Orthopedics

Feature Article Free

Return to Work After Diskogenic Fusion in Workers' Compensation Subjects

Joshua T. Anderson, BS; Arnold R. Haas, BS, BA; Rick Percy, PhD; Stephen T. Woods, MD; Uri M. Ahn, MD; Nicholas U. Ahn, MD

Abstract

Lumbar fusion for degenerative disk disease (DDD) is associated with variable clinical outcomes. Patients with workers' compensation claims often have worse fusion outcomes than the general population. Few studies have evaluated the risk factors for poor outcomes within this clinically distinct population. The goal of this study was to identify preoperative predictors of return to work status after fusion for DDD in a workers' compensation setting. The authors used International Classification of Diseases, Ninth Revision (ICD-9), diagnosis and Current Procedural Terminology (CPT) procedural codes to identify 1037 subjects from the Ohio Bureau of Workers' Compensation database who underwent fusion for DDD between 1993 and 2013. Of these subjects, 23.2% (n=241) made a sustained return to work within 2 years after fusion. To identify preoperative predictors of postoperative return to work status, the authors used multivariate logistic regression analysis, adjusting for many important covariates. These included prolonged time out of work (P<.001; odds ratio [OR], 0.24), psychiatric history (P<.001; OR, 0.14), prolonged use of opioid analgesics (P<.001; OR, 0.46), male sex (P=.014; OR, 0.65), and legal representation (P=.042; OR, 0.67). The return to work rates associated with these risk factors were 10.4%, 2.0%, 11.9%, 21.1%, and 20.7%, respectively. Of the study subjects, 76.8% (n=796) did not return to work and had considerably worse postoperative outcomes, highlighted by chronic opioid dependence and high rates of failed back syndrome, additional surgery, and new psychiatric comorbidity. The low return to work rates and other generally poor outcomes reported in this study may indicate a more limited role for lumbar fusion among patients with DDD who have workers' compensation claims. More studies are needed to determine whether fusion for DDD can improve function and quality of life in these patients. [Orthopedics. 2015; 38(12):e1065–e1072.]

Abstract

Lumbar fusion for degenerative disk disease (DDD) is associated with variable clinical outcomes. Patients with workers' compensation claims often have worse fusion outcomes than the general population. Few studies have evaluated the risk factors for poor outcomes within this clinically distinct population. The goal of this study was to identify preoperative predictors of return to work status after fusion for DDD in a workers' compensation setting. The authors used International Classification of Diseases, Ninth Revision (ICD-9), diagnosis and Current Procedural Terminology (CPT) procedural codes to identify 1037 subjects from the Ohio Bureau of Workers' Compensation database who underwent fusion for DDD between 1993 and 2013. Of these subjects, 23.2% (n=241) made a sustained return to work within 2 years after fusion. To identify preoperative predictors of postoperative return to work status, the authors used multivariate logistic regression analysis, adjusting for many important covariates. These included prolonged time out of work (P<.001; odds ratio [OR], 0.24), psychiatric history (P<.001; OR, 0.14), prolonged use of opioid analgesics (P<.001; OR, 0.46), male sex (P=.014; OR, 0.65), and legal representation (P=.042; OR, 0.67). The return to work rates associated with these risk factors were 10.4%, 2.0%, 11.9%, 21.1%, and 20.7%, respectively. Of the study subjects, 76.8% (n=796) did not return to work and had considerably worse postoperative outcomes, highlighted by chronic opioid dependence and high rates of failed back syndrome, additional surgery, and new psychiatric comorbidity. The low return to work rates and other generally poor outcomes reported in this study may indicate a more limited role for lumbar fusion among patients with DDD who have workers' compensation claims. More studies are needed to determine whether fusion for DDD can improve function and quality of life in these patients. [Orthopedics. 2015; 38(12):e1065–e1072.]

Lumbar fusion for degenerative disk disease and diskogenic low back pain, also called diskogenic fusion, is associated with variable clinical outcomes.1–9 It has been associated with decreased improvement in Oswestry Disability Index and Short Form Health Survey scores as well as higher reoperation rates compared with fusion for more definable instability or deformity.1,3 Rates of fusion for degenerative disk disease and diskogenic low back pain have increased approximately 220% since 1990, and these are now estimated to be the most common indications for fusion in the Unites States.5,10

In addition, US patients with workers' compensation claims often have worse outcomes than the general population.6,8,11–17 Reports of return to work rates among subjects receiving workers' compensation are typically low, ranging from 26% to 36% in several studies.6,8,9,12,18 Rates of lumbar fusion within this clinically distinct population have been rising.19 Even though workers' compensation is a well-described risk factor for worse outcomes, relatively few studies have evaluated this population for predictors of worse outcomes.6,9,12,13,20,21 Among those receiving workers' compensation, the ability to return to work is of particular clinical importance, especially when considering that the estimated cost of treating chronic low back pain is greater than $100 billion per year in the United States, and two-thirds of this cost is believed to result from lost wages and reduced productivity.22,23 Therefore, the authors wanted to identify preoperative factors that predicted return to work status after diskogenic lumbar fusion in patients with workers' compensation claims. The authors also compared population characteristics and secondary fusion outcomes between those who returned to work after fusion and those who did not.

