According to this article, Health IT projects require hefty investments. I'm going to have to disagree with Dr. Caitlin Cusack. Maybe current Health IT projects do, but that doesn't mean they have to. Your two cost factors are the actual hardware / software you use and the cost of training your people.
I believe that the customers should invest nothing for the software. Just like you can get "pay as you go" mobile phones, Health IT should be delivered as "pay as you go". How?
Build it in the cloud and scale horizontally. It's not hard or expensive to build a system for a 3,000 patient clinic. But instead of designing a massive silo (aks Salesforce.com), for 10,000 customers, deploy 10,000 cloud servers. And then just charge when the customer actually uses the thing. Which will keep you on your toes at all times.
As for hardware, it doesn't take much. A great netbook or desktop costs $300-$400 and you can set up secure wireless for $100 per building. So you can automate a 5 physician office with a total investment of $5-$10K. Because all the heavy lifting is done in the cloud (aka Amazon Web services, my favorite). And you can run a dedicated server 24/7 with a lot of storage for less than $1,000 a year with zero installation costs.
And deployment (where the real costs are) should be done in an iterative fashion. Prototype, pilot, repeat until satisfied. Then expand.
This is a different way of looking at things. When I was at Baxter Healthcare, conventional wisdom said all IT projects were heavily invested in, took many months (or over a year) to get implemented and then get everyone trained. Working in the divisions I was rolling out new applications in 6 weeks. And that was 25 years ago.
Prototype, pilot, repeat.