Medical Policy

This document addresses the use of a robotic arm that is mounted to a wheelchair or located on another surface such as a table, and is intended to assist individuals with upper extremity disability and mobility limitations due to neurologic conditions, trauma, or other problems.

This document does not address the use of devices worn by the individual (see OR-PR.00005 Upper Extremity Myoelectric Orthoses).

Note: Please see the following related documents for additional information:

• CG-DME-10 Durable Medical Equipment
• CG-DME-34 Wheeled Mobility Devices: Wheelchair Accessories
• OR-PR.00005 Upper Extremity Myoelectric Orthoses

Note: For a high-level overview of this document, please see “Summary for Members and Families” below. 

Investigational and Not Medically Necessary:

The use of a robotic feeding assistive device or wheelchair mounted robotic arm is considered investigational and not medically necessary for all indications.

This summary describes clinical studies and expert recommendations and explains why robotic arm assistive devices are not clinically appropriate. The following summary does not replace the medical necessity criteria or other information in this document. The summary may not contain all relevant criteria or information. This summary is not medical advice. Please check with your healthcare provider for any advice about your health.

Key Information

Robotic arm assistive devices include robotic feeding devices and wheelchair-mounted robotic arms that are proposed to help people with limited arm and hand function perform daily tasks such as eating, picking up objects, opening containers, or controlling parts of their environment. They may be used by people with spinal cord injury, amyotrophic lateral sclerosis (ALS), muscular dystrophy, cerebral palsy, stroke, or other conditions that affect movement.

These types of devices may be controlled in different ways, including by use of joysticks, switches, voice commands, eye-gaze systems, or sip-and-puff controls. Some devices are attached to a wheelchair, while others are mounted on a table or other stable surface. Their goal is to help people with severe upper-body weakness perform tasks that might otherwise require assistance from another person.

What the Studies Show

Several small studies found that some people were able to complete tasks using these devices and reported positive experiences. Some studies showed faster task completion or improved ease of use in laboratory settings. However, most studies involved very small numbers of participants, lacked comparison groups, or took place in controlled environments rather than everyday settings. Some devices also required significant setup and support from caregivers. Current research has not shown that these devices improve long-term health, increase independent function in daily life, or reduce caregiver burden. Better studies are needed to know if these devices improve health.

Is this clinically appropriate?

Robotic feeding assistive devices and wheelchair-mounted robotic arms are not clinically appropriate because they have not been proven to improve health.

(Return to Description/Scope)

Summary

A robotic arm device is an electromechanical assistive technology engineered to replicate the movement capabilities of a human upper extremity. It is typically mounted on a fixed surface, such as a power wheelchair or stationary base, and controlled through user-operated interfaces (for example, joystick, switch, sip-and-puff, eye gaze, or voice control). These devices are intended to assist individuals with upper extremity impairments in performing daily tasks. Robotic arm devices are designed to facilitate independent performance of tasks such as feeding (for example, utensil use, lifting food to mouth, object manipulation (for example, grasping household items, opening containers), or environmental control (for example, light switch operation, door handling). These devices are primarily used by individuals with upper limb motor impairments due to spinal cord injury, neuromuscular diseases (for example, ALS, muscular dystrophy), or other neurologic or musculoskeletal conditions.

Current evidence addressing the clinical utility of wheelchair mounted robotic arms (WMRAs) is insufficient to permit reasonable conclusions concerning the clinical utility of robotic arm assistive devices for individuals with tetraplegia, including independent feeding devices and WMRAs. To date, available published evidence does not demonstrate that the independent eating assistive device or WMRA use leads to improvement in net health outcome, facilitates independent function related to activities of daily living (ADLs) or overall caregiver burden for individuals with tetraplegia. Further studies are needed to establish their effectiveness in improving ADLs, reducing caregiver burden, or enhancing overall health outcomes.

Discussion

Tetraplegia and quadriplegia are equivalent terms referring to weakness or paralysis of all four extremities. It can be caused by trauma, stroke, cerebral palsy, or other conditions affecting the nervous system. People with tetraplegia face significant challenges in all ADLs. Rehabilitation engineers have developed a variety of devices to assist these individuals. The robotic feeding assistive device, also known as an independent eating assistive device, has been proposed to assist with feeding independence for individuals with upper extremity limitations. Makers of a (WMRA propose that this device can help people with upper-extremity weakness to do such things as picking up objects and opening doors.