Materials and Methods

This retrospective cohort study included 1037 subjects from the Ohio Bureau of Workers' Compensation database who underwent diskogenic fusion with at least 3 years of follow-up. The authors initially identified 14,640 subjects from the Ohio Bureau of Workers' Compensation administrative database who were diagnosed with lumbar comorbidities after a workplace injury that qualified for workers' compensation between 1993 and 2013 with International Classification of Diseases, Ninth Revision (ICD-9), codes. Those who underwent lumbar fusion surgery were identified with Current Procedural Terminology (CPT) codes. A CPT coding schema published by Nguyen et al8 was used to characterize all fusion surgeries. The authors based the study and outcome measures on each subject's index fusion surgery after injury. This study did not require institutional review board approval because the authors worked directly with the legal department of the Ohio Bureau of Workers' Compensation and all subject data were deidentified.

From the initial population, 13,603 subjects were excluded. First, the authors excluded those who did not undergo lumbar fusion; those who underwent fusion for indications other than degenerative disk disease and diskogenic low back pain; those who underwent fusion with an approach other than anterior, posterior, or 360°; and those who had follow-up of less than 3 years. Then the authors excluded those with a history of other lumbar surgery, smoking, or failed back syndrome because these factors have been reported to negatively influence outcomes.4,12,24–27 The authors identified lumbar comorbidities, failed back syndrome, and smoking history with ICD-9 codes and the use of prescription smoking deterrents and previous lumbar surgery with CPT codes.

The final study population consisted of 1037 subjects who underwent anterior, posterior, or 360° lumbar fusion for degenerative disk disease and diskogenic low back pain with at least 3 years of follow-up. Next the authors determined which subjects returned to work within a reasonable amount of time after fusion. Specifically, the authors classified subjects as returning to work if they returned within 2 years of fusion and remained working for longer than 6 months of the following year. Only 241 (23.2%) subjects in the study population met these criteria, thus forming the return to work group. The remaining 796 (76.8%) subjects formed the out of work group. Figure 1 shows the subject selection process, and Table 1 lists the codes used to form the study population.

The authors used a combination of International Classification of Diseases, Ninth Revision, and Current Procedural Terminology codes to arrive at the final study population. Abbreviations: DDD, degenerative disk disease; LBP, low back pain.

Figure 1:

The authors used a combination of International Classification of Diseases, Ninth Revision, and Current Procedural Terminology codes to arrive at the final study population. Abbreviations: DDD, degenerative disk disease; LBP, low back pain.

Coding Used for the Study Population

Table 1:

Coding Used for the Study Population

The primary outcome of this study was whether subjects met the authors' criteria for return to work. The authors measured the following secondary outcomes 3 years after index fusion: days out of work, medical costs paid by the Ohio Bureau of Workers' compensation, opioid analgesic use, use of physical therapy and chiropractic care, and rates of new psychiatric comorbidity, failed back syndrome, pseudarthrosis, newly awarded permanent disability benefits, all-cause mortality, and additional lumbar surgery.

Data on prescription opioid analgesic use were collected for all subjects before index fusion and within 3 years after index fusion. The authors determined the total number of days that each subject was supplied with opioids. All prescriptions were converted to morphine equivalents, and the total morphine equivalents supplied were calculated. The authors also determined the average daily morphine equivalents to quantify daily opioid loads for subjects.

Additionally, data on a number of important covariates were obtained, including age at index fusion, percentage of subjects older than 50 years at fusion, sex, approximated income, and rate of obesity. Specific incomes for each subject were unavailable. The authors correlated each subject's zip code of residence to a mean per capita income value based on 2010 US Census data. The authors determined the number of days out of work before fusion, the percentage of subjects out of work for longer than 1 year, the percentage of subjects receiving total or partial permanent disability benefits, and preoperative rates of psychiatric disorders, lumbar diskography, physical therapy, chiropractic care, and legal representation. For each subject, the authors characterized the fusion approach, graft types, and instrumentation.

