Abstract
Background:
In recent years, the issue of out-of-network billing for privately insured patients has been highlighted as a source of unexpected out-of-pocket charges for patients, even in the setting of an in-network primary surgeon. Congress recently passed the No Surprises Act to curtail these practices, but the new law contains exceptions, and its regulatory system has yet to be established. As one of the most frequently performed major surgical procedures, hysterectomy represents a significant exposure to out-of-network bills among non-elderly females in the United States.
Objective:
To describe the extent and nature of out-of-network bills at the time of hysterectomy among privately insured patients in the context of the recently passed No Surprises Act.
Study Design:
We performed a retrospective cohort study of women ages 18-64 who underwent simple hysterectomy from 2008 to 2018 with in-network primary surgeon in the IBM Watson Marketscan claims database, which includes data from over 350 different payers. We identified out-of-network claims for facility or professional services, and analyzed the frequency, size, and source of the payments. We used multivariable logistic regression to assess for patient, procedure, and facility characteristics associated with risk of out-of-network claims.
Results:
We identified 585,223 hysterectomy cases meeting all inclusion criteria, evenly split between inpatient (49.6%) and outpatient (50.4%) procedures. Overall, 8.8% of cases included at least 1 out-of-network claim, with median out-of-network expenditures of $553 for inpatient procedures and $438 for outpatient procedures. Compared to professional out-of-network claims, facility out-of-network claims were less common (2.3% vs. 7.4%) but far greater in amount billed (median $8,307 vs. $400 inpatient, $3,281 vs. $407 outpatient). Among professional claims, those from midlevel surgical assistants were most frequently out-of-network when present (13.8% inpatient; 20.0% outpatient), while out-of-network claims from anesthesia were most common overall, and largest (median $890 inpatient, $1,021 outpatient) when present. In a multivariable model, older age, increasing comorbidity, and complications during the stay were associated with higher odds of any out-of-network claim. In contrast, risk of facility out-of-network claims was more strongly associated with facility region and surgical approach, with highest odds for cases in the North Central region and those using robotic approach.
Conclusion:
Out-of-network bills for privately insured patients at the time of hysterectomy occurred in 8.8% of cases. Approximately one-quarter of these included out-of-network facility claims, which tended to have higher payments than out-of-network professional claims, and may not be prevented by the No Surprises Act. Gynecologic surgeons should be aware of potential out-of-network charges for ancillary services at the time of surgery, and particularly the network status of the facility, in order to provide maximal transparency and financial protection to our patients.
Keywords: hysterectomy, out-of-network billing, out-of-pocket costs, health policy, financial toxicity, private health insurance
Introduction
Patients with private health insurance sometimes face unexpected charges after inadvertently or unknowingly receiving care from an out-of-network provider or facility, in what is often referred to as a “surprise” medical bill.1,2 Depending on plan details, out-of-network claims can increase patient spending via balance billing, where the provider directly bills the patient for the remaining charges after insurance company reimbursement, or via higher cost sharing.3,4 The No Surprises Act was signed into law on December 27, 2020 with the aim of curtailing this practice through new regulations.5 However, the legislation will not be implemented until 2022 and continues to face intense opposition from powerful lobbying groups attempting to influence the specific rules and systems behind enforcement.6,7 Additionally, the law includes certain exceptions, particularly for out-of-network facility claims, which may continue to leave patients exposed to unexpected and burdensome charges.8
Hysterectomy is one of the most common major surgical procedures performed in the United States (US), with upwards of 500,000 procedures performed annually, 90% of which are among non-elderly (under 65) Americans who lack universal access to Medicare coverage.9,10 While the vast majority of hysterectomies are performed in a non-emergent setting where the patient has the flexibility to plan for the procedure and select their primary surgeon, numerous other parties may bill for ancillary care at the time of the procedure, including the facility, and anesthesia services, pathology, surgical assistant, and medical/surgical consultant services. A prior study found that 25% of 67,000 included hysterectomies included out-of-network claims.11 However, this work had several limitations: (1) the database included only one large insurer, (2) the study excluded cases with out-of-network facility claims that will not be prevented by the recently passed No Surprises Act, and (3) it assessed multiple surgical procedures, among which hysterectomy is relatively unique. Hysterectomy is performed exclusively by gynecologists, with significant variety in the surgical technique utilized, including open, laparoscopic, robotic, and vaginal approaches. Additionally, known biases in reimbursement that lead to lower payments for female-specific procedures could also impact out-of-network billing patterns,12 and the role of indication and route of surgery on the risk of out-of-network billing remains unexplored.
Our objective in this study was to use a large commercial claims database to describe the frequency, size, and source of out-of-network billing for hysterectomy cases in the privately insured population. We also sought to assess for patient, facility, and surgical characteristics associated with increased risk of out-of-network bills, and to estimate the extent to which the No Surprises Act will prevent this practice in this setting.
