Advancements in surgical and medical therapies during the past several decades have led to expanded treatment options for many gynecologic conditions. Specifically, medical management of conditions such as abnormal bleeding and leiomyoma has led to a decrease in hysterectomies over the past 30 years.1–3 Despite this, hysterectomy remains one of the most common surgical procedures performed in the United States, with more than 400,000 performed annually and contributing to more than $5 billion in health care costs.4
Given recent changes in access to health care services and the aging populations in the United States, significant attention is being paid to understanding what factors contribute to the cost of health care. Recently the U.S. Congress petitioned the Institute of Medicine, now the National Academy of Medicine, to conduct an assessment of Medicare costs across the country and found significant geographic variation in cost.5 Researchers in other disciplines have similarly identified significant geographic cost variation in commonly performed procedures. For instance, Goz et al6 found that there is significant regional cost variance by more than $4,000 for the same cervical spine procedure.
Previous cost analyses for hysterectomy primarily focused on the effect of mode of hysterectomy has on cost (Borahay MA, Amagwula TM, Driver TD, Achjian T, Kilic G. Trends of the cost of hysterectomy: analysis of tertiary care center data . Obstet Gynecol 2015;125(suppl):17S).7,8 However, these analyses did not specifically evaluate regional variations in cost. There is evidence that higher spending does not necessarily lead to better surgical outcomes nor are lower health care costs associated with rationed or limited care.9 To better control health care spending and provide value-based care to patients, it is important to first understand the reasons behind cost variations and to identify regions with low-cost and high-quality practices to create health policy that promotes efficient and value of health care delivery.10 In this study we have attempted to estimate whether there is a geographic cost variation for hysterectomy in the United States.
MATERIALS AND METHODS
This study was a cross-sectional, population-based study using data from the 2013 National Inpatient Sample from the Health Care Utilization Project, sponsored by the Agency for Health Care Research and Quality.12 The writing of this article followed the Strengthening the Reporting of Observational Studies in Epidemiology statement.13 The National Inpatient Sample is the largest all-payer database and represents a 20% random sample of discharges from hospitals in the United States. It receives information from all states participating in the Health Care Utilization Project representing approximately 95% of the U.S. population. The 2013 National Inpatient Sample contains data from 44 states and 4,034 hospitals. The National Inpatient Sample obtains hospital location data from the U.S. Census Division, which divides the United States into four regions: Midwest, West, Northeast, and South. Each region is further split into nine divisions: New England, mid-Atlantic, East North Central, West North Central, South Atlantic, East South Central, West South Central, Mountain, and Pacific.12 The East and West divisions of the North and South Central regions were combined into simply South Central and North Central so that a total of seven U.S. regions were analyzed in this study. Table 1 provides a breakdown of each division and the states included.
Each individual patient record contains 15 procedure and diagnostic codes based on the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM). The National Inpatient Sample also contains U.S. Census demographic data and information about total charges billed by the hospital. This study did not use any personal health information and was exempt by the institutional review board.
Patients 18 years old or older undergoing inpatient hysterectomy were identified using the following ICD-9-CM procedure codes for hysterectomy: subtotal abdominal (683.9), total abdominal (684.9), subtotal laparoscopic (683.1), total laparoscopic (684.1, 686.1), laparoscopic-assisted vaginal (685.9), vaginal (685.9), and laparoscopic robotic-assisted procedure (174.2, 174.4). Patients with a primary diagnosis of malignancy of female genitalia (ICD-9-CM: 179.0, 180.0–184.9) were excluded as were those who underwent radical hysterectomy (686.0, 687.0). Records with incomplete procedure, diagnosis, or charge data were also excluded from the analysis. Patients younger than 18 years were excluded from the analysis as a result of the fact hysterectomy in children and adolescents likely represents an unusual circumstance such as malignancy and severe pelvic trauma and may be associated with factors than can significantly affect the cost of surgery and inpatient care, which may be difficult to evaluate in this study.