To identify predictors of postoperative return to work, multivariate logistic regression analysis was used. The dependent variable was whether the criteria for return to work were met. In the regression model, the authors adjusted for the following preoperative and operative binary variables: out of work for longer than 1 year, single-level vs multilevel fusion, decompression with fusion, permanent disability benefits, age older than 50 years at fusion, sex, obesity, approximated income above or below the population mean, individual lumbar comorbidities, psychiatric comorbidity, lumbar diskography, physical therapy, chiropractic care, obesity, use of opioid analgesics for longer than 1 year, daily opioid load above or below the population mean, and legal representation. The authors adjusted for the following categorical variables: type of fusion surgery, instrumentation, and graft. The authors compared secondary outcomes and baseline population characteristics between the return to work and out of work groups with chi-square tests for binary and categorical variables and Student's t tests for continuous variables. P≤.05 was considered statistically significant. For all analyses, Statgraphics Centurion XVI version 16.2.04 software (Statpoint Technologies, Inc, Warrenton, Virginia) was used.

Results

Table 2 shows important preoperative characteristics for the return to work and out of work groups. Compared with the out of work group, subjects in the return to work group were 1.8 years younger (P=.005), were out of work for 421.7 fewer days (P<.001), and were supplied with opioids for 296.2 fewer days (P<.001), which equated to 18,916.9 fewer morphine equivalents (P<.001), with similar daily opioid loads. The return to work group had an 8.0% lower rate of being male (P=.020), a 40.7% lower rate of being out of work for longer than 1 year before fusion (P<.001), an 11.5% lower rate of psychiatric disorders (P<.001), an 11.5% lower rate of legal representation (P<.001), a 1.6% lower rate of permanent total disability (P=.046), and a 23.7% lower rate of preoperative opioid supply for longer than 1 year (P<.001).

Preoperative Baseline Characteristics of the Population

Table 2:

Preoperative Baseline Characteristics of the Population

Table 3 shows important surgical data. The return to work and out of work groups differed only in rates of different fusion approaches (P=.037). The return to work group underwent posterior lumbar interbody fusion at a 3.9% higher rate, underwent posterior lumbar fusion and posterior lumbar interbody fusion at a 9.3% lower rate, and underwent 360° fusion at a 5.2% higher rate. Rates for the other fusion approaches differed less than 2%.

Characterization of Index Fusion

Table 3:

Characterization of Index Fusion

Only 23.2% (241 of 1037) of the subjects met the criteria for return to work. Of these subjects, 69.7% returned to work within the first year after fusion. Significant preoperative predictors of return to work status included being out of work for longer than 1 year (P<.001; odds ratio [OR], 0.24), history of a psychiatric disorder (P<.001; OR, 0.14), opioid use for longer than 1 year (P<.001; OR, 0.46), male sex (P=.014; OR, 0.65), and legal representation (P=.042; OR, 0.67). Decompression with fusion (P=.060; OR, 0.70) and lumbar sprain (P=.055; OR, 0.72) were negative predictors that nearly reached statistical significance (Table 4). Figure 2 shows the return to work rates associated with each of these predictors as well as the resulting return to work rate if that variable was excluded. The strongest negative preoperative predictor was psychiatric disorders, closely followed by prolonged work loss and long-term opioid use. If all subjects associated with these 3 variables were eliminated, then the overall return to work rate increased to 43.8% (154 of 352).

Predictors of Return to Work Status

Table 4:

Predictors of Return to Work Status

Rates for return to work (RTW) associated with each significant prognostic factor from the regression model. Also shown are the RTW rates if patients with each individual predictive factor were removed from the population. The greatest difference in RTW rate was between those who were out of work for longer than 1 year before fusion and those who were not out of work for longer than 1 year (29.1%).

Figure 2:

Rates for return to work (RTW) associated with each significant prognostic factor from the regression model. Also shown are the RTW rates if patients with each individual predictive factor were removed from the population. The greatest difference in RTW rate was between those who were out of work for longer than 1 year before fusion and those who were not out of work for longer than 1 year (29.1%).

A chi-square goodness of fit test determined that there was no reason to reject the fitted logistic regression model at the 95% confidence level. No serious multicollinearity existed in the regression model because no correlation coefficient absolute value was greater than 0.5. Regarding the predictive capability of the regression model, with an outcome cutoff value of 0.55, where 1=will return to work and 0=will not return to work, the model achieved its highest overall predictive performance, with 77.3% of the observations in this study predicted accurately. This equated to a 13.9% prediction rate of true-positives (ie, those who would later return to work) and a 96.7% rate of true-negatives (ie, those who would not return to work). At cutoff values of greater than 0.55, the model approached 100% ability to predict those who will not return to work. Figure 3 shows the predictive performance of the model. The fitted regression model more effectively predicted those who will not return to work than those who will return to work.