Materials and Methods
We performed a retrospective cohort study of the IBM Watson Marketscan Commercial Claims and Encounters database, which includes approximately 350 different insurers.13 This database includes claims for inpatient and outpatient services, and includes an indicator for each service to designate it as an in-network or out-of-network claim. Marketscan data is de-identified and the Columbia University Institutional Review Board deemed this study exempt from review.
Within Marketscan, we identified female patients ages 18 to 64 who underwent simple hysterectomy (with or without additional procedures) between 2008 and 2018 (see Appendix A: Table A.1). We limited the study cohort to patients with continuous coverage and from 3 months prior to 1 month after the date of surgery. We additionally excluded cases with obstetric indications, given that these hysterectomies were likely unplanned and complicated by additional care related to the pregnancy/delivery. We further excluded patients who were on capitated plans due to the different reimbursement and incentive structure for providers, and patients with missing data for facility fees, professional fees, or in-network versus out-of-network indicators. Lastly, we excluded patients with an out-of-network or missing primary surgeon. In most cases, patients will have information on the network status of their primary surgeon; therefore, many of these cases likely represent an intentional choice by the patient to seek care from a particular surgeon, understanding the potential of higher cost. We included all claims from the dates of inpatient stay for inpatient cases, and the date of surgery for outpatient cases.
We described the demographic and surgical characteristics of the final cohort, including year of the procedure, age (18-39, 40-49, 50-59, 60-64), Elixhauser comorbidity score (0, 1, ≥2),14 region (northeast, north central, south, west), metropolitan statistical area residence (MSA defined by US Office of Management and Budget, vs. non-MSA, unknown),15 facility type (inpatient, outpatient), indication for surgery (cancer, endometrial hyperplasia, fibroids, abnormal uterine bleeding, dysmenorrhea, endometriosis, benign mass, or prolapse; see Table A.2 for International Classification of Diseases (ICD), 9th and 10th revision, clinical modification code categorization), route of surgery (open/abdominal, laparoscopic, robotic, or vaginal; see Table A.1 for code categorization), and complications (categorized as intraoperative, surgical site, medical, or transfusion; see Table A.3 for code categorization).
Our primary outcome was the percentage of cases that included at least one out-of-network claim. For secondary outcomes, we described the breakdown of claim payments (total, out-of-pocket, and out-of-network). We also assessed outcomes by claims category (facility versus professional), and by service (e.g. anesthesia, pathology, co-surgeon, midlevel surgical assistant, etc.). The presence of out-of-network facility claims was considered of particular relevance, as such bills will not be banned by the No Surprises Act when implemented in 2022. We compared the presence of out-of-network claims between subgroups with χ2 test, and compared median expenditures with the Wilcoxon Rank Sum test. As expenditure data was highly skewed, we did not report on, or perform statistical comparisons of, means, due to violations of assumptions of normality.
In order to explore the contribution of various patient, facility, and surgical characteristics to the risk of out-of-network claims in cases of simple hysterectomy, we created a multivariable logistic regression model. In the model, we included all available variables that we a priori felt may be predictive of the risk of out-of-network expenditures: age, comorbidity, facility location and type, hysterectomy type and indication, complications, and year. For this assessment, we simplified the specification of hysterectomy indication to malignant vs. benign. We assessed two different outcomes with this model. In the first model, we used the outcome of an indicator for the presence of at least one out-of-network claim of any kind to describe risk factors in at present. We then ran the same model with the alternative outcome of an indicator for an out of network facility claim specifically, as this is the primary notable exception to the protections in the No Surprises Act and represents risk factors that may persist if the law is implemented as planned in 2022.
All costs were adjusted for inflation to 2018 US dollars by consumer price index.16 We considered statistical significance by p<0.05. STROBE guidelines for observational research were followed. All analyses were conducted with SAS Studio version 3.71 (SAS Institute, Cary, NC, USA). Figures were created with SAS and Microsoft PowerPoint/Excel v14.7.7 (Microsoft Corporation, Redmond, WA, USA).
Results
We identified an initial cohort of 1.01 million cases of simple hysterectomy among females ages 18-64 in the Marketscan database between 2008 and 2018. After applying exclusions to limit the cohort with in-network primary surgeon and complete data on both facility and inpatient claims, we identified 585,223 cases meeting all inclusion criteria. Among included cases, 49.6% were performed inpatient and 50.4% were performed and billed as an outpatient procedure (see Appendix A: Figure A.1 for CONSORT diagram). Overall, 51,512 cases, representing 8.8% of the included cohort, included at least one out-of-network claim. Out-of-network claims were slightly more frequent in outpatient vs. inpatient cases (9.0% vs. 8.6%; p<0.001). While the proportion of cases with out-of-network facility claims has shown some decline over time from 3.3% in 2008 to 1.2% in 2018 (p=0.001), the proportion of cases with any out-of-network claim has fluctuated somewhat over time, with a slight rise from 7.5% to 9.0% from 2016 to 2018 (p=0.001; see Figure 1).
Figure 1.
Time trends in proportion of cases with ≥1 out-of-network claim for hysterectomy, IBM Watson Marketscan database, 2008-2018. Error bar indicates 95% CI.
Clinical characteristics of included patients are shown in Table 1. Although the majority of hysterectomies (72.1%) in the cohort were performed among patients under the age of 50, patients above 50 showed slightly higher rates of out-of-network claims (9.1% vs. 8.7%; p<0.001). The most common coded diagnoses for indications for hysterectomy included fibroids (61.9%), abnormal uterine bleeding (58.3%), and endometriosis/adenomyosis (36.6%). Among indications, those with a diagnosis of prolapse (14.3% of included cases) showed slightly increased risk of out-of-network claims (9.3%). Robotic hysterectomy was the least common surgical approach in the study cohort (7.4% of included cases), but had the highest rate of out-of-network claims (10.9%). Cases with complications or length of stay of 5 days or longer had greater exposure to out-of-network claims (12.9% for stay ≥5 days, 10.0% for intraoperative complication, 11.6% for surgical site complication, and 11.2% for medical complication).
Table 1.
Clinical characteristics of hysterectomy cases for non-elderly adults, with and without out-of-network claims, IBM Watson Marketscan database, 2008-2018.
| Characteristic | No out-of- network claim |
≥1 out-of- network claim |
Percent ≥1 out-of- network claim |
P Value |
|---|---|---|---|---|
| Total Claims (N) | 533,711 | 51,512 | 8.8% | - |
| Age | <0.001 | |||
| 18-39 | 124,998 | 11,690 | 8.6% | |
| 40-49 | 260,406 | 24,970 | 8.7% | |
| 50-59 | 117,767 | 11,724 | 9.1% | |
| 60-64 | 30,540 | 3,128 | 9.3% | |
| Elixhauser Comorbidities | <0.001 | |||
| 0 | 269,911 | 25,597 | 8.7% | |
| 1 | 146,564 | 14,048 | 8.7% | |
| ≥2 | 117,236 | 11,867 | 9.2% | |
| Indication(s) | ||||
| Gynecologic cancer | 27,266 | 2,729 | 9.1% | 0.06 |
| Endometrial Hyperplasia | 25,169 | 2,373 | 8.6% | 0.26 |
| Fibroids | 329,920 | 32,365 | 8.9% | <0.001 |
| Abnormal Uterine Bleeding | 310,919 | 30,006 | 8.8% | 0.98 |
| Dysmenorrhea | 90,210 | 8,311 | 8.4% | <0.001 |
| Endometriosis | 195,916 | 18,148 | 8.5% | <0.001 |
| Benign (non-fibroid) mass | 133,621 | 12,844 | 8.8% | 0.61 |
| Prolapse | 76,182 | 7,838 | 9.3% | <0.001 |
| Metropolitan Statistical Area | <0.001 | |||
| Metropolitan | 417,187 | 42,326 | 9.2% | |
| Non-Metropolitan | 98,676 | 8,478 | 7.9% | |
| Unknown | 17,848 | 708 | 3.8% | |
| Region | <0.001 | |||
| Northeast | 61,822 | 6,164 | 9.1% | |
| North Central | 127,448 | 12,246 | 8.8% | |
| South | 260,245 | 25,025 | 8.8% | |
| West | 72,680 | 7,748 | 9.6% | |
| Unknown | 11,516 | 329 | 2.8% | |
| Surgery Site | <0.001 | |||
| Inpatient | 265,557 | 24,901 | 8.6% | |
| Outpatient | 268,154 | 26,611 | 9.0% | |
| Hysterectomy Type | <0.001 | |||
| Open | 175,068 | 16,477 | 8.6% | |
| Laparoscopic | 250,422 | 24,403 | 8.9% | |
| Robotic | 38,703 | 4,748 | 10.9% | |
| Vaginal | 69,518 | 5,884 | 7.8% | |
| Length of stay | <0.001 | |||
| 0-1 days | 354,964 | 34,300 | 8.8% | |
| 2-4 days | 167,619 | 15,571 | 8.5% | |
| ≥5 days | 11,128 | 1,641 | 12.9% | |
| Surgical Complications | ||||
| Intraoperative | 14,908 | 1,665 | 10.0% | <0.001 |
| Surgical Site | 7,358 | 966 | 11.6% | <0.001 |
| Medical complication | 20,286 | 2,571 | 11.2% | <0.001 |
| Transfusion | 26,338 | 2,808 | 9.6% | <0.001 |
Excludes claims with out-of-network primary surgeon; non-elderly = ages 18-64; P value by χ2 test
The breakdown of the source of out-of-network expenditures is shown in Table 2. The pattern of risk by claims category did not notably differ between inpatient and outpatient cases. Professional claims from out-of-network were more commonly present than out-of-network facility claims (7.4% vs. 2.3%), and 0.9% of cases included out-of-network claims from both the facility and professional services. Aside from primary surgeon claims, professional claims from particular service categories were not present for all cases. Unlike anesthesia and pathology claims, which were identified for the vast majority of hysterectomies (89.7% and 86.0% of cases), claims for other services were present in only a minority of cases, lessening the exposure to potential out-of-network claims. Among professional services, out-of-network claims for anesthesia services had the highest median expenditures (see Figure 2), and were the most common (15,500 cases, 3.0% of cases with an anesthesia claim), followed by pathology (10,300 cases; 2.0% of cases with a pathology claim). However, one service categories showed a particularly high risk of claims being out-of-network when present; claims for midlevel surgical assistants (nurse practitioners and physician assistants); while only present in 5.4% of cases, over 18.1% of these claims were from out-of-network. Other claims types at high-risk of being out-of-network when billed included: medical consultants (4.4% of claims out-of-network), co-surgeon (3.8% of claims out-of-network), and other (not otherwise categorized; 15.2% of claims out-of-network).
