, insurance, severity of illness, length of stay, and elective admission (P < .01). Hospitals in the western United States and those with higher wage indices or smaller bed sizes were significantly more expensive (P < .05). CONCLUSION The rate of adult spinal deformity surgery and the mean case cost increased from 2001 to 2013, exceeding the rate of inflation. Both patient and hospital factors are important contributors to cost variation for spinal deformity surgery. Cost of surgery, Spinal fusion, Spinal deformity, Scoliosis, National inpatient sample (NIS) database, BMP, Interbody fusion devices ABBREVIATIONS ABBREVIATIONS BMP bone morphogenic protein NIS National Inpatient Sample SOI severity of illness The United States spent $86 billion on the treatment of back and neck conditions in 2005.1 National spending on spine disease increased at an average rate of 7.0% per year from 1997 to 2006 (adjusted for inflation) and has continued to climb.2 Several studies demonstrate significant variation in cost for spinal surgery.3,4 There are many factors that affect the cost of spinal procedures, including patient factors such as comorbidities and severity of illness (SOI),5 hospital factors like teaching status and bed size,5 and, of course, surgery-specific factors such as the number of levels fused. A handful of studies also suggest that geography impacts the cost and utilization of spinal surgery.6,7 Goz et al found that posterolateral fusion procedures were significantly more expensive in the northeastern United States than the Midwest based on 2012 Medicare data.6 Another study reported lumbar fusion rates to be highest in Idaho, Montana, Iowa, Florida, and Wyoming and lowest in Maine, Kentucky, Indiana, North Dakota, and New Jersey.8 Despite this analysis of cost variation for spine surgery in general,3,4,6,7 there has been relatively little research in the area of spinal deformity surgery. Recent work shows that spinal deformity is more prevalent than previously thought: approximately 32% of the adult population and more than 60% of the elderly US population suffer from scoliosis.9,10 Anecdotally, the number of spinal deformity surgeries in the United States appears to be rising, as more surgeons are trained in this subspecialty, and research reveals additional indications for deformity surgery. In addition, spinal deformity surgeries represent some of the most costly spinal surgeries,11-13 with single-center studies showing average primary surgery costs over $100 00011 and revision surgery costs of ∼$55 000 per case.14 Work from the International Spine Study Group suggests that physician preferences may significantly impact operating room costs for spinal deformity surgery, potentially by driving implant and biological choices.11 To the best of our knowledge, however, there has not been a national analysis of the rates of spinal deformity surgery or the factors affecting its cost. The goal of our study was therefore to determine the utilization rates, total hospital costs, and factors contributing to cost of spinal fusion for adult spinal deformity using the 2001 to 2013 National Inpatient Sample (NIS) database, the largest all-payer inpatient care database in the United States. METHODS Study Population All patients ≥18 yr old who underwent spinal fusions for spinal deformity from 2001 to 2013 in the Healthcare Cost and Utilization Project NIS database were analyzed in this study. IRB approval was not required for this study. Patient informed consent was not obtained, as the NIS is a national database, and patients cannot be identified or contacted. The NIS approximates a 20% stratified sample of discharges from US community hospitals and represents 95% of the US population.15 A total of 55 599 patients met the inclusion and exclusion criteria specified below. Inclusion Criteria Spinal fusions for scoliosis/spinal deformity were defined using one of the following procedure codes and one of the following diagnostic codes (associated with the hospitalization for the given surgery). Inclusion ICD-9 procedure codes: 81.00 to 81.08 (spinal fusion), 81.62 (fusion or refusion of 2-3 vertebrae), 81.63 (fusion or refusion of 4-8 vertebrae), 81.64 (fusion or refusion of 9 or more vertebrae) Inclusion ICD-9 diagnosis codes: 737.0 to 737.9 (scoliosis and curvature of the spine), 738.2 (acquired deformity of neck), 738.5 (other acquired deformity of back or spine) Exclusion Criteria To ensure a homogeneous patient population undergoing primary fusion surgery for adult spinal deformity only, we excluded patients with the following diagnosis codes. Exclusion ICD-9 procedure codes: 81.30 to 81.39 (refusions) Exclusion ICD-9 diagnosis codes: 140 to 239.9 (all neoplasms), 324.1 (intraspinal abscess), 630 to 676 (pregnancy-related diagnoses), 720.0 to 720.9 (inflammatory spondyloarthropathies), 730 to 730.99 (osteomyelitis), 733.1, 733.10, 733.13 (pathological fractures), 733.8, 733.81 to 733.82 (nonunion/mal-union of fracture), 805 to 806.9 (fractures of spinal column), 839 to 839.59 (all vertebral dislocations), E800 to E849.9 (vehicular accidents), pseudarthrosis (733.82), 741.00 to 741.03 (spina bifida with hydrocephalus), 741.90 to 741.93 (spina bifida without mention of hydrocephalus), and 742.59 (other specified congenital anomalies of spinal cord). Collected Variables Patient Variables Patient variables included age, gender, race, insurance status, mean annual income of patient's zip code (<$39 000; $39 000-47 999; $48 000-62 999; >$63 000), population of patient's county of residence (central metropolitan county with population >1 million, fringe county of metro area with >1 million population, county in metro area with 250 