Predictive ability of the function derived from the multivariate logistic regression model compared with actual observations in the study at all outcome cutoff values. The best overall predictive performance was achieved with a return to work outcome cutoff value of 0.55, where 1=will return to work and 0=will not return to work. At and above this value, the model accurately predicted subjects who would not return to work at a rate greater than 96%.

Figure 3:

Predictive ability of the function derived from the multivariate logistic regression model compared with actual observations in the study at all outcome cutoff values. The best overall predictive performance was achieved with a return to work outcome cutoff value of 0.55, where 1=will return to work and 0=will not return to work. At and above this value, the model accurately predicted subjects who would not return to work at a rate greater than 96%.

Table 5 shows data on secondary outcomes. As expected, subjects in the out of work group had considerably worse postoperative outcomes than those in the return to work group. Within 3 years after fusion, subjects in the return to work group were, on average, out of work for 611.6 fewer days than those in the out of work group (P<.001), and this was associated with $28,777.3 lower medical costs paid by the Ohio Bureau of Workers' compensation per subject (P<.001). In addition, subjects in the return to work group were supplied with opioid analgesics for 498.8 fewer days (P<.001), which equated to 43,110.8 fewer total morphine equivalents (P<.001) and a daily opioid load that was 8.4 morphine equivalents lower (P=.041). Subjects in the return to work group had a 12.0% lower rate of failed back syndrome (P<.001), a 2.1% lower rate of dying of any cause (P=.022), and an 18.4% lower rate of additional lumbar surgery (P<.001). Among subjects who did not receive permanent disability benefits before fusion, those in the return to work group had a significantly higher rate of newly awarded permanent disability benefits after fusion. However, further analysis showed that most of these subjects received partial disability benefits, whereas the out of work group had a 9.4% higher rate of new permanent total disability benefits after fusion (P<.001).

Secondary Outcomes After Fusion

Table 5:

Secondary Outcomes After Fusion

The rate of diagnosis of new psychiatric disorders was 19.8% lower in the return to work group (P<.001). It is concerning that 26.5% (185 of 796) of subjects in the out of work group were diagnosed with new psychiatric comorbidity after fusion, and 90.8% of these were diagnosed with depression. Only 11.0% (114 of 1037) of all subjects, 4.6% (11 of 241) of the return to work group, and 12.9% (103 of 796) of the out of work group underwent a psychiatric diagnostic examination before fusion. After fusion, 27.8% (288 of 1037) of all subjects, 10.8% (26 of 241) of the return to work group, and 32.9% (262 of 796) of the out of work group underwent psychiatric diagnostic examination. The return to work rate for all subjects who were diagnosed with psychiatric disorders, either before or after fusion, was 6.0% (18 of 301). This rate increased to 30.3% (223 of 736) if these subjects were excluded.

Discussion

Diskogenic fusion has been associated with variable clinical outcomes.1–9 Patients with workers' compensation claims often have worse outcomes after fusion.6,8,11–17 Relatively few studies have evaluated the risk factors for worse outcomes within this clinically distinct population.6,9,12,13,20,21 Among the 1037 subjects identified in the Ohio Workers' Compensation database who underwent diskogenic fusion, the authors identified a number of preoperative factors that predicted postoperative return to work status. The strongest negative predictor of return to work status was a history of a psychiatric disorder before fusion, with an associated return to work rate of only 2.0%. This was followed closely by being out of work for longer than 1 year and prolonged opioid use before fusion, with associated return to work rates of only 10.4% and 11.9%, respectively.

Overall return to work outcomes were poor, with only 23.2% of the study population returning to work after fusion. In addition, 69.7% of subjects who returned to work did so within the first year after fusion, suggesting that the longer that patients with workers' compensation claims are out of work postoperatively the less likely they are to return to work. Furthermore, even after subjects with the 3 strongest negative outcome predictors were removed from analysis, the resulting return to work rate was only 43.8%. Therefore, it cannot necessarily be concluded that subjects without these risk factors had good return to work outcomes. Subjects who did not return to work had considerably worse postoperative outcomes, as shown in Table 5. It is concerning that the vast majority of subjects (76.8%) in this study did not return to work within a reasonable amount of time after fusion and thus had poor outcomes.

