Most office software today is locally hosted – the program runs on an in-house computer (desk tops, notebooks or laptops) and stores the data either on a hard drive or on a network server in the premises. This offers the greatest control, since the data and the program reside on in-house servers. Further, there is the bonus that because the data is local, overall application speed tends to be higher, especially when dealing with large amounts of data.
Client-server proponents point out that EMR applications are generally quite stable and require little maintenance. There is a capacity to create scheduled backups that copy data to off-site computers fairly easily.
It does however require the practice or clinic to be responsible for upgrades, database maintenance, security, backups, and generally making sure the system works as it’s supposed to, all the time. If there is no internal IT staff and data centre, it is quite likely that a practice manager or doctor will find themselves rolling up their sleeves pretending top be IT specialists. Alternatively they will be spending quite a bit on consultants and hardware.
If the client-server route is selected, then it is more likely to pay a lump sum to buy the software outright and then a smaller monthly or annual fee for support. If the practice or clinic has multiple users, there will be need for a high-capacity server to run the application, which can mean a hardware bill of several thousand dollars, or more.