If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Jeremy E. Raducha, MD, Department of Orthopaedic Surgery, Brown University, The Warren Alpert Medical School of Brown University, 593 Eddy Street, Providence, RI 02903.
The annual high volume of carpal tunnel releases (CTRs) has a large financial impact on the health care system. Validating the cost drivers related to CTR in a large, diverse patient population may aid in developing cost reduction strategies to benefit health care systems.
Methods
Adult patients with carpal tunnel syndrome who underwent CTR were identified in the New York Statewide Planning and Research Cooperative System database from 2016 to 2017. The Statewide Planning and Research Cooperative System is a comprehensive all-payer database that collects all inpatient and outpatient preadjudicated claims in New York. A multivariable mixed model regression with random effects was performed for the facility to assess the variables that contributed significantly to the total charge. The variables included were patient age, sex, anesthesia method, whether the surgery took place in an ambulatory surgery center or a hospital outpatient department, operation time in minutes, primary insurance type, race, ethnicity, Charlson Comorbidity Index, and categories for billed procedure codes.
Results
During the period of 2016 to 2017, 8,717 claims were included, with a mean charge per claim of $4,865. General anesthesia was associated with higher charges than local anesthesia. A procedure at a hospital outpatient department was associated with an approximately 48.2% increase in the total charge compared with that at an ambulatory surgery center. A 1-minute increase in the operation time was associated with a 0.3% increase in the total charge. Claims with antiemetics, antihistamines, benzodiazepines, intravenous fluids, narcotic agents, or preoperative antibiotics were associated with higher total charges than claims that did not bill for these. Compared with endoscopic procedures, open procedures had a 44.3% decrease in the total charges.
Conclusions
This comprehensive multivariable model has validated that general anesthesia, hospital-based surgery, the use of antibiotics and opioids, longer operative times, and endoscopic CTR significantly increased the cost of surgery.
Although CTS is initially treated nonsurgically, many patients ultimately undergo surgical management to relieve their symptoms. Over 600,000 carpal tunnel releases (CTRs) are performed annually in the United States, creating a large financial and economic impact.
When considering the Medicare population alone, estimates of the economic impact in the United States are nearly $5 billion per year, with release providing close to $1.6 billion in economic benefit.
There has been a recent emphasis by payers on minimizing costs in the setting of value-based health care models. Given the proportion of patients undergoing CTR, reducing the total costs can have substantial financial implications. Understanding the cost drivers associated with CTR is imperative in minimizing the cost of the procedure.
Several recent studies have been published regarding the costs of CTR.
Reduced costs were associated with open CTR, surgeries performed under local anesthesia, and surgeries performed in the ambulatory surgery centers (ASCs). However, most of these studies were conducted with smaller, single institution cohorts and may not be generalizable to the US population. Foster et al
investigated CTR costs using the PearlDiver database, but they only considered the costs associated with the surgical technique, procedure setting, and anesthetic modality. They did not assess other potential cost drivers such as materials and medications used during surgery, and their database was limited to a single payer. Although several studies have characterized the cost drivers of CTR, none have evaluated them on a granular multivariable level.
The purpose of the present study was to provide a contemporary update and validate the cost drivers of CTR set forth by previous studies using a comprehensive multivariable model in New York state that may be generalizable to the United States overall.
Using the New York Statewide Planning and Research Cooperative System (SPARCS) database, we aimed to simultaneously analyze a variety of factors that could have an impact on costs, including surgical venue, the type of anesthesia, method of CTR, and antibiotic administration. By defining the cost associated with each factor, we aimed to propose strategies to reduce the total health care dollars spent on treating CTS.
Materials and Methods
Adult patients (18 years and older) were identified in the New York SPARCS database from 2016 to 2017. SPARCS is a comprehensive all-payer database that collects all inpatient and outpatient preadjudicated claims in New York. This includes all International Classification of Diseases diagnosis codes and International Classification of Diseases/Current Procedural Terminology (CPT) procedure codes associated with all visits.