The high rate of newly diagnosed psychiatric disorders among subjects in the out of work group (25.6%) may have been related to poor outcomes after fusion, increased time away from work, disability, or postoperative complications. In addition, fusion may have precipitated a previously undetected psychiatric disorder. Only 11.0% of subjects in the out of work group underwent psychiatric diagnostic examination before fusion, compared with 32.9% who underwent examination after fusion. Return to work rates increased 5-fold when subjects who were diagnosed with any psychiatric disorder, either before or after fusion, were excluded. Although this change was only from 6.0% to 30.3%, the findings indicate the potential impact of better screening and treatment for psychiatric disorders before lumbar fusion. It also cannot be concluded that those without psychiatric disorders had good return to work outcomes.

Limitations

The study was limited by its observational and retrospective design. The authors adjusted for many relevant covariates in this analysis, but the study subjects may have varied in unobserved ways, and the findings may include some unmeasured confounding factors. The generalizability of the results to patients with workers' compensation claims in other states may be affected by different legislative regulations, treatment guidelines, and potential geographic effects on outcomes. For example, a recent study reported differences in outcomes between subjects receiving workers' compensation in California and Washington State because of policy and reimbursement differences.28 In addition, the completeness of the data may have been limited by the use of the Ohio Bureau of Workers' Compensation administrative database. However, this database was used in earlier studies.8,9 Also, 2 recent studies reported high sensitivity and specificity of administrative billing data when compared with actual medical records, proving that such databases can provide accurate information, especially when identifying diagnoses, indications for surgery, and the type of procedure performed.29,30

Conclusion

The authors identified a number of significant predictors for worse return to work outcomes within a large population of subjects with workers' compensation claims. Psychiatric disorders, legal representation, extensive work loss, and prolonged opioid use before fusion negatively affected the likelihood of returning to work after fusion. The low return to work rates and other generally poor outcomes reported in this study may suggest a more limited role for spinal fusion in patients with degenerative disk disease and diskogenic low back pain who are receiving workers' compensation. Although the authors did not include a nonoperative control group for comparison with the study population, it is difficult to conclude that the study subjects had favorable outcomes. This study focused largely on return to work rates and work loss. It is possible that these patients did improve after fusion, but not to the extent that they were able to return to work. More studies are needed to determine whether diskogenic fusion can improve function and quality of life in the workers' compensation population. The authors hope that this study can help to identify which patients receiving workers' compensation may present a greater clinical challenge after diskogenic fusion.

References

  1. Ahn DK, Park HS, Choi DJ, Kim KS, Yang SJ. Survival and prognostic analysis of adjacent segments after spinal fusion. Clin Orthop Surg. 2010; 2(3):140–147. doi:10.4055/cios.2010.2.3.140 [CrossRef]
  2. Buttermann GR, Garvey TA, Hunt AF, et al. Lumbar fusion results related to diagnosis. Spine (Phila Pa 1976). 1998; 23(1):116–127. doi:10.1097/00007632-199801010-00024 [CrossRef]
  3. Glassman SD, Carreon LY, Djurasovic M, et al. Lumbar fusion outcomes stratified by specific diagnostic indication. Spine J. 2009; 9(1):13–21. doi:10.1016/j.spinee.2008.08.011 [CrossRef]
  4. Hanley EN Jr, . The indications for lumbar spinal fusion with and without instrumentation. Spine (Phila Pa 1976). 1995; 20(suppl 24):143S–153S. doi:10.1097/00007632-199512151-00012 [CrossRef]
  5. Lee CK, Langrana NA. A review of spinal fusion for degenerative disc disease: need for alternative treatment approach of disc arthroplasty?Spine J. 2004; 4(suppl 6):173S–176S. doi:10.1016/j.spinee.2004.07.002 [CrossRef]
  6. Maghout Juratli S, Franklin GM, Mirza SK, Wickizer TM, Fulton-Kehoe D. Lumbar fusion outcomes in Washington State workers' compensation. Spine (Phila Pa 1976). 2006; 31(23):2715–2723.
  7. Mirza SK, Deyo RA. Systematic review of randomized trials comparing lumbar fusion surgery to nonoperative care for treatment of chronic back pain. Spine (Phila Pa 1976). 2007; 32(7):816–823. doi:10.1097/01.brs.0000259225.37454.38 [CrossRef]
  8. Nguyen TH, Randolph DC, Talmage J, Succop P, Travis R. Long-term outcomes of lumbar fusion among workers' compensation subjects: a historical cohort study. Spine (Phila Pa 1976). 2011; 36(4):320–331.
  9. Anderson JT, Haas AR, Percy R, Woods ST, Ahn UM, Ahn NU. Single-level lumbar fusion for degenerative disc disease is associated with worse outcomes compared with fusion for spondylolisthesis in a workers' compensation setting. Spine (Phila Pa 1976). 2015; 40(5):323–331. doi:10.1097/BRS.0000000000000734 [CrossRef]
  10. Deyo RA, Gray DT, Kreuter W, Mirza S, Martin BI. United States trends in lumbar fusion surgery for degenerative conditions. Spine (Phila Pa 1976). 2005; 30(12):1441–1445. doi:10.1097/01.brs.0000166503.37969.8a [CrossRef]
  11. Carreon LY, Glassman SD, Kantamneni NR, Mugavin MO, Djurasovic M. Clinical outcomes after posterolateral lumbar fusion in workers' compensation patients: a case-control study. Spine (Phila Pa 1976). 2010; 35(19):1812–1817. doi:10.1097/BRS.0b013e3181c68b75 [CrossRef]
  12. DeBerard MS, Masters KS, Colledge AL, Schleusener RL, Schlegel JD. Outcomes of posterolateral lumbar fusion in Utah patients receiving workers' compensation: a retrospective cohort study. Spine (Phila Pa 1976). 2001; 26(7):738–746. doi:10.1097/00007632-200104010-00007 [CrossRef]
  13. Franklin GM, Haug J, Heyer NJ, McKeefrey SP, Picciano JF. Outcome of lumbar fusion in Washington State workers' compensation. Spine (Phila Pa 1976). 1994; 19(17):1897–1903. doi:10.1097/00007632-199409000-00005 [CrossRef]
  14. Gum JL, Glassman SD, Carreon LY. Is type of compensation a predictor of outcome after lumbar fusion?Spine (Phila Pa 1976). 2013; 38(5):443–448. doi:10.1097/BRS.0b013e318278ebe8 [CrossRef]
  15. Mayer TG, Gatchel RJ, Brede E, Theodore BR. Lumbar surgery in work-related chronic low back pain: can a continuum of care enhance outcomes?Spine J. 2014; 14(2):263–273. doi:10.1016/j.spinee.2013.10.041 [CrossRef]
  16. Wheeler AJ, Smith AL, Gundy JM, Sautter T, DeBerard MS. Predicting back pain treatment outcomes among workers' compensation patients: important information for clinical neuropsychologists. Clin Neuropsychol. 2013; 27(1):49–59. doi:10.1080/13854046.2012.750686 [CrossRef]
  17. Trief PM, Ploutz-Snyder R, Fredrickson BE. Emotional health predicts pain and function after fusion: a prospective multicenter study. Spine (Phila Pa 1976). 2006; 31(7):823–830. doi:10.1097/01.brs.0000206362.03950.5b [CrossRef]
  18. DeBerard MS, Wheeler AJ, Gundy JM, Stein DM, Colledge AL. Presurgical biopsychosocial variables predict medical, compensation, and aggregate costs of lumbar discectomy in Utah workers' compensation patients. Spine J. 2011; 11(5):395–401. doi:10.1016/j.spinee.2011.03.018 [CrossRef]
  19. Rajaee SS, Bae HW, Kanim LE, Delamarter RB. Spinal fusion in the United States: analysis of trends from 1998 to 2008. Spine (Phila Pa 1976). 2012; 37(1):67–76. doi:10.1097/BRS.0b013e31820cccfb [CrossRef]
  20. Hodges SD, Humphreys SC, Eck JC, Covington LA, Harrom H. Predicting factors of successful recovery from lumbar spine surgery among workers' compensation patients. J Am Osteopath Assoc. 2001; 101(2):78–83.
  21. Wheeler AJ, Gundy JM, DeBerard MS. Predicting compensation and medical costs of lumbar fusion patients receiving workers' compensation in Utah using presurgical biopsychosocial variables. Spine (Phila Pa 1976). 2012; 37(7):605–611. doi:10.1097/BRS.0b013e318226abf7 [CrossRef]
  22. Crow WT, Willis DR. Estimating cost of care for patients with acute low back pain: a retrospective review of patient records. J Am Osteopath Assoc. 2009; 109(4):229–233.
  23. Katz JN. Lumbar disc disorders and low-back pain: socioeconomic factors and consequences. J Bone Joint Surg Am. 2006; 88(suppl 2):21–24. doi:10.2106/JBJS.E.01273 [CrossRef]
  24. Miller B, Gatchel RJ, Lou L, Stowell A, Robinson R, Polatin PB. Interdisciplinary treatment of failed back surgery syndrome (FBSS): a comparison of FBSS and non-FBSS patients. Pain Pract. 2005; 5(3):190–202. doi:10.1111/j.1533-2500.2005.05304.x [CrossRef]
  25. Jenkins LT, Jones AL, Harms JJ. Prognostic factors in lumbar spinal fusion. Contemp Orthop. 1994; 29(3):173–180.
  26. Andersen T, Christensen FB, Laursen M, Høy K, Hansen ES, Bünger C. Smoking as a predictor of negative outcome in lumbar spinal fusion. Spine (Phila Pa 1976). 2001; 26(23):2623–2628. doi:10.1097/00007632-200112010-00018 [CrossRef]
  27. Cobo Soriano J, Sendino Revuelta M, Fabregate Fuente M, Cimarra Díaz I, Martínez Ureña P, Deglané Meneses R. Predictors of outcome after decompressive lumbar surgery and instrumented posterolateral fusion. Eur Spine J. 2010; 19(11):1841–1848. doi:10.1007/s00586-010-1284-2 [CrossRef]
  28. Martin BI, Franklin GM, Deyo RA, Wickizer TM, Lurie JD, Mirza SK. How do coverage policies influence practice patterns, safety, and cost of initial lumbar fusion surgery? A population-based comparison of workers' compensation systems. Spine J. 2014; 14(7):1237–1246. doi:10.1016/j.spinee.2013.08.018 [CrossRef]
  29. Singh JA, Ayub S. Accuracy of VA databases for diagnoses of knee replacement and hip replacement. Osteoarthritis Cartilage. 2010; 18(12):1639–1642. doi:10.1016/j.joca.2010.10.003 [CrossRef]
  30. Kazberouk A, Martin BI, Stevens JP, McGuire KJ. Validation of an administrative coding algorithm for classifying surgical indication and operative features of spine surgery. Spine (Phila Pa 1976). 2015; 40(2):114–120. doi:10.1097/BRS.0000000000000682 [CrossRef]

