Care Optimizer: How Capgemini Delivered a Web and Windows 8 App with Mendix
on February 19, 2014
Today, we have a guest post from Christiaan Tick, one of the Lead Mendix Consultants in our partner Capgemini’s fast-growing Productivity Platforms Centre of Excellence. Christiaan is a Change as a Strategy evangelist and works intimately together with Capgemini’s clients to help them transform. In this post, he shares his experience using the Mendix App Platform to quickly take Capgemini’s Care Optimizer application from demo to production, including multi-language support and a native Windows 8 app. It’s great to see Mendix not only improving the speed and ease of application delivery, but also supporting a project that’s making a difference by improving the quality of healthcare.
It’s not often that you get the opportunity to help make the world a better place. My grandmother had Alzheimer’s disease and we saw her decline, needing more care at each step. Eventually she went from home care to a nursing home where she passed away at the beautiful age of 91. Although all of the healthcare organizations involved tried their best to give her the most optimal care possible, we never knew how much they really cared. I mean, she looked okay, but was it enough? Was it the right type of care? Could we take over some of the care activities? We had a lot of questions and most of them were answered with ‘we think it’s enough care, although we’re not sure’.
In 2011 we met Kajsa Thorsell, a former care nurse wanting to make a difference. She studied the delivery of care in Swedish healthcare organizations for 10 years, which resulted in a method to optimize the delivery of care: Care Optimizer. The basic concepts behind the model are the Time in Care – need (TiC-n) calculations and Time in Care – time (TiC-t) calculations. TiC-n ‘calculates’ the care needs (care activities) of a patient, while TiC-t calculates the time that is spent on care activities by nurses by measuring the start and stop time of each activity. Analyzing and comparing these numbers delivers insight into how care is delivered in the real world and how it can be optimized.
In 2011, Kajsa met colleagues of mine from Capgemini Sweden’s Healthcare sector and they instantly saw the value of the Care Optimizer method, combined with a cloud-based web environment. So, the method was there, the drive was there, but nobody had any idea what the application should look like. Fortunately, there was a buzz that Capgemini The Netherlands had a technology that was perfect for quickly creating demos and sophisticated applications: The Mendix App Platform!
From Demo to Production with Mendix
With Mendix, we could easily mature the Care Optimizer application from demo to our first production version. In the first stage, we created a demo in the Netherlands with the product owner in Sweden, just to show the platform’s capabilities and the responses we got from Sweden were very encouraging. In Sweden, several nursing homes were found who were happy to join a pilot, so we made the application production ready. Also, we added HTML5 mobile forms to support the measuring of activities using tablets, since nurses are always on the road and a lot of caretakers are adopting a tablet strategy. The only issue was that we didn’t speak a word of Swedish.
Fortunately, using the built-in multi-language support of the Mendix App Platform, we could develop the application in any language (we used English and Dutch). So, extending to Swedish was a matter of sending some Excel files and reading back in the translation. In the pilot, the nurses could use Care Optimizer in their own language, so we minimized the chance that wrong care activities were measured. The results were insightful, since it was shown that about 20% of all time was spent on direct care activities (compared to indirect care activities and organization-related activities). Furthermore, previously used intake models incorrectly categorized patients. Application-wise, it was up to the nurses to use the mobile site so we decided that native apps were necessary to support the nurses in the most efficient way possible.
Developing a Native Windows 8 App
Because Windows adoption in our launching region (Sweden) was so high, we chose to develop the first native app on the Windows 8 platform. It seemed logical since the initial customers all were planning on investing in Windows 8 tablets. We also needed an environment where we had more control over scaling up and down resources, so we installed it all in a Microsoft Azure cloud environment. It took us about a day. Installing Mendix in an Azure cloud environment using Virtual Machines that is. The development of the Windows 8 app took a bit longer. We had workshops with end-users in Sweden, and besides fun, we had a lot of discussions on how to make the interface as non-interruptive as possible. Combined with input from our experienced user interface designers, we created a seemingly simple user interface that you can clearly see from a distance and use very intuitively. We adopted the O-AUTH authentication to minimize the burden of having to log-in all the time. Furthermore, we kept the native app as lightweight as possible. Hardly any information is stored locally; everything is retrieved from the Mendix Business Server using REST web services.
So combining the ease of use and flexibility of the Mendix App Platform with a flexible cloud platform (Azure), we quickly built the Care Optimizer app perfectly in harmony with the needs and workflow of the end-users. Now if healthcare organizations find themselves answering ‘we think it’s enough care although we’re not sure ‘ we can let them experience Care Optimizer, saying ‘see how much you really care’!
For more info, see http://www.slideshare.net/gwikstrand/care-optimizer-transforms or contact me at firstname.lastname@example.org.