Claims in the outpatient setting were identified using the International Classification of Diseases, Ninth Revision, Clinical Modification and International Classification of Diseases, Tenth Revision, Clinical Modification diagnosis codes for CTS (G56.01, G56.02, G56.03). The patients included in the analysis had a primary diagnosis of CTS to avoid more complex cases that would have increased charges. Claims were then further filtered for carpal tunnel surgery (CPT codes 64721, 29848).
Statistical analyses
A multivariable mixed model regression with random effects was performed for the facility to assess the variables that contributed significantly to the total charge of the claim. The total charge was modeled as the natural logarithm of the total charge. The variables included in the regression were patient age, sex, anesthesia method, whether the surgery took place in an ASC or a hospital outpatient department, operation time in minutes, primary insurance type, race, ethnicity, Charlson Comorbidity Index (CCI), and categories for billed procedure codes. For example, the coding for penicillin use was categorized as “antibiotic” in the model. The CCI was calculated using the method described by Deyo et al
and extended to International Classification of Diseases, Tenth Revision, Clinical Modification. The CCI was dichotomized to a score of 0 versus ≥1.
Outliers for operation time in minutes were removed. A lower limit was set at 5 minutes, and an upper limit was set at Q3 + 1.5 ∗ IQR, where the equation is the calculation for the upper limit of estimation; Q3 is the third quartile (75th percentile); and IQR is the interquartile range (75th–25th percentile). Once these outliers were removed, to avoid modeling outlier charges, we used 80% the average charge reported by Medicare for CPT 29848 and 64721 in each year in New York.
These lower bounds were applied to the charges before they were adjusted for inflation. After charges were adjusted for inflation, an upper bound was set at Q3 + 1.5 ∗ IQR for the billed charges. All claims that had a procedure code category with less than 10 occurrences were excluded because of the lesser likelihood that those categories can be reliably modeled.
P values were adjusted using the false discovery rate method. An adjusted P value of <.05 was considered significant for all statistical analyses.
Results
After searching for a primary diagnosis of CTS, 8,717 CTR claims were included from 2016 to 2017. The median charge per claim was $4,162, with a mean of $4,865. The maximum total charge amount included in the analysis was $14,386. Women significantly outnumbered men (P < .0001). Most patients did not have a qualifying comorbidity at the time of surgery. The median time in the operating room was 28 minutes, with a mean of 30 minutes and a maximum of 72 minutes (Table 1).
Figure 1 shows that the health service areas in southern New York have average charges that differ by upward of $6,000 compared with those in western New York. The most expensive region was the health service area 8 at $8,606 per average claim, whereas the least expensive was health service area 1 at $2,378 (Fig. 1).
Figure 1Average carpal tunnel surgery charge by health service area in New York state. Breakdown of the cities in each health service area is included in Table E1 (available online on the Journal’s website at www.jhandsurg.org).
The CPT code 64721 was the dominant procedure with approximately 66.9% cases (Table 1). Besides the primary CPT codes, the top 3 most common procedure code categories were the administration of benzodiazepines (36.2% of cases), antiemetic agents (24.1% of cases), and preoperative antibiotics (20.3% of cases). Most cases were performed at a hospital outpatient department (54.7%) as opposed to an ASC. Local anesthesia was the most common method of anesthesia accounting for 47.7% of cases (Table 2).
Table 2Predictors for Total Billed Amount for Carpal Tunnel Surgery
General anesthesia was associated with higher charges than local or regional anesthesia. A procedure performed at a hospital’s outpatient department was associated with an approximately 48.2% increase in the total charge amount compared with that performed at an ASC. A 1-minute increase in the operation time was associated with a 0.3% increase in the total charge. In terms of the drugs used, claims with antiemetics, antihistamines, benzodiazepines, intravenous fluids, narcotic agents, or preoperative antibiotics were associated with higher total charge amounts than the claims that did not account for them. Compared with endoscopic procedures, open procedures had significantly less charges at a 44.3% reduction. Figure 2 demonstrates that an open CTS has total charge amounts distributed toward lower charges than that of an endoscopic CTS (Fig. 2).