The Obi^® (DESIN LLC, Jacksonville, FL) is an independent eating assistive device intended for individuals with disability or disease affecting upper extremity strength and mobility limitations. The Obi is intended to increase independence and confidence by providing choice among four compartments of food and delivering the food to a region in front of the user’s mouth. The device requires a caregiver to prepare food, position the user and device for optimal use, determine the appropriate customizable accessibility switches, power on the device, teach the appropriate food delivery location in front of the user’s mouth, monitor use, and clean the device. To date, there is no published literature addressing the clinical utility of the independent eating assistive device. The evidence addressing WMRA is limited to outcomes reported in 1 retrospective uncontrolled study involving 31 participants (Maheu, 2011), 1 case series of 7 participants reported by Beaudoin (2019), and case reports. While several participants in these studies were able to use the WMRA to perform some tasks, extensive set-up support continued to be needed. There is insufficient evidence to evaluate long-term durability, tolerability, or to show improvements in net health outcomes. Further research is needed to see if the long-term use of the device could reduce caregiver assistance and increase user autonomy.

Maheu (2011) reported findings from a retrospective uncontrolled study of 31 users aged 18 to 64 years of age. The study evaluated the ability of users with upper extremity disabilities to manipulate the robotic arm in a controlled setting with the KINOVA JACO^® Assistive robot (Kinova Inc., Boisbriand, Quebec, Canada). There were 2 participants who were unable to complete basic tasks due to technical issues. The authors reported that 79% of participants were able to accomplish JACO’s 16 movements (all possible actions of the robotic arm) twice (test #1), and 93% of participants accomplished test #2 JACO involving 6 tasks. Participants were then asked to complete a questionnaire regarding caregiver support during ADL, their perception of ability to complete ADL tasks with the JACO arm system, current use of assistive devices to accomplish tasks, and sociodemographic profile. Although the authors estimated that the use of the JACO arm system could potentially reduce caregiver time by 41%, this was based on self-reported estimates by the study participants. The study did not objectively or directly observe reductions in caregiver time.

Downey (2016) evaluated a shared-control system that combined a brain-machine interface (BMI) with vision-guided robotic assistance in 2 individuals with tetraplegia. The system blended user-generated BMI commands with automated assistance for object recognition, hand positioning, and grasp stabilization during reaching and grabbing tasks. Compared with BMI control alone, shared control improved task performance, including higher success rates on object transfer tasks, shorter completion times, more efficient and stable movement trajectories, and lower perceived task difficulty. In a multiple-object selection task, 1 participant successfully grasped the correct object in 92% of trials with shared control compared with 46% without assistance. The authors concluded that integrating autonomous robotic assistance with BMI control may improve functional performance while preserving user intent and high-level control. Several limitations affect interpretation of the findings. The study included only 2 participants with different neurologic conditions, implant locations, and neural signal characteristics, limiting generalizability. Testing was conducted in a highly controlled laboratory environment using predefined objects and structured tasks rather than real-world home or community settings. In addition, the system relied on surgically implanted intracortical electrode arrays and specialized research equipment. The study evaluated the feasibility of a shared-control neuroprosthetic platform rather than the clinical effectiveness of a commercially available wheelchair-mounted assistive robotic arm. The study also did not evaluate long-term outcomes, ADLs, quality of life, caregiver burden, or real-world functional independence. Further development and evaluation are needed to establish generalizability, portability, long-term usability, and effectiveness in everyday environments.

In 2019, Beaudoin reported findings from a case series of 7 JACO robotic arm users 14 years of age or older who used the device for at least 6 months. The study also reported results for 5 main caregivers for these users. User performance was evaluated with a measurement developed for this study based on an upper extremity performance test (Test d'Evaluation des Membres Supérieurs de Personnes Agées [TEMPA]). Three tasks taken from the TEMPA included picking up and moving a jar, handling coins, and picking up and moving small objects. The authors reported JACO’s impact for users and their family caregivers after 6 months or more:

Participants reported positive impacts from using JACO, even though some difficulties were encountered….Users’ increased participation in their life habits may decrease the amount of caregiver assistance required, if only slightly. Users reported being generally satisfied with their device and that using JACO has positive psychosocial impacts.

