Cost Analysis of Robotic versus Standard Laparoscopic Hysterectomies: A Preliminary Analysis

Abstract

Objective: The purpose of this study was to compare the costs and net revenue associated with robotic versus traditional laparoscopic hysterectomies. Design: The study was designed as a retrospective chart review. Methods: The study group consisted of patients who underwent laparoscopic hysterectomies using the robotic versus standard laparoscopy from January 2005 to March 2009. Data were collected as the mean for the cohort and included total charges, total collected, total direct costs, and operative margin (OM). Total costs were divided into operating room costs (labor, supplies, and equipment), hospital costs (anesthesia, pharmacy, room, and board) and additional costs. Statistical analysis was performed comparing the two groups by the Wilcoxon rank sum test, two sided t test, and Fisher exact test. Results: Twenty-eight robotic cases were matched to 48 laparoscopic cases. Demographics were similar between groups. Total charges where higher with the robot, ($30,725 vs. $22,526, p<0.001) as was operating room cost ($4,767 vs. $2,874, p<0.001), labor ($1,830 vs. $1,386, p<0.001), and supplies ($2,637 vs. $1,387, p<0.001). Total collections ($8,651 vs. $7,946, p=0.6), hospital cost ($912 vs. $913, p=0.4), and anesthesia costs ($275 vs. $276, p=0.8) were not different. Total direct costs were higher in the robotic group ($6,630 vs. $4,491, p<0.001). The median operating margin was $3,646 less for robotics (p=0.2). Conclusions: In Carilion Clinic, even though charges are higher for robotic hysterectomies, the collections are similar to those for laparoscopic hysterectomies. Operating room costs are higher for the robotic hysterectomy group, which has a negative impact on the operative margin. These preliminary data have potential implications for those establishing gynecologic robotic programs and budgets involving surgical productivity. (J GYNECOL SURG 28:255)

Introduction

Ever since the first case report of a laparoscopic assisted hysterectomy in 1989, there have been many advances in the field of minimally invasive surgery for gynecology.¹ Several studies have shown that laparoscopic surgery offers improved quality of life with shorter hospitalizations, faster recovery, improved cosmesis, decreased blood loss, and less postoperative pain.^2,3 The disadvantages of laparoscopy have been cited to be a steep learning curve, longer operative times associated with advanced procedures, two-dimensional visualization, poor ergonomics, amplified surgical tremor, and limited and counterintuitive instrumentation movements.⁴ These limitations have slowed the widespread adoption of minimally invasive surgery in all but the simplest of gynecologic cases.^5–7

The DaVinci^® Robot system (Intuitive Surgical, Inc., Sunnyvale, CA) was first introduced in 1999 and gained United States Food and Drug Administration (FDA) approval for gynecologic procedures in March, 2005. This system sought to improve minimally invasive surgery acceptance with many significant advances over conventional laparoscopic techniques. Besides improved ergonomics, three-dimensional (3-D) visualization, and removal of hand tremor, one of the biggest differences is the replication of full range of surgical hand motion. The fulcrum effect of laparoscopy is eliminated and the robotic instrument's wrist joint increases the degrees of freedom for movement from three to seven. This increased motion allows for a more natural movement of instruments and facilitates dissection, suturing, and knot tying. The main disadvantages of robotics are increased cost, lack of tactile feedback, limited vaginal access with the older robotic models, and decreased resident and fellow training with only one surgeon seated at the console. Despite these disadvantages, robotic procedures have tripled since 2007 and Intuitive Surgical, Inc. has seen a 75% increase in sales with >1400 units sold in the United States alone.⁷

New technology is often adopted in medicine before its true strengths, weaknesses, and economic impact are fully understood. Robotics is no different, with a void in the literature of randomized clinical trial data comparing robotic versus traditional laparoscopy in gynecologic surgery. The aim of this pilot study was to provide insight into the controversial issue of whether robotics is cost effective and worth the investment as it relates to performing hysterectomies. With this information, hospital administrators and gynecologists will have a better understanding as to the impact of robotic hysterectomies on the hospital's overall operating room budget.