We identified patients having the primary diagnosis of abnormal uterine bleeding (621.0, 621.2, 626.0, 626.2, 626.3, 626.44, 626.5, 626.6, 626.7, 626.8, 626.9, 627.0, 627.1), uterine leiomyomas (218.0, 218.1, 218.2, 218.9), cervical dysplasia (622.10, 622.11, 622.12), pelvic prolapse (618.0–618.05, 618.09, 618.1–618.4, 618.6–618.8, 618.81–618.83, 618.89, 618.95), endometriosis (617, 617.0–617.9), and pelvic pain (6250, 625.2–625.5). The following concomitant procedures commonly performed at the time of hysterectomy were also included in the analysis: unilateral salpingo-oophorectomy (656.2, 654.9, 656.4, 654.1, 653.9, 664), bilateral salpingo-oophorectomy (656.1, 656.3, 653.1), bilateral salpingectomy (665.1), pelvic reconstructive procedures (692.3, 701.8, 705.0–705.5, 706.3, 706.4, 707.7–707.89, 709.3–709.5), anti-incontinence procedures (597, 593, 597.9, 594, 595), cystoscopy (573.2), adhesiolysis (545, 545.1, 545.9, 574.1, 658.9), and blood transfusion (990.4–990.6).
Additional patient variables included in the analysis were: length of stay, age at the time of hysterectomy, race, primary insurance payer, and median household income. Hospital characteristics included hospital region, population density of area where the hospital was located, and total hospital charges for each individual hospitalization. Total charges are reported to the Health Care Utilization Project by the participant hospitals; values are rounded to the nearest dollar and zero charges are set to “missing.” Generally, the participant institutions do not include professional fees and noncovered charges; if this information is provided, it is removed from the total charges during Health Care Utilization Project processing. Hospital charges can be converted into actual hospital-specific costs by using the cost-to-charge ratio files developed by the Agency for Healthcare Research and Quality. These ratios are calculated by dividing inpatient costs by inpatient charges and both values are reported to the Centers for Medicare and Medicaid Services Healthcare Reporting Information System. The National Inpatient Sample tracks each participant hospital using a unique identifier, HOSP_NIS, which is linked to both total charges and the cost-to-charge ratios; the ratios can then be applied directly to charges to calculate costs. This technique has been validated by the Agency for Healthcare Research and Quality as well as prior studies.2,12
Stepwise multivariate regression analysis was used to evaluate the effect of covariables on hysterectomy costs. Descriptive statistics are reported as medians and interquartile ranges. Statistical significance was set at P<.05. All statistical analysis was performed using JMP 10.
A total of 38,414 individual patients undergoing hysterectomy for benign indications were identified. Figure 1 shows the median cost and interquartile ranges of hysterectomy cost by geographic region. The median cost of hysterectomy was $13,981 (interquartile range $9,075–29,770). The mid-Atlantic region demonstrated the lowest median cost $9,661 (interquartile range $6,243–15,335), whereas the South Atlantic and Pacific regions were associated with the highest median cost of hysterectomy $17,697 (interquartile range $11,844–27,700) and $22,534 (interquartile range $15,380–33,797), respectively. These differences persisted for both regions when stratified by mode of hysterectomy (Table 2). For all regions, vaginal hysterectomy was the cheapest and robotic hysterectomy was the most expensive.
Tables 3 and 4 demonstrate the patient and procedure characteristics in our cohort. The majority of patients were white, lived in large metropolitan areas, and had private insurance; the median age was 46 years (interquartile range 41–52 years). Abdominal hysterectomy was the most common (58.4%) and robotic was the least common (11.4%) route of hysterectomy performed. The most common indication for hysterectomy was leiomyoma (38.9%) and bilateral salpingo-oophorectomy was the most common concomitant procedure (39.6%).