Although the JACO is commercially available, its potential impact in individuals with upper-extremity disabilities is still poorly understood (Beaudoin, 2018). An interventional clinical trial that studied the benefits of the JACO2 mechanical arm in 20 individuals with Muscular Dystrophy over the age of 10 years was completed in Italy in 2019, (NCT04313049); no results have been published at this time.

Lebrasseur (2021) reported a prospective experimental study that evaluated the impact of 3 intelligent control algorithms (predefined position, fluidity filter, and drinking mode) integrated into the JACO in 14 adults with upper-body disabilities who used powered wheelchairs. Participants completed a series of simulated ADLs, including pressing a button, eating from a plate, grasping objects, and drinking from a glass. The predefined position and drinking mode algorithms improved user performance on the specific tasks for which they were designed, resulting in reductions in task completion times of up to 72% for repetitive positioning tasks, improved ease-of-use ratings, and higher usability scores across measures of perceived usefulness, ease of use, emotions, and attitude. Improvements associated with the drinking mode algorithm were primarily reflected in ease-of-use and usability assessments rather than objective efficiency measures. The fluidity filter algorithm did not significantly improve task completion times or ease of use, although participants generally rated it more favorably than a traditional filtering approach. No meaningful differences were observed between proportional and non-proportional control modes. The study was limited by its small sample size (n=14), short-term laboratory-based testing, heterogeneous participant population, short-term laboratory-based testing, brief device familiarization period, and reliance on subjective usability measures. Additionally, few participants had significant tremors or spasticity, limiting evaluation of the fluidity filter algorithm in its intended target population. The findings suggest that task-specific automation may improve efficiency and user satisfaction in a controlled setting but do not establish long-term clinical effectiveness, real-world functional outcomes, or sustained benefits in home and community environments.

Ding (2022) evaluated a vision-guided shared-control (VGS) system designed to help people with upper-limb disabilities use an assistive robotic arm for a multi-step drinking task. In this system, the user controlled the robot's gross movements, while the robot automatically performed more precise actions such as grasping objects and aligning them for use. The drinking task included opening a cabinet, retrieving a cup, filling the cup, bringing it to a drinking position, and returning it to the table. Bench testing of the robot's autonomous functions showed success rates ranging from 80% to 100% depending on the task. The highest success rate was for placing the cup on the table (100%), while opening the cabinet from arbitrary starting positions had the lowest success rate (80%). Common causes of failure included loss of grip, collisions with surrounding objects, path-planning failures, and errors in fiducial-marker localization. The system was subsequently evaluated by 3 research team members without disabilities and 1 person with a cervical spinal cord injury. Compared with standard joystick teleoperation, VGS generally improved performance, with overall task success rates increasing from 60%-80% to 80%-100% among the test users, task completion times decreasing by approximately 14%-51%, and substantially fewer joystick mode switches required. The participant with a spinal cord injury also completed tasks more quickly and used fewer joystick mode switches when using VGS. In addition, this participant reported lower mental demand, effort, frustration, and perceived workload with the VGS system. However, usability ratings were similar between VGS and teleoperation. The authors concluded that VGS may improve the efficiency of assistive robotic arm operation and reduce user burden during complex tasks. Interpretation of these findings is limited by the proof-of-concept design, very small sample size, inclusion of only 1 individual with a disability, lack of statistical analysis, use of a simulated cup-filling task, testing in a controlled laboratory environment with the robot mounted on a table instead of a wheelchair, evaluation of only 1 task, and technical issues such as navigation and software errors. Larger studies in more realistic settings are needed to confirm effectiveness and usability.

