Objective
The objective of the study was to determine total hospital costs and net hospital income for different types of minimally invasive hysterectomy and financial impact if a subset of patients underwent total vaginal hysterectomy (TVH) instead of their selected procedure.
Study Design
A retrospective chart review was performed of patients who underwent hysterectomy for benign disease by TVH, laparoscopic assisted vaginal hysterectomy (LAVH), total laparoscopic hysterectomy (TLH), and robotic hysterectomy (RH) between Jan. 1, 2007, and April 30, 2010, at Thomas Jefferson University Hospital. The hospital decision support database was used to calculate net hospital income. A subset of patients with at least 1 prior vaginal delivery, no more than 1 laparotomy, and a uterine size less than 14 weeks who had undergone RH, TLH, or LAVH was identified as potential TVH candidates. The financial impact of performing TVH over the selected hysterectomy was calculated.
Results
Three hundred thirty-four cases of minimally invasive hysterectomy were identified. Fifty-five percent were TVH, 33% LAVH, 3% TLH, and 9% RH. Mean total hospital costs for TVH were $7903, $10,069 for LAVH, $11,558 for TLH, and $13,429 for RH ( P < .0001). Net hospital income was $1260 for TVH. The hospital incurred losses of $–1306 for LAVH, $–4049 for TLH, and $–4564 for RH ( P = .03). Our criteria to determine the mode of hysterectomy increased TVH from 57% to 76% of all minimally invasive hysterectomy.
Conclusion
Hospital costs were greater with LAVH, TLH, and RH than for TVH. The hospital incurred financial losses with LAVH, TLH, and RH. TVH was the only minimally invasive modality of hysterectomy that generated net hospital income. Our conservative criteria to determine the route of hysterectomy would increase the number of TVHs by more than 30%.
Approximately 600,000 hysterectomies are performed annually in the United States, mostly for benign conditions. The American College of Obstetricians and Gynecologists (ACOG) supports the use of the minimally invasive hysterectomy, and in particular vaginal hysterectomy, as the surgical approach of choice when planning a hysterectomy. Minimally invasive surgeries result in shorter hospital stays, less blood loss, and faster return to normal activities. A Cochrane review found no benefit of laparoscopic hysterectomy over vaginal hysterectomy. In this same review, total vaginal hysterectomy (TVH) was associated with shorter operating times and less bleeding than laparoscopic hysterectomy.
The rate of laparoscopic hysterectomy increased from 0.3% in 1990 to 11.8% in 2003, whereas the incidence of total vaginal hysterectomy decreased from 24.4% to 21.8%. The incidence of robotic hysterectomy increased from 0.5% of all hysterectomies in 2007 to 9.5% in 2010. Laparoscopic hysterectomy increased from 24% to 30% in the same time period. In hospitals in which robotic hysterectomy is performed, it now accounts for 22% of all hysterectomies. An ACOG position statement recently stated that health care costs would increase by an estimated $960 million to $1.9 billion annually if robotic surgery were used for all hysterectomies.
In our study, we compared hospital costs and net hospital income for the various modalities of minimally invasive hysterectomy for benign disease including TVH, laparoscopic-assisted vaginal hysterectomy (LAVH), total laparoscopic hysterectomy (TLH), and robotic hysterectomy (RH). We then evaluated the financial impact of preferentially performing a TVH on a subgroup of patients.
Materials and Methods
The study proposal received approval from the Thomas Jefferson University Institutional Review Board. All the authors were on staff at the institution at the time of data collection.
We performed a retrospective chart review for patients who had undergone TVH, LAVH, TLH, or RH for benign indications between Jan. 1, 2007, and April 30, 2010, at Thomas Jefferson University Hospital (TJUH), using International Classification of Diseases , ninth revision, Clinical Modification, procedure codes.