Coding Used for the Study Population

VariableCode
Fusion technique (CPT code)
  Posterior lumbar fusion
    Single-level fusion22612
    Multilevel fusion22612+22614
  Posterior lumbar interbody fusion
    Single-level fusion22630 or 22633
    Multilevel fusion22630 or 22633+22632 or 22634
  Anterior lumbar interbody fusion
    Single-level fusion22558
    Multilevel fusion22558+22585
Lumbar comorbidity (ICD-9 code)
  Degenerative disk disease722.52, 722.73, or 722.93
  Disk herniation722.10
  Spondylosis721.3
  Lumbar sprain847.2
  Radiculopathy724.4 or 729.2

Preoperative Baseline Characteristics of the Population

CharacteristicGroupP

Return to WorkOut of Work
Subjects, No.241796
Age at index fusion, mean±SD, y43.1±8.444.9±8.7.005
Age >50 y at index fusion, No.50 (20.7%)209 (26.3%).083a
Sex, No..020
  Male144 (59.8%)540 (67.8%)
  Female97 (40.2%)256 (32.2%)
Out of work, mean±SD, d257.6±315.8679.3±646.5<.001
Out of work >1 y, No.58 (24.1%)516 (64.8%)<.001
Psychiatric history, No.2 (0.8%)98 (12.3%)<.001
  Depression1 (0.4%)88 (11.1%)<.001
  Anxiety0 (0.0%)9 (1.1%).097a
  Adjustment reaction1 (0.4%)9 (1.1%).319
Permanent disability benefits,b No.104 (43.2%)317 (39.8%).357
  Permanent partial disability104 (43.2%)307 (38.6%).202
  Permanent total disability0 (0.0%)13 (1.6%).046
Lumbar diskography, No.103 (42.7%)387 (48.6%).109
Physical therapy, No.213 (88.4%)723 (90.8%).252
  Sessions, mean±SD, No.72.7±111.680.7±104.4.305
Chiropractic care, No.89 (36.9%)324 (40.7%).294
  Sessions, mean±SD, No.24.7±51.124.4±47.7.933
Legal representation, No.173 (71.8%)663 (83.3%)<.001
Opioid analgesic use
  Supplied with opioids, mean±SD, d219.2±371.1515.4±768.2<.001
  Supplied for >1 y, No.46 (19.1%)341 (42.8%)<.001
  Net morphine equivalents supplied, mean±SD13,114.6±31,129.632,031.5±73,071.2<.001
  Daily morphine equivalents, mean±SD55.1±64.754.0±46.5.771
Lumbar comorbidity, No.
  Disk herniation126 (52.3%)404 (50.8%).678
  Radiculopathy29 (12.0%)123 (15.5%).189
  Spondylosis21 (8.7%)99 (12.4%).113
  Lumbar sprain119 (49.4%)461 (57.9%).019