Table 5 demonstrates the multiple logistic regression model of variables contributing to cost of hysterectomy. After adjusting for the following confounders: age, race, length of stay, concomitant procedure, primary diagnosis, mode of hysterectomy, population density, median household income, and primary insurer, compared with the mid-Atlantic region, the Pacific region was still associated with the highest probability that hysterectomy costs would be above the median (adjusted odds ratio [OR] 10.43, 95% confidence interval [CI] 9.44–11.45). New England was not associated with a statistically significant increase in cost (adjusted OR 1.05, 95% CI 0.91–1.18). All other regions were associated with an increased probability of costs above the median with the South Atlantic (adjusted OR 5.39, 95% CI 4.95–5.86) and South Central (adjusted OR 2.40, 95% CI 2.21–2.62) regions having the highest risk.
Compared with vaginal hysterectomy, both laparoscopic (adjusted OR 2.86, 95% CI 2.61–3.15) and robotic hysterectomy (adjusted OR 5.66, 95% CI 5.11–6.26) were associated with an increased probability of costs being greater than the median. Interestingly, abdominal hysterectomy was not associated with a statistically significant increase in cost (adjusted OR 1.01, 95% CI 0.91–1.09). Compared with leiomyoma, which was the most common indication for hysterectomy, all other primary indications for hysterectomy that were evaluated were associated with a statistically significant decrease in cost below the median.
Compared with micropolitan counties with populations between 10,000 and 50,000 people, central metropolitan counties, with populations exceeding 1 million, were associated with increased odds of higher cost (adjusted OR 2.09, 95% CI 1.93–2.27). Outlying counties surrounding urban centers and counties with populations between 250,000 and 1 million people had similar odds of increased costs compared with medium-sized metropolitan (adjusted OR 1.48, 95% CI 1.36–1.61 and adjusted OR 1.58, 95% CI1.46–1.72, respectively). Small metropolitan counties had a low but significant association with increased costs (adjusted OR 1.14, 95% CI 1.04–1.26). Compared with patients who self-pay, patients with Medicare or private insurance were associated with a risk of increased costs (adjusted OR 1.17, 95% CI 1.05–1.32 and adjusted OR 1.18, 95% CI 1.05–1.31, respectively), whereas patients with Medicaid were not statistically more likely to have higher costs (adjusted OR 1.02, 95% CI 0.92–1.13).
Compared with white women, black (adjusted OR 1.04, 95% CI 1.01–1.16) and Hispanic women– (adjusted OR 1.36, 95% CI 1.26–1.49) were slightly more likely to have higher costs of hysterectomy. Other minority groups did not have an increased risk of above median costs (adjusted OR 1.09, 95% CI 0.98–1.20). Compared with women with an annual median household income of less than $38,000, women with higher annual median household incomes were not associated with higher costs. For every 10-year increase in age, there was a 0.3% increase in probability of having a higher than median cost. Similarly, there was an 88% incrementally increased risk of higher costs for every day of hospitalization.
Most concomitant procedures at the time of hysterectomy were associated with an increased cost, with anti-incontinence procedures (adjusted OR 2.03, 95% CI 1.83–2.24) and blood transfusion (adjusted OR 1.91, 95% CI 1.72–2.14) having the highest risk of costs above the median. Bilateral salpingectomy was not associated with an increased risk of higher cost (adjusted OR 1.03, 95% CI 0.95–1.12).