We extracted cost and reimbursement data from the TJUH hospital decision support database using the Allscripts Decision Support System. This system takes input feeds from numerous sources (payroll, general ledger, patient billing, and health information management) and transforms the inputs into a cost for each charge. Charges at this hospital, and many other large hospitals, are based on labor time units (eg, room and bed charge, operating room [OR] times), supplies (medications, implantable devices or blood products based on vendor contracts and a markup for the labor associated with processing the item and transportation and other miscellaneous costs), or tests (x-ray, blood work). The charges are not modified for a specific payer.
Costs per charge take into account the job category and average expected time assigned to perform an individual task based on internal time studies as well as the direct supply expense associated with the charges and other allocated expenses including depreciation if appropriate for the charge. Physician costs were not included in these calculations. The cost per charge can then be applied to the patient level charges to determine a cost per patient.
Itemized data on total hospital costs including nursing costs and OR/postanesthesia care unit (PACU) costs and net hospital revenue, which consists of payments received from insurers and copayments by patients, were extracted from the decision support system. Net hospital income was calculated by subtracting net hospital revenue from total hospital costs. The examiner’s clinical estimate of uterine size at the time of the preoperative evaluation as documented in the preoperative history and physical was used. If there was also a clinical estimate of size in the operative report or there was a size range recorded, the larger size was used. If no clinical estimates were recorded, we used calculations validated by Cantuaria et al, who found a relationship between the examiners’ clinical estimate of uterine size, the length of the uterus on transvaginal ultrasound, and the length of the pathologic specimen.
Using the length of the pathological specimen (PL) to calculate the ultrasound length (USL; 2.94 + [PL*0.75]), we then used this value to give an estimate of size in weeks of gestation ([USL – 3.68]/0.68) rounded to the nearest whole number. If there was no recorded clinical estimate of size and no pathological specimen length (for example, if the specimen had been morcellated during the procedure), then the patient was excluded from our analysis. We calculated the averages for all of these data points for each type of minimally invasive hysterectomy.
We did a secondary analysis of patients who underwent LAVH, TLH, or RH to identify all patients who were likely to have been reasonable candidates for TVH ( Figure ). Using a history of vaginal delivery as a marker for descensus, we included patients with at least 1 vaginal delivery. If information on parity was not available in the inpatient hospital chart, the electronic outpatient record was reviewed. If there was still no history regarding parity, the patient was excluded. Additionally, we excluded all patients who had a history of more than 1 laparotomy.
It was assumed that cholecystectomy, appendectomy, and tubal ligation were performed laparoscopically unless it was stated otherwise. Previous myomectomies were assumed to have been performed via laparotomy and excluded, unless it was specifically stated that they were performed hysterscopically or laparoscopically.
Additionally, we excluded all patients who had estimated uterine sizes of more than 14 weeks. We chose these criteria based on the consensus of the authors’ combined training and expertise. We then calculated the potential savings in hospital costs if these cases had been performed as a TVH over the actual minimally invasive procedure performed.
Analysis of variance (ANOVA), assuming unequal group variances, was performed to compare different hysterectomy modalities with respect to total hospital costs, hospital net income, OR and PACU costs, nursing costs, OR time, and uterine size. The software package SAS version 9.3 (SAS Institute, Cary, NC) was used. Data that were not normally distributed were log transformed to meet criteria for parametric testing. Geometric means and 95% confidence intervals are presented by hysterectomy types for log-transformed variables. Post-hoc pairwise comparisons were adjusted using Dunnett’s method after the ANOVA to compare TVH directly with each minimally invasive procedure.
Results
Our search resulted in a total of 339 minimally invasive hysterectomies performed at our institution during the study period. Four cases were excluded because no data existed on uterine size. One case was subsequently discovered to be an adenocarcinoma of the uterus and excluded. This left a total of 334 cases in our initial analysis.
We identified 185 total vaginal hysterectomies, 55% of all minimally invasive hysterectomies, 110 LAVHs (33%), 9 TLHs (3%), and 30 RHs (9%).