Figure 2Distribution of total claim charges and average charges in endoscopic (above) and open (below) carpal tunnel releases.
Carpal tunnel release surgery is one of the most common hand procedures performed in the United States; therefore, it is associated with a substantial financial impact on the health care system, given the sheer volume of surgeries.
We determined a mean $4,865 per procedure in a large, diverse population in the New York SPARCS database. The mean cost varied tremendously based on the region of New York, from $2,378 in western New York to $8,606 in the area surrounding Manhattan. This is likely because of the higher cost of living in New York City than in the rest of the state and the country overall. Services and goods are typically more expensive in the northeastern region, and medical costs are not exempt from this trend.
Using a comprehensive multivariable approach, we found several statistically significant cost drivers in our database evaluation that serve as validation of previous studies regarding CTS cost, including the use of general anesthesia, hospital-based surgery, operative time, endoscopic release, antibiotic use, narcotic use and the additional use of adrenergic agonists, antiemetics, antihistamines, benzodiazepines, and intravenous fluids (Table 2). We believe that by using a comprehensive multivariable approach simultaneously accounting for all materials, the operation room time, and procedures used in the operating room, we have been able to validate, update, and more accurately describe the marginal effects of these cost drivers of CTR as the accuracy of the magnitude of cost savings is likely to play an integral role in economic decision-making. Furthermore, this provides the opportunity to ordinally rank cost drivers by their magnitude of effect with respect to one another. This study validates the choice of anesthesia to be a significant cost driver for CTR as the use of general anesthesia has previously been shown to cost more than local or regional anesthesia due to the additional costs of medications and the time and expertise of the anesthesiologist.
In addition to the lower cost, there is evidence that the use of wide-awake local anesthesia no tourniquet technique has similar postoperative pain control and patient satisfaction compared with general or monitored anesthesia care anesthesia.
The use of antiemetics, antihistamines, benzodiazepines, and intravenous fluids are also associated with the use of general anesthesia; thus, these costs could be eliminated with the decreased use of general anesthetic. With the lower cost and equivalent outcomes, an increase in local-only anesthesia use beyond the 47.7% rate we found could help decrease costs.
Nevertheless, we understand that there are barriers to achieving this, such as patient expectations and tolerance of wide-awake local anesthesia no tourniquet, so this decision will likely still need to be made on a patient-to-patient basis.
Hospital-based surgery was found to cost 48.2% more on average than ASC-based surgery and had the largest marginal effect on cost of all the variables modeled, thus providing the greatest opportunity to reduce costs in CTR. This also serves as validation of prior studies on CTR and other small hand surgeries.
Along that trend, office-based procedures have shown an even greater reduction in cost than those in ASCs. Using time-driven activity-based costing, White et al
demonstrated that ASC procedures using monitored anesthesia care + local anesthesia cost approximately 3.5 times more and took approximately 3 times longer than office-based procedures with wide-awake local anesthesia no tourniquet technique, with equivalent postoperative pain scores. Hospital-based surgery is associated with higher costs for several reasons, including slower turnover time, more operating room and perioperative staff, and more equipment used. Our study explicitly characterizes cost as a function of operating room time and found that increased time in the operating room increased the overall costs, which corresponds to the higher costs of hospital-based surgery.
Additionally, we found open CTR was associated with 44.3% lower charges than endoscopic procedures on average, which is consistent with most of the literature.
Both open and endoscopic CTRs are viable options for CTR, with long-term outcomes showing equivalent symptomatic relief, pinch and grip strength, complications, and outcomes scores for both open and endoscopic releases in several studies.
Nevertheless, each approach has significantly different costs. Recent estimates of average annual reimbursement are over $1,700 in open CTR and over $2,500 in endoscopic CTR, which is consistent with our findings.
Endoscopic releases require more equipment which must be set up and cleaned, increasing time and costs. The disposable endoscopic blade assembly alone costs $217 on average.
