Wireless Data: Challenges in the New New Economy

Wireless computing may be the way of the future, but several obstacles currently prevent us from achieving it.

Although the Internet spawned the new economy, its been said that wireless computing will spawn the new new economy. With the degree of travel and the high demand for remote information access that exists in clinical trials, mobile computing provides the timeliest mechanism for accessing, and in some cases capturing, clinical trial data. Two categories of wireless platforms are poised to affect clinical trials: cellular wide-area data networks and wireless local-area networks (WLAN).

The vision of mobile computing is easily imagined today; however, significant challenges impede progress toward ubiquitous wireless computing in clinical trials. Consider six of these challenges: high, perhaps inflated, expectations; still waiting for an indispensible application; dead spots in cellular coverage; unsettled standards; insecure airwaves; and a still-uncertain technological future.

Great expectations
The acceptance of Internet technologies into clinical trials with the great promise of cleaner data faster, real-time trial visibility, and disintermediated processes makes the notion of wireless computing in clinical trials readily imaginable. We already understand the efficiencies promised by Internet technologies. Therefore, wireless = Internet + mobility. Our expectations for new technologies have been set by the impact of the Internet.

But having such heady expectations for wireless clinical trial automation is more likely a liability than a strength. The wild success of the Internet grew from our extraordinarily low expectations for the technology. In 1994, we surfed the Web using our Mosaic browsers with minimal expectations. The quirky and often homegrown sites we visited were just for fun. At that time none of us anticipated e-commerce transactions to the tune of billions of dollars.

Geography also affects expectations. While Americans wait for cable modems and DSL to fulfill their need for high-speed browsing, Europeans tend to be more interested in bandwidth for their mobile phones. In Europe, the number of mobile phone users is more than double that of home Internet users.

Solid business drivers support extending existing infrastructures to provide mobile access to clinical trial data. However, the limited wireless data infrastructure demands that expectations be carefully managed. Rapidly evolving technology, a fragmented device base, sparse network coverage, uncertain standards efforts, and very limited device capabilities make wireless clinical information systems challenging.

No indispensable application
Successful technology platforms have historically benefited from the existence of an application that generated such tremendous demand that it justified the existence of the entire platform. The IBM PC had Lotus 1-2-3 and word processing, the Apple Macintosh had desktop publishing, and the Web had commerce.

Wireless computing in clinical trials has not found its indispensable applicationyet. Mobile voice technology is ubiquitous, but significant mobile data communication is still the exception rather than the rule. So, is there real demand for wireless applications in clinical trials today, or do we just hear noise generated by the marketing engines of a few technology pioneers? Acknowledging the fact that no indispensable application is readily apparent, it is fair to say that some obvious pockets of demand do exist today. For instance, wireless electronic subject diaries and the wireless delivery of clinical trial portal content are in use today.

Cellular coverage deficiencies
Cellular data networks provide true mobilitysort of. Nevertheless, cellular network coverage deficiencies create reliability issues, as weve all experienced with voice coverage. Reliability is the foundation upon which any communication technologys ubiquity is built. In the United States, POTSplain old telephone serviceand its ever-present dial tone has set the expectation level. A recent Consumer Reports survey found that only half of the respondents were completely or very satisfied with their cell phone service.

Testing coverage and reliability a priori is neither simple nor easy, especially when users might be moving in and out of coverage areas. This significantly complicates wireless electronic diary deployment. What assumptions have been made about coverage, and what contingencies accommodate those without coverage? Expectation management is critical, because a real-time data collection system cannot deliver on its promise without adequate network coverage. Subjects outside covered areas may need to use a different system that requires separate processes and procedures.

Further complicating matters, those networks with the best coverage also tend to be the slowest. Many of the most popular data service networks run at 9600 bps, extraordinarily limited by wired standards.

Shifting standards
Growth of the wired Internet benefited greatly from relatively stable standards. Competing and often incompatible standards have limited the effectiveness of wireless data applications. Differences between U.S. and European cellular technologies complicate global rollouts. Furthermore, a technology standard cannot assume survival; many will not become mainstream.