Table 2.
Professional claims for non-primary surgeon, total and out-of-network, for non-elderly adults undergoing inpatient and outpatient hysterectomy, IBM Watson Marketscan database, 2008-2018.
| Claim Category | Inpatient Hysterectomy | Outpatient Hysterectomy | ||||
|---|---|---|---|---|---|---|
| Total Claims | Out-of-Network Claims | Total Claims | Out-of-Network Claims | |||
| n (overall %) | n (overall %) | % by Service* | n (overall %) | n (overall %) | % by Service* | |
| Facility Claim | 290,458 (100.0%) | 6,372 (2.2%) | - | 294,765 (100.0%) | 6,805 (2.3%) | - |
| Professional Claim | 290,458 (100.0%) | 21,380 (7.4%) | - | 294,765 (100.0%) | 22,156 (7.5%) | - |
| Co-Surgeon | 132,588 (45.6%) | 4,080 (1.4%) | 3.1% | 118,922 (40.3%) | 5,542 (1.9%) | 4.7% |
| Surgical Assistant | 9,782 (3.4%) | 1,350 (0.5%) | 13.8% | 21,878 (7.4%) | 4,378 (1.5%) | 20.0% |
| Anesthesia | 239,090 (82.3%) | 9,327 (3.2%) | 3.9% | 285,647 (96.9%) | 6,173 (2.1%) | 2.2% |
| Pathology | 227,434 (78.3%) | 5,329 (1.8%) | 2.3% | 275,847 (93.6%) | 4,971 (1.7%) | 1.8% |
| Radiology | 28,838 (9.9%) | 608 (0.2%) | 2.1% | 7,412 (2.5%) | 153 (0.1%) | 2.1% |
| Surgical Consultants | 77,108 (26.5%) | 1,412 (0.5%) | 1.8% | 92,489 (31.4%) | 1,299 (0.4%) | 1.4% |
| Medical Consultants | 55,653 (19.2%) | 2,398 (0.8%) | 4.3% | 34,375 (11.7%) | 1,598 (0.5%) | 4.6% |
| Other | 17,488 (6.0%) | 1,781 (0.6%) | 10.2% | 20,430 (6.9%) | 3,976 (1.3%) | 19.5% |
Excludes claims with out-of-network primary surgeon; non-elderly = ages 18-64
Percentage by (n out-of-network claims / n total claims) within each professional service claim category
Figure 2.
Bar chart depicting median out-of-network expenditure ($) when present, by service, IBM Watson Marketscan database, 2008-2018. Error bar indicates IQR
The value of total expenditures, out-of-pocket spending, and out-of-network payments for cases with and without out-of-network claims are described in Table 3. Again, patterns of expenditures for inpatient and outpatient cases were generally similar (median total expenditures $13,490 inpatient, $13,463 outpatient). Among hysterectomy cases with at least one out-of-network claim, median out-of-network expenditures were $553 for inpatient cases and $438 for outpatient cases. Notably, the expenditures for out-of-network facility claims were far greater in value than for professional claims (median $8,307 vs. $400 inpatient; $3,281 vs. $407 outpatient; see Figure 3). Median out-of-network expenditures were the largest in the 0.9% of cases with both out-of-network facility and professional claims (median $10,448 inpatient, $6,673 outpatient).
Table 3.