000 to 999 000 population, county in metro area with 50 000 to 249 000 population, micropolitan county with <50 000 population, not metropolitan or micropolitan county), elective vs nonelective case, length of stay, number of comorbidities, All Patient Refined Disease Related Group (APR DRG) SOI (minor, moderate, major, extreme), and APR DRG risk of mortality (minor, moderate, major, extreme). APR DRG comorbidities include AIDS, alcohol abuse, chronic pulmonary disease, liver disease, obesity, valvular disease, etc. APR DRG SOI and risk of mortality scores are calculated using a proprietary formula from the principal and secondary diagnoses, procedure codes, age, sex, and discharge disposition of the patient. For example, a minor SOI would be uncomplicated diabetes, moderate is diabetes with renal manifestations, major is diabetes with ketoacidosis, and extreme SOI is diabetes with hyperosmolar coma. Hospital Variables Hospital variables included geographic region (Northeast, Midwest, South, West), location (rural/urban nonteaching, urban teaching), bed size (small, medium, large), and year-specific wage index. Surgical Variables Surgical variables included use of interbody fusion device (ICD-9 84.51 or 84.59) and use of recombinant bone morphogenic protein (BMP; ICD-9 84.52). Primary Outcome The primary outcome was total hospital cost (in US dollars) for each case, calculated by multiplying the total hospital charges by each hospital's year-specific unique all-payer inpatient cost-to-charge ratio. Costs were then adjusted for inflation (ie, all costs were expressed in 2013 dollars) using the US Consumer Price Index rates.16 Statistical Analysis We collected and sorted data in Stata 13 (StataCorp, College Station, Texas) and performed statistical analyses in R (v. 3.2.3). We performed univariate and then multivariate least-squares regressions (general linear model) to evaluate the effect of each patient and hospital variable on total cost for spinal deformity surgery. To analytically handle the complex design of the NIS, we applied the survey package in R (v. 3.30-3).17 For the general linear model, we used the gamma distribution because cost is a positive continuous variable that is right skewed. This distribution assumes that the standard deviation is proportional to the mean. A logistic link function improved the stability of the model and avoided the problem of negative predicted cost. Due to the multiplicative nature of our model, coefficients were interpreted as percent changes in cost. For example, a coefficient of 1.25 indicates the category is 25% more expensive than the reference category; a coefficient of 0.75 indicates this category is 25% less expensive than the reference. RESULTS From 2001 to 2013, 55 599 fusions were performed for adult spinal deformity in the NIS database. The average patient age was 59.9 (±15.4) yr old, and the majority of patients were white (72.6%) and female (67.2%; Table 1). Of the patients, 73.9% had a minor mortality risk, although the mean number of chronic medical conditions was 5.56 (±2.8; Table 1). Of spinal deformity patients, 53.9% had a moderate SOI, and the mean length of stay was 5.07 (±4.9) d (Table 1). The majority of fusions for spinal deformity were performed at hospitals with large bed sizes (64.7%) and at urban teaching hospitals (56.5%; Table 1). The greatest number of fusions for spinal deformity was performed in the southern United States (n = 22 012; Table 1). TABLE 1. Patient and Hospital Characteristics for 55 599 Fusions for Spinal Deformity in NIS Database from 2001 to 2013   Spinal deformity  Variables  (n = # of cases, % of total)  Patient and clinical characteristics    Age (Mean ± SD)  59.9 (±15.4)  Gender     Male  18 172 (32.7%)   Female  37 383 (67.2%)  Race     White  40 379 (72.6%)   Black  2240 (4.0%)   Hispanic  1742 (3.1%)   Other  1678 (3%)  Median income in patient zipcode     1st quartile: <$39k  8657 (15.6%)   2nd quartile: $39k-47 999  10 837 (19.5%)   3rd quartile: $48k-62 999  11 563 (20.8%)   4th quartile: >$63k  11 302 (20.3%)  Insurance     Private including HMO  23 880 (43.0%)   Medicare  26 103 (46.9%)   Medicaid  2220 (4.0%)   Other (including self-pay)  3313 (6.0%)  Mortality risk     Minor  41 080 (73.9%)   Moderate  8687 (15.6%)   Major/Extreme  3798 (6.8%)  Severity of illness     Minor  10 852 (19.5%)   Moderate  29 990 (53.9%)   Major/Extreme  12 723 (22.9%)  Patient transferred in     No  35 883 (64.5%)   Yes  294 (0.5%)  Length of Stay, days (± SD)  5.07 (±4.9)  # Chronic conditions, mean (± SD)  5.56 (±2.8)  Admission type     Nonelective  3502 (6.3%)   Elective  50 221 (90.3%)  Hospital characteristics    Bedsize     Small  7594 (13.7%)   Medium  11 804 (21.2%)   Large  35 949 (64.7%)  Wage index, mean (± SD)  1.02 (±0.15)  Hospital type     Urban nonteaching/Rural  23 907 (43.0%)   Urban teaching  31 440 (56.5%)  Hospital region     Northeast  7466 (13.4%)   Midwest  12 779 (23.0%)   South  22 012 (39.6%)   West  13 342 (24.0%)    Spinal deformity  Variables  (n = # of cases, % of total)  Patient and clinical characteristics    Age (Mean ± SD)  59.9 (±15.4)  Gender     Male  18 172 (32.7%)   Female  37 383 (67.2%)  Race     White  40 379 (72.6%)   Black  2240 (4.0%)   Hispanic  1742 (3.1%)   Other  1678 (3%)  Median income in patient zipcode     1st quartile: <$39k  8657 (15.6%)   2nd quartile: $39k-47 999  10 837 (19.5%)   3rd quartile: $48k-62 999  11 563 (20.8%)   4th quartile: >$63k  11 302 (20.3%)  Insurance     Private including HMO  23 880 (43.0%)   Medicare  26 103 (46.9%)   Medicaid  2220 (4.0%)   Other (including self-pay)  3313 (6.0%)  Mortality risk     Minor  41 080 (73.9%)   Moderate  8687 (15.6%)   