The promise of robotic assistance in rehabilitation is moving closer to reality, but not without a need for careful consideration. New guidelines drafted by researchers at Nelson Mandela University are aiming to address the practical and legal hurdles to implementing socially assistive robots – machines designed to guide and motivate patients through physical therapy – in South Africa’s public healthcare system. The work, completed in December 2025, comes at a critical time, as the country grapples with shortages of qualified therapists and disparities in access to rehabilitation services.
For Xolile Zepe, the impetus behind this research is deeply personal. His own family’s experience navigating the challenges of recovery after his sister’s accident highlighted the significant gaps in support available, particularly within the public health sector. “If you have medical aid, then there’s a chance that you can walk even if you have a stroke,” Zepe explained. “But if you rely on the public health system it often becomes a challenge to recover.” This disparity fueled his determination to explore innovative solutions to expand access to quality rehabilitation care.
Zepe’s research revealed a striking lack of awareness and implementation of socially assistive robots in South African rehabilitation facilities. A survey of therapists across four provinces found that 80% were unfamiliar with the technology. This finding underscored the need for a comprehensive framework to guide the responsible adoption of these tools. The resulting guidelines, based on a design science research methodology, address everything from regulatory compliance and training to infrastructure readiness and cost-sharing models.
Addressing a Critical Need: Rehabilitation Access in South Africa
South Africa faces a well-documented shortage of physiotherapists and biokineticists, especially outside of major metropolitan areas. The country’s National Health Strategic Plan identifies human resource constraints as a major challenge to achieving universal health coverage. Lower limb disability, resulting from accidents, stroke, or complications from conditions like diabetes, often requires sustained, supervised exercise to restore function. Socially assistive robots offer a potential solution by providing consistent, data-driven support, freeing up therapists to focus on more complex cases and extending the reach of limited resources.
These robots aren’t the exoskeletons often depicted in science fiction. Instead, they act as interactive coaches, guiding patients through exercises, demonstrating proper form, recording performance data, and providing feedback. Crucially, therapists retain control, programming the exercises and supervising the robot’s interaction with patients. “The attraction of the robot assistant…is less about novelty than scalability,” Zepe argues. A robot can consistently deliver repetitive exercises without fatigue or bias, providing valuable data to clinicians.
Navigating Legal and Ethical Concerns
Despite the potential benefits, the introduction of AI-enabled healthcare technologies raises significant legal and ethical questions. Zepe’s research highlighted a strong preference among therapists for “supervised autonomy,” meaning they want to maintain control over the robot’s operation. This reflects broader concerns about liability, data protection, and professional accountability.
Existing South African legislation, including the National Health Act and the Protection of Personal Information Act (POPIA), were not designed with these technologies in mind. “If the robot makes a mistake or injures someone, who is responsible, is that going to be the therapist? Is it the manufacturer?” Zepe asks, pointing to the need for clear regulatory frameworks. The guidelines he developed aim to proactively address these concerns, providing a roadmap for responsible implementation.
From Backyard Rehabilitation to National Policy Implications
Zepe’s journey began with a simple act of support: walking with his sister in their backyard, helping her regain her mobility after her accident. That experience instilled in him a deep understanding of the power of sustained rehabilitation and the challenges faced by those without access to formal therapy. Today, his sister walks independently, a testament to the family’s perseverance and the potential for recovery even without specialized care.
Dr. Sue Petratos, Zepe’s supervisor and Senior Lecturer in the School of Information Technology at Nelson Mandela University, emphasized the novelty of the research. “We couldn’t discover any studies that actually focused specifically on lower limb therapy, and also specifically in South Africa,” she said. “We think we’ve uncovered something new here.”
Key Guidelines for Implementation
The nine evidence-based guidelines developed by Zepe cover a range of critical areas:
- Regulatory Compliance: Ensuring adherence to existing and emerging healthcare regulations.
- Training: Providing adequate training for therapists and technicians on the use and maintenance of the robots.
- Infrastructure Readiness: Assessing the necessary infrastructure, including space, power, and internet connectivity.
- Cost-Sharing Models: Exploring sustainable funding mechanisms for robot acquisition and maintenance.
- Human-Robot Interaction: Designing interfaces and protocols that promote safe and effective interaction between patients and robots.
- Data Security and Privacy: Protecting patient data in accordance with POPIA.
- Liability and Risk Management: Establishing clear lines of responsibility and procedures for addressing potential risks.
- Ethical Considerations: Addressing ethical concerns related to autonomy, bias, and patient consent.
- Awareness and Education: Raising awareness among healthcare professionals and the public about the benefits and limitations of socially assistive robots.
Zepe, now pursuing doctoral research, plans to further refine these guidelines and address the remaining regulatory gaps. The project positions Nelson Mandela University at the forefront of a crucial conversation: how to integrate intelligent systems into public healthcare in a way that enhances, rather than erodes, trust, accountability, and professional expertise.
The next step involves disseminating these guidelines to key stakeholders within the South African healthcare system and initiating pilot programs to evaluate their effectiveness in real-world settings. The potential to improve rehabilitation outcomes for countless individuals is significant, but requires a collaborative and cautious approach.
What are your thoughts on the role of robotics in healthcare? Share your comments below, and please share this article with your network.
Related reading
