OR WAIT null SECS
Digital dashboards will allow you to pilot your clinical trial from the captain's chair.
Does sitting on a plane get you thinking about life as a jet pilot? The hours I used to spend trying to keep my 747 from crashing into Lake Michigan in Microsoft's Flight Simulator had that effect on me. At the time, though, I was blind to parallels between flying jets and managing clinical trials.
On a transatlantic trip, a pilot and his or her co-pilot spend up to 12 hours in a small, essentially sealed room while a huge plane flies almost entirely on automatic functions. Yet, they must keep alert for any signs of danger or deviation from normal function or course, analyze the issue if it occurs, and respond quickly and effectively to it. These very specific needs are met by a cockpit display that manages almost every function of the plane. In fact, the cockpit display is really the same concept as a car dashboard, but more complex. But you already know all about car dashboards, don't you?
Monitoring an airplane is a sophisticated feat, but it comes down to principles that are applicable in monitoring any process, including clinical trials. Let's think about ways that we can classify airplane monitoring, and then see how these apply to clinical trial management.
I know some of the basic instruments of a 747, again through the magic of Flight Simulator. If you think about your glimpses of the cockpit of a 747 as you "deplane," there are a number of rows of switches surrounding the pilots, a variety of dials, and many computer screens. The old 747 had three people in the cockpit-two pilots and an engineer. They each monitored a separate set of mechanical and computer dials and displays. Today, the 747 is manned by only two pilots, and the displays are almost entirely computer screens. There are fewer monitoring displays, as well, since computers now monitor many engine-related functions.
The most important aspect of an airplane cockpit, or car dashboard, is the view. Operators need to monitor most, if not all, of the vital functions of highly complex, multifunctional vehicles without moving from their chairs.
The pilots have to keep their eyes on dozens of different functions throughout the plane, always mindful that any one failure could be dangerous for all aboard. These can be categorized into several different monitoring classes. First of all are the key indicators about the plane and its direction. This snapshot includes the altimeter, the airspeed indicator, and the horizon indicator.
While the stakes may not be quite as high as piloting a jet plane, running a clinical trial can be every bit as demanding. A large, multicenter clinical trial often involves tens or hundreds of people whose activities are coordinated across international borders in multiple languages. As with a plane, there are many different points of failure that can cause minor "discomfort," and many that can lead to significant damage or even complete disruption of the trial. These failure points can appear suddenly, or can be identified early as trends in the data. Wouldn't it be nice to have a "cockpit" for running your clinical trials? Well, the concept of a "digital dashboard" is currently quite popular in technology circles (and has been for at least five years), and is being incorporated broadly into software design. It is also finding its way into clinical trial software design.
You are probably already quite familiar with digital dashboards. The digital dashboard concept is based on the modular construction of Web pages. A digital dashboard Web page can be constructed with data and content obtained entirely from within one company, or from a dozen or more data sources scattered around the globe. When viewing the page, the users are unaware that they are not looking at one data source.
The most well known example of a digital dashboard is "My Yahoo." (my.yahoo.com -if you don't have one, it is worth it to set one up). This commercial dashboard allows a user to create a customized home screen that monitors many different concerns at once. One dash can monitor general and specific news, stock portfolios (with alerts for earning surprises, for example), weather, diet trackers, recipes, mortgage monitors, currency converters, personal address books, calendars, interactive maps, sport scores, email searches, local movies schedules, or even clinical trials.
Many of these can be displayed on a single page with key information. Each alert or information item is typically linked to a more in-depth display of related data, often with the ability to expand well beyond the original data, using the Web. Many people have gravitated to this as their home page, customizing it to suit their needs.
The management of a company may have a "digital dashboard" that provides front-end access to recent emails, voice messages, a calendar, search options for contacts, and many other personal functions. In addition there could be quarterly financial statements, the stock price and volume, the status of various projects, and production reports from factories. Thus, from a single seat the manager may be able to tell a great deal about the company's status.
