An Economic Analysis of Robot-Assisted Therapy for Long-Term Upper-Limb Impairment After Stroke
Background and Purpose—Stroke is a leading cause of disability. Rehabilitation robotics have been developed to aid in recovery after a stroke. This study determined the additional cost of robot-assisted therapy and tested its cost-effectiveness.
Methods—We estimated the intervention costs and tracked participants' healthcare costs. We collected quality of life using the Stroke Impact Scale and the Health Utilities Index. We analyzed the cost data at 36 weeks postrandomization using multivariate regression models controlling for site, presence of a prior stroke, and Veterans Affairs costs in the year before randomization.
Results—A total of 127 participants were randomized to usual care plus robot therapy (n=49), usual care plus intensive comparison therapy (n=50), or usual care alone (n=28). The average cost of delivering robot therapy and intensive comparison therapy was $5152 and $7382, respectively (P<0.001), and both were significantly more expensive than usual care alone (no additional intervention costs). At 36 weeks postrandomization, the total costs were comparable for the 3 groups ($17 831 for robot therapy, $19 746 for intensive comparison therapy, and $19 098 for usual care). Changes in quality of life were modest and not statistically different.
Conclusions—The added cost of delivering robot or intensive comparison therapy was recuperated by lower healthcare use costs compared with those in the usual care group. However, uncertainty remains about the cost-effectiveness of robotic-assisted rehabilitation compared with traditional rehabilitation.
Rehabilitation robotics have been developed to aid in rehabilitation, alter the physical burden on a therapist, and potentially improve a clinic's productivity. Such technologies have led to modest improvements in functioning among patients with chronic upper-extremity disability related to stroke.1,2 This study determined whether cost differences would favor robot-assisted treatment compared with more conventional therapies for patients with stroke with a moderate to severe upper-extremity impairment.
In a 3-arm randomized controlled trial (VA ROBOTICS), we enrolled patients whose stroke caused moderate to severe upper-extremity impairment and predated the study entry by >6 months.3 Patients were randomized to usual care plus robot-assisted therapy, usual care plus intensive comparison therapy, or usual care alone. Intensive comparison therapy mirrored the intensity, frequency, and type of movements of robot-assisted therapy.4 Both robot and intensive comparison therapy involved 3 1-hour sessions per week for 12 weeks. Details on the study design are presented elsewhere.5
We tracked participants' use of Veterans Affairs (VA) inpatient care, outpatient care, and costs in the Patient Treatment File and National Patient Care Database and the Decision Support System, respectively. Non-VA use and use of formal and informal caregiving were tracked with case report forms. We calculated the travel distance from the patient's home zip code to the medical center's zip code. Quality of life was assessed using the Health Utilities Index,6 a Feeling Thermometer, and the Stroke Impact Scale.7
We used multivariate regression models controlling for site, presence of a prior stroke, VA costs in the year before randomization, and Charlson Comorbidity Index in the year before randomization.8 We compared costs and quality-adjusted life-years, based on the Health Utilities Index, using an incremental cost-effectiveness ratio based on the 36-week data.
There were no significant differences among the 3 groups in baseline demographics or the use of the VA system in the year before enrollment, except time from the index stroke to randomization, which was significantly longer in the usual care group.2
The cost per session of robot and intensive comparison therapy was estimated at $140 and $218, respectively. The unadjusted average cost of the intervention over the 12-week treatment period was $5152 for the robot and $7382 for intensive comparison therapy (P<0.001; see http://stroke.ahajournals.org for the detailed results).
Costs and Outcomes at 36 Weeks
The analyses identified no significant differences among the 3 groups' VA use, non-VA use, or patient-incurred costs at 36 weeks. Total healthcare costs (excluding intervention costs) at the end of the 36 weeks averaged $12 679 for the robot group, $12 364 for the intensive comparison therapy group, and $19 098 for the usual care group. Although the average costs for the usual care group were higher than the other groups, the data were skewed, which is evident when we compare the median costs of the robot ($5541), intensive comparison therapy ($6912), and usual care group ($6799; see http://stroke.ahajournals.org for the detailed results).
Self-reported functioning and quality of life did not significantly differ among the 3 groups over time, except for the Stroke Impact Scale, in which the robot group had a significantly higher score than the usual care group at 12 weeks (P=0.04). This difference decreased over time and was not significant at 24 or 36 weeks.
VA Costs After 36 Weeks
After the trial ended at 36 weeks, we tracked use of VA health care and mortality through the end of Fiscal Year 2009 (September 30, 2009) for participants in the 2 active control groups (following usual care patients was not possible because at 36 weeks, they could elect robot or intensive comparison therapy). For the robot group, VA health care averaged $7777, whereas the intensive comparison therapy group averaged $14 513 (P=0.04). When healthcare and travel costs were combined, the difference was not statistically significant (P=0.06). The total cost from randomization through the end of Fiscal Year 2009 averaged $25 608 for the robot group and $34 259 for the intensive comparison therapy group (P=0.21; see http://stroke.ahajournals.org).
Incremental Cost-Effectiveness Ratio
At 36 weeks, the robot group had a lower average cost ($1267) and an increase in quality-adjusted life-years (0.049) relative to the usual care group. However, the large standard errors around effectiveness and costs yielded an incremental cost-effectiveness ratio with a wide bootstrapped confidence region (−$450 255, +$393 356).
Improvements as measured by the Stroke Impact Scale were significantly better at 12 weeks for the robot versus usual care, but these differences were not maintained over 24 and 36 weeks. At 36 weeks, the average costs were not significantly different among the 3 groups. After 36 weeks, the robot group used less VA care and had an average cost that was less than the intensive comparison therapy group.
One limitation of this trial is the sample size. Although larger than many other robot studies, it is small for analyzing cost data. Also, the predominantly male sample, although representative of the VA, may limit the generalizability of the results to women.
The use of robotic technology for stroke rehabilitation is a rapidly growing area. The VA ROBOTICS study demonstrated modest clinical benefit for robot-assisted therapy compared with usual care at 36 weeks. Although providing additional care using new technology can be expensive, the total costs, which include therapy and healthcare costs, were not greater for the robot group than the usual care group. Patients in the robot and intensive comparison groups had lower average costs than patients in usual care group. From a healthcare system perspective, this suggests an offset in the system, which will require studies with larger samples to fully understand.
Sources of Funding
This study was funded by the Department of Veterans Affairs Cooperative Studies Program and the Rehabilitation Research and Development Service.
H.I.K. is a co-inventor in the Massachusetts Institute of Technology (MIT)-held patents for the robotic device used in this work. He holds equity positions in Interactive Motion Technologies, the company that manufactures this technology under license to MIT.
The online-only Data Supplement is available at http://stroke.ahajournals.org/cgi/content/full/STROKEAHA.110.606442/DC1.
The views, opinions, and content of this publication are those of the authors and do not necessarily reflect the views, opinions, or policies of the Department of Veterans Affairs.
- Received October 25, 2010.
- Revision received March 7, 2011.
- Accepted March 29, 2011.
- © 2011 American Heart Association, Inc.
- Volpe BT,
- Lynch D,
- Rykman-Berland A,
- Ferraro M,
- Galgano M,
- Hogan N,
- et al.
- Lo AC,
- Guarino P,
- Krebs HI,
- Volpe BT,
- Bever CT,
- Duncan PW,
- et al.