Materials and Methods

A retrospective analysis of consecutive patients undergoing laparoscopic hysterectomies with the DaVinci Robot system versus standard laparoscopy from January 2005 through March 2009 was performed at Carilion Clinic, Roanoke, Virginia. Data were extracted from the medical record, and data on charges, cost, and collections were provided by the accounting department. This study was approved by Carilion Clinic's Institutional Review Board. The patients who had successful completion of their robotic or laparoscopic hysterectomies were used in this analysis. A matched cohort of consecutive patients undergoing laparoscopic hysterectomies without the DaVinci Robot during this time frame by the principle investigator was used as the comparative group. The two groups were not randomized.

Demographic data were collected including age, body mass index (BMI), and pre- and postoperative diagnosis. Surgical data collected included procedure, operative findings, total operating room time, intraoperative complications, estimated blood loss (EBL), uterine weight, and pathologic diagnosis. Postoperative data included acute complications as well as length of stay. All patients were admitted for 23 hour observation and then formally admitted if complications required a longer hospital stay. All data reported are the mean for each variable within each cohort. Accounting data included insurance status (Medicaid, Medicare, private, uninsured, health maintenance organization [HMO] or preferred provider network [PPO]), total charges, percentage of charges collected, total collected, total direct cost, and net operative margin. Total costs were divided into operating room costs (labor, supplies, and equipment), hospital costs (anesthesia, pharmacy, room, and board) and additional costs. Additional costs were predominately ancillary services including physical, occupational, and respiratory therapy. Total direct cost was the sum of the operating room costs, hospital costs, and additional costs. The indirect costs included, primarily, the cost of purchasing, marketing, and maintaining the robot. Because this trial was preliminary, there was no attempt to estimate these indirect costs. The net operating margin was defined as the total money collected less the total direct costs. Net revenue per insurance was also calculated along with the percentage of patients who generated negative revenue per insurance status. All procedure codes were cross referenced to the procedure to verify accuracy.

Procedures were defined as “simple,” which included a laparoscopic hysterectomy +/− uni/ bilateral salpingo-oophorectomy, or “complex,” if additional procedures were performed in conjunction with the simple procedure. All patients in this study received preoperative antibiotics, general endotracheal anesthesia, and sequential compression devices to their lower extremities to prevent thromboembolic disease. Patients were placed in the dorsal lithotomy position in steep Trendelenburg position with their upper extremities tucked to their sides and their lower extremities positioned in Allen stirrups (Allen Medical Systems, Acton, MA). To decrease the patient's torso shifting, the robotics group received foam pads placed across the chest at shoulder level and tape across the pads, which was attached to the operating room table. The RUMI^® uterine manipulator was used along with a balloon tip, pneumo-occluder balloon, and the KOH ring (Cooper Surgical, Trumbell, CT) for all cases. The Veress needle was used to create a pneumoperitoneum, and then a 12 mm bladeless trocar (extra-long in the robot cases) was used for the camera port infraumbilically. Two robotic reusable 8 mm trocars or two 5 mm bladeless trocars (for laparoscopic hysterectomies) were placed 8–10 cm lateral and 2–3 cm inferior to the camera trocar. A 12 mm accessory trocar was also used in the robot cases for additional retraction, suction/irrigation, bowel manipulation, and transferring suture. This was placed in the midclavicular line in the left upper quadrant. An oral gastric tube was placed prior to this trocar to decompress the stomach. All robotic hysterectomies were performed using monopolar EndoWrist^® scissors in the right arm and bipolar fenestrated graspers in the left arm. These instruments were changed to two needle drivers to suture the vaginal cuff closed once the uterus was removed vaginally. In all of the laparoscopic hysterectomy cases, the cuff was closed vaginally. In all cases, the uterus was removed vaginally without morcellation. All operations were performed with the principle investigator as the primary surgeon and the first assistant being an obstetrics and gynecology resident. A second assistant was placed between the legs to manipulate the uterus during the procedure. Complications were noted to be intraoperative if they occurred in the operating room, and acute if they occurred within 30 days from surgery.

Tests for normality were done by using the Shapiro–Wilks test, as well as by visual examination of the distribution of the data. This has been done for each of the outcome variables. In most cases the Shapiro–Wilks test showed the data to be non-normal. As a result, all of the analyses were done using a t test on rank-transformed data. This is equivalent to a Wilcoxon rank sum test, a nonparametric test. As ranks were used for the analysis, median costs/revenues are reported for these categories. A p value of <0.05 was considered statistically significant.