Another interventional clinical trial (NCT04323449) compared two control methods (new vision-guided control vs. the default control) for WMRA in 12 individuals with spinal cord injury/disorder was completed in June 2024. The National Library of Medicine (NLM) Clinical Trials website notes that trial updates were posted in June 2025, however, as of this writing, the study was not found on the NLM website and no results have been published.

The Rehabilitation Engineering and Assistive Technology Society of North America (RESNA) has published several position papers concerning manual wheelchairs, power wheelchairs, wheelchair components and accessories, and other assistive technologies. RESNA’s key position papers do not include recommendations for accessories such as a WMRA.

The U.S Department of Veterans Affairs (VA) Rehabilitation and Prosthetic Services clinical practice recommendations for motorized wheeled mobility devices do not provide guidance on use of WMRAs.

According to the Centers for Disease Control and Prevention (2020) and the World Health Organization (WHO) (2020) there are three dimensions of disability: impairment, activity limitations, and participation restrictions.

According to the U.S. Census Bureau's American Community Survey (ACS), about 45.8 million people in the United States had a disability in 2024, which is about 13.7% of the population, or roughly 1 in 7 Americans. It is estimated that about 5.5 to 6 million U.S. adults use a wheelchair, which is approximately 2.3% of adults (Census, 2025).

A wheelchair mounted robotic arm (WMRA), such as the JACO Assistive Robot, is intended to allow individuals with loss of upper limb functions to improve independence as well as quality of life. The JACO device is mounted on a motorized wheelchair. The user can control the arm using the chair’s joystick, head control, sip-and-puff, or head array system. The robotic arm features 6-axis movement corresponding to shoulder, elbow, and wrist, 16 movements in all, to mimic a fully functioning human hand.

The JACO robotic arm is a six-degree-of-freedom robotic manipulator designed to help in the performance of tasks that the user is unable to do independently. The device is an FDA-listed Class I medical device regulated under 21 CFR §890.3420 (External Limb Prosthetic Component). The JACO robotic arm is exempt from FDA premarket notification 510k requirements.

Activities of daily living (ADLs): Basic self-care activities necessary for personal independence, including eating, bathing, dressing, grooming, toileting, and transferring (for example, moving between a bed and a chair).

Assistive Technology: Any item, equipment, system, or product used to increase, maintain, or improve the functional capabilities and independence of individuals with disabilities.

Caregiver Burden: The physical, emotional, social, or time-related demands experienced by individuals who provide assistance to another person.

Disability: Defined by the World Health Organization (WHO, 2025) as having the following 3 dimensions:

1. Impairment in a person's body structure or function, or mental functioning
2. Activity limitation, such as difficulty seeing, hearing, walking, or problem solving.
3. Participation restrictions in normal daily activities, such as working, engaging in social and recreational activities, and obtaining health care and preventive services.

Functional mobility: The ability to move safely and effectively within home, community, and other environments in order to perform daily activities, with or without the use of assistive devices.

Independent Eating Assistive Device: A powered assistive device designed to help individuals with limited upper-extremity function feed themselves by selecting, transporting, and presenting food to the mouth.

Quality of Life: An individual's perception of physical, psychological, and social well-being and ability to participate in desired activities.

Shared Control: A method of operating a robotic device in which control is divided between the user and the robotic system, allowing the user to direct overall actions while the device performs certain automated functions such as object recognition, positioning, or grasping.

Tetraplegia (Quadriplegia): Partial or complete paralysis or weakness affecting all four extremities and the trunk resulting from injury or disease involving the cervical spinal cord or other parts of the nervous system.

The following codes for treatments and procedures applicable to this document are included below for informational purposes. Inclusion or exclusion of a procedure, diagnosis or device code(s) does not constitute or imply member coverage or provider reimbursement policy. Please refer to the member's contract benefits in effect at the time of service to determine coverage or non-coverage of these services as it applies to an individual member.

When services are Investigational and Not Medically Necessary:
For the following procedure code, or when the code describes a procedure indicated in the Position Statement section as investigational and not medically necessary.

