Reimbursement for Prostate Cancer Treatment

Introduction

Prostate cancer (PCa) is the most common solid organ malignancy in men with over 200,000 new cases diagnosed each year. The long life expectancy of most men diagnosed with PCa has led to a burgeoning prevalence of the disease. An estimated 2.7 million men currently live with PCa in the United States, and the prevalence is only likely to increase in the future with the growth of the aging population. The healthcare resource implications associated with the diagnosis and treatment of PCa are vast. PCa accounts for approximately 10% ($12 billion) of the total annual budget for cancer care in the United States. Most cases are diagnosed by prostate-specific antigen (PSA) screening, which results in significant overdetection of indolent, low-risk PCa. The overdetection and overtreatment of PCa are estimated to account for around 40% of the total national cost of PCa.

PCa patients are confronted with multiple possible treatment options with comparable outcomes but with widely varying cost. The American Urologic Association guidelines recommend either surgery or radiation therapy for localized disease; however, there is currently no level I evidence demonstrating superiority of one modality over another. The guidelines recommend active surveillance as an option for patients with low-risk disease, yet studies have shown that it is rarely utilized with even the lowest-risk patients most often treated with definitive surgery or radiation therapy. The last decade has also witnessed the introduction of ever more expensive technologies, such as robotic surgery and intensity-modulated radiation therapy (IMRT) as first-line therapies in PCa. These novel technologies have disseminated widely and with little consideration to cost-effectiveness over traditional treatments such as open radical prostatectomy and conformal radiation therapy.

Several factors can influence the treatment decision in PCa patients. A significant referral bias exists with urologists more likely to recommend surgery, and radiation oncologists more likely to recommend radiation therapy. In the United States, Medicare reimbursement for medical services can also influence decision-making and the choice of therapy. An AUA survey reported that 16% of urologists had discontinued tests or procedures due to a reduction in reimbursement. In another study, Escarce et al. demonstrated that a reduction in Medicare reimbursement was associated with a decrease in volume and complexity of the services provided. Elliott et al. demonstrated the effect of reduced reimbursement for androgen deprivation therapy (ADT) following the 2003 Medicare Modernization act. They found that there was a significant reduction in nonindicated ADT use (defined as ADT given without surgery or radiation in low-risk disease) one year after implementation of the reimbursement changes.

Medicare expenditure for treating PCa continues to rise. Between 1992 and 2003, expenditure increased 20% per capita, mostly due to higher usage of physician services (rather than rising cost of services). This increase in cost to the health system was prior to the introduction of expensive technologies such as robotic-assisted laparoscopic prostatectomy (RALP) and IMRT, which have led to further rises in healthcare expenditure for PCa. With the unsustainable growth of national healthcare expenditure, a comprehensive analysis of the cost effectiveness of various treatment options for prostate cancer is imperative.

Economics of open versus robotic radical prostatectomy

There has been an exponential rise in the uptake of robotic technology by the urologic community over the last decade. In 2004 only 8% of prostatectomies were performed with robotic assistance, compared to 67% in 2010. The introduction of the robot may have even led to an increase in the overall proportion of men receiving surgery compared to nonsurgical treatments for PCa. This is evidenced by a 60% increase in the number of hospital discharges for radical prostatectomy between 2005 and 2008 based on the National Inpatient Sample. During the same period, prostate cancer incidence had remained stable. Several factors have driven this dramatic increase in RALP over the last decade. Aggressive (and often inaccurate) direct-to-consumer marketing by the manufacturer has been reported. Patient preferences for the most innovative and minimally invasive procedure further fuel the drive for robotic surgery. From the surgeon’s perspective, the robotic approach offers many benefits such as stereoscopic visualization, wristed instruments, tremor filtration, and improved ergonomics. Surgeons transitioning to RALP do not experience as steep a learning curve as that associated with pure laparoscopic prostatectomy; however, experience has still been shown to impact results. Medical facilities have embraced the technology due to patient and physician demand and possible financial incentives. However, the marketing drive to robotic surgery can lead to unrealistic patient expectations, with one study demonstrating increased patient dissatisfaction and regret following RALP as compared to patients undergoing open radical prostatectomy.

Despite the perceived advantages, there are no prospective randomized trials that demonstrate improved oncologic outcomes for RALP compared to open prostatectomy. Retrospective studies have not demonstrated an improvement in biochemical recurrence or survival for RALP compared to open or laparoscopic prostatectomy. The only consistently demonstrated benefit of RALP over open retropubic prostatectomy has been reduced blood loss and shorter hospital stay. The effect of RALP on positive surgical margin rates is unclear. A recent study demonstrated lower positive surgical margin rates for RALP compared to open prostatectomy (HR 0.75, p < 0.001), which goes against the findings of the meta-analysis by Ficarra et al. that demonstrated no difference in PSM between open radical prostatectomy and minimally invasive approaches. Recent results also suggest improved 12-month potency and continence rates following RALP compared to open surgery, although the use of nerve-sparing approaches, surgeon experience, and cancer stage influence these endpoints. In a recent systematic review comparing RALP to standard laparoscopic prostatectomy, RALP was associated with significantly reduced risk of organ injury and improved positive surgical margins. However, a study of Medicare claims of 5923 men comparing RALP and standard laparoscopic prostatectomy did not find a difference in bowel complications, or other general medical and surgical complications. The lack of standardized reporting is a major limitation of these comparative retrospective studies and partly explains the inconsistencies in outcomes between studies. Furthermore, the reports include surgeons with different levels of experience and at different points in the learning curve of RALP.