Results

A total of 28 consecutive robotic hysterectomy cases for benign and malignant conditions were compared with a matched cohort of 48 standard laparoscopic hysterectomies. The characteristics of the cohorts are listed in Table 1. The average age of the robot and laparoscopic groups were 49 and 52 years, p=0.8. The average BMI were 31.5 and 29.4, p=0.2, with 53.6% and 40.0% obese (BMI≥28), p=0.8 and 25% and 16.7% were subcategorized as morbidly obese (BMI≥35), respectively. The most common indications for surgery were uterine cancer, adnexal masses, and fibroids. Other indications included endometriosis, cervical dysplasia, endometrial hyperplasia, pelvic pain, abnormal uterine bleeding, genetic risks of ovarian cancer, and trophoblastic disease. In addition to total hysterectomy +/− uni/bilateral salpingo-oophorectomy, complex procedures included complete pelvic and para-aortic lymph node dissections, lysis of adhesions, cystoscopy, upper vaginectomy, and cystotomy repair. Complex procedures occurred in each group, robot 13/28 (46.4%) and laparoscopy 16/48 (33.3%), p=0.3. Lysis of adhesions occurred in each group, robot 2/28 (7.1%) and laparoscopy 1/48 (2.1%), p=0.2. Uterine weight was 125.5 g for the robot group and 149.0g for the laparoscopy group, p=0.2. EBL was robot 104.4 mL and laparoscopy 83.9 mL, p=0.3, and total operating room time was robot 170.0 (range 68–260) minutes and laparoscopy 124.0 (range 74–201) minutes, p<0.001. A multiple regression model determined robotic (p<0.0001) and complex procedure (p=0.008) were both significant predictors of total operating room time. Length of stay was robot 1.3 days and laparoscopy 1.0 days, p=0.005.

Table 1.

Patient Characteristics

	Robotic group	Laparoscopic group	p value
n	28	48
Age	49 (29–77)	52 (21–80)	0.8
BMI (kg/m²)	31.5 (17.5–53.8)	29.4 (19.4–58.4)	0.2
% Obese (BMI≥28)	15/28 (53.6%)	19/48 (40.0%)	0.8^a
Indication for OR:^b
Uterine cancer	28.6%	29.2%
Adnexal mass	21.4%	25%
Fibroids	10.7%	14.6%
Endometriosis	17.9%	4.2%
Cervical dysplasia	10.7%	2.1%
Endometrial hyperplasia	3.6%	12.5%
Pelvic pain	3.6%	6.3%
Other	3.6%	6.3%
Total OR Time	170.0 (68–260)	124.0 (74–201)	<0.001
Complex procedures:	13/28 (46.4%)	16/48 (33.3%)	0.3
Uterine cancer staging	8/28 (28.6%)	14/48 (29.2%)	0.8
TLH, BSO, LOA	2	1
TLH, BSO, Cystoscopy	1	1
TLH, Upper vaginectomy	1	0
TLH, BSO, Cystotomy repair	2	0
Uterine weight (g)	125.5	149.0	0.2
EBL (mL)	104.4 (25–250)	83.9 (25–450)	0.3
Length of stay	1.3	1.0	0.005

Statistical analysis was performed using χ² testing; all other analyses were performed using t tests.

No statistical significance was seen on any of these variables.

BMI, body mass index; OR, operating room; TLH, total laparoscopic hysterectomy; BSO, bilateral salpingo-oophrectomy; LOA, lysis of adhesions.

There were two intraoperative complications in the robotic group and none in the laparoscopic group. Both robotic patients had significant pelvic adhesive disease and cystotomies were created. The cystotomies were repaired robotically with two layers of running 3-0 polyglactin suture. Acute postoperative complications occurred in 4 robotic patients and none occurred in the laparoscopic group. All acute complications were considered not serious and included urinary retention (requiring an indwelling bladder catheter for 7 days), an ileus (prolonging hospital stay for 1 day), and 2 patients requiring transfusions of 2 units of packed red blood cells.

