How can Health Technology Assessment (HTA) support the optimal use of high-cost devices? A case study of robotic-assisted surgery in Scotland

Lai, Tzu-Jung (2025) How can Health Technology Assessment (HTA) support the optimal use of high-cost devices? A case study of robotic-assisted surgery in Scotland. PhD thesis, University of Glasgow.

Abstract

Background and aims: Health technology assessment is a multidisciplinary process that evaluates the safety, efficacy, and cost-effectiveness of healthcare interventions, guiding evidence-based decisions and addressing social, organisational, and ethical issues. Traditionally, health technology assessment has been effectively applied to new drugs for reimbursement decisions post regulatory approval, leveraging clear evidence on costs and consequences. However, high-cost medical devices present unique challenges that complicate the health technology assessment process, such as incremental development, context dependency (i.e. organisational impact), quality variation (i.e. evidence uncertainties) and physical mode of action (i.e. device-operator interaction). Additionally, health technology assessment is often overlooked at the time of equipment purchase, leading to uncertainties in assessing long-term value and impact.

Surgical robotics, specifically robotic-assisted surgery, exemplifies these challenges. Robotic-assisted surgery systems are expensive, complex, and have been rapidly adopted despite ongoing debates about their evidence base. In Scotland, significant investments in surgical robots were made in 2021, aiming to improve access to minimally invasive procedures and reduce health inequalities. Decision-makers now face the challenge of expanding robotic assisted surgery services amidst these uncertainties.

Given the ongoing adoption and expansion of robotic-assisted surgery in Scotland, this thesis proposes there could be a role for health technology assessment methodology in addressing these associated uncertainties and helping decisionmakers to prioritise the expansion.

This thesis aims to investigate whether health technology assessment can support the optimal use and implementation decisions for high-cost devices by taking a case study of robotic-assisted surgery in Scotland. This study seeks to determine how health technology assessment can guide future investment and expansion decisions, ultimately informing strategies for integrating innovative technologies into healthcare systems effectively.

Methods: To explore the role of health technology assessment in optimising the use of robotic-assisted surgery for Scottish decision-makers, this research employed a multi-step approach.

First, an overview review of clinical effectiveness evidence was conducted to identify which specialties and procedures were most likely to benefit from the expansion of robotic-assisted surgery. This review aimed to map the current landscape of evidence across intra-cavity procedures and pinpoint areas with the most robust comparative outcomes, thereby guiding where the use of robotic-assisted surgery may be most clinically appropriate.

Secondly, a scoping review of economic evaluations of robotic-assisted surgery was undertaken to identify what economic methods have been used in analysing robot-assisted surgery, to investigate how they addressed the challenges of robotic-assisted surgery in economic evaluation research and to explore what opportunity there is to improve methods of evaluation. Insights gained from this analysis informed the design of a tailored approach for economic evaluation of high-cost devices.

Third, a two-stage economic model was developed, informed by both the evidence reviews and stakeholder consultations which helped shape the model's scope and assumptions to ensure it addressed relevant policy questions and practical constraints. The first-stage model was a short-term, procedure-specific costutility analysis comparing robotic-assisted surgery with laparoscopic and open surgery across selected procedures: prostatectomy, colorectal resection, hysterectomy, and pancreaticoduodenectomy. The second-stage model integrated this into a system-level platform model, allowing for exploration of case-mix strategies, annual procedure volumes, and surgical replacement proportions across specialties. This experimental model enables decision- makers to simulate various utilisation scenarios and identify more efficient strategies for shared robotic-assisted surgery platform use under different capacity and investment constraints.

Results: From the overview of systematic reviews, most evidence was available in urology, colorectal, hepatopancreaticobiliary and gynaecology. A total of 165 systematic reviews were included comparing robotic-assisted surgery to laparoscopic and open surgery. In my developed novel evidence map, it presented the strength of evidence and its orientation. Within the selected procedures, the evidence (such as conversion rate, estimated blood loss, length of hospital stay, and postoperative complication) was largely neutral or positive for robotic-assisted surgery compared to both laparoscopic and open approaches with the exception of operative time. Evidence was more positive compared with open surgery. I found that most systematic reviews were of low quality due to a failure to deal with the inherent bias in observational evidence.