Expenditure breakdown for inpatient and outpatient hysterectomy claims for non-elderly adults, with and without any out-of-network claim, IBM Watson Marketscan database, 2008-2018.
| Outcome ($) | No Out-of-Network Claims median (IQR) |
≥1 Out-of-network Claim median (IQR) |
|---|---|---|
| Inpatient Hysterectomy | ||
| Overall Expenditures | ||
| Total | $13,431 (10,233-18,357) | $14,203 (10,442-20,112) |
| Out-of-pocket | $1,146 (269-2,157) | $1,042 (126-2,271) |
| Out-of-network | - | $553 (114-2,665) |
| Facility Expenditures | ||
| Total | $10,145 (7,236-14,612) | $10,483 (7,173-15,548) |
| Out-of-pocket | $720 (115-1,547) | $496 (0-1,449) |
| Out-of-network | - | $8,307 (0-12,990) |
| Professional Expenditures | ||
| Total | $3,092 (2,464-4,035) | $3,411 (2,653-4,763) |
| Out-of-pocket | $269 (0-606) | $307 (0-762) |
| Out-of-network | - | $400 (127-1,107) |
| Outpatient Hysterectomy | ||
| Overall Expenditures | ||
| Total | $13,561 (9,204-19,678) | $12,449 (8,447-18,939) |
| Out-of-pocket | $1,208 (173-2,262) | $1,424 (296-2,621) |
| Out-of-network | - | $438 (108-1,698) |
| Facility Expenditures | ||
| Total | $10,111 (5,876-15,864) | $8,154 (4,817-14,454) |
| Out-of-pocket | $673 (56-1,556) | $616 (0-1,568) |
| Out-of-network | - | $3,281 (0-12,271) |
| Professional Expenditures | ||
| Total | $3,215 (2,640-4,078) | $3,574 (2,845-4,715) |
| Out-of-pocket | $333 (0-$708) | $518 (40-1,038) |
| Out-of-network | - | $407 (146-1,028) |
Excludes claims with out-of-network primary surgeon; non-elderly = ages 18-64; IQR = interquartile range
Risk factors for presence of out-of-network bills, overall and facility-specific claims, were assessed with multiple logistic regression (see Table 4). The odds of out-of-network claims was lower in more recent years (aOR=0.68, 95%CI 0.65-0.72, for any out-of-network claim, aOR=0.35, 95%CI 0.31-0.39 for facility out-of-network claim, 2018 vs. 2008). After adjustment, outpatient cases had slightly higher odds of out-of-network claims, overall (aOR=1.19, 95%CI 1.16-1.23) and specific to facility (aOR=1.06, 95%CI 1.01-1.12). Older age, increased comorbidity, and presence of complications were associated with increased odds of any out-of-network claim, but these findings were not consistent for facility claims. Facility out-of-network claims were most strongly driven by the facility region, with the lowest odds in Northeast (referent), and significantly greater odds in the North Central (aOR=4.36, 95%CI 4.03-4.72), South (aOR=1.41, 95%CI 1.30-1.52) and West (aOR=1.32, 95%CI 1.20-1.45), as well as the surgical technique, with the lowest risk for open cases (referent) versus higher risk for robotic (aOR=1.91, 95%CI 1.78-2.06), laparoscopic (aOR=1.36, 95%CI 1.29-1.43), and vaginal cases (aOR=1.14, 95%CI 1.07-1.22). When also considering professional claims, robotic cases remained at elevated odds (aOR=1.31, 95%CI 1.26-1.36), while vaginal cases were at overall lower odds (aOR=0.86, 95%CI 0.84-0.89).
Table 4.
Characteristics associated with presence of out-of-network claims for non-elderly adults undergoing hysterectomy, multiple logistic regression, IBM Watson Marketscan, 2008-2018.