Charges and cost data are presented in Table 2; all values reported in median and range. Total charges were higher in the robotic group than in the laparoscopic group, with higher collection rates in the laparoscopy group that approached statistical significance. Total collections for both groups were similar.

Table 2.

Median (Range) Charges and Cost Data

	Robotic group	Laparoscopy group	p value
n	28	48
Total charges	$30,725	$22,526	<0.001
	(22,386–54,268)	(16,023–34,015)
% Charges collected	28.9%	39.4%	0.07
Total collections	$8,651 (2,552–41,021)	$7,946 (2,846–23,745)	0.6
Costs
Total OR cost	$4,767 (3,265–11,972)	$2,874 (1,557–4,821)	<0.001
OR labor	$1,830 (997–2,573)	$1,386 (803–2,784)	<0.001
OR supplies	$2,637 (1,252–8,145)	$1,387 (466–1,902)	<0.001
OR equipment	$192 (64–291)	$144 (34–291)	0.001
Total hospital cost	$912 (649–4,467)	$913 (616–1,882)	0.4
Anesthesia non-drug	$275 (161–3,856)	$276 (135–599)	0.8
Pharmacy	$231 (127–371)	$210 (87–1,162)	0.9
Med/Surg unit	$324 (283–1,332)	$283 (283–830)	0.1
Additional cost	$338 (96–2,393)	$541 (118–5,353)	0.2
Total direct cost	$6,630 (4,653–15,936)	$4,491 (2,925–6,515)	<0.001
Net op. margin	$-120	$3,526	0.2
	($-2,471–25,085)	($-6515–17965)

OR, operating room.

The total operating room costs were higher in the robotic group. The total hospital costs were similar in both groups. The total direct costs including the operating room, hospital, and additional costs were higher in the robotics group. The net operating margin was not statistically significant between groups.

The insurance status and median collections per insurance status in both groups are presented in Table 3. There were no significant differences in the distribution of insurance status within both groups. The majority of patients were a part of a PPO. Carilion Clinic is a not-for-profit teaching institution that accepts all patients regardless of their insurance status, hence uninsured patients made up 10.7% and 8.5% (p=0.9) of both groups, respectively. The median collections from each insurance type were not significantly different in each group except for Medicare, through which the robotic group collected more than the traditional laparoscopy group. Of those that were uninsured, the hospital collected from only 1/7 patients who paid $20,615 out of pocket.

Table 3.

Median (Range) Collections by Insurance

	Robotic group	Laparoscopy group
	n (%) / $ Collected	n (%) / $ Collected	p value
PPO	9 (32.1)/$11,881	17 (36.2)/$10,145	0.8
	($8,981–21,252)	($5,035–16,946)
Medicare	6 (21.4)/$7,042	9 (19.1)/$2,929	0.003
	($4,693–10,166)	($129–5,243)
HMO	5 (17.9)/$11,110	11 (23.4)/$8,959	0.6
	($2,552–23,472)	($2,915–18,562)
Medicaid	4 (14.3)/$4,477	2 (4.3)/$6,258	0.7
	($3,690–9,374)	($6,048–6,468)
Uninsured	3 (10.7)/$0	4 (8.5)/$0^a	—
		($0–20,615)
Private	1 (0.4)/$41,021	4 (8.5)/$8,876	—
		($8,577–23,745)

One patient paid the hospital $20,615 out of pocket.

HMO, health maintenance organization; PPO, preferred provider network.

Data on the net operative margin per insurance status and the percent of patients who had negative or zero net revenue for both groups is presented in Table 4. In the robotic group, the hospital showed a positive net operative margin for patients with PPO, HMO, and private insurance status. With Medicare, Medicaid, and uninsured patients, the hospital saw a negative operative margin. These trends were not statistically different from those of the laparoscopy group, except for Medicaid. In the robotic group, only 2/13 showed a positive net operative margin with Medicare, Medicaid, or uninsured patients.

Table 4.