In the scoping review of economic evaluations, a total of 50 studies addressing the economic analysis of robotic-assisted surgery were identified. Cost-utility analysis (46%) was the most commonly applied economic evaluation method, followed by cost-consequence analysis (32%). Generally, I found the evidence on the costeffectiveness of robotic-assisted surgery compared to open and/or laparoscopic surgery was mixed, with evaluations having a high degree of heterogeneity including multiple indications, outcomes, comparators, time horizons, perspectives and settings. Distinctive features related to the assessment of robotic-assisted surgery were under-addressed in economic evaluations. Only 40% of the included studies considered learning curve and organisational impact including capital cost investment, annual volume of procedures and platform sharing, and less than 12% of the included studies reflected on incremental innovation and dynamic pricing. Only two studies addressed the fact that the surgical platform was shared. Overall, a large proportion of economic evaluations did not explicitly account for the specific characteristics of robotic-assisted surgery. It is clear that to have a more realistic assessment of the costeffectiveness of robotic-assisted surgery, economic analysis should consider these distinctive features to ensure its optimal utilisation in clinical practice.

In the two-stage economic model evaluation, stage one demonstrated that while robotic-assisted surgery consistently offered higher utility gains compared to laparoscopic and open surgery, its cost-effectiveness varied significantly by procedure. Among the procedures studied, robotic-assisted surgery was not cost-effective against laparoscopic surgery, but showed more favourable results when replacing open surgery, particularly in prostatectomy. Scenario analyses indicated that removing capital costs, representing settings where surgical robots are donated or externally funded, substantially improved the cost-effectiveness of robotic-assisted surgery. However, even in such cases, the opportunity cost of the capital investment must still be considered, especially when viewed from a system or national perspective. Sensitivity analyses identified utility values, length of stay, and operative time as the most influential drivers of cost-effectiveness. These variables also helped explain the findings of the threshold analyses, which showed that increasing the proportion of open surgery replaced by robotic-assisted surgery consistently reduced incremental cost-effectiveness ratios. Stage two extended the analysis to the system level, showing that cost-effectiveness depends on both procedural mix and surgical volume. Economies of scale were critical, with most strategies only becoming cost-effective at ≥350 cases annually, or when focused on high-impact procedures. These findings highlight that RAS can represent value for money if strategically deployed at sufficient volumes and targeted to procedures with the greatest marginal benefit.

Conclusion: This thesis highlights the critical role of health technology assessment in supporting the optimal adoption and utilisation of high-cost medical devices, with a specific focus on robotic-assisted surgery. The research demonstrates that health technology assessment provides a vital tool for decision-makers, facilitating a structured approach to assess the clinical effectiveness, costeffectiveness, and broader implications of innovative technologies.

The overview of clinical effectiveness narratively summarises the evidence and maps it into a novel evidence spectrum. It showed that evidence for robotic-assisted surgery is largely neutral or positive compared to laparoscopic and open approaches. This suggests that selective adoption of robotic-assisted surgery could improve patient outcomes through strategic replacement of more invasive techniques.

The scoping review of economic evaluations revealed that key features unique to robotic-assisted surgery, such as the learning curve, platform-sharing potential, volume sensitivity, and dynamic pricing, are often neglected in existing models. Incorporating these elements can offer a more realistic and comprehensive understanding of robotic-assisted surgery’s value, guiding more efficient decisions around adoption and scale-up.

Building on these insights, a two-stage system-level economic model was developed, offering a novel approach to guide resource allocation and utilisation strategies post-acquisition. Stage one assessed procedure-level cost-effectiveness; stage two allowed decision-makers to test alternative configuration scenarios, such as case-mix, annual volumes, and replacement strategies, based on their local context. The framework provides decision-makers with a practical tool for planning, emphasising that the role of RAS lies not only in clinical outcomes but also in enabling broader access to minimally invasive surgery and guiding resourceefficient service delivery.

While the model provides recommended prioritisation strategies, successful implementation depends on operational realities such as workforce capacity, procedural demand, and existing infrastructure. Nonetheless, its adaptability allows for iterative refinement as new data and service constraints emerge. Ultimately, this thesis demonstrates that HTA can be applied not only at the point of adoption but throughout the technology’s lifecycle, from early evaluation to post-investment optimisation. Though centred on RAS, the insights and methods presented here are generalisable to other high-cost, cross-specialty platform technologies. The research provides a robust and adaptable framework for ensuring that such innovations are integrated into healthcare systems in a costeffective, evidence-informed, and context-sensitive manner.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: R Medicine > R Medicine (General); T Technology > T Technology (General)
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Health & Wellbeing > Health Economics and Health Technology Assessment
• Supervisor's Name: Boyd, Professor Kathleen and Heggie, Dr. Robert
• Date of Award: 2025
• Depositing User: Theses Team
• Unique ID: glathesis:2025-85567
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 06 Nov 2025 15:48
• Last Modified: 06 Nov 2025 15:49
• Thesis DOI: 10.5525/gla.thesis.85567
• URI: https://theses.gla.ac.uk/id/eprint/85567

Related URLs

• Article DOI
• Article DOI
• Conference item
• Conference item