| Characteristic | ≥1 Out-of-network claim (any) aOR (95%CI) |
≥1 Out-of-network facility claim aOR (95%CI) |
|---|---|---|
| Age | ||
| 18-39 | ref. | ref. |
| 40-49 | 1.020 (0.997-1.044) | 0.99 (0.94-1.03) |
| 50-59 | 1.06 (1.03-1.09)* | 1.04 (0.99-1.09) |
| 60-64 | 1.11 (1.06-1.16)* | 1.12 (1.03-1.21)* |
| Elixhauser Comorbidities | ||
| 0 | ref. | ref. |
| 1 | 1.01 (0.99-1.03) | 0.962 (0.922-1.002) |
| ≥2 | 1.07 (1.04-1.09)* | 0.96 (0.92-1.01) |
| Cancer Indication (vs. benign) | 0.956 (0.916-0.998)* | 1.10 (1.02-1.20)* |
| Non-MSA residence† (vs. MSA) | 0.86 (0.84-0.88)* | 1.03 (0.99-1.08) |
| Region † | ||
| Northeast | ref. | ref. |
| North Central | 0.94 (0.91-0.97)* | 4.36 (4.03-4.72)* |
| South | 0.96 (0.93-0.99)* | 1.41 (1.30-1.52)* |
| West | 1.09 (1.05-1.13)* | 1.32 (1.20-1.45)* |
| Outpatient (vs. inpatient) | 1.19 (1.16-1.23)* | 1.06 (1.01-1.12)* |
| Hysterectomy Type | ||
| Open | ref. | ref. |
| Laparoscopic | 1.00 (0.97-1.03) | 1.36 (1.29-1.43)* |
| Robotic | 1.31 (1.26-1.36)* | 1.91 (1.78-2.06)* |
| Vaginal | 0.86 (0.84-0.89)* | 1.14 (1.07-1.22)* |
| Complications | ||
| Intraoperative | 1.10 (1.04-1.16)* | 1.00 (0.90-1.11) |
| Surgical Site | 1.27 (1.19-1.37)* | 1.51 (1.31-1.73)* |
| Medical complication | 1.30 (1.24-1.36)* | 1.02 (0.93-1.13) |
| Transfusion | 1.07 (1.03-1.12)* | 0.71 (0.64-0.79)* |
| Year | ||
| 2008 | ref. | ref. |
| 2009 | 0.90 (0.87-0.94)* | 1.068 (1.001-1.139)* |
| 2010 | 0.69 (0.66-0.72)* | 0.74 (0.69-0.80)* |
| 2011 | 0.64 (0.62-0.67)* | 0.65 (0.60-0.70)* |
| 2012 | 0.65 (0.63-0.68)* | 0.53 (0.49-0.57)* |
| 2013 | 0.66 (0.63-0.69)* | 0.60 (0.55-0.65)* |
| 2014 | 0.63 (0.60-0.66)* | 0.60 (0.56-0.65)* |
| 2015 | 0.65 (0.62-0.68)* | 0.44 (0.40-0.48)* |
| 2016 | 0.55 (0.52-0.58)* | 0.37 (0.34-0.41)* |
| 2017 | 0.63 (0.59-0.66)* | 0.37 (0.33-0.41)* |
| 2018 | 0.68 (0.65-0.72)* | 0.35 (0.31-0.39)* |
Excludes claims with out-of-network primary surgeon; non-elderly = ages 18-64; aOR = adjusted odds ratio; MSA = metropolitan statistical area, defined by US Office of Management and Budget; see supplemental tables for coding definitions for hysterectomy type and complications
p<0.05
Missing data were classified as unknown and included in the models (not reported).
Comment
Principal Findings
In this retrospective analysis of claims data for privately insured individuals across multiple employers and health plans, we found that patients undergoing simple hysterectomy were billed for out-of-network claims as a part of their care in 9% of cases, despite the primary surgeon being in-network. While patients and primary surgeons may often confirm in-network status prior to surgery, it may be difficult to confirm that all billing parties will be in-network for the patient. Median out-of-network expenditures were approximately $500 when present, though facility out-of-network expenditures were notably several fold higher, ranging from $3,300 for outpatient cases to $8,300 for inpatient cases. The proportion of cases with out-of-network claims has declined slightly over time, and the incidence of out-of-network facility claims were at the lowest in 2018.
Results in Context
Our finding of out-of-network bills in 9% of cases is a much lower estimate than the 25% for hysterectomy from one of the first published studies to describe this issue by Chhabra, et al.11 This prior work utilized the Optum Clinformatics claims database. While similar to Marketscan in structure, Clinformatics represents a single large private insurer and includes a smaller overall sample size. The discrepancy in findings highlights that the risk of out-of-network claims is related to the network structure of the particular insurer and insurance plan, and therefore risk may vary extensively between different privately insured individuals. Even for an individual, there will be some variation in risk of out-of-network claims across different surgical procedures; a similar Marketscan study of total joint arthroplasty found over 10% of patients with out-of-network facility claims (vs. 2.3% in our study of hysterectomy).17
Clinical Implications
While Chhabra, et al. found a higher overall rate of out-of-network claims than our study, their study excluded facility out-or-network claims from the primary analyses. Cases with out-of-network facility claims are of particular importance in light of the recently passed No Surprises Act.5 Effective in 2022, the No Surprises Act will ban patient exposure to out-of-network bills for professional services at in-network facilities without a patient waiver, but it will not protect against out-of-network claims for non-emergent care at out-of-network facilities.6,8 In our study, we found that only 2.3% of hysterectomy cases included an out-of-network facility claim. While small in percentage, this still represents thousands of cases annually, and the expenditures for out-of-network facility claims are much larger than professional fees, on the order of several thousand dollars at the median. Additionally, the details of regulation and enforcement have yet to be finalized and will impact the ability of profit maximizing providers and facilities to use patient waivers or other mechanisms to facilitate charging higher, un-negotiated rates for out-of-network services.
Our analysis is also novel in its assessment of details of out-of-network billing specific to hysterectomy. While infrequently utilized, midlevel surgical assistant claims were out-of-network in nearly one in five cases when billed. We also found that robotic surgery and outpatient facility were both associated with increased risk of out-of-network claims. Hysterectomy for gynecologic cancer had lower risk of any out-of-network claim, but higher risk of facility out-of-network claims in comparison to cases for benign indications. We hope that, in identifying such risk factors, providers can work with medical practices and associated hospitals to establish systems to avoid such out-of-network claims for his or her patients.
Research Implications
As a major surgical procedure, cases of hysterectomy commonly include bills from numerous ancillary services including anesthesia and pathology, but also surgical assistants and consultants. However, assuming implementation of the No Surprises Act as intended, the facility charge that will be the primary exposure to out-of-network charges in the future. As there will be a facility charge associated with even minor procedures, such as hysteroscopy, dilation and curettage, or salpingectomy, it is possible that patients may more commonly incur unexpected charges for these minor surgical procedures. Future research should continue to investigate the impact of these charges, in the context of implementation of the No Surprises Act. Additionally, direct assessment of patients may facilitate a better understanding of the actual charges incurred and the associated financial burden.
Strengths and Limitations
The strength of this study is in the large sample provided by Marketscan, including a diversity of large employers and health plans. While the sample is not perfectly representative of all privately insured patients, the overall size and diversity of included plans improves the generalizability of our findings relative to prior work on this topic. However, there are limitations to these data as well. As with any large database study, some findings may be statistically significant but represent small absolute differences between groups. Marketscan lacks information on patient race and ethnicity for identification of disparities in exposure to out-of-network bills. Marketscan also contains limited information on provider characteristics (including academic versus private hospital, or facility size/volume) to relate to the risk of out-of-network bills. Additionally, we are limited in our ability to assess for the particular clinical context of out-of-network bills due to the nature of claims data, which is designed for billing, rather than clinical, purposes. Lastly, Marketscan does not provide data on actual payments from patients related to out-of-network bills, or usual and customary rates for procedures from which to estimate the balance billing exposure of the patient in each case. Therefore, out-of-network expenditures from the patient’s perspective are underestimated in cases that included balance billing.
Conclusions
While there has been promising movement in preventing out-of-network bills, the laws around these sometimes large and unexpected medical bills remain an important issue for privately insured patients moving forward. The No Surprises Act makes important progress in protecting patients from financially catastrophic charges, but details of implementation and the law’s effectiveness in practice remain uncertain. As the regulatory systems to enforce this new law are designed, they should maximize patient protection to prevent circumvention of the rules to impose unexpected bills. Given the exception in the law for non-emergent care at out-of-network facilities, gynecologists and patients should carefully screen insurance coverage prior to surgery to ensure in-network status for both the primary surgeon and the surgical facility. Furthermore, this issue is not limited to major surgery, with even outpatient laboratory tests billed as out-of-network in >5% of cases.18,19 While the complexities of the US healthcare system can be difficult to navigate for both patients and providers, we should continue to advocate for reform that leads to transparency and fairness in the out-of-pocket cost burden to patients.
Supplementary Material
Condensation:
Among privately insured hysterectomies with in-network primary surgeon, 8.8% included ≥1 out-of-network claim, approximately 25% of which were facility claims, not impacted by pending legislation.
AJOG at a Glance:
A. Why was the study conducted?
Out-of-network billing for privately insured patients represents a source of unexpected financial burden for patients. Hysterectomy is a common exposure to out-of-network bills, even with in-network primary surgeon.
B. What are the key findings?
We found that 8.8% of cases of hysterectomy from 2008-2018 included ≥1 out-of-network claim. Anesthesia was the most frequent source, while midlevel surgical assistant claims were highest risk to be out-of-network when present. Facility claims were out-of-network in 2.2% of cases and had higher charges than professional out-of-network claims when present.
C. What does this study add to what is already known?
Using data from over 350 payers, we found a lower risk of out-of-network bills than a previous estimate from a single large insurer. The No Surprises Act should prevent the majority of professional out-of-network charges, but facility charges would remain as an uncommon but significant source of financial burden for patients.
Financial Support:
HAM is supported by early career funding from the National Institutes of Health (BIRCWH K12HD043446). Dr. Wright has served as a consultant for Clovis Oncology, has received research funding from Merck, and has received royalties from UpToDate. Dr. Havrilesky has received research funding from Tesaro/GSK and Astra Zeneca.
Footnotes
Disclosures: The authors report no conflicts of interest. This study is not a clinical trial. This work has not been previously presented at a meeting.