Despite these increased costs and similar long-term outcomes, there are prospective, randomized trials showing an earlier return to work and better early (<3 months) improvement in scar tenderness, pinch strength, and subjective satisfaction in those undergoing endoscopic release.
These results evened out after 3 months, and both approaches maintained their clinical improvements at a mean of 12.8 years after surgery in the study by Atroshi et al.
These early benefits may be important for particular subsets of patients, and the increased costs associated with endoscopic release need to be weighed against these benefits.
By controlling for all materials used, our study characterized increased costs of perioperative antibiotics in CTR. There is abundant evidence that perioperative antibiotics do not affect surgical site infection rates in clean hand surgery, even in patients with diabetes.
Despite this, 20.3% of the patients in this study received perioperative antibiotics, which is higher than those in other studies of clean hand surgery (11% and 13.6%).
Given the nature of our study, we were unable to evaluate the presence of immunocompromised status or other circumstances surrounding each decision to give antibiotics; however, the routine use of perioperative antibiotics in clean hand surgery does not improve outcomes and increases costs and potential complications related to medication side effects (eg, Clostridium difficile or selection of drug-resistant bacteria). The decision to administer preoperative antibiotics should weigh the risks and benefits for each patient.
The use of opioid medications was also associated with increased costs, likely due to the physical cost of the medication in the operating room and perioperative period. Although the direct costs of opioid administration during CTR could be estimated in the current study, the emergence of public health and more widespread societal costs of chronic opioid dependence in the United States are important to consider. The routine use of opioid medications for postoperative pain is losing favor because of the risks of dependence and emerging state regulations.
There is clear evidence that opioids do not improve postoperative pain control compared with acetaminophen and ibuprofen. In a prospective randomized double-blinded trial, Ilyas et al
demonstrated postoperative pain control was not significantly different between patients receiving an opioid, acetaminophen, or ibuprofen following both open and endoscopic CTR, with patients receiving opioids actually trending toward higher pain scores. Four of the 5 complications also occurred in the opioid group with minor side effects. Even if opioids are prescribed, other studies show that the number of pills actually taken by patients is less than that prescribed and for a shorter number of days.
Overall, this is an area of potential cost reduction and public health improvement by decreasing the amount of prescribed opioids.
The strengths of our study are that this is a large, economically diverse patient population that allowed for the evaluation of numerous cost variables in open and endoscopic CTRs. The limitations are that it is a retrospective study, and the nature of the database prevents the evaluation of the details and circumstances surrounding each case and decision. There is also the possibility that the data documented in the database are incomplete. For example, the recorded percentage of patients with a CCI of >1, excluding age, was only 16.2% in a population, with a median age of 58 years, which is low, given that the age of more than 50 years itself is worth at least 1 point. It is possible that noncontributory comorbidities were not completely documented for all CTRs. Also, given the nature of the data used in this investigation, we could not evaluate the effect of surgery costs on patient-centered outcomes. Additionally, this study evaluated only patients in New York state, which could limit its generalizability to other states with different payor mixes or different billing and payment structures. The data used in this study showed variability by region (Fig. 1). Nevertheless, our random effects modeling approach for facility not only helps control for the variances in cost by health service area but also provides greater control of these cost variances than controlling for health service areas would. Finally, this patient population was also predominantly White, and although we did not anticipate that the costs of the operating room and anesthesia would vary based on ethnicity, the costs associated with other patient ethnicities cannot be evaluated fairly. Future studies could investigate the effect of ethnicity on carpal tunnel surgery costs, utilization, and outcomes.
Overall, the New York SPARCS database provided a large, economically diverse patient population to evaluate and validate the costs of CTS. Open surgeries, with local or regional anesthesia, in an ambulatory setting were found to be significant cost savers in this patient population. The additional use of medications, many associated with general anesthesia, was also a separate significant cost driver. Most of these factors are modifiable, and with evidence-based practices, the costs of carpal tunnel surgery can be reduced in the US population as a whole.
Appendix
Table E1Definitions of Health Service Areas in New York State
00695-X&id=gr1.jpg){kind=link}
00695-X&id=gr2.jpg){kind=link}