As with the jet cockpit, the display may differ depending on the urgency of the information being displayed. For example, certain corporate information may be urgent in nature-a more than
$1 fall in stock price; an urgent email; a delay of more than 30 days in a project. These events could trigger an alarm on the screen, which can be manifest in many ways-a red light, a flashing word, etc. The CEO would get a different set of alarms than the Chief Information Officer. They would share some alarms, but have others customized just for them.
Again, as in the jet, much of the data needs to be presented in a more detailed and complete fashion than an alarm, which is just an on-or-off binary state. In a jet these may be represented by analog indicators of pitch and yaw. Corporate dashboards, in turn, may be presented in tabular form, or as graphics-a pie chart, histogram, or point and line chart. Ideally, a mouse click on a particular graphic display of data would "drill down" to a more detailed level of information.
Finally, some of the digital dashboard would be the raw data itself-email messages or links to text documents. As with the jet's radio communications, sometimes the pilot needs to have direct access to the data itself. A summary alone will not suffice.
Many software products are including the concept of a digital dashboard. Microsoft Outlook has a built-in digital dashboard that can be highly customized for individual users. In addition, many customized enterprise software programs are including digital dashboards for reporting on enterprise functions. A call center program might report waiting calls, time waiting, average waiting time, resolution status of last 24 hours, etc. As more and more data warehouses get built, the digital dashboard creates a perfect window to view aggregated business performance data. It allows managers to closely follow progress and develop forecasts for sales or production.
Again, as with the jet cockpit's systems view, business data is collected through business processes, and different analytic engines use the data to feed reporting engines. As in the jet cockpit, these reports can be binary on/off alarms or highly complex displays.
The ideal for the digital dashboard is for it to be the first thing that you look at each day: it's where you get your email, check your calendar, look at bulletins, identify problem areas for you and immediate action items, and dig into items of interest. It should also be a digital "destination" to which you return between meetings or phone calls and at the end of the day. In addition, it should synchronously communicate with your cell phone or PDA, sending along alerts automatically as they occur. As you can see, the digital dashboard can take on major significance and could be the most hotly sought-after "real estate" for software companies since the spreadsheet.
Clinical trial management is a perfect application for the digital dashboard. It is a complex and prolonged process with multiple participants taking many different roles. Investigator management occurs through communications and information collection, which is highly amenable to being incorporated into a dashboard. Furthermore, the clinical trial process has multiple steps where delays can occur, and the progress needs to be monitored against planned milestones. Clinical trial management systems have some rudimentary digital dashboards incorporated into some of them, but the sophisticated, "jet cockpit" digital dashboard hasn't yet been developed.
The digital dashboard for clinical trial management should be highly optimized for the individual role of the user. The project manager will have one view, while the data manager has another. One will be focused on milestones and budget, the other on data collection activities. Each of the participants from medical monitor to CRA would also have a view of the project through their dashboard. Many of the reports would be the same for all users-for example, typical office productivity tools like email, contacts, and calendar. In addition, some of the project information would be shared: project phone lists, recruitment forecasts, and many others.
The clinical trial digital dashboard would also be the window into the document repository for a clinical trial. One could envision a central repository where he/she might have the ability to read, modify, or even delete the document, all with a full audit trail.
The true ideal for a clinical trial dashboard doesn't come all that close to reality. Certainly, there are many that would claim the mantle of a clinical trial digital dashboard. However, the integration of all the different activities in a clinical trial, available to all involved, has not yet been realized. Like many good ideas, developing a robust clinical trial dashboard may take a convergence of workflow, new participants, technology empowerment, and process change.
A particularly interesting challenge in clinical trials is the "one investigator-many sponsor" problem. One investigator is likely to do clinical trials with several or even many sponsors. They don't want to have to use multiple digital dashboards for each trial, or even each sponsor. They would much prefer one dashboard that accommodated all of their workflow, from all of their sponsors. This would require the unlikely collaboration of all sponsors, or the emergence of one dominant design to work across sponsors. The most likely scenario is a dashboard standard that would accept "snap-in" modules from individual clinical trials or sponsors-the "killer application" still to come.