Archive for October, 2006
I have seen a lot of project heartache that could have been avoided with better testing. However, testing is an art form and very few organizations do it well. Most people believe the extent of testing is to make sure that the system behaves as expected. But that is the easy part.
There are many aspects of testing, including unit testing (just testing the particular aspect of functionality), integration testing (making sure all of the components work end-to-end), interface testing, conversion testing, stress testing (simulating real world volumes), and exception testing (because users don’t always do what you expect). This is far from complete. The type and amount of testing will vary by project.
Most project plans that I see need to triple the amount of testing that they plan. The development of a comprehensive test plan is a key component of any IT project.
Also, testing should be auditable and signed off by the business. When I worked for SMS in the 80s we created folders for screen prints. The outside of the folder had the scenario stapled and the expected results were checked off as they were validated. Today we typically document the testing in a QuickBase application.
I feel so strongly about testing we have created a new position in our organization for a testing guru. This person will be the consultant to the various projects helping the project teams to develop effective test plans. This person will also develop our internal methodologies which we will improve over time.
In general Microsoft makes some great software. I think the office suite may be the exception. I really find Word difficult to use and needlessly complex. It is a classic example of “bloatware.”
The other thing that drives me crazy about Microsoft is their pricing strategies. They typically price their products below market at the beginning of their life cycle. Then, once we are dependent on them, they raise their prices well above market. What really makes their software expensive is their client pricing. When we install a new server with Microsoft software we pay for the software on that server AND an additional fee for a all of the PCs that connect to that server.
At my organizations, we pay more for Microsoft software than GE/IDX and MEDITECH combined.
Well, we are going to start to bring some sanity to our Microsoft spending. Recently a Ministry/Affinity team completed a project to develop Ministry and Affinity’s strategy for managing Microsoft costs, and I think they have come up with an outstanding plan. Not all PCs will have Microsoft Office. In clinical areas we will only install the free viewers. So, people can read Word and Excel documents, but won’t have the expensive software needed to create those documents. This requires extra work on our behalf, but the money we will save justifies the added effort 10 times over.
Also, we have made a decision to NOT deploy Microsoft SharePoint servers. we are using an ASP collaboration tool called QuickBase, which is much more intuitive and will be significantly less expensive.
Recently Microsoft rolled out their Reporting Services tool. We will not use that product unless a full analysis of the options supports that direction.
In general, I only use Word or Excel in rare situations. I never type a message in Word and attach it to an email. That only perpetuates Word usage, and it wastes the user time sine they have to open the email, then open the Word attachment.
We will not be installing Microsoft Access on any PCs unless there is a specific justification accepted by an IT manager. Usually QuickBase is a much better choice than Access since it is natively multi-user, easier to use, and much less expensive.
There is a new web service called bidrx.com. They aspire to be a website that links consumers with pharmacies, manufacturers, prescribers and payers so all can make better decisions when purchasing prescription drugs. I am familiar with them because they are based in my part of Northeast Wisconsin.
I have long thought that we could really provide our patients with a service by providing them information regarding what local pharmacies are charging for the medications that we are about to prescribe for them. Because electronic prescribing systems fax/transmit the prescriptions directly to the pharmacy, we are forcing patients to choose a pharmacy without having the information they need to make a good buying decision.
Wouldn’t it be cool if we could imbed bidrx.com’s technology/information in an electronic prescribing system? So instead of choosing that pharmacy based upon convenience, the patient could also consider price (which could vary by $100 for something like a dose of lamisil).
That was the idea that I floated past an e-prescribing developer. But he pointed out that their SureScripts agreement from steering patients to a particular pharmacy. The road to remaking healthcare is filled with obstacles like this. The pharma industry is especially good at putting up roadblocks. Health plans don’t even know the unit cost of the medications that the pay for. They just get one big bill from the Pharmacy Benefits Manager (PBM). There is a lot of trust and little transparency.