Email-based telemedicine has been reported to be an efficient method of delivering online health services to patients at a distance and is often described as a low-cost form of telemedicine. The service may be low-cost if the healthcare organisation utilise their existing email infrastructure to provide their telemedicine service. Many healthcare organisations use commercial-off-the-shelf (COTS) email applications. COTS email applications are designed for peer-to-peer communication; hence, in situations where multiple clinicians need to be involved, COTS applications may be deficient in delivering telemedicine. Larger services often rely on different staff disciplines to run their service and telemedicine tools for supervisors, clinicians and administrative staff are not available in COTS applications. Hence, some organisations may choose to develop a purpose-written email application to support telemedicine. We have conducted a cost-minimisation analysis of two different service models for establishing and operating an email service. The first service model used a COTS email application and the second used a purpose-written telemedicine application.
The actual costs used in the analysis were from two organisations that originally ran their counselling service with a COTS email application and later implemented a purpose-written application. The purpose-written application automated a number of the tasks associated with running an email-based service. We calculated a threshold at which the higher initial costs for software development were offset by efficiency gains from automation. We also performed a sensitivity analysis to determine the effect of individual costs on the threshold.
The cost of providing an email service at 1000 consultations per annum was $19,930 using a COTS email application and $31,925 using a purpose-written application. At 10,000 consultations per annum the cost of providing the service using COTS email software was $293,341 compared to $272,749 for the purpose-written application. The threshold was calculated at a workload of 5216 consultations per annum. When more than 5216 email consultations per annum are undertaken, the purpose-written application was cheaper than the COTS service model. The sensitivity analysis showed the threshold was most sensitive to changes in administrative staff salaries.
In the context of telemedicine, we have compared two different service models for email-based communication – purpose-written and COTS applications. Under the circumstances described in the paper, when workload exceeded 5216 email consultations per annum, there were savings made when a purpose-written email application was used. This analysis provides a useful economic model for organisations contemplating the use of an email-based telemedicine system.