is scattered in different places it will be overlooked or missed, and doctors will spend too much time looking at the chart rather than listening and developing relationships. Likewise, an interface cluttered with too many buttons or tabs will make it hard and frustrating to find things. Intuitive design is an art, and we need some artists in the field.

The summary page needs to be viewed at the same time the encounter is being built to minimize jumping back and forth between pages. A left-scrollable summary panel and a right encounter panel could easily accomplish this.

The summary page also needs to limit the presence of non-important clinical data such as ICD-9 codes and timestamps. Let them live in the background. They have no use for providers.

A quick view of a well-maintained problem list can allow providers to briefly review the issues without spending a lot of time looking through the chart (and ignoring the patient). Problem lists are the ultimate summary and chronology of a patient's health issues. EHRs currently are all over the place on problem-list design. Many only allow archaic ICD-9 terminology. Most don't allow clarifying information or addition of events that have occurred on a problem.

Electronic systems can be so powerful here, yet most fall short of being useful and don't even function as well as paper. Part of the problem is we are trying to duplicate paper workflows in an electronic environment. Working as we do in hospital notes, updating each problem with new information at each encounter (rather than hiding it in encounter notes) is a much better workflow and way to find data.

Another area that needs attention is the documentation process. First, let me say that I believe we are being held back from better approaches by having to write complete notes with each visit. This satisfies insurance companies but detracts from better attention to the problems, as I mentioned above. But since we have to write notes, the process needs some help.

For those who like to type in front of patients, this function should be preserved, but beware that it is hard for patients to feel listened to when you are typing in front of them. Additional ways of entering data should include drag-and-drop of SH, FH, common phrases, etc. The ability to touch each section of the encounter and dictate a sound clip for a transcriptionist or insert voice-to-text would also help slower typists. Finally, handwriting recognition, while not the fastest, can be used to pull in phrases or write notes to self that can be expounded on later. The patient, not the documentation, should be the focus of the visit.

Finally, templates need to be smart. Copying text from one note to another or from an all-text template propagates errors and often states things that either aren't true or were not done, or leaves out things that were done. A note should reflect what was done as closely as possible. Templates can have structured elements to serve as ticklers to provide consistency, but should also allow variation to be documented. This is a large subject, but there are many ways new technology could be implemented to improve templates.

I am a believer in the ability of EHRs to help us provide more consistent care. But the patient, not the data or the documentation, should be the focus. The EHR is a tool. The tool shouldn't prevent us from doing the job. More attention to details, usability, and use practices can help make the products excel.

Apply now for the 2015 InformationWeek Elite 100, which recognizes the most innovative users of technology to advance a company's business goals. Winners will be recognized at the InformationWeek Conference, April 27-28, 2015, at the Mandalay Bay in Las Vegas. Application period ends Jan. 9, 2015.