Our findings demonstrate that significant regional variation in cost, which have previously been described in nongynecologic literature, also exist for hysterectomy.14–17
In our analysis, there was a 233% difference in cost between the lowest cost mid-Atlantic region, which includes New York, New Jersey, and Pennsylvania, and the highest cost Pacific region, which includes California, Oregon, Washington, Alaska, and Hawaii. The South Atlantic, South Central, and Midwest regions were also associated with higher costs compared with the mid-Atlantic, but the differences were smaller, whereas the costs in New England were not significantly different. The probability of higher costs in these regions persisted after adjusting for confounding variables. This is consistent with previous studies that have reported the higher cost of inpatient care and surgical procedures in these regions.13,15,18
Regional cost variation can in part be explained by differences in practice patterns; however, these variables are not accurately captured within the National Inpatient Sample database. For instance, some surgeons may choose to use disposable devices, uterine manipulators, and hemostatic agents during hysterectomy, which have all been shown to lead to substantial differences in cost.19 Additionally, duration of surgery, which varies significantly with surgeon volume and experience as well as use of postoperative services such as laboratory testing and imaging can markedly affect the cost of a surgery. However, practice variations cannot fully account for the marked price differences seen in this and other studies.
The reported costs in this study reflect what hospitals are reimbursed by insurance companies. These costs are negotiated annually for individual hospitals, physicians, and health systems. The same insurance company may negotiate different reimbursements for the same diagnosis or procedure with individual hospitals.20 The Blue Cross and Blue Shield Association recently released an analysis regarding variations in the reimbursement of angioplasty and total knee and hip replacements. The most extreme variations were 512% for angioplasty and 313% for knee replacement with significant price differences noted in the same city.21,22 The regions with higher costs in these analyses were similar to what was found in this study.
Our findings reaffirm that vaginal hysterectomy remains the least costly method of hysterectomy and robotic-assisted hysterectomy, the most costly, with no significant difference between vaginal and abdominal hysterectomy.23,24 The contemporary view is that as a result of shorter hospital stays hospital stays compared with abdominal hysterectomy, hysterectomy using minimally invasive techniques actually leads to lower costs. However, after adjusting for length of stay, it is likely that shorter surgical times and use of nondisposable instruments used in abdominal hysterectomy actually result in lower surgical costs compared with minimally invasive modes of hysterectomy.
We also found that compared with leiomyoma, all other primary diagnoses were associated with decreased risk of higher costs. This may be attributable to the fact that women undergoing surgery for leiomyomas typically have larger uteri, which are associated with longer surgical times. In contrast, all concomitant procedures were associated with increased costs with the exception of bilateral salpingectomy. This may be the result of the fact that of all the other procedures evaluated, salpingectomy is the technically simplest and shortest and does not require the use of additional instruments or implants. However, these results should be viewed critically, because some of the diagnostic and procedure codes involved relatively small total or regional populations such as blood transfusion and cervical dysplasia. Additionally, although both unilateral salpingo-oophorectomy and bilateral salpingo-oophorectomy were significantly associated with higher costs, the adjusted ORs were close to 1.0, suggesting that these may not associations may not be clinically significant.
In our cohort, black and Hispanic women were slightly more likely to have higher associated costs compared with white women, with Hispanic women having a stronger association. Race has previously been reported as associated with higher costs for other surgical procedures, but it has not been specifically evaluated for hysterectomy. These studies have found that women belonging to these minority groups are more likely to have complications and longer hospital stays, both of which are associated with increased costs, compared with white women.25,26 Even after adjusting for length of stay, the increased risk of higher costs persisted for this population. Interestingly, women who were on Medicaid or were listed as “self-pay” were less likely to incur higher hysterectomy costs. This is likely a reflection of what the hospital was able to collect from these patients rather than what was charged.
Similar to other studies, private insurance and Medicare were associated with higher costs compared with self-pay patients. Conversely, Medicaid patients were not significantly more likely to have higher costs relative to this group. This is, again, likely the result of what hospitals and various payers have negotiated for reimbursement. Additionally, studies have shown that patients with private or Medicare insurance are likely to use more health care services than patients without insurance or those with Medicaid and self-pay patients, leading to increased costs in these groups.27,28
Increasing population density was associated with increasing probability of higher hysterectomy costs. Specifically, patients in core urban areas were more likely to be subject to higher hysterectomy costs compared with those in outlying, rural counties and smaller metropolitan areas. Previous studies have shown that patients in these areas are more likely to be self-pay, have Medicaid, and have less access to care, all of which are associated with lower cost.29
The main strength of this study is the use of the National Inpatient Sample, which allows for the analysis of a large cohort of patients undergoing hysterectomy. Given that these women represent a cross-section of the entire United States, these results are applicable to the entire population.