Characterization of Index Fusion

Independent VariableGroup, No.P

Return to Work (n=241)Out of Work (n=796)
Fusion technique.037
  Anterior lumbar interbody fusion16 (6.6%)62 (7.8%)
  Posterior lumbar fusion49 (20.3%)151 (19.0%)
  Posterior lumbar interbody fusion40 (16.6%)101 (12.7%)
  Posterior lumbar fusion + posterior lumbar interbody fusion98 (40.7%)398 (50.0%)
  360° fusion38 (15.8%)84 (10.6%)
Decompression with fusion145 (60.2%)515 (64.7%).200
Instrumentation.827
  Instrumented59 (24.5%)199 (25.0%)
  Intervertebral biomechanical device (eg, cage)44 (18.3%)125 (15.7%)
  Instrumented + intervertebral biomechanical device127 (52.7%)435 (54.6%)
  Uninstrumented11 (4.6%)37 (4.6%)
Graft.778
  Allograft17 (7.1%)57 (7.2%)
  Allograft + autograft18 (7.5%)65 (8.2%)
  Autograft125 (51.9%)383 (48.1%)
Bone morphogenic protein or other graft material81 (33.6%)291 (36.6%)

Predictors of Return to Work Status

Independent VariableOdds Ratio95% Confidence IntervalP
Out of work >1 y before index fusion0.240.16–0.34<.001
History of psychiatric disorder0.140.03–0.60<.001
Supplied with opioid analgesics >1 y before index fusion0.460.31–0.69<.001
Male sex0.650.47–0.92.014
Legal representation before index fusion0.670.45–0.98.042
Decompression with fusiona0.700.49–1.01.060
Diagnosis of lumbar sprain before index fusiona0.720.52–1.01.055

Secondary Outcomes After Fusion

Independent VariableGroupP

Return to Work (n=241)Out of Work (n=796)
Out of work after index fusion,a mean±SD, d450.3±258.11061.9±165.2<.001
Medical costs paid by Bureau of Workers' Compensation, mean±SD$59,523.7±$26,478.0$88,301.0±$40,251.8<.001
Opioid analgesic use, mean±SD
  Days supplied295.8±408.4794.6±665.8<.001
  Net morphine equivalents prescribed17,786.2±38,696.560,897.0±88,163.5<.001
  Daily morphine equivalents58.2±54.166.6±56.5.041
Newly diagnosed psychiatric disorder,b No.16 (6.7%)185 (26.5%)<.001
  Depression12 (5.0%)168 (24.1%)<.001
  Anxiety2 (0.1%)22 (3.2%).051c
  Adjustment reaction2 (0.1%)11 (1.6%).399
  Otherd0 (0.0%)4 (0.6%).241
Failed back syndrome, No.4 (1.7%)109 (13.7%)<.001
Pseudarthrosis, No.0 (0.0%)11 (1.4%).067c
New permanent disability benefits,e No.82 (59.9%)151 (31.5%)<.001
  Permanent partial disability80 (58.4%)105 (21.9%)<.001
  Permanent total disability2 (1.5%)52 (10.9%)<.001
All-cause mortality, No.0 (0.0%)17 (2.1%).022
Physical therapy, No.217 (90.0%)730 (91.7%).420
Sessions, mean±SD, No.52.2±60.272.8±81.4
Chiropractic care38 (15.8%)203 (25.5%).800
Sessions, mean±SD, No.5.6±17.56.7±23.2
Additional lumbar surgery, No.15 (6.2%)196 (24.6%)<.001
  Additional lumbar fusion13 (5.4%)150 (18.8%)<.001
  Additional decompression6 (2.5%)114 (14.3%)<.001
Authors

The authors are from the Department of Orthopaedics, University Hospitals Case Medical Center (JTA, NUA), Cleveland, Case Western Reserve University School of Medicine (JTA), Cleveland, and the Ohio Bureau of Workers' Compensation (ARH, RP, STW), Columbus, Ohio; and the New Hampshire NeuroSpine Institute (UMA), Bedford, New Hampshire.

Mr Anderson, Mr Haas, and Dr Percy have no relevant financial relationships to disclose. Dr Woods holds stock in OrthoNeuro, Inc. Dr U. M. Ahn is a paid consultant for, is on the speaker's bureau of, and receives travel expenses from Alphatec, Spine 360, and K1; receives royalties from Alphatec and Spine 360; and holds stock in K1. Dr N. U. Ahn receives grants and research support from Stryker.

Correspondence should be addressed to: Joshua T. Anderson, BS, 2477 Overlook Rd, Apt 3, Cleve-land Heights, OH 44106 ( Jta38@case.edu).

Received: January 21, 2015
Accepted: April 02, 2015

10.3928/01477447-20151120-02

Sign up to receive

Journal E-contents