Major/Extreme  3798 (6.8%)  Severity of illness     Minor  10 852 (19.5%)   Moderate  29 990 (53.9%)   Major/Extreme  12 723 (22.9%)  Patient transferred in     No  35 883 (64.5%)   Yes  294 (0.5%)  Length of Stay, days (± SD)  5.07 (±4.9)  # Chronic conditions, mean (± SD)  5.56 (±2.8)  Admission type     Nonelective  3502 (6.3%)   Elective  50 221 (90.3%)  Hospital characteristics    Bedsize     Small  7594 (13.7%)   Medium  11 804 (21.2%)   Large  35 949 (64.7%)  Wage index, mean (± SD)  1.02 (±0.15)  Hospital type     Urban nonteaching/Rural  23 907 (43.0%)   Urban teaching  31 440 (56.5%)  Hospital region     Northeast  7466 (13.4%)   Midwest  12 779 (23.0%)   South  22 012 (39.6%)   West  13 342 (24.0%)  Note that percentages do not always equal 100% due to missing data. SD = standard deviation. View Large The number of cases of fusions for spinal deformity increased 273%, from 1803 in 2001 to 6728 in 2013 (Table 2). Similarly, the rate of fusions for spinal deformity rose 234%, from 4.16 per 100 000 adults in 2001 to 13.9 per 100 000 adults in 2013 (Figure 1; Table 3). This significant increase in the number and rate of spinal deformity surgery cases mirrors the increase in lumbar fusions for degenerative conditions also seen in the NIS data and reported in our prior work18 (from 53.15 in 2001 to 77.47 per 100 000 adults in 2013; Figure 1; Table 3). During the same time period, the rate of lumbar laminectomy/discectomy procedures for degenerative conditions decreased from 34.06 to 19.74 per 100 000 adults (Figure 1; Table 3). FIGURE 1. View largeDownload slide Temporal trends in spinal surgery rates in National Inpatient Sample database from 2001 to 2013: lumbar fusion for degenerative conditions (green), lumbar laminectomy for degenerative conditions (blue), and spinal fusion for spinal deformity (red). FIGURE 1. View largeDownload slide Temporal trends in spinal surgery rates in National Inpatient Sample database from 2001 to 2013: lumbar fusion for degenerative conditions (green), lumbar laminectomy for degenerative conditions (blue), and spinal fusion for spinal deformity (red). TABLE 2. Mean Cost (standard error) in US Dollars for Spinal Fusion for Deformity in NIS Database from 2001 to 2013 Year  # Cases  US cost, mean  Northeast cost, mean  Midwest cost, mean  South cost, mean  West cost, mean  Spinal fusion    2001  1803  $32 671 ($1283)  $33 746 ($3004)  $30 660 ($2529)  $28 203 ($1681)  $39 402 ($2523)  2002  2102  $33 594 ($1330)  $32 482 ($2538)  $29 334 ($2785)  $31 316 ($1830)  $41 280 ($2481)  2003  2471  $33 652 ($1736)  $26 401 ($4387)  $34 343 ($4852)  $30 694 ($1688)  $41 378 ($2556)  2004  2820  $37 052 ($1470)  $37 196 ($3608)  $35 806 ($3916)  $34 269 ($1917)  $42 367 ($2542)  2005  3258  $41 588 ($1790)  $38 839 ($2533)  $45 000 ($3970)  $36 521 ($2646)  $45 851 ($3365)  2006  3321  $39 627 ($1370)  $43 265 ($2396)  $35 095 ($2433)  $36 061 ($1379)  $49 753 ($3337)  2007  3543  $41 272 ($1578)  $38 881 ($2721)  $43 691 ($4455)  $36 544 ($1455)  $47 523 ($2714)  2008  4893  $44 299 ($1617)  $48 395 ($5359)  $44 673 ($3750)  $38 800 ($2160)  $50 490 ($2459)  2009  5109  $45 666 ($2352)  $42 029 ($3191)  $41 391 ($2467)  $44 585 ($5223)  $54 855 ($3438)  2010  6281  $46 272 ($1795)  $45 768 ($4465)  $39 513 ($1915)  $48 068 ($2981)  $50 857 ($4292)  2011  6980  $44 758 ($1654)  $42 666 ($4230)  $42 499 ($4209)  $40 406 ($1828)  $54 538 ($3571)  2012  6290  $43 300 ($911)  $39 841 ($2028)  $39 896 ($1722)  $39 795 ($1240)  $53 871 ($2222)  2013  6728  $43 433 ($882)  $38 937 ($2240)  $40 573 ($1850)  $41 625 ($1338)  $51 490 ($1802)  Year  # Cases  US cost, mean  Northeast cost, mean  Midwest cost, mean  South cost, mean  West cost, mean  Spinal fusion    2001  1803  $32 671 ($1283)  $33 746 ($3004)  $30 660 ($2529)  $28 203 ($1681)  $39 402 ($2523)  2002  2102  $33 594 ($1330)  $32 482 ($2538)  $29 334 ($2785)  $31 316 ($1830)  $41 280 ($2481)  2003  2471  $33 652 ($1736)  $26 401 ($4387)  $34 343 ($4852)  $30 694 ($1688)  $41 378 ($2556)  2004  2820  $37 052 ($1470)  $37 196 ($3608)  $35 806 ($3916)  $34 269 ($1917)  $42 367 ($2542)  2005  3258  $41 588 ($1790)  $38 839 ($2533)  $45 000 ($3970)  $36 521 ($2646)  $45 851 ($3365)  2006  3321  $39 627 ($1370)  $43 265 ($2396)  $35 095 ($2433)  $36 061 ($1379)  $49 753 ($3337)  2007  3543  $41 272 ($1578)  $38 881 ($2721)  $43 691 ($4455)  $36 544 ($1455)  $47 523 ($2714)  2008  4893  $44 299 ($1617)  $48 395 ($5359)  $44 673 ($3750)  $38 800 ($2160)  $50 490 ($2459)  2009  5109  $45 666 ($2352)  $42 029 ($3191)  $41 391 ($2467)  $44 585 ($5223)  $54 855 ($3438)  2010  6281  $46 272 ($1795)  $45 768 ($4465)  $39 513 ($1915)  $48 068 ($2981)  $50 857 ($4292)  2011  6980  $44 758 ($1654)  $42 666 ($4230)  $42 499 ($4209)  $40 406 ($1828)  $54 538 ($3571)  2012  6290  $43 300 ($911)  $39 841 ($2028)  $39 896 ($1722)  $39 795 ($1240)  $53 871 ($2222)  2013  6728  $43 433 ($882)  $38 937 ($2240)  $40 573 ($1850)  $41 625 ($1338)  $51 490 ($1802)  View Large TABLE 3. Rates of spinal fusion for deformity, lumbar laminectomy for degenerative disease, and lumbar fusion for degenerative disease in the NIS database from 2001 to 2013   Rate of surgical cases per 100 000 adult population  Year  Spinal deformity  Lumbar laminectomy  Lumbar fusion  2001  4.16  34.06  53.15  2002  4.49  33.03  55.17  2003  5.25  35.27  58.61  2004  6.02  32.88  58.66  2005  6.94  33.00  63.58  2006  6.95  29.45  65.81  2007  7.27  29.88  65.97  2008  10.02  29.70  78.47  2009  10.68  28.830  80.82  2010  12.82  28.35  85.43  2011  13.69  24.93  82.25  2012  13.08  21.78  77.39  2013  13.85  19.74  77.47    Rate of surgical cases per 100 000 adult population  Year  Spinal deformity  