The retrospective design and the fact that the National Inpatient Sample relies on ICD-9-CM and Current Procedural Terminology codes, which could lead to erroneous inclusion or omission of clinical information, are limitations of this study. Another limitation is that, although the cost information within the National Inpatient Sample includes factors such as medications, ancillary services, and supplies used during hysterectomy and postoperative hospitalization, it does not provide a breakdown of all of these elements. Furthermore, the database does not include the physician professional fees, which would further affect cost.
As previously mentioned, the results of this study must be interpreted carefully, because some of the variables assessed such as cervical dysplasia and pelvic pain have small sample sizes leading to possible overestimation of statistical significance. Conversely, variables with larger sample sizes such as unilateral salpingo-oophorectomy and bilateral salpingo-oophorectomy showed relatively low adjusted ORs, and although they may be statistically significant, they may not hold clinical relevance. The results of this analysis suggest that there is significant geographic variation for costs related to inpatient hospitalization after hysterectomy that persist after controlling for confounders. Additional studies evaluating practice patterns of surgeons and use of postoperative services as well as what factors play a role in how hospitals and insurance companies negotiate reimbursements may help further delineate these differences in cost.
1. Farquhar CM, Steiner CA. Hysterectomy rates in the United States 1990–1997. Obstet Gynecol 2002;99:229–34.
2. Wright JD, Herzog TJ, Tsui J, Ananth CV, Lewin SN, Lu YS, et al. Nationwide trends in the performance of inpatient hysterectomy in the United States. Obstet Gynecol 2013;122:233–41.
3. Jacoby VL, Autry M, Jacobson G, Domush R, Nakagawa MS, Jacoby A. Nationwide use of laparoscopic hysterectomy compared with abdominal and vaginal approaches. Obstet Gynecol 2009;114:1041–8.
4. Agency for Healthcare Research and Quality. Health services research on hysterectomy and alternatives. Available at: http://archive.ahrq.gov/consumer/uterine2.htm
. Retrieved September 27, 2016.
5. Newhouse JP, Garber AM. Geographic variation in Medicare services. N Engl J Med 2013;368:1465–8.
6. Goz V, Rane A, Abtahi AM, Lawrence BD, Brodke DS, Spiker WR. Geographic variations in the cost of spine surgery. Spine 2015;40:1380–9.
7. Herling SF, Palle C, Møller AM, Thomsen T, Sørensen J. Cost-analysis of robotic-assisted laparoscopic hysterectomy versus total abdominal hysterectomy for women with endometrial cancer and atypical complex hyperplasia. Acta Obstet Gynecol Scand 2016;95:299–308.
8. Sculpher M, Manca A, Abbott J, Fountain J, Mason S, Garry R. Cost effectiveness analysis of laparoscopic hysterectomy compared with standard hysterectomy: results from a randomised trial. BMJ 2004;328:134.
9. Gordon TA, Burleyson GP, Tielsch JM, Cameron JL. The effects of regionalization on cost and outcome for one general high-risk surgical procedure. Ann Surg 1995;221:43–9.
10. Fisher ES, Bynum JP, Skinner JS. Slowing the growth of health care costs—lessons from regional variation. N Engl J Med 2009;360:849–52.
11. Ransom SB, McNeeley SG, White C, Diamond MP. A cost analysis of endometrial ablation, abdominal hysterectomy, vaginal hysterectomy, and laparoscopic-assisted vaginal hysterectomy in the treatment of primary menorrhagia. J Am Assoc Gynecol Laparosc 1996;4:29–32.
