Whenever you visit a medical practice, depending on the question he or she needs answered, there's a good chance the sample is sent to a commercial laboratory that actually runs the diagnostic. The lab reports the results back to the doctor and bills the patient's insurance separately for the work. An entirely new dimension to this has emerged in recent years with the growth in labs that specialize in testing on the genetic level. Software for labs have existed for a long time, but at Ovation we set out to develop a next-generation version. I spent a long time at labs, with everyone from the CEO to bench technicians. I'll admit, it was hard wading through so much scientific expertise. But the key came one day when I saw technicians doing something very human and low-tech: recording the position and progress of samples on a simple clipboard. By capturing the whole process according to a batch paradigm much like in a manufacturing setting, we were able to completely reinvent laboratory information management software.
Even if I feel there are bigger questions at play, I generally like to start where the pain feels the greatest and the solution's success is more easily measurable. The requisition process, in which samples are received, information about the patient and their insurance is collection, and samples are stored, is a key moment in the process. It not only takes up a lot of time if not done correctly, but is required in order to calculate the lab's turn-around time on processing a sample, one of the lab's key metrics. I put my own spin on the classic UX problem of forms by adding clear completion feedback in the form of messaging and disabling and enabling buttons.
Labs had always recorded the results of relevant activities....that's what they're there to do. But activities were very much treated in isolation of one another. Inputs were were recorded separately and therefore redundantly, and recording output data was a very manual process. By associating a batch of samples to an activity, we were able not only to capture activity data, but tie it to the activity data that came before and record data on the more granular, sample level without any extra work.
No one had ever provided a visualization of an entire workflow from end-to-end before, let alone allowed labs to author their own workflows based on activities chained together. Labs could move a whole batch through or, ands this was very common, split batches or combine them depending on the equipment in use. Sample data was always recorded, and very importantly the insurance billing process was made far easier than anyone had ever accomplished before.