Median (Range) Net Operative Margin per Insurance Status and Percent Zero or Negative Revenue

	Robotic group	Laparoscopy group	p value
PPO	$2,899	$6,395	0.09
	($-2094–10,972)	($-4719–16,219)
% neg	2/9 (22.2%)	2/17 (11.8%)
Medicare	$-634	$-2,022	0.1
	($-1365–2023)	($-137–1770)
% neg	4/6 (66.7%)	8/9 (88.9%)
HMO	$5,070	$5,163	0.9
	($-2471–12,472)	($-1354–13,997)
% neg	1/5 (20.0%)	1/11 (9.1%)
Medicaid	$-946	$1,905	0.004
	($1739–2071)	($-736–2220)
% neg	4/4 (100%)	0/2 (0%)
Uninsured	$-5,904	$-4,525	—
	($-5777–6820)	($-6515–16,098)
% neg	3/3 (100%)	3/4 (75.0%)^a
Private	$25,085	$5,031	—
		($4446–17,965)
% neg	0/1 (0%)	0/4 (0%)

One patient paid the hospital $20,615 out of pocket.

HMO, health maintenance organization; PPO, preferred provider network.

Discussion

Despite surgical advances within the last decade, the abdominal approach to hysterectomies is still more common than the vaginal or laparoscopic route.⁶ Robotics was approved for use in gynecology in 2005. Robotics has the main advantages of intuitive surgical movements with greater degrees of freedom than standard laparoscopy. This allows for more advanced procedures to potentially be completed by the minimally invasive route. Despite these significant advantages, hospital administrators and operating room directors have been opponents of this technology because of the perspective that this new technology is a large capital expense without significant published data on its return for their investment. Barbash et al. estimated that robotics increased the cost of a procedure by 6% ($1,600) in 2007, and if the original cost of the robotic system is amortized and factored into this calculation, the cost increased to 13% ($3,200) per case.⁸ As hospital administrators continue to face significant economic hardships, large capital expenses such as robotics platforms become more highly scrutinized.

As robotic surgery becomes more widely used across and within surgical specialties, we gain more information regarding the costs and benefits of this tool. A review of the current literature details the costs of robotics for urologic, cardiothoracic, colorectal, general surgical, and gynecologic procedures. In urology, it is controversial as to whether radical cystectomy performed robotically is cost effective.^9,10 Robotic radical prostatectomy data, however, have shown robotics to be more costly than laparoscopic or open prostatectomies because of higher operating room costs.¹¹ Kam et al. reported no difference in overall costs in robotic versus open mitral valve repair, and Park et al. showed the cost of robotic pulmonary lobectomy was increased greater than that for video-assisted thoracic surgery, but cost effective if compared with open thoractomy.^12,13 In general and colorectal surgery, Nissen fundoplications and rectopexies were shown to cost more than laparoscopic and open procedures, respectively. This increased total cost was predominately the result of increased operating room costs.^14,15

In gynecologic surgery, there have been published data on cost comparisons for tubal reanastomosis, sacral colpopexies, myomectomies, and hysterectomies. Dhari Patel et al. showed that tubal reanastomosis performed robotically is cost effective compared with an open procedure, with comparable outcomes.¹⁶ Judd et al. reported that the cost of robotic sacral colpopexies was higher than that for both laparoscopic and open approaches; however, if operating room time and costs were decreased, it would become cost effective.¹⁷ Advincula et al. showed that robotic myomectomies are more costly than open procedures, but resulted in decreased blood loss, complication rates, and length of stay, showing a significant societal benefit.¹⁸ The first report of cost comparisons for hysterectomies was in gynecologic cancer patients, comparing uterine cancer staging robotically, laparoscopically, and through open procedures. Bell et al. showed that the total cost of robotics approached statistical significance compared with laparoscopy, $8,212.00 versus $7,569.80, p=0.06; however, both cost significantly less than an open procedure, which cost $12,943.60.¹⁹ An analysis from Switzerland showed that robotic hysterectomies accounted for an average of $2,563 more than traditional laparoscopy. The length of stay in that series was 3.3 versus 3.9 days, respectively.²⁰ Barnett et al. published a cost-modeling analysis that built in societal costs, and estimated robotics to be more costly than laparoscopy. The model also showed that if hospital costs were estimated by subtracting societal benefits and including indirect robotic costs, robotics again was more costly. It was only when operating room equipment was reduced by $1,496 per case that robotics became cost effective.²¹ The largest series using the Premier hospital database, 1661 robotic cases from 358 hospitals, showed that total hospital costs were higher for robotics than for laparoscopy, $9,640 versus $6,973, p<0.01 for inpatient procedures, and $7,898 versus $5,949, p<0.01 for outpatient procedures. This difference, as for other surgical specialties, was predominately the result of increased operating room time and cost.²²