12. HCUP Nationwide inpatient sample [NIS]. Healthcare Cost and Utilization Project [HCUP]. Available at: https://www.hcup-us.ahrq.gov/nisoverview.jsp. Retrieved August 5, 2016.
13. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg 2014;12:1495–9.
14. Welch WP, Miller ME, Welch HG, Fisher ES, Wennberg JE. Geographic variation in expenditures for physicians' services in the United States. N Engl J Med 1993;328:621–7.
15. Yu HY, Hevelone ND, Lipsitz SR, Kowalczyk KJ, Hu JC. Use, costs and comparative effectiveness of robotic assisted, laparoscopic and open urological surgery. J Urol 2012;187:1392–8.
16. Volinn E, Diehr P, Ciol MA, Loeser JD. Why does geographic variation in health care practices matter? (And seven questions to ask in evaluating studies on geographic variation). Spine (Phila Pa 1976) 1994;19(suppl):2092S–100S.
17. Huntington CR, Cox TC, Blair LJ, Prasad T, Lincourt AE, Heniford BT, et al. Nationwide variation in outcomes and cost of laparoscopic procedures. Surg Endosc 2016;30:934–46.
18. Mushinski M. Hysterectomy charges: geographic variations United States, 1994. Stat Bull Metrop Insur Co 1995;77:2–12.
19. Schaer GN, Koechli OR, Haller U. Single-use versus reusable laparoscopic surgical instruments: a comparative cost analysis. Am J Obstet Gynecol 1995;173:1812–5.
20. Reinhardt UE. The pricing of U.S. hospital services: chaos behind a veil of secrecy. Health Aff (Millwood) 2006;25:57–69.
21. Blue Cross and Blue Shield Association. The Health of America Report: a study of cost variation for percutaneous coronary interventions (angioplasties) in the U.S. Available at: https://www.bcbs.com/about-us/capabilities-initiatives/health-america/health-of-america-report/study-cost-variation
. Retrieved December 15, 2016.
22. Blue Cross and Blue Shield Association. The Health of America Report: a study of cost variations for knee and hip replacement surgeries in the U.S. Available at: https://www.bcbs.com/about-us/capabilities-initiatives/health-america/health-of-america-report/study-cost-variations
. Retrieved December 15, 2016.
23. Rosero EB, Kho KA, Joshi GP, Giesecke M, Schaffer JI. Comparison of robotic and laparoscopic hysterectomy for benign gynecologic disease. Obstet Gynecol 2013;122:778–86.
24. Paparella P, Sizzi O, Rossetti A, De Benedittis F, Paparella R. Vaginal hysterectomy in generally considered contraindications to vaginal surgery. Arch Gynecol Obstet 2004;270:104–9.
25. Feng W, Nietert PJ, Adams RJ. Influence of age on racial disparities in stroke admission rates, hospital charges, and outcomes in South Carolina. Stroke 2009;40:3096–101.
26. Vaid S, Tucker J, Bell T, Grim R, Ahuja V. Cost analysis of laparoscopic versus open colectomy in patients with colon cancer: results from a large nationwide population database. Am Surg 2012;78:635–41.
27. Hershman DL, Richards CA, Kalinsky K, Wilde ET, Lu YS, Ascherman JA, et al. Influence of health insurance, hospital factors and physician volume on receipt of immediate post-mastectomy reconstruction in women with invasive and non-invasive breast cancer. Breast Cancer Res Treat 2012;136:535–45.
28. Hsia RY, MacIsaac D, Baker LC. Decreasing reimbursements for outpatient emergency department visits across payer groups from 1996 to 2004. Ann Emerg Med 2008;51:265–74.
29. Larson SL, Fleishman JA. Rural-urban differences in usual source of care and ambulatory service use: analyses of national data using Urban Influence Codes. Med Care 2003;41(suppl):III65–74.