Lumbar laminectomy  Lumbar fusion  2001  4.16  34.06  53.15  2002  4.49  33.03  55.17  2003  5.25  35.27  58.61  2004  6.02  32.88  58.66  2005  6.94  33.00  63.58  2006  6.95  29.45  65.81  2007  7.27  29.88  65.97  2008  10.02  29.70  78.47  2009  10.68  28.830  80.82  2010  12.82  28.35  85.43  2011  13.69  24.93  82.25  2012  13.08  21.78  77.39  2013  13.85  19.74  77.47  Note that lumbar laminectomy and lumbar fusion data are from Zygourakis et al, Neurosurgery 2016 (in press).18 View Large The mean inflation-adjusted total cost per spinal deformity surgery increased 25%, from $32 671 in 2001 to $43 433 in 2013 (Table 2). The western United States was the most expensive region across all years (Table 2; Figure 2). In 2001, the mean cost per case was $39 402 in the West compared to $28 203 in the least expensive region (the South). By 2013, the mean cost per spinal deformity case was $51 490 in the western United States vs $38 937 in the least expensive region, this time the Northeast (Table 2; Figure 2). FIGURE 2. View largeDownload slide Regional trends in inflation-adjusted mean cost for spinal fusion for deformity (n = 55 599) in National Inpatient Sample database from 2001 to 2013. FIGURE 2. View largeDownload slide Regional trends in inflation-adjusted mean cost for spinal fusion for deformity (n = 55 599) in National Inpatient Sample database from 2001 to 2013. In our univariate analyses, we found the following patient variables were associated with higher cost for spinal deformity surgery: Hispanic and “other” race (as compared to white), patient county of residence (central county metro area >1 million population vs all others), higher median income in patient zip code, Medicaid insurance (vs private insurance), higher patient mortality risk, higher patient SOI, greater number of chronic conditions, and longer length of stay (Table 4; all P < .01). In addition to these patient and clinical factors, the following hospital factors were also associated with higher cost for spinal deformity fusion: higher wage index, urban teaching (vs urban nonteaching/rural hospital), and the western region (vs the Northeast; Table 4; P < .01). TABLE 4. Univariate Analysis of Predictors of Total Cost for Spinal Fusion   Spinal deformity (n = 55 559)  Variables  Mean cost, $ (SE)  P-value  Patient and clinical factors    Age (yr, Quartiles)     Q1: <52  41 456 (853)     Q2: 52–62  44 371 (863)  <.01**   Q3: 62-71  44 504 (707)  <.01**   Q4: >71  39 013 (560)  <.01**  Gender     Male  41 708 (587)     Female  42 546 (732)  .08  Race     White  42 380 (728)     Black  40 438 (897)  .02*   Hispanic  46 297 (1839)  .02*   Other  47 870 (1333)  <.01**  Patient county of residence     Central county metro area >1 million pop  47 907 (1497)     Fringe county of metro area >1 million  42 277 (1310)  <.01**   County in metro area 250-999k  41 185 (1318)  <.01**   County in metro area 50-250k  45 098 (1539)  .16   Micropolitan county (<50k)  40 184 (1394)  <.01**   Not metropolitan or micropolitan  42 801 (1618)  .01*  Median income in patient zipcode     Q1: <$39k  41 089 (834)     Q2: $39k-47 999  42 393 (735)  .02*   Q3: $48k-62 999  44 681 (838)  <.01**   Q4: >$63k  46 765 (986)  <.01**  Insurance     Private  42 621 (900)     Medicare  41 657 (605)  .16   Medicaid  48 333 (1218)  <.01**   Other  40 604 (766)  .04*  Mortality risk     Minor  38 032 (513)     Moderate  49 455 (937)  <.01**   Major/Extreme  76 270 (1786)  <.01**  Severity of illness     Minor  31 399 (428)     Moderate  36 871 (449)  <.01**   Major/Extreme  65 547 (1395)  <.01**  Patient transferred in     No  44 402 (707)     Yes  66 846 (15 243)  .07  Length of stay (d, quartiles)     Q1: <3  26 751 (304)     Q2: 4  37 539 (536)  <.01**   Q3: 5–6  46 041 (849)  <.01**   Q4: >6  69 961 (1216)  <.01**  Elective admission     No  41 717 (1207)     Yes  42 679 (684)  .45  Chronic conditions (number, quartiles)     Q1: ≤4  40 986 (810)     Q2: 5  43 187 (847)  <.01**   Q3: 6–7  45 270 (846)  <.01**   Q4: >7  51 591 (1062)  <.01**  Hospital factors    Bedsize     Small  41 332 (1301)     Medium  43 012 (1723)  .43   Large  42 220 (771)  .56  Wage index (quartiles)     Q1: <0.9138  36 658 (1144)     Q2: 0.9151-0.9801  41 597 (1018)  <.01**   Q3: 0.9801-1.0945  42 560 (1371)  <.01**   Q4: >1.0945  48 648 (1175)  <.01**  Hospital type     Urban nonteaching/rural  40 083 (1027)     Urban teaching  43 968 (830)  <.01**  Hospital region     Northeast  39 962 (1235)     Midwest  39 861 (1390)  .96   South  39 803 (1108)  .92   West  49 978 (1224)  <.01**    Spinal deformity (n = 55 559)  Variables  Mean cost, $ (SE)  P-value  Patient and clinical factors    Age (yr, Quartiles)     Q1: <52  41 456 (853)     Q2: 52–62  44 371 (863)  <.01**   Q3: 62-71  44 504 (707)  <.01**   Q4: >71  39 013 (560)  <.01**  Gender     Male  41 708 (587)     Female  42 546 (732)  .08  Race     White  42 380 (728)     Black  40 438 (897)  .02*   Hispanic  46 297 (1839)  .02*   Other  47 870 (1333)  <.01**  Patient county of residence     Central county metro area >1 million pop  47 907 (1497)     Fringe county of metro area >1 million  42 277 (1310)  <.01**   County in metro area 250-999k  41 185 (1318)  <.01**   County in metro area 50-250k  45 098 (1539)  .16   Micropolitan county (<50k)  40 184 (1394)  <.01**   Not metropolitan or micropolitan  42 801 (1618)  .01*  Median income in patient zipcode     Q1: <$39k  41 089 (834)     Q2: $39k-47 999  42 393 (735)  .02*   Q3: $48k-62 999  44 681 (838)  <.01**   Q4: >$63k  46 765 (986)  <.01**  Insurance     Private  42 621 (900)     Medicare  41 657 (605)  .16   Medicaid  48 333 (1218)  <.01**   Other  40 604 (766)  .04*  Mortality risk     