The national movement for comparative effectiveness research recognizes the necessity of information to help providers choose the best treatment for each patient, and comparative studies of prostate cancer treatment options are a high priority on the agenda.26
Mounting evidence from observational studies suggests that for men with early-stage prostate cancer, MRP and ORP are more similar than different with respect to surgical complications, functional impairment and oncologic outcomes.3-5, 27
Results of our population-based analysis suggest that the two procedures also have similar economic outcomes, at least in the first year following surgery.
Estimates of health care costs associated with different interventions can be strongly influenced by both the perspective and time horizon of analysis. Prior comparative studies of costs associated with MRP and ORP have primarily focused on the perioperative period and assumed a hospital or surgeon perspective. Most of these studies found that the robotic cases cost more than the open ones, and differences between the two procedures were primarily attributable to the high fixed costs associated with purchasing and maintaining the robot.9, 10, 12, 28
The da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA, USA) costs between $1 million and $2.5 million with a maintenance fee of $120,000 per year after the first year, and the average cost of disposable supplies is approximately $1,500 per case. The purchase and maintenance costs of the robot are estimated to add almost $2,700 per patient to the cost of a prostatectomy,9
although this incremental cost would be expected to decline with increasing case volume.12
A recent review of several different robotic procedures found that use of the robot increased costs from as little as $400 to as much as $4,800 per case.8
Considering amortized equipment costs, utilization of the robot increased total procedure costs by more than $3,000 per case, compared with the open version of a procedure.8
We assessed cost from the perspective of the payer, rather than a hospital or provider, using Medicare payment as an approximation of cost. For radical prostatectomy, hospital reimbursement formulas within the Medicare prospective payment system do not vary by surgical approach, while the professional fees billed by surgeons and reimbursed under the resource-based relative value scale methodology do vary. Without pelvic lymphadenectomy, the difference in average physician reimbursement is about $100, favoring MRP; with lymphadenectomy the difference is closer to $1,000.29
Although we saw a difference in Medicare payments associated with all care received during the initial surgical stay, this difference was less than $1,000 – smaller than the differences of $1,726 to $2,698 estimated from a hospital or provider perspective.9, 10
In addition to examining costs associated with the initial surgical procedure, we also estimated costs in the first year following surgery. If there were a true difference in clinical outcomes between MRP and ORP, we might expect to see differences in the use and cost of health care services beyond the perioperative period. We found that in the first year, mean total costs of care did not differ, but costs within specific categories did. For example, average costs associated with surgery, inpatient care after the initial surgical hospitalization and other medical care were greater in the MRP group, while the average cost of other cancer therapy was greater in the ORP group. However, average costs are a function of both service utilization and cost per service. While a slightly greater proportion of ORP patients had any inpatient care costs after the initial hospitalization, the average cost of inpatient care among users was more than $6,000 greater for those who had MRP. Conversely, while a substantially greater proportion of MRP patients had any home health care costs, the average cost of home health care among users was more than $800 greater in the ORP group.
Trends in the use and cost of prostate cancer surgery are likely to influence the overall cost of treating this disease. Between 1991 and 2002, the initial costs of prostate cancer treatment declined, largely due to a decrease in the proportion of men receiving surgery.16
Since then, however, aggressive marketing by hospitals and the robot manufacturer, combined with Americans’ appetite for new technology, has fueled a dramatic increase in the number of MRPs performed annually in the US. Intuitive Surgical, manufacturer of the da Vinci robotic surgical system, reports that in 2009 approximately 90,000 robotic radical prostatectomies were performed in the US compared to just under 20,000 in 2005.30
The number of da Vinci robotic systems purchased by US hospitals has increased from approximately 300 in 2005 to over 1,150 in 2010.30
From 2005 to 2008, hospital discharges for radical prostatectomy increased 60% in the US, largely because of the increase in robotic procedures, 8
and analysis of hospital discharges in six states found that acquisition of a robot was associated with an increase in the total number of prostatectomies, at both the hospital and regional levels.31
Whether or not increased use of MRP represents overtreatment (i.e., surgeries that would not have been performed or indicated in the absence of the robot), the popularity of robotic surgery has likely driven up the overall cost of prostate cancer care. Technological advances in other treatment modalities, such as the use of intensity-modulated radiation therapy instead of three-dimensional conformal radiation therapy may be increasing prostate cancer treatment costs as well.32
Some advantages of robotic surgery may not be evident from standard economic and clinical endpoints. Robotic surgery may improve visualization and dexterity, reducing the technical complexity of the procedure compared with traditional laparoscopic prostatectomy. However, there is little evidence that these benefits have influenced clinical and economic outcomes, other than reductions in blood loss and length of stay with MRP. From the payer’s perspective over a one-year period, these benefits did not translate to net savings in our analysis, compared with ORP.
Several limitations of our analysis should be noted. Our findings are generalizable to the US population of older prostate cancer patients covered by Medicare, but costs associated with prostate cancer surgery and health service utilization after surgery may differ in younger men, due to both their age and insurance coverage. Although we were able to control for numerous important demographic and clinical characteristics, other unmeasured factors, such as the patient’s functional status or his surgeon’s experience, could potentially confound the relationship between surgical approach and costs. Because all laparoscopic prostatectomies shared a single HCPCS code (55866) during the study period, we could not distinguish procedures performed with and without robotic assistance. However, information regarding the uptake and use of the da Vinci surgical system suggests that the majority of MRPs in our analysis were robot-assisted. Finally, the use of a payer perspective ignores fixed costs borne by hospitals (e.g. purchase and maintenance costs of robot and disposable instrument costs) and out-of-pocket costs borne by patients and their families, although it is not clear whether those out-of-pocket costs would vary by surgical procedure, particularly if clinical outcomes are similar.
As the population ages and the incidence of early-stage prostate cancer rises with continued use of prostate-specific antigen (PSA) screening, total expenditures for prostate cancer, and the portion associated with surgery, will grow as well. If cancer control and functional recovery are similar between MRP and ORP, then cost considerations become even more important for payers, providers, and patients choosing between open and robotic surgical approaches. Our findings, in a population-based cohort of older men, suggest that there is no obvious cost advantage to either approach in the first year following surgery.