Implementation of a spine triage program and its effect on outpatient radiology utilization.

OBJECTIVE: Clinical care pathways designed to triage spinal disorders have been shown to reduce wait times and improve patient satisfaction. The goal of this study was to perform an analysis of outpatient radiology costs before and after the implementation of a spine care triage pathway. METHODS: All imaging orders and surgical procedures were captured in a prospective spine registry for patients referred to the department of neurosurgery within a single academic center between July 1, 2017, and November 3, 2020. A spine triage algorithm was developed and implemented January 1, 2018. Healthcare utilization was recorded for 1 year after the first appointment in the department of neurosurgery. Imaging costs were estimated using publicly available data from the Centers for Medicare and Medicaid Services. Statistical analysis consisted of an independent sample t-test or randomization test for continuous variables and a chi-square test for categorical variables. RESULTS: A total of 3854 patients were included in this study. The mean age was 60 years (50.8% female) and 89.8% had undergone advanced imaging before being referred to the department of neurosurgery. In total, 12.6% of patients were referred with a specific surgical diagnosis (i.e., spinal stenosis, lumbar spondylolisthesis, etc.). During the pretriage phase 1810 patients were enrolled, and there were 2044 patients enrolled after the triage algorithm was implemented. Advanced imaging (CT or MRI) was ordered more frequently by providers before the triage program was initiated, with imaging ordered in 34% (617/1810) of patients pretriage versus 14.8% (302/2044) after the triage pathway was implemented (p < 0.001). The authors calculated a significant reduction in cost associated with reduced radiology utilization. Before triage, the cost of radiology utilization was $85,475/1000 patients compared with $40,107/1000 patients afterward (p < 0.001). The triage program did not change the utilization of surgery (14.6% before, 13.6% after). CONCLUSIONS: Among patients treated after a spinal triage program was implemented in a single neurosurgery department, there was a substantial reduction in the use of advanced imaging and a 50% reduction in cost associated with outpatient radiology utilization. The triage program did not change the rate of spine surgery being performed.

Economic Burden of Hospitalizations Associated with Opioid Dependence Among Patients Undergoing Spinal Fusion.

BACKGROUND: The current study seeks to examine the association between chronic opioid use and postoperative outcomes for patients undergoing anterior cervical discectomy and fusion (ACDF) and posterior lumbar fusion (PLF). METHODS: The National Inpatient Sample was queried for patients with and without chronic opioid use undergoing ACDF or PLF for degenerative disc disease between 2012 and 2015 using ICD-9 diagnosis and procedure codes. Multivariable conditional logistic regression was performed to assess the association between chronic opioid use and length of stay (LOS), nonhome discharge, and hospital charge. RESULTS: A total of 391 patients undergoing ACDF and 644 patients undergoing PLF with opioid dependence were identified. On multivariable regression analysis, opioid dependence was significantly associated with an increased LOS (mean, 3.09 days vs. 2.16 days; odds ratio (OR) for prolonged LOS (>3 days), 2.11; 95% confidence interval [CI], 1.43-3.14; P < 0.001). Although on unadjusted analyses, patients with opioid dependence undergoing ACDF were found to have higher hospital charges (mean, U.S. $18,698.42 vs. $11,378.61; P < 0.001) and higher rates of nonroutine discharge (19.18% vs. 10.21%; P < 0.001), the multivariable regression analyses found no significant association between opioid dependence and odds of hospital charges >75th percentile (OR, 1.44; 95% CI, 0.84-2.47; P = 0.188) or nonroutine discharge (OR, 1.48; 95% CI, 0.93-2.34; P = 0.098). For those undergoing PLF, opioid dependence was significantly associated with increased hospital charges (mean, U.S. $37,712.98 vs. $30,475.43, P < 0.001; OR for hospital charge >75th percentile, 1.78, 95% CL, 1.23-2.58, P = 0.002), LOS (mean, 3.42 days vs. 2.30 days; OR for prolonged LOS, 1.53; 95% CI, 1.16-2.00; P = 0.003), and nonroutine discharge (46.89% vs. 36.47%; OR, 1.74; 95% CI, 1.34-2.26; P < 0.001) on both unadjusted and adjusted multivariable regression analyses. CONCLUSIONS: Our analysis using a national administrative database showed that opioid dependence may be associated with worse economic outcomes for patients undergoing ACDF and PLF.

Prevalence and Trends in Management of Idiopathic Normal Pressure Hydrocephalus in the United States: Insights from the National Inpatient Sample.

BACKGROUND: Over the past 2 decades, management of idiopathic normal pressure hydrocephalus (iNPH) has evolved significantly. In the current study, we sought to evaluate the national prevalence and management trends of iNPH in the United States using a national database. METHODS: The National Inpatient Sample was queried for patients with an International Classification of Diseases diagnosis code for iNPH from 2007 to 2017. Trends in prevalence and procedure type were evaluated per 100,000 discharges and as a percentage of discharges, using weighted discharges. Utilization of procedure type across U.S. regions and hospital types was also compared. RESULTS: From 2007 to 2017, 302,460 weighted discharges with any diagnosis code for iNPH, aged ≥60 years, were identified. Prevalence ranged from 0.04% to 0.20% (41/100,000 to 202/100,000) among admitted patients ≥60 years old, giving an average prevalence during the study duration of 0.18% (179/100,000). Of 66,759 weighted discharges with a primary diagnosis code of iNPH undergoing surgical management, ventriculoperitoneal shunt (72.0% of discharges, n = 48,977) was most commonly used; of these, 9.3% (n = 4567) were performed laparoscopically. This result was followed by lumbar peritoneal shunt (15.1% of discharges, n = 10,441). Up to 15.1% (n = 9990) of discharges reported only a lumbar puncture, assumed to be only diagnostic, for screening, or part of serial cerebrospinal fluid removal procedures. Significant discrepancies in procedure utilization were also identified among hospitals in the Western, Southern, Northeast and Midwest regions, as well as between urban and rural hospitals (P < 0.05). CONCLUSIONS: We have summarized the national prevalence of iNPH, trends in its management over the previous decade and trends by region and hospital type.

Predictors of adverse outcomes and cost after surgical management for idiopathic normal pressure hydrocephalus: Analyses from a national database.

INTRODUCTION: We utilized a national administrative database to investigate drivers of immediate adverse economic and hospital outcomes, including non-routine discharge, prolonged length of stay (LOS), and admission costs among patients undergoing surgery for idiopathic normal pressure hydrocephalus (iNPH). METHODS: The National Inpatient Sample (NIS) was queried from 2007 to 2017 for patients aged ≥60 with a diagnosis code for iNPH undergoing surgery. Multivariable logistic-regression models and Wald χ2 were used to identify drivers of non-routine discharge, prolonged length of stay (LOS) (>75th percentile) and higher admission costs (>90th percentile). RESULTS: A total of 13,363 patients with iNPH undergoing surgical management were identified. The most common comorbidity reported in the cohort was a cardiovascular pathology (56.9 %, n = 7,787), followed by urinary pathology (37.2 %, n = 5,084), osteoarthritis (7.8 %, n = 1,071), Alzheimer's disease (4.6 %, n = 626) and cerebrovascular pathology (4.2 %, n = 569). The most frequently employed procedure was ventriculo-peritoneal (VP) shunt placement (65.6 %, n = 8,942) of which 89.8 % (n = 8,027) were performed open and 10.2 % (n = 915) laparoscopically. This was followed by lumbo-peritoneal (LP) shunting (15.5 %, n = 2,115), lumbar puncture alone (screened, serial CSF removal) (14.8 %, n = 2,013), endoscopic third ventriculostomy (ETV) (2%, n = 274), ventriculo-atrial (VA) shunt (0.95 %, n = 130) and ventriculo-pleural (Vpleural) shunt (0.46 %, n = 64). The median (IQR) LOS was 3 days (2-5), the rate of non-routine discharge was 37.3 % and median (IQR) cost was $11,230 ($7,735-15,590). On multivariable-analysis, emergent-admission (OR 2.91), older age (76-90: OR 1.55; 90+: OR 2.66), VP shunt (open: OR 3.09; laparoscopic: OR 2.32), ETV (OR 3.16), VA/VPleural shunt (OR 2.73) and hospital admission in Northeast-region compared to Midwest (OR 1.27) were found to be associated with increased risk of non-routine discharge. Some of the highly significant associated factors for prolonged LOS included emergent-admission (OR 11.34), ETV (OR 10.92), VA/VPleural shunt (OR 7.79) and open VP shunt (OR 8.24). For increased admission costs, some of the highly associated factors included VA/VPleural shunt (OR 18.48), laparoscopic VP shunt (OR 9.92), open VP shunt (OR 12.72) and ETV (OR 9.34). Predictor importance analysis revealed emergent admission, number of diagnosis codes (comorbidities) open VP shunt, hospital region, age] and revision or removal of shunt to be the most important drivers of these outcomes. CONCLUSION: Analyses from a national database indicate that among patients with iNPH, an emergent-admission may be the most significant risk-factor of adverse economic outcomes and higher costs.

Comparison of Costs and Postoperative Outcomes between Vertebroplasty and Kyphoplasty for Osteoporotic Vertebral Compression Fractures: Analysis from a State-Level Outpatient Database.

BACKGROUND: Vertebral cement augmentation techniques are routinely used to treat osteoporotic vertebral compression fractures (VCFs). In the current study, we used a state-level outpatient database to compare costs and postoperative outcomes between vertebroplasty and kyphoplasty. METHODS: We queried the 2016 Florida State-Ambulatory Surgery Database of the Healthcare Cost and Utilization Project for patients undergoing thoracolumbar vertebroplasty or kyphoplasty for osteoporotic VCFs. Demographic and clinical characteristics, as well as postoperative outcomes were compared between the 2 groups. RESULTS: A total of 105 patients (11.6%) who underwent vertebroplasty and 801 patients (88.4%) who underwent kyphoplasty were identified. Patients undergoing kyphoplasty were more likely to stay overnight or longer, with the P value trending toward significance (kyphoplasty with >1 day stay: 7.4% vs. vertebroplasty with >1 day stay: 1.9%; P = 0.086). Patients undergoing vertebroplasty had a significantly higher rate of discharge to home routine compared with patients undergoing kyphoplasty (97.1% [n = 102] vs. 94.1% [n = 754]; P < 0.001). Undergoing kyphoplasty was also associated with higher index admission costs ($40,706 vs. $18,965; P < 0.001) and higher readmission costs ($27,038 vs. $11,341; P = 0.046). The rates of 30-day and 90-day readmission were similar between the 2 groups (all P > 0.05). The rates of 30-day, 90-day, and overall readmission because of a new-onset fracture were also similar (all P > 0.05). However, vertebroplasty had a higher rate of readmissions associated with a procedure within a year (21.9% [n = 23] vs. 14.5% [n = 116]; P = 0.047). CONCLUSIONS: Our analyses from a state-level database of patients undergoing vertebroplasty and kyphoplasty for osteoporotic VCFs show similar postoperative outcomes for the 2 procedures but a higher cost for kyphoplasty.

Cost of Readmissions Following Anterior Cervical Discectomy and Fusion: Insights from the Nationwide Readmissions Database.

BACKGROUND: Postoperative readmissions are a significant driver of variation in bundled care costs associated with cervical spine surgery. OBJECTIVE: To determine the factors predicting the cost of readmission episodes following elective anterior cervical discectomy and fusion (ACDF). METHODS: We queried the Healthcare Cost and Utilization Project Nationwide Readmissions Database for patients undergoing elective ACDF during 2012 to 2015. Multivariable linear regression was performed to establish the factors associated with the cost of each 30-/90-d readmission episode. RESULTS: A total of 139 877 and 113 418 patients met inclusion criteria for the evaluation of 30- and 90-d readmissions, respectively. Among these, the national rates of 30- and 90-d readmission after an elective ACDF were 3% and 6%, respectively. The median cost of a 30- and 90-d readmission episode was $6727 (IQR: $3844-$13 529) and $8507 (IQR: $4567-$17 460), respectively. Relative predictor importance analysis revealed that the number of procedures at index admission (IA), length of stay at IA, and time elapsed between index surgical admission and readmission were the top predictors of both 30- and 90-d readmission costs (all P < .001). Although cervical myelopathy accounted for only 3.6% of all 30-d readmissions, it accounted for the largest share (8%) of 30-d readmission costs. CONCLUSION: In this analysis from a national all-payer database, we determined the factors associated with the cost of readmissions following elective ACDF. These results are important in assisting policymakers and payers with a better risk adjustment in bundled care payment systems and for surgeons in implementing readmission cost-reduction efforts.