This study provides a detailed description of the median cost per RAS patient for multiple specialties including benign gynaecology, cardiothoracic, colorectal and urology, and outlines the main cost drivers specifically related to RAS such as the implementation, maintenance and consumable costs and considers how these would change depending on case volume.
Substantial variation was found between specialties in line with the different procedure types with the median cost per patient being $19,269 (IQR: $15,445 to $32,199) whereby the overall specific RAS costs contributed almost half of this (46%) A further breakdown of the specific RAS costs found the contribution of the implementation, maintenance and consumable costs to the overall per patient cost were 24%, 12% and 10% respectively. It should be highlighted the fixed implementation and maintenance costs per patient were derived based on the volume of cases undertaken during the study period, which was relatively low. Further investigation is required to determine the reason for this, which could include case selection with an overall robotic theatre utilisation of 37% (ranging from 27% for cardiothoracic to 41% for colorectal). In considering the impact of case volume, it was shown the implementation and maintenance cost can be reduced considerably if the robotic theatre is utilised to its full capacity. This is in keeping with previous research, which has suggested that an increase in the number of cases was associated with cost reduction. 13,19 Accordingly, implementation costs, which are rarely factored into robotic costing studies, should also be considered from a cash flow perspective regarding when and over what time period the capital costs are incurred by an organization, along with attention being given to how the costs may be further reduced such as not requiring or postponing theatre refurbishments or the negotiation of annual maintenance contracts.
Overall the main cost driver for RAS cases were those incurred within the operating theatre (44%), which is consistent with the findings of other studies.10 Given it is well known operating theatres are high cost environments regardless of the technology being utilised, an examination of the RAS consumable costs against the overall operating theatre expenditure found it to be contributing to 22% of these costs.
Consideration of the costings described in this study should be made within the context of case selection in terms of the contribution that RAS makes to the overall cost per case. The experience across the four surgical specialties involved in this public hospital was mixed. Within cardiothoracic and colorectal, the cases were complex with patients staying in hospital around 6 days with a median cost of $44,656 and $32,656 per case resulting in the contribution of the additional RAS expenditure of $7,787 and $9,212 being approximately 17% and 28%, respectively. It is important to note that for some of these cases, usual care would not have been an option due to their complexity. This was in comparison to benign gynaecology and urology where the patients stayed in hospital less than two days with the overall cost per case being much lower at $15,067 and $18,000 so the overall contribution of the additional RAS expenditure was much greater at 60% and 49%, respectively.
Within the current Activity Based Funding (ABF) environment, it is also important to highlight the coding system does not recognize RAS resulting in the additional costs of the technology having to be absorbed within the hospital’s general budget. This was evident in the average difference between the iFRAC expenditure data (incorporates all expenditure) and the NWAU costing data (which does not yet recognise robotics) demonstrating an average funding deficit of $4,162 per RAS case, which is similar to the specific RAS costs described at $4,138 per case (when excluding implementation costs). Although the nationally applied coding system has resulted in this barrier being experienced within RAS programs in other Australian states, 9,10 the funding model has longer term implications for the introduction of new surgical technology into the public sector. Arguably the lack of responsiveness and considerable lag time for the coding system to be updated to incorporate new technology or procedures, does not readily support innovation and potentially hinders the public sector having access to the latest technological advances. It should be noted that New South Wales does have a mechanism for supporting models of care involving high cost and low volume patient care; however, surgical robotics is currently not included within this scheme.
The costings described within this study are comparable to other costing studies investigating RAS both nationally and internationally (when converted to Australian dollars). Within benign gynaecology, the per patient cost in this study (AU$15,067) corresponds to the experience in the US for robotic-assisted hysterectomies costed at $AU15,11720 and $AU14,906. 21 Within cardiothoracic, the per patient cost in this study ($AU44,655) for partial robotic-assisted CABG was higher compared to a study in the US where it was found to be $AU18,34222; however, the partial nature of the local RAS cases makes comparison difficult. For RAS within colorectal, the consumable costs were found to be $AU2,728 higher in a Victorian based study, 8 which is similar to the $AU2,231 of specific RAS consumables found for rectal resections in this study. Finally within urology, the per patient cost in this study of $18,000 for robotic-assisted prostatectomy was similar to the cost of $AU17,582 found in a Queensland study, 10 and $AU13,860 in the US. 23
This study has a number of limitations. Firstly, it is based on the experience of a single centre during a set period of time and there may be some variation in the application of costing methodologies used. Similarly, there are known constraints to using administrative datasets including they have been shown to under-report patient complexity.24 Although the overall cohort of RAS patients examined is large, the numbers within each specialty are relatively small and may result in the findings not being truly representative. Finally, this study is purely descriptive and does not compare the cost drivers of RAS to conventional laparoscopic or open procedures, or explore the opportunity costs associated with low case numbers. As such the potential cost effectiveness of RAS is not able to be determined from this dataset. However, given the dearth of literature describing the detailed costings of RAS cases in the public sector across multiple specialties it was still felt this study offers a valuable contribution. Future economic studies from this RAS program will be focusing on cost-effectiveness analyses to compare the cost and consequences of RAS versus usual care.
In conclusion, the cost of RAS is substantial contributing between 17–60% of the overall cost of the patient’s surgical treatment depending on inclusions, volume and the surgical specialty. It is important for local hospital administrators, health ministries and governing bodies to be aware of the cost components and drivers when establishing a RAS program, and to highlight the importance of new technology being incorporated into standardised funding systems.