REFERENCES
- 1.Rosenthal E. Costs Can Go Up Fast When E.R. Is in Network but the Doctors Are Not. New York Times. September 28, 2014. [Google Scholar]
- 2.Cooper Z, Scott Morton F. Out-of-Network Emergency-Physician Bills — An Unwelcome Surprise. N Engl J Med. 2016;375(20):1915–1918. doi: 10.1056/nejmp1608571 [DOI] [PubMed] [Google Scholar]
- 3.Cooper Z, Nguyen H, Shekita N, Morton FS. Out-of-network billing and negotiated payments for hospital-based physicians. Health Aff. 2020;39(1):24–32. doi: 10.1377/hlthaff.2019.00507 [DOI] [PubMed] [Google Scholar]
- 4.Pollitz K, Claxton G, Cox C, Levitt L. An Examination of Surprise Medical Bills and Proposals to Protect Consumers from Them.; 2020. https://www.healthsystemtracker.org/brief/an-examination-of-surprise-medical-bills-and-proposals-to-protect-consumers-from-them-3/ [Google Scholar]
- 5.116th Congress. H.R.133 - Consolidated Appropriations Act, 2021.; 2020.
- 6.Chhabra KR, Fuse Brown E, Ryan AM. No More Surprises — New Legislation on Out-of-Network Billing. N Engl J Med. 2021;384(15):1381–1383. doi: 10.1056/nejmp2035905 [DOI] [PubMed] [Google Scholar]
- 7.Luthi S. Biden faces health industry fight over new ‘surprise’ billing ban. POLITICO. Published 2021. Accessed April 28, 2021. https://www.politico.com/news/2021/04/23/biden-health-industry-surprise-billing-ban-484507 [Google Scholar]
- 8.Hoadley J, Keith K, Lucia K. Unpacking the No Surprises Act: An Opportunity To Protect Millions. Health Affairs Blog. Published 2020. Accessed April 3, 2021. https://www.healthaffairs.org/do/10.1377/hblog20201217.247010/full/ [Google Scholar]
- 9.Wright JD, Herzog TJ, Tsui J, et al. Nationwide trends in the performance of inpatient hysterectomy in the United States. Obstet Gynecol. 2013;122(2 Pt 1):233–241. doi: 10.1097/aog.0b013e318299a6cf [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cohen SL, Ajao MO, Clark NV, Vitonis AF, Einarsson JI. Outpatient Hysterectomy Volume in the United States. Obstet Gynecol. 2017;130(1):130–137. doi: 10.1097/AOG.0000000000002103 [DOI] [PubMed] [Google Scholar]
- 11.Chhabra KR, Sheetz KH, Nuliyalu U, Dekhne MS, Ryan AM, Dimick JB. Out-of-Network Bills for Privately Insured Patients Undergoing Elective Surgery with In-Network Primary Surgeons and Facilities. In: JAMA - Journal of the American Medical Association. Vol 323. American Medical Association; 2020:538–547. doi: 10.1001/jama.2019.21463 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Benoit MF, Ma JF, Upperman BA. Comparison of 2015 Medicare relative value units for gender-specific procedures: Gynecologic and gynecologic-oncologic versus urologic CPT coding. Has time healed gender-worth? Gynecol Oncol. 2017;144(2):336–342. doi: 10.1016/j.ygyno.2016.12.006 [DOI] [PubMed] [Google Scholar]
- 13.The Truven Health MarketScan databases for life sciences researchers. Accessed October 1, 2020. https://truvenhealth.com/Portals/0/Assets/2017-MarketScan-Databases-Life-Sciences-Researchers-WP.pdf [Google Scholar]
- 14.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–1139. doi: 10.1097/01.mlr.0000182534.19832.83 [DOI] [PubMed] [Google Scholar]
- 15.Office of Management and Budget. Office of Management and Budget Standards. Census.gov. Published 2021. Accessed April 1, 2021. https://www.census.gov/programs-surveys/metro-micro/about/omb-standards.html
- 16.CPI Inflation Calculator. Published online 2020. Accessed August 25, 2020. https://www.bls.gov/data/inflation_calculator.htm
- 17.Varady NH, Amen TB, Chopra A, Freccero DM, Chen AF, Smith EL. Out-of-Network Facility Charges for Patients Undergoing Outpatient Total Joint Arthroplasty. J Arthroplasty. Published online March 5, 2021. doi: 10.1016/j.arth.2021.03.001 [DOI] [PubMed] [Google Scholar]
- 18.Sen AP, Meiselbach MK, Wang Y, Eisenberg MD, Anderson GF. Frequency and Costs of Out-of-Network Bills for Outpatient Laboratory Services Among Privately Insured Patients. JAMA Intern Med. Published online April 26, 2021. doi: 10.1001/jamainternmed.2021.1422 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Song Z, Lillehaugen T, Wallace J. Out-of-Network Laboratory Test Spending, Utilization, and Prices in the US. JAMA. 2021;325(16):1674. doi: 10.1001/jama.2021.0720 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.