Total hospital costs, as well as OR and PACU costs, were lowest for TVH compared with other minimally invasive hysterectomies. TVH also generated the highest hospital net income, with a mean income of $1260 compared with losses of $4000 for TLH and RH ( Table 1 ). A post-hoc pair wise comparison using Dunnett’s method demonstrated that each of the other minimally invasive hysterectomy modalities resulted in higher hospital costs and generated less hospital income when compared directly with TVH.
| Variable | TVH (n = 185) | LAVH (n = 110) | TLH (n = 9) | RH (n = 30) | P value |
|---|---|---|---|---|---|
| Hospital net income, $ | 1260 (544.3−1976) | −1306 (−2232 to −379) | −4049 (−5975 to −2123.6) | −4563.9 (−7657 to −1471) | < .0001 |
| Total hospital costs, $ | 7903 (7613−8203) | 10,068.52 (9715−10,435) | 11,558 (10,115−13,206) | 13,429 (12,061−14,952) | < .0001 |
| Nursing costs, $ | 1440 (1365−1518) | 1950 (1666−2281) | 1429 (1075−1900) | 2245 (1771−2844) | < .0001 |
| OR/PACU costs, $ | 4980 (4777−5191) | 6399 (6166−6642) | 8190 (6830−9819) | 8974 (8188−9835) | < .0001 |
| Other costs, $ | 1426 (1376−1477) | 1444 (1374−1517) | 1680 (1400−2016) | 1783 (1568−2029) | .006 |
| OR time, h | 3.39 (3.26−3.53) | 3.37 (3.18−3.56) | 4.74 (3.87−5.81) | 4.68 (4.23−5.19) | < .0001 |
| Uterine size, wks | 9 (8.7−9.3) | 11 (10.3−11.5) | 8.78 (7.3−10.3) | 10.3 (9.3−11.2) | < .0001 |
In our secondary analysis, 60 patients with no history of vaginal delivery were excluded. If our criteria for the subanalysis had been applied to all patients who actually underwent TVH, there would have been 15 fewer vaginal hysterectomies of the 185 actually performed in the TVH group, which underscores the conservative nature of our parameters.
Of the women who underwent other forms of minimally invasive hysterectomy rather than TVH, 6 patients were excluded from being in the potential TVH group for having more than 1 laparotomy, an additional 6 patients were excluded because the estimated uterine size was greater than 14 weeks, and 7 patients were excluded because their obstetrical history was unavailable. This left a total of 253 patients who potentially could have undergone a TVH rather than the procedure they had performed. This meant that of all patients undergoing minimally invasive hysterectomy, 76% could have undergone a TVH applying our criteria. The hospital net income of this subgroup of patients demonstrated losses of $1521, $4714, and $4791 for LAVH, TLH, and RH respectively ( P = .0002; Table 2 ).
| Variable | LAVH (n = 64) | TLH (n = 6) | RH (n = 13) | P value |
|---|---|---|---|---|
| Hospital net income, $ | −1520.9 (−2579 to −463) | −4714 (−7239 to −2189) | −4791 (−8850 to −732.5) | .0002 |
| Total hospital costs, $ | 9905.6 (9470.4−10,361) | 12,048 (10,379−13,984) | 13,425 (11,246−16,026) | < .0001 |
| Nursing costs, $ | 1841 (1419−2387) | 1299.5 (918−1840) | 2772 (1861−4129) | .007 |
| OR/PACU costs, $ | 6242 (5944−6555) | 8802 (7443.1−10,409) | 8291 (7351−9351) | < .0001 |
| Other costs, $ | 1452 (1365−1544) | 1713.5 (1317−2230) | 1808 (1818−2305) | .11 |
| OR time, h | 3.3 (3.1−3.5) | 5.4 (4.4−6.7) | 4.2 (3.7−4.8) | < .0001 |
| Uterine size, wks | 10 (9.6−10.7) | 9.7 (8−11.4) | 10.2 ( 9−11.3) | .007 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