The widespread adoption of RALP as the primary surgical treatment modality has occurred with little consideration for the cost implications. The main costs to consider when comparing different modalities for prostatectomy are equipment costs, operative time, and length of hospital stay. Several studies have highlighted the considerable increase in cost of surgery associated with RALP compared to open prostatectomy. Bolenz et al. reviewed a number of studies reporting direct costs for RP, and demonstrated that the costs varied from center to center but on average was higher for RALP (range $5058–$11,806) than open prostatectomy (range $4075–$6296). Lotan et al. demonstrated that RALP was $1155 more costly than open prostatectomy, largely due to the high cost of the instruments. This analysis did not take into account the fixed fee for purchasing the robot, amounting to 1–2 million dollars. In addition to the large initial outlay of purchasing the robot, there is a yearly service maintenance fee of $100,000–$200,000. In some centers where the robot is received as a gift, a maintenance cost of $150,000 would still add $1000 per case if 150 cases were performed per year. The current lack of competition in the robotic market, with only one manufacturer producing the surgical robot means that equipment costs are likely to remain high.

Operative time and length of stay are important factors to consider in the cost of surgical procedures. With the inclusion of robotics into many urology residency programs, the issue of the learning curve is becoming less of a factor for urologists starting independent practice. Studies vary widely in their estimates of number of RALP cases required to master the procedure. Guillonneau and Vallancien reported that it took 100 cases for an expert laparoscopist to reduce their RALP operating time from 4 h to 3 h; however, others have reported that it can take up to 200 cases to master the procedure. The cost of the learning curve based on differences in operating time has been estimated to be approximately $215,000. Davis et al. examined a nationwide database of over 71,000 prostatectomies and demonstrated longer operative time for RALP compared to open (mean 4.4 h vs. 3.4 h, p < 0.0001), but also a shorter inpatient stay with RALP (mean 2.2 days vs. 3.2 days, p < 0.0001). Surgery time, hospital stay, and complication rates improved significantly with surgeon experience.

Medicare reimbursement for radical prostatectomy is the same regardless of surgical approach. Thus, the added cost of robotic prostatectomies falls on the provider. Lotan et al. demonstrated that each robotic case resulted in a loss of over $4000 to the hospital. However, in a recent analysis of hospital reimbursement for RALP from mixed payers, reimbursement was higher for RALP compared with open prostatectomy (approximately $2000 per case). RALP is also potentially advantageous to the surgeon due both to increased volume and additional reimbursement for the S2900 code (use of robot). However, reimbursement for the S2900 code is inconsistent since Medicare does not pay for the S2900 code and payment is variable between insurance companies. In addition, the performance of laparoscopic pelvic lymph node dissection is bundled with open radical prostatectomy but not with RALP so further physician payment can be obtain for performing a laparoscopic node dissection at time of RALP.

The additional cost of RALP could be justified if the procedure provided improved outcomes or quality of life for patients with PCa. To date, given the minimal benefit associated with RALP and the additional cost of the robot, most studies have not demonstrated the cost-effectiveness of RALP. Cost-effectiveness analyses employ the incremental cost-effectiveness ratio (ICER) as a measure of cost-effectiveness of one health intervention over another. Although not an absolute cut-off, an ICER of <$50,000 per quality adjusted life year (QALY) is regarded as a cut-off indicative of cost-effective intervention. Markov models and decision tree analyses are often employed in cost-effectiveness studies to model patient outcomes over a specified time horizon. In a study comparing the various surgical and radiation treatment modalities, Cooperberg et al. compared cost-effectiveness based on literature estimates of outcome probabilities over a lifetime horizon. They determined that QALY was not significantly different between different surgical modalities. Costs accounted for included direct costs of medical resource utilization (procedures, office visits, imaging). The mean lifetime costs were comparable for robotic and open prostatectomy ($19,901 vs. $20,245). Since cost estimates were derived from Medicare Fee schedules, which reimburse similarly for RALP and open prostatectomy, the study did not demonstrate a significant difference in cost-effectiveness between the two surgical modalities. This study took the perspective of the payer (Medicare) and did not account for the additional costs of RALP, which Medicare does not cover.

A cost-effectiveness analysis of RALP versus laparoscopic prostatectomy from the European perspective estimated the incremental cost per QALY gained from robotic compared to laparoscopic prostatectomy at £18,329, well below the threshold of £30,000 for cost-effectiveness set by the National Institute for Health and Care Excellence (NICE) in the United Kingdom. This analysis was conducted with the assumption that >150 prostatectomies per year would be performed, and the sensitivity analysis demonstrated that when only 50 procedures per year were performed, the ICER rises considerably to £106,839/QALY. RALP was deemed cost-effective despite higher costs due to the model assumption that RALP offered an advantage in terms of fewer organ injuries and lower rates of positive surgical margins, which over a lifetime horizon result in higher QALY with RALP. This study, like the Cooperberg study, did not take into account the capital and maintenance costs of the robotic procedure. Despite geographic variations, these studies highlight the potential cost-effectiveness of RALP, and emphasize the dependence of cost on surgical case volume.