Peer Reviewed Publications:

1. Beaudoin M, Lettre J, Routhier F, et al. Impacts of robotic are use on individuals with upper extremity disabilities: a scoping review. Can J Occup Ther. 2018; 85(5)397-407.
2. Beaudoin M, Lettre J, Routhier F, et al. Long-term use of the JACO robotic arm: a case series. Disabil Rehabil Assist Technol. 2019; 14(3):267-275.
3. Ding D, Styler B, Chung CS, et al. Development of a vision-guided shared-control system for assistive robotic manipulators. Sensors (Basel). 2022; 22(12):4351.
4. Downey JE, Weiss JM, Muelling K, et al. Blending of brain-machine interface and vision-guided autonomous robotics improves neuroprosthetic arm performance during grasping. J Neuroeng Rehabil. 2016; 13:28.
5. Lebrasseur A, Lettre J, Routhier F, et al. Assistive robotic arm: evaluation of the performance of intelligent algorithms. Assist Technol. 2021; 33(2):95-104.
6. Maheu VS, Archambault P, Frappier J, et al. Evaluation of the JACO robotic arm: clinico-economic study for powered wheelchair users with upper-extremity disabilities. IEEE Int Conf Rehabil Robot. 2011; 2011:5975397.

Government Agency, Medical Society, and Other Authoritative Publications:

1. Centers for Disease Control and Prevention. Disability and health overview. April 2, 2025. Available at:   https://www.cdc.gov/disability-and-health/about/?CDC_AAref_Val=https://www.cdc.gov/ncbddd/disabilityandhealth/disability.html. Accessed on August 3, 2026.
2. Centers for Medicare & Medicaid Services. National Coverage Decision (NCD) for Mobility Assistive Equipment (MAE) NCD# 280.3. Effective May 5, 2005. Available at: https://www.cms.gov/medicare-coverage-database/search.aspx. Accessed on August 3, 2026.
3. CGS Administrators, LLC. Jurisdiction J-C. Local Coverage Determination for Wheelchair Seating (L33312). Revised April 1, 2025. Available at: https://www.cms.gov/medicare-coverage-database/search.aspx. Accessed on August 3, 2026.
4. National Census Bureau. Facts for Features: Anniversary of Americans with Disabilities Act: July 26, 2026. May 19, 2026. Available at: https://www.census.gov/newsroom/facts-for-features.html. Accessed on August 3, 2026.
5. National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR). Last modified May 5, 2025. Available at: https://www.acl.gov/about-acl/about-national-institute-disability-independent-living-and-rehabilitation-research.Accessed on August 3, 2026.
6. Rehabilitation Engineering and Assistive Technology Society of North America. Benefits of JACO Robotic Arm on independent living and social participation: an exploratory study. 2014. Available at: https://www.resna.org/sites/default/files/conference/2014/PDF%20Versions/Robotics/Routhier.pdf. Accessed on August 3, 2026.
7. Rehabilitation Engineering and Assistive Technology Society of North America. RESNA position on the application of power mobility devices for pediatric users- update 2017. Available at: https://www.resna.org/Portals/0/Documents/Position%20Papers/RESNA%20Ped%20Power%20Paper%2010_25_17%20-BOD%20approval%20Nov2_2017.pdf. Accessed on August 3, 2026.
8. Veterans Administration. Department of Defense. Clinical practice guideline for the management of upper limb amputation rehabilitation. March 2022. Available at: https://www.healthquality.va.gov/guidelines/Rehab/ULA/VADoDULACPG_Final_508.pdf. Accessed on August 3, 2026.

Independent Eating Assistive Device
KINOVA JACO Assistive robot
Obi
Robotic Feeding Assistive Device
Wheelchair Mounted Robotic Arm (WMRA)

The use of specific product names is illustrative only. It is not intended to be a recommendation of one product over another, and is not intended to represent a complete listing of all products available.

Federal and State law, as well as contract language, including definitions and specific contract provisions/exclusions, take precedence over Medical Policy and must be considered first in determining eligibility for coverage. The member’s contract benefits in effect on the date that services are rendered must be used. Medical Policy, which addresses medical efficacy, should be considered before utilizing medical opinion in adjudication. Medical technology is constantly evolving, and we reserve the right to review and update Medical Policy periodically.

No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, or otherwise, without permission from the health plan.

© CPT Only – American Medical Association