Technical Report 41: Implementation Costs of the Cooperative Pain Education and Self-Management (COPES) Intervention to Deliver In-Home Self-Management Support to Veterans with Chronic Pain

Suggested Citation

So AY, LaChappelle KM, Heapy AA, Wagner TH. Implementation Costs of the Cooperative Pain Education and Self-Management (COPES) Intervention to Deliver In-Home Self-Management Support to Veterans with Chronic Pain. Technical Report 41. Health Economics Resource Center, VA Palo Alto Health Care System, U.S. Department of Veterans Affairs. 

 

For a list of VA acronyms, please visit the VA acronym checker on the VA intranet at http://vaww.va.gov/Acronyms/fulllist.cfm.

1. Introduction

As the number of Veterans with chronic pain has increased, VA’s capacity to provide the recommended evidence-based pain self-management support services to all Veterans who could benefit has been strained, with access to these services becoming a significant challenge. In-person pain support is resource intensive, often inaccessible to Veterans, and delivered unevenly across patients and facilities. VA researchers found that cognitive behavioral therapy (CBT) delivered primarily via an automated Interactive Voice Response (IVR) system leads to patient-centered outcomes that are comparable to standard VA approaches in which CBT is delivered in-person by a therapist over ten weekly sessions.¹ This program, known as Cooperative Pain Education and Self-management (COPES), is advantageous in that it uses scalable, automated messaging to deliver in-home self-management support. Thus, COPES has the potential to reach Veterans with chronic pain who have limited access to specialty pain management services.

In this study, we estimated the total cost of the COPES intervention. We also estimated the cost of adding COPES to additional VA medical centers as well as the cost of replicating the intervention at a non-VA site.

return to top 

2. Methods

The COPES intervention requires an automated case-finding dashboard, a computerized display of eligible and enrolled patients, based on data from the VA’s electronic medical record. To compute the costs of COPES, we used micro-costing methods to estimate labor, supplies, office space, and contracts.²

Labor Costs

To measure cost of dashboard development, we used activity logs to track staff time. We gathered the number of full-time equivalent (FTE) staff involved in dashboard development and the total number of hours spent on activities related to creating the dashboard. We separated staff into different job titles/categories: co-principal investigator, research assistant, and project coordinator.

To estimate VA staff costs, we linked the staff’s grade and step to the annual salary from the 2019 Federal General Schedule (GS) as reported by the Office of Personnel Management. We also used the geographic adjustment of wages to calculate the staff’s wages. To then determine raw hourly employment costs from VA wages, we divided the annual cost by 2088, the number of hours in a 52-week work year. Employers also pay taxes, contribute to insurance premiums, and make other benefit payments. Researchers at the Palo Alto VA medical center estimate that these additional costs equal to 30% of an employee's salary. Accordingly, we adjusted the gross hourly cost to include this 30% by multiplying the raw hourly cost by 1.3 to get an estimate of hourly employment costs.

The five clinics in the COPES program also had higher than average wages. We, therefore, estimated costs at the national average wage rate. For national average calculations, we obtained 2019 national annual median and hourly wages for equivalent roles from the Bureau of Labor Statistics (BLS). For estimates of national hourly averages, we multiplied national median hourly wages by 1.3 to include 30% fringe benefits.

To assign a cost to each staff involved in dashboard development, we multiplied their total number of hours spent on activities by their hourly wages. This new cost was totaled for all the staff involved to come up with a total cost for creating the dashboard.  

Supplies

Patients enrolled in COPES were sent a patient handbook and a pedometer. We estimated the costs of these based on their purchase price plus mailing.  

Office Space

We excluded the cost of office space for any of the COPES staff. We assumed that the staff had existing office space that they could use or that they could work remotely. These costs could be included, if so desired.

Contracts

The Interactive Voice Response system has a set-up cost and an annual maintenance cost. We included the annual maintenance costs. The set-up costs were excluded from some analyses and included in others; if VA wanted to extend COPES to another site, it would not need to include the set-up costs. However, if another health care system wanted to adopt COPES, then it should budget for the set-up costs.

Analysis

We computed three cost estimates: 1) the total cost of the COPES intervention, 2) the cost of adding COPES to additional VA medical centers, and 3) the cost of replicating the intervention at a non-VA site. Computing these costs required that we identify whether the cost component was variable, fixed, or sunk.

Variable Costs

Variable costs are costs that vary with the scale of production (i.e., the number of patients). COPES incurred variable costs, including the therapist time to review the patient data, IVR maintenance, dashboard maintenance, and supplies (e.g., patient handbooks, pedometers, printed materials). These costs varied by the number of participants.