This preliminary study was designed not only to shed light on Carilion Clinic's costs for these two procedures, but also to be the first to show net operative margin as it relates to charges, total collections, and insurance status. These additional variables, although not previously reported, are very important when attempting to estimate the cost effectiveness of establishing a robotics program. The robotic group was compared with a well matched cohort of patients who had had standard laparoscopic hysterectomies with vaginal cuff closure. There were no differences in age, BMI, uterine weight, and EBL. The principle investigator is a gynecologic oncologist who had extensive laparoscopic experience prior to using robotics. This experience is reflected in the incidence of more challenging cases with obesity rates being 53.6% and 40.0%, p=0.2, in the robotic and laparoscopic groups, respectively. There was no difference in complexities between the groups with the robotic group averaging 46 more minutes of operating room time. Even though the increased time with the robotics group can be a reflection of a learning curve, the data are consistent with the published literature. The intraoperative and acute complications were not significant for either group.

In Carilion Clinic, even though total charges for robotics were significantly higher than for the laparoscopic group, the amount collected was no different. Collection rates are dependent upon several variables including insurance status, coding, and whether collection strategies are aggressive. As laparoscopic codes are used for both groups, it is predictable that collections for a robotic procedure will be similar to those for a laparoscopic procedure. The advantage to charging more is seen only with payers who pay as a percentage of total charges. In most instances, however, these increased charges act to notify the payers that the hospital has increased costs associated with this procedure. With operating room costs being higher in the robotic group because of increased labor, supplies, and equipment as shown in the data in this study, cost efficiency will not be proven in a comparison with traditional laparoscopy. Where robotics becomes cost efficient is when the robotic approach is successful and the patient is spared a laparotomy. This, however, is dependent upon several factors, including pathology, patient body habitus, previous surgical history, and the surgeon's laparoscopic experience and confidence. Lapartomies have lesser operating room costs, but this is offset by increased hospital costs with longer lengths of stay and higher societal costs with longer return to normal activity. Bell et. al. reported 24.1 days for robotic procedures versus 52 days for laparotomy before the patient could return to normal activity.¹⁹

Hospital costs are similar when comparing robotics with laparoscopy. This is not surprising, as the length of stay is similar. In the data presented here, the length of stay was 1.3 days versus 1.0 (p=0.005), however this is not clinically significant as this variable was measured not in hours but in days. Additional costs including other hospital services such as physical, occupational, and respiratory therapy were the same in both groups. Indirect costs of the robot including the purchase, marketing, and the maintenance costs were not included in this analysis. The indirect costs were estimated by Judd et al. to be $581, $865, or $1,724 per procedure if 60, 40, or 20 robotic procedures are done monthly, respectively.¹⁷ These costs, however, may be offset by several factors, including recruitment of increased patient referrals for robotic procedures, preservation of market share, surgeon satisfaction, surgeon retention, and surgeon recruitment. These variables are hard to objectively measure, but need to be considered in the decision to have a robotics program.

Total direct costs to the hospital were higher in the robotics group. This was primarily because of operating room costs. Operating room costs are potentially higher because of the need for a robotics team, which requires specialized training in setting up, maintaining, and working with robotics, in addition to needing more costly supplies and equipment, which adds to the total surgical costs. Each robotic instrument is limited to a set number of procedures. The more robotic instruments used, the higher the cost of the procedure. This increased cost had a negative impact on the hospital's net operative margin, reducing the net revenue. In this preliminary study, this did not approach statistical significance, because of the small sample size and a wide variance in net operative margin. This variance is explained by the large differences in collections with difference insurance statuses. Collections ranged from $0 to $41,021 for robotics and from $3,451 to $12.519 for laparoscopy.

With the insurance distribution as shown in Table 3, the hospital showed a profit from robotic procedures. This profit predominately came from those patients who had private insurance, PPOs, or HMOs. Collections were significantly higher in the robotics group from Medicare: $7,051 versus $3,451, p=0.003. This most likely is a reflection of the additional procedures coded along with the hysterectomy in this subgroup of patients, and not the reimbursement from the hysterectomy alone. The total collections were not sufficient enough with the uninsured, Medicare, or Medicaid patients to cover the total direct costs. Table 4 shows the net revenue per insurance status and the percent of zero or negative revenue in both cohorts. The hospital collected only from 1 of the 7 uninsured patients. This patient paid $20,615 out of pocket, >$10,000 from the median collections the hospital receives for this procedure from private insurance, PPO, and HMO patients. This is a clear example of the economic hardships our patients and hospital administrators have to face. Even though the numbers of patients are small in these insurance subsets, the data shed light on how hospital administrators increasingly have to rely on private insurance to cover the cost of those who are uninsured, or have Medicaid or Medicare. Patients without insurance are stuck with hospital bills two to three times the amount that hospitals routinely collect from insurance.

This preliminary analysis is limited by reporting only on one surgical procedure, hysterectomies, whereas in reality, multiple surgical specialties and multiple procedures are being performed robotically in all institutions that have the technology. There is selection bias in that the cost analysis was done for only those patients during this time frame who had successful completion of their minimally invasive surgery, and it was nonrandomized. These data may also be limited only to Carilion Clinic, Carilion Roanoke Memorial Hospital, Roanoke, Virginia and may not be representative of a non-teaching or for-profit hospital system.

Based on the data in this study, a sample size of 147 patients per group would have been required to achieve 80% power to detect the difference between the net operating margin of robotics versus laparoscopy when α=0.05, and it is one of the limitations of this study. Even though the numbers of patients are not large enough to show a statistically significant difference in net operative margin between groups, the data are consistent with previously reported data on costs. The operating room costs are significantly higher for robotics than for laparoscopy, and this higher cost negatively impacts net operative margin. Indirect costs were not included in this analysis, because there are many variables that affect them. The distribution of insurance status is shown to have a significant impact on the net operative margin. In this cohort, 15/28 (53.6%) of the robotic group and 32/48 (66.7%) of the laparoscopy group had an insurance status that provided higher collection rates, and a positive net operating margin. These data also exemplify that serving Medicare, Medicaid, and uninsured patients is not as cost effective for the U.S. healthcare system, with the majority showing a negative net operative margin irrespective of the cohort they were in.

Conclusions

Hospital administrators are under increased pressure to purchase robotic platforms and establish robotic programs, with little to no data on the impact of these programs on their operative margin. The data from this study, although limited, shed some light on the financial impact of performing robotic hysterectomies compared with laparoscopy, with the potential for robotics to still achieve a positive net operating margin. Future analyses would benefit from incorporating the impact of other robotic procedures as well as indirect costs to the operative margin. Ideally, a large population-based cost analysis should help establish formulas to properly estimate the operative margin for different proportions of patients undergoing robotic versus laparoscopy versus laparotomy procedures with differing proportions of insurance status. These analyses would then allow hospital administrators to make a more educated decision about investing in establishing a robotics program.

Footnotes

Acknowledgments

The authors thank Robert K. Herbertson, Jr., from the Carilion Clinic Statistics Department for his assistance in the analysis of this manuscript. No external financial support was received for this project. Presented as a poster presentation at the 41^st Annual Meeting of the Society of Gynecologic Oncologists, San Francisco, CA, March, 2010.

Disclosure Statement

Dennis Scribner is a paid consultant for Intuitive Surgical, Inc. No competing financial interests exist for the other authors.

References

Reich

, Decaprio

, McGlynn

. Laparoscopic hysterectomy. J Gynecol Surg, 1989; 5:213.

Mais

, Ajossa

, Guerriero

, Mascia

, Solla

, Melis

. Laparoscopic versus abdominal myomectomy: A prospective randomized trial to evaluate benefits in early outcome. Am J Obstet Gynecol, 1996; 174:654.

Paraiso

, Wlaters

, Rackley

, Melek

, Hugney

. Laparoscopic and abdominal sacral colpopexies: A comparative cohort study. Am J Obstet Gynecol, 2005; 192:1752.