Minor  38 032 (513)     Moderate  49 455 (937)  <.01**   Major/Extreme  76 270 (1786)  <.01**  Severity of illness     Minor  31 399 (428)     Moderate  36 871 (449)  <.01**   Major/Extreme  65 547 (1395)  <.01**  Patient transferred in     No  44 402 (707)     Yes  66 846 (15 243)  .07  Length of stay (d, quartiles)     Q1: <3  26 751 (304)     Q2: 4  37 539 (536)  <.01**   Q3: 5–6  46 041 (849)  <.01**   Q4: >6  69 961 (1216)  <.01**  Elective admission     No  41 717 (1207)     Yes  42 679 (684)  .45  Chronic conditions (number, quartiles)     Q1: ≤4  40 986 (810)     Q2: 5  43 187 (847)  <.01**   Q3: 6–7  45 270 (846)  <.01**   Q4: >7  51 591 (1062)  <.01**  Hospital factors    Bedsize     Small  41 332 (1301)     Medium  43 012 (1723)  .43   Large  42 220 (771)  .56  Wage index (quartiles)     Q1: <0.9138  36 658 (1144)     Q2: 0.9151-0.9801  41 597 (1018)  <.01**   Q3: 0.9801-1.0945  42 560 (1371)  <.01**   Q4: >1.0945  48 648 (1175)  <.01**  Hospital type     Urban nonteaching/rural  40 083 (1027)     Urban teaching  43 968 (830)  <.01**  Hospital region     Northeast  39 962 (1235)     Midwest  39 861 (1390)  .96   South  39 803 (1108)  .92   West  49 978 (1224)  <.01**  *Indicates statistically significant with P-value < .05; **indicates statistically significant with P-value < .01. SE = standard error; Q = quartile. View Large After adjusting for covariates in regression analyses, the western region was 29% more expensive than the Northeast for spinal deformity surgery (Table 5). Higher hospital wage index and smaller bed size were also correlated with higher cost for spinal deformity surgery (Table 5; P < .05). In addition to these hospital factors, the following patient variables were significantly associated with cost for spinal deformity surgery: age, race, insurance, SOI, length of stay, and elective admission status (Table 5). Of these, SOI was the most significant driver of cost, where the cost for patients with major/extreme SOI was 52% greater than that for patients with minor SOI (P < .01; Table 4). TABLE 5. Multivariate Analysis of Predictors of Total Cost for Spinal Fusion for Deformity   Estimate  P-Value  Significance  Patient/Clinical factors        Age (10 yr increments)  0.97  <.01  **  Gender (female vs male)  1.00  .77    Race (vs white)     Black  0.95  <.01  **   Hispanic  0.99  .46     Other  1.01  .63    Insurance (vs private)     Medicare  0.96  <.01  **   Medicaid  0.93  <.01  **   Other  0.93  <.01  **  Mortality risk (vs minor)     Moderate  0.98  .04  *   Major/Extreme  0.97  .09    Severity of Illness (vs minor)     Moderate  1.15  <.01  **   Major/Extreme  1.52  <.01  **  Length of stay (d)  1.08  <.01  **  Elective admission (vs nonelective)  1.15  <.01  **  Hospital factors    Bedsize (vs small)     Medium  1.01  .72     Large  0.95  .04  *  Wage index (0.1 increments)  1.04  <.01  **  Hospital type (vs urban nonteaching/rural)     Urban teaching  1.00  .8    Hospital region (vs northeast)     Midwest region  1.17  <.01  **   South region  1.16  <.01  **   West region  1.29  <.01  **    Estimate  P-Value  Significance  Patient/Clinical factors        Age (10 yr increments)  0.97  <.01  **  Gender (female vs male)  1.00  .77    Race (vs white)     Black  0.95  <.01  **   Hispanic  0.99  .46     Other  1.01  .63    Insurance (vs private)     Medicare  0.96  <.01  **   Medicaid  0.93  <.01  **   Other  0.93  <.01  **  Mortality risk (vs minor)     Moderate  0.98  .04  *   Major/Extreme  0.97  .09    Severity of Illness (vs minor)     Moderate  1.15  <.01  **   Major/Extreme  1.52  <.01  **  Length of stay (d)  1.08  <.01  **  Elective admission (vs nonelective)  1.15  <.01  **  Hospital factors    Bedsize (vs small)     Medium  1.01  .72     Large  0.95  .04  *  Wage index (0.1 increments)  1.04  <.01  **  Hospital type (vs urban nonteaching/rural)     Urban teaching  1.00  .8    Hospital region (vs northeast)     Midwest region  1.17  <.01  **   South region  1.16  <.01  **   West region  1.29  <.01  **  *Indicates statistically significant with P-value < .05; **P-value < .01. View Large Finally, because implants represent a considerable portion of spinal deformity costs, we analyzed the utilization of interbody fusion devices and BMP over this period of time. Use of interbody fusion devices increased significantly, from 6.42% in 2002 to 61.58% of spinal deformity cases in 2013 (Table 6; Figure 3). BMP utilization, however, rose from 2002 to 2010 (0.9-40.58%), then declined to 22.76% of spinal deformity cases in 2013 (Table 6; Figure 3). FIGURE 3. View largeDownload slide Temporal trends of utilization of recombinant bone morphogenic protein (BMP) and interbody fusion devices. FIGURE 3. View largeDownload slide Temporal trends of utilization of recombinant bone morphogenic protein (BMP) and interbody fusion devices. TABLE 6. Utilization Rates of BMP and Interbody Fusion Devices in Spinal Deformity Cases in NIS Database, 2002 to 2013   BMP Utilization  Interbody Fusion Utilization  Year  # Cases  % Cases  # Cases  % Cases  2002  19  0.9  135  6.42  2003  241  9.75  676  27.36  2004  604  21.42  963  34.15  2005  905  27.78  1290  39.59  2006  1120  33.72  1459  43.93  2007  1280  36.13  1651  46.6  2008  1863  38.07  2507  51.24  2009  2139  41.87  2851  55.8  2010  2549  40.58  3513  55.93  2011  2181  31.25  4271  61.19  2012  1608  25.56  3734  59.36  2013  1531  22.76  4143  61.58    BMP Utilization  Interbody Fusion Utilization  Year  # Cases  % Cases  # Cases  % Cases  2002  19  0.9  135  6.42  2003  241  9.75  676  27.36  2004  604  21.42  963  34.15  2005  905  27.78  1290  39.59  2006  1120  33.72  1459  43.93  2007  1280  36.13  1651  46.6  2008  1863  38.07  2507  51.24  