After the dashboard development process was completed, the dashboard still required some routine maintenance and troubleshooting with site coordinators to maintain functionality. To measure the cost of dashboard maintenance, we used activity logs to track staff time. We gathered the number of full-time equivalent (FTE) staff involved in dashboard maintenance and the total number of hours spent on activities related to maintaining the dashboard. We separated staff into different job titles/categories: co-principal investigator, research assistant, and project coordinator. To assign a cost to each staff involved in dashboard maintenance, we multiplied their total number of hours spent on activities by their hourly wages. This new cost was totaled for all the staff involved to come up with a total cost for maintaining the dashboard.  

The cost estimates for the CBT-CP therapist were adjusted to 20% effort to reflect the actual percentage of time they spent working on this project.

The costs of supplies and equipment were gathered through project manager surveys. Two factors should be kept in mind when obtaining supply costs: supply and equipment costs may fall if a new intervention is widely adopted, and the list price of a good may greatly overstate the cost of supplies and equipment because large providers like VA frequently negotiate substantial discounts.

Fixed Costs

The cost of employing an IVR coordinator and scheduler, and the cost of training staff on the dashboard were considered fixed because they are constant across the scale of production. These employees were situated at one VA medical center and provided services across five clinics. These costs would not increase if more VA medical centers or participants were added. To measure the cost of dashboard training, we used activity logs to track staff time. We gathered the number of full-time equivalent (FTE) staff involved in dashboard training and the total number of hours spent on activities related to training staff on the dashboard. We separated staff into different job titles/categories: co-principal investigator, research assistant, and project coordinator. To assign a cost to each staff involved in dashboard training, we multiplied their total number of hours spent on activities by their hourly wages. This new cost was totaled for all the staff involved to come up with a total cost for dashboard training.  

The cost estimates for the scheduler were adjusted to 20% effort to reflect the actual percentage of time they spent working on this project.

Sunk Costs

The one-time cost of setting up the IVR system and the cost of creating a dashboard for the program are considered “sunk” and would not be incurred if the program was replicated at other VA medical centers. These costs are excluded from the program cost estimates. However, if another health care organization wanted to implement this program, then it should budget for one-time set-up costs, if appropriate.

return to top 

3. Results

Program Costs

The total cost of the COPES intervention, as implemented, was $309,155 over three years (see Table 2). At each site, the added cost of implementing COPES was $240,877. This includes therapist time to review the patient data and provide weekly asynchronous feedback messages ($52,588), IVR maintenance ($40,000), patient handbooks ($3,729), pedometers ($2,558), and printed materials ($158). One of the program components included building a dashboard, which cost $10,128 to maintain. The COPES intervention lasted 3 years, and the annual cost of running the program was $180,132 (Table 2). Included in the program costs was $130,000 for employing an IVR coordinator and scheduler, and $1,715 for dashboard training, which were fixed. These employees were situated at one VA medical center and provided services across the five clinics.

Excluded from this cost estimate are the one-time cost of setting up the IVR system ($30,000) and creating the dashboard ($38,279). We excluded these costs because they were “sunk” and would not be incurred if the program was replicated at other VA sites. If another health care organization wanted to implement this program, then it should budget for one-time set up costs, if appropriate. We also excluded any costs related to office space for the staff.

Person-level Costs and Replication Costs

COPES enrolled 102 persons in the program, which lasted for three years. The average cost of COPES per person per year was $1,766. Some of the components of COPES varied in proportion to the number of patients. Other costs were fixed and did not vary with the number of patients involved in the program. We estimate that 59.5% of the costs varied in proportion to the number of patients involved, while the remainder was fixed. 

The expected total three-year cost of the intervention would be $296,454 at the national average, or $2,906 per person. If VA added COPES to more VA medical centers, the added cost would be $1,034 per person per year. To replicate the intervention at a non-VA site, the cost would be $1,705 per person per year, because of the added cost of having to create the dashboard with the non-VA site.

return to top 

4. Discussion

Health care providers are eager to find ways to help patients who are suffering from chronic pain. Some alternatives are labor intensive and expensive. Other programs, such as COPES, rely more on automation and virtual care, thereby making care more accessible. We estimated the costs of implementing COPES, the costs to add COPES to an additional VA medical center, and the costs of replicating the intervention at a non-VA site. Future work will evaluate COPES to place the costs in context relative to a control group.

return to top 

5. References

1. Heapy AA, Higgins DM, Goulet JL, LaChappelle KM, Driscoll MA, Czlapinski RA, Buta E, Piette JD, Krein SL, Kerns RD. Interactive Voice Response-Based Self-management for Chronic Back Pain: The COPES Noninferiority Randomized Trial. JAMA Intern Med. 2017 Jun 1;177(6):765-773

2. Wagner TH, Yoon J, Jacobs JC, et al. Estimating Costs of an Implementation Intervention. Medical Decision Making. 2020;40(8):959-967.