Stylopoulos

, Rattner

. Robotics and ergonomics. Surg Clin North Am, 2003; 83:1321.

Agency for Health Research and Quality. Three characteristics double the likelihood of hysterectomy for women with common non-cancerous pelvic conditions. www.ahrq.gov/research/sep07/0907RA13.htm. 2010 January 18.

, Wechter

, Geller

, Nguyen

, Visco

. Hysterectomy rates in the United States, 2003. Obstet Gynecol., 2007; 110:1091.

Farquhar

, Steiner

. Hysterecotmy rates in the United States 1990–1997. Obstet Gynecol., 2002; 99:229.

Barbash

, Flied

. New technology and health care costs: The case of robotic-assisted surgery. N Engl J Med, 2010; 363:701.

Martin

, Nunez

, Castel

. Robot-assisted radical cystectomy versus open radical cystectomy: A complete cost analysis. Urology, 2010; 77:621.

10.

Smith

, Kurpad

, Lal

, Nielsen

, Wallen

, Pruthi

. Cost analysis of robotic versus open radical cystectomy for bladder cancer. J Urol, 2010; 183:505.

11.

Bolenz

, Gupta

, Hotze

, Ho

, Cadeddu

, Roehrborn

, Lotan

. Cost comparison of robotic, laparoscopic, and open radical prostatectomy for prostate cancer. Eur Urol, 2010; 57:453.

12.

Kam

, Cooray

, Kam

, Smith

, Almeida

. A cost-analysis study of robotic versus conventional mitral valve repair. Heart Lung Circ, 2010; 19:413.

13.

Park

, Flores

. Cost comparison of robotic, video-assisted thoracic surgery and thoracotomy approaches to pulmonary lobectomy. Thorac Surg Clin, 2008; 18:297.

14.

Heemskerk

, van Gemert

, Greve

, Bouvy

. Robotic-assisted versus conventional laparoscopic Nissen fundoplication: A comparative retrospective study on costs and time consumption. Surg Laprosc Endosc Percutan Tech, 2007; 17:1.

15.

Heemskerk

, de Hoog

, van Gemert

, Baeten

, Greve

, Bouvy

. Robotic-assisted versus conventional laparoscopic rectopexy for rectal prolapse: A comparative study on costs and time. Dis Colon Rectum, 2007; 50:1825.

16.

Dhari Patel

, Steinkampf

, Whitten

, Malizia

. Robotic tubal anastomosis: Surgical technique and cost effectiveness. Fertil Steril, 2008; 90:1175.

17.

Judd

, Siddiqui

, Barnett

, Visco

, Havrilesky

, Wu

. Cost-minimization analysis of robotic-assisted, laparoscopic, and abdominal sacrocolpopexy. J Minim Invasive Gynecol, 2010; 17:493.

18.

Advincula

, Xu

, Goudeau

, Ransom

. Robot-assisted laparoscopic myomectomy versus abdominal myomectomy: A comparison of short-term surgical outcomes and immediate costs. J Minim Invasive Gynecol, 2007; 14:698.

19.

Bell

, Torgerson

, Seshadri-Kreaden

, Wierda-Suttle

, Hunt

. Comparison of outcomes and cost for endometrial cancer staging via traditional laparotomy, standard laparoscopy and robotic techniques. Gynecol Oncol, 2008; 111:407.

20.

Sarlos

, Kots

, Stevanovic

, Schaer

. Robotic hysterectomy versus conventional laparoscopic hysterectomy: Outcome and cost analyses of a matched case-control study. Eur J Obstet Gynecol Reprod Biol, 2010; 150:92.

21.

Barnett

, Judd

, Wu

, Scales Jr

, Myers

, Havrilesky

. Cost comparison among robotic, laparoscopic, and open hysterectomy for endometrial cancer. Obstet Gynecol, 2010; 3:685.

22.

Pasic

, Rizzo

, Fang

, Ross

, Moore

, Gunnarsson

. Comparing robot-assisted with conventional laparoscopic hysterectomy: mpact on cost and clinical outcomes. J Minim Invasive Gynecol, 2010; 17:730.