2009  2139  41.87  2851  55.8  2010  2549  40.58  3513  55.93  2011  2181  31.25  4271  61.19  2012  1608  25.56  3734  59.36  2013  1531  22.76  4143  61.58  View Large DISCUSSION Our analyses show the number of spinal fusions for deformity increased 273%, and the rate of spinal deformity surgery increased 234% from 2001 to 2013. The increasing rates of spinal deformity surgeries may be due to several factors, including the training of more surgeons in this subspecialty, a heightened focus on spinal alignment parameters in neurosurgical training,19 and the evolving body of research showing improvement in health-related quality of life with spinal deformity surgery.20-22 Due to the significant attention paid to adult spinal deformity in the scientific literature and at recent meetings and conferences, it is also possible that the increased rate of deformity surgery reflects the growing use of scoliosis/deformity diagnosis codes, which would previously have been coded differently. We also find that the mean cost per spinal deformity case increased 25% over this time period, similar to the cumulative US inflation rate of 31.6%.16 In addition, there may be factors specific to spinal deformity surgery that help explain the rise in costs. Spinal deformity cases have some of the highest revision rates in all of spine surgery,23 and the main causes include proximal junctional kyphosis, pseudarthrosis, and infection. To prevent such complications, deformity surgeons have recently added costly items like increased interbody fusions, BMP, vertebroplasty, and ligament augmentation to their armamentarium. In particular, BMP utilization has recently increased in most types of spine surgery,24 with the exception of anterior cervical fusion, in which BMP utilization decreased after the 2008 US Food and Drug Administration safety warning.25 The frequent use of BMP in spinal deformity may be largely driven by its association with lower rates of pseudarthrosis necessitating reoperation.26 However, while we see an increase in BMP utilization from 2002 to 2010, its use in spinal deformity cases decreased from 2010 to 2013, which may reflect backlash from the aforementioned safety concerns. In contrast, interbody fusion utilization has increased steadily from 2002 to 2013 and may contribute partially to increased surgical costs for these cases. Another significant finding is that several patient factors (namely, patient age, race, insurance, SOI, length of stay, and elective admission status) are significantly associated with cost for spinal deformity surgery. Hospitals with smaller bed sizes are more expensive for spinal deformity fusions, supporting the principle of economy of scale. This also suggests that a system that aims to deliver the highest value care may consider focusing its spinal deformity surgeries in large, urban hospitals with high volumes. Our analysis also reveals that the western United States is the most expensive region across all studied years, even after controlling for covariates. This confirms prior studies showing large geographic variation in cost for spine surgery in general.3,5-7 Although one group describes the highest costs for two spinal fusion procedures in the northeastern United States,6 several papers report the highest costs in the West, as we have found.7,27 We hypothesize that systems delivery issues, rather than implant price or surgeon practice patterns alone, may contribute to the higher costs for spinal deformity surgery in the western United States. For example, the wage index in western states like California may not adequately account for the significantly higher cost of labor, and regional policies like the minimum nurse-to-patient ratios mandated in California may further increase costs in these western locations.28 We used the NIS database for this study because it represents the most comprehensive national database of patients from all payers, with data from more than 7 million hospital stays a year, representing 95% of the US population.15 It is specifically designed to enable researchers and policymakers to make national estimates of health care utilization (with complex statistical design to avoid sampling error from different states) and to analyze trends over time, which is the goal of this paper. Because the government and insurers are making payment and coverage decisions based on these databases, it is essential that we as front-line neurosurgical providers understand this data, even if it is imperfect. Limitations The NIS database has several important limitations. As a national administrative database, it relies on ICD-9 coding data, which may be unreliable29,30 and at times heterogeneous. In particular, three of the codes utilized in this study (81.63, 81.63, and 81.64) include both primary fusions and refusions, which may reflect the fact that many deformity operations are performed after initial smaller fusion operations and are sometimes required to treat iatrogenic deformity. In addition, as with many databases, there is some missing information, particularly with regards to race and income, but the amount of missing data is fairly low. Although we convert charges into costs using cost-to-charge ratios, NIS costs still inherently rely on charge data, which may not be an accurate reflection of true cost. Finally, we cannot extract data on the number of levels of spinal surgery, which directly impacts the cost of spinal deformity surgery. However, we do not have any reason to believe that surgeons in the western United States are routinely operating on more levels than those in other parts of the country, although studies suggest different rates of utilization of spinal fusions in various parts of the country.8 Despite these limitations, our study is unique in that it represents, to the best of our knowledge, the first national analysis of the utilization rates and the cost drivers of spinal deformity surgery. Despite its limitations, the NIS database represents the most comprehensive national database of patients from all payers and is specifically designed to enable researchers and policymakers to analyze national health care utilization rates. This type of research is an essential step towards providing the most cost-effective and highest value neurosurgical care. CONCLUSION The number and rate of fusions for spinal deformity rose substantially from 2001 to 2013 in the NIS database. Utilization of interbody fusion rates increased steadily during this time period, while BMP usage peaked in 2010 and declined thereafter. The increased rate of spinal deformity surgery is likely influenced by several factors, including the expanding body of research demonstrating improvement in health-related quality of life measures with spinal deformity surgery. In addition, the mean cost per spinal deformity case increased at a pace similar to inflation. Both patient and hospital factors are important contributors to the variation in cost for spinal deformity surgery. After adjusting for covariates, hospitals in the western United States and those with higher wage indices or smaller bed sizes were significantly more expensive for adult spinal deformity surgery. Disclosures Dr Zygourakis has received travel grants from Nuvasive and Globus to attend resident education courses. Ms Keefe is a consultant for DePuy Spine. Dr Ratliff is a consultant and receives royalties and travel funding from Stryker. Dr Mummaneni has received grants and honoraria from AOSpine, honoraria from Globus and DePuy, royalties from Thieme, Springer, and Taylor and Francis publishers, and is a stock holder in Spinicity/ISD. Dr Ames has received grants and research funding from DePuy Synthes Spine, is a consultant for DePuy, Medtronic, and Stryker, is a stock/shareholder with Baxano and Doctor's Research Group, and receives royalties from Aesculap and Biomet Spine. Dr Zygourakis is supported by a research fellowship from the UCSF Center for Healthcare Value. The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Martin BI, Deyo RA, Mirza SK et al.   Expenditures and health status among adults with back and neck problems. JAMA . 2008; 299( 6): 656- 664. Google Scholar CrossRef Search ADS PubMed  2. Martin BI, Turner JA, Mirza SK, Lee MJ, Comstock BA, Deyo RA. Trends in health care expenditures, utilization, and health status among US adults with spine problems, 1997-2006. Spine (Phila Pa 1976) . 2009; 34( 19): 2077- 2084. Google Scholar CrossRef Search ADS PubMed  3. Ugiliweneza B, Kong M, Nosova K et al.   Spinal surgery: variations in health care costs and implications for episode-based bundled payments. Spine (Phila Pa 1976) . 2014; 39( 15): 1235- 1242. Google Scholar CrossRef Search ADS PubMed  4. Schoenfeld AJ, Harris MB, Liu H, Birkmeyer JD. Variations in Medicare payments for episodes of spine surgery. Spine J . 2014; 14( 12): 2793- 2798. Google Scholar CrossRef Search ADS PubMed  5. Missios S, Bekelis K. Hospitalization cost after spine surgery in the United States of America. J Clin Neurosci . 2015; 22( 10): 1632- 1637. Google Scholar CrossRef Search ADS PubMed  6. 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. Google Scholar CrossRef Search ADS PubMed  7. Cook C, Santos GC, Lima R, Pietrobon R, Jacobs DO, Richardson W. Geographic variation in lumbar fusion for degenerative disorders: 1990 to 2000. Spine J . 2007; 7( 5): 552- 557. Google Scholar CrossRef Search ADS PubMed  8. 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. Google Scholar CrossRef Search ADS PubMed  9. Robin GC, Span Y, Steinberg R, Makin M, Menczel J. Scoliosis in the elderly: a follow-up study. Spine (Phila Pa 1976) . 1982; 7( 4): 355- 359. Google Scholar CrossRef Search ADS PubMed  10. Schwab F, Dubey A, Gamez L et al.   Adult scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population. Spine (Phila Pa 1976) . 2005; 30( 9): 1082- 1085. Google Scholar CrossRef Search ADS PubMed  11. McCarthy IM, Hostin RA, Ames CP et al.   Total hospital costs of surgical treatment for adult spinal deformity: an extended follow-up study. Spine J . 2014; 14( 10): 2326- 2333. Google Scholar CrossRef Search ADS PubMed  12. McCarthy IM, Hostin RA, O’Brien MF et al.   Analysis of the direct cost of surgery for four diagnostic categories of adult spinal deformity. Spine J . 2013; 13( 12): 1843- 1848. Google Scholar CrossRef Search ADS PubMed  13. McCarthy I, Hostin R, O’Brien M, Saigal R, Ames CP. Health economic analysis of adult deformity surgery. Neurosurg Clin N Am . 2013; 24( 2): 293- 304. Google Scholar CrossRef Search ADS PubMed  14. Theologis AA, Miller L, Callahan M et al.   The economic impact of revision surgery for proximal junctional failure after adult spinal deformity surgery: a cost analysis of 57 operations in a 10-year experience at a major deformity center. Spine (Phila Pa 1976) . 2016; 41( 16): E964- E972. Google Scholar CrossRef Search ADS PubMed  15. Houchens RL, Ross DN, Elixhauser A, Jiang J. Nationwide Inpatient Sample (NIS) Redesign Final Report . 2014. HCUP NIS Related Reports ONLINE. U.S. Agency for Healthcare Research and Quality. Available at: http://www.hcupus.ahrq.gov/db/nation/nis/nisrelatedreports.jsp. Published April 4, 2014. Accessed March 27, 2017. 16. US Inflation Calculator. Consumer price index data from 1913 to 2016. Available at: http://www.usinflationcalculator.com/inflation/consumer-price-index-and-annual-percent-changes-from-1913-to-2008/. Accessed March 27, 2017. 17. Houchens RL, Ross D, Elixhauser A. Using the HCUP National Inpatient Sample to Estimate Trends . 2015. HCUP Methods Series Report # 2006-05 ONLINE. Available at: http://www.hcup-us.ahrq.gov/reports/methods/methods.jsp. Published January 4, 2016. Accessed March 27, 2017. 18. Zygourakis CC, Liu CY, Wakam G et al.   Geographic and hospital variation in cost of Lumbra laminectomy and lumbar fusion for degenerative conditions [published online ahead of print March 10, 2017]. Neurosurgery . 2017. doi: 10.1093/neuros/nyx047. 19. Ames CP, Smith JS, Scheer JK et al.   Impact of spinopelvic alignment on decision making in deformity surgery in adults: a review. J Neurosurg Spine . 2012; 16( 6): 547- 564. Google Scholar CrossRef Search ADS PubMed  20. Smith JS, Shaffrey CI, Glassman SD et al.   Risk-benefit assessment of surgery for adult scoliosis: an analysis based on patient age. Spine (Phila Pa 1976) . 2011; 36( 10): 817- 824. Google Scholar CrossRef Search ADS PubMed  21. Schwab F, Lafage V, Patel A, Farcy JP. Sagittal plane considerations and the pelvis in the adult patient. Spine (Phila Pa 1976) . 2009; 34( 17): 1828- 1833. Google Scholar CrossRef Search ADS PubMed  22. Schwab F, Patel A, Ungar B, Farcy JP, Lafage V. Adult spinal deformity-postoperative standing imbalance: how much can you tolerate? An overview of key parameters in assessing alignment and planning corrective surgery. Spine (Phila Pa 1976) . 2010; 35( 25): 2224- 2231. Google Scholar CrossRef Search ADS PubMed  23. Smith JS, Klineberg E, Lafage V et al.   Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery. J Neurosurg Spine . 2016; 25( 1): 1- 14. Google Scholar CrossRef Search ADS PubMed  24. Louie PK, Hassanzadeh H, Singh K. Epidemiologic trends in the utilization, demographics, and cost of bone morphogenetic protein in spinal fusions. Curr Rev Musculoskelet Med . 2014; 7( 3): 177- 181. Google Scholar CrossRef Search ADS PubMed  25. Singh K, Nandyala SV, Marquez-Lara A, Fineberg SJ. Epidemiological trends in the utilization of bone morphogenetic protein in spinal fusions from 2002 to 2011. Spine (Phila Pa 1976) . 2014; 39( 6): 491- 496. Google Scholar CrossRef Search ADS PubMed  26. Paul JC, Lonner BS, Vira S, Kaye ID, Errico TJ. Use of recombinant bone morphogenetic protein is associated with reduced risk of reoperation after spine fusion for adult spinal deformity. Spine (Phila Pa 1976) . 2016; 41( 1): E15- E21. Google Scholar CrossRef Search ADS PubMed  27. Thakore RV, Greenberg SE, Bulka CM, Ehrenfeld JM, Obremskey WT, Sethi MK. Geographic variations in hospital charges and Medicare payments for major joint arthroplasty. J Arthroplasty . 2015; 30( 5): 728- 732. Google Scholar CrossRef Search ADS PubMed  28. Coffman JM, Seago JA, Spetz J. Minimum nurse-to-patient ratios in acute care hospitals in California. Health Aff (Millwood) . 2002; 21( 5): 53- 64. Google Scholar CrossRef Search ADS PubMed  29. Faciszewski T, Johnson L, Noren C, Smith MD. Administrative databases' complication coding in anterior spinal fusion procedures. What does it mean? Spine (Phila Pa 1976) . 1995; 20( 16): 1783- 1788. Google Scholar CrossRef Search ADS PubMed  30. Gologorsky Y, Knightly JJ, Chi JH, Groff MW. The Nationwide Inpatient Sample database does not accurately reflect surgical indications for fusion. J Neurosurg Spine . 2014; 21( 6): 984- 993. Google Scholar CrossRef Search ADS PubMed  COMMENTS The authors propose an interesting study on the variation in cost for spinal deformity surgery in the United States. For this purpose, they utilized the National Inpatient Sample (NIS) database from 2001–2013 to determine the factors that contribute to the cost of spinal deformity surgery. The authors found that there was a significant increase in the number and rate of spinal deformity surgery cases and a parallel increase in the cost per case over the reported years. The western region was found to be 29% more expensive than the northeastern region for spinal deformity surgery in the United States. The authors subdivided the factors that affected cost into hospital factors and patient factors. After adjusting for covariates in regression analyses, hospitals of higher wage indices and smaller bed sizes were mostly associated with an increase in cost. Patients with higher severity of illness were also identified as a significant driver of cost. This article is well-written and novel in terms of using a national registry, the NIS database, to report the regional variation of cost per case of spinal deformity surgery. The advantage of using such database is that it represents a 20% stratified sample of all US community hospitals. The authors should be commended for this important addition to our understanding of the most significant factors that affect the cost of spinal deformity procedures. Moving forward, it will be interesting to utilize the findings of this study in future cost-effectiveness reports. Seba Ramhmdani Ali Bydon Baltimore, Maryland The authors of this analysis present further evidence on the increasing social burden that the United States has experienced in the first decade of this century. While there are deep inherent limitations to subject selection and the analysis of cost in these patients, it must be acknowledged that the major trends reported here are consistent with previous smaller scale evaluations and that an enhanced focus on the trends of this clinical problem is required. Peter G. Passias New York, New York Copyright © 2017 by the Congress of Neurological Surgeons

Journal

NeurosurgeryOxford University Press

Published: Mar 1, 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