Bluetooth, for example, is in danger of becoming the ISDN of the wireless worldirrelevant by the time it arrives in any practical way. Bluetooth is a short-range standard that provides always-on connectivity for PDA-laptop synchronization, wireless Web surfing, and downloading music. But support for the standard seems to be waning in light of the significant popularity of the WLAN (802.11b) standard. Microsoft and Intel have both recently expressed concern and limited their support for Bluetooth.

Many argue that 802.11b WLANs provide all the functionality offered by Bluetooth at greater speeds and distances. 802.11b provides a 100- to 300-foot bubble of wireless access, operating at performance levels similar to wired networks. WLAN technology promises immediate impact by running the same applications supported in the wired world. The advantages that Bluetooth enjoys in small, low-powered devices will go away with the newer implementations of 802.11b.

Deploying WLANs in Phase 1 units seems to create obvious value. Assuming no restrictions on wireless deployments (as there are likely to be in a hospital), WLANs enable laptops to be positioned where they are most useful within the clinic, without necessitating a wired network connection. Internet-based electronic data capture (EDC) systems and data reporting tools can be fully accessible anywhere within the clinic.

Undefended airwaves
Security issues are paramount in wireless information systems, because most wireless technologies involve broadcasting signals over the airwaves. Because broadcast signals can be intercepted by anyone, the trick is to secure the privacy of messages through encryption, as well as to authenticate the parties exchanging information. Securing wireless data is possible, but not simple.

WAP (Wireless Application Protocol) is the most common protocol used to access Internet-based information via a wireless platform. The wireless equivalent of TCP/I, WTLS (Wireless Transport Layer Security) secures WAP communications and is the functional equivalent of SSL (Secure Socket Layer) in the wired world. Unfortunately, although WTLS and SSL provide similar functionality, they are incompatible with one another.

These incompatible security models create a vulnerability commonly referred to as the WAP gap. WAP communications begin in the wireless world, but eventually reach the wired Internet to access Web applications. Because WTLS and SSL are incompatible, the gateway bridging the wireless and wired world must convert the incoming message from WTLS to SSL. At the time of the conversion, the message is left unencrypted, in plain text, for a microsecondthe WAP gapjust long enough to cause serious concerns among security experts. Many companies consider the vulnerability negligible. For example, Ameritrade permits WAP-based trading, while others like Fidelity consider the WAP gap an unacceptable risk.

The 802.11b WLANs have also been plagued by security shortcomings. 802.11b uses WEP (Wired Equivalent Privacy) to ensure that intruders are not intercepting network traffic. Earlier this year, UC Berkeley and others published theoretical shortcomings in WEP. Just recently, theory became reality when Airsnort was released on the Internet. Airsnort enables nearly anyone with some technical savvy to detect, intercept, and decrypt traffic, including passwords, on 802.11b networks. Hackers could easily intercept information while outside of a building, because only access to the bubble is required.

Ironically, the same companies spending millions to protect their networks from the Internet have WLANs installed, creating significant exposure. In fact, the Gartner Group stated that while 50% of all organizations currently plan to deploy WLANs, 20% of large enterprises already have rogue WLANs in place. Rogue WLANs are installed by non-IT users, often have no security enabled at all, and are typically connected to other corporate networks.

WLAN deployments are expected to grow rapidly over the next five years, and their availability will have an impact on clinical trials. To make them secure, however, requires additional layers of security. Virtual private networks (VPNs) are one mechanism for improving security, and Cisco has announced products that add additional security to WLANs, making them appropriate for use with sensitive information. WLANs alone, without additional security measures deployed, should not be considered for use with sensitive or regulated information.

Future considerations
Of course, the future of clinical trials includes wireless information systems. Of that there is little doubt. The timing, however, is open to debate. There are many technologies either available today or on the horizon that will play a role in the future wireless IT infrastructure.

Better interoperability and higher bandwidth will solve many of the current limitations with wireless applications. New standards like 3G (3rd Generation), or even 2.5G, wireless have been developed to provide enough bandwidth for an interactive multimedia wireless experience. These high-speed wireless data services will enable a new generation of applications to be deployed on every imaginable type of device. But 3G has experienced significant delays. In fact, skeptics suspect that the delays indicate that the technology, in its current state, doesnt work. The skeptics claim the high-speed 3G devices will experience battery life problems and become unbearably hot. Much like high-speed wired Internet access for the home, it appears that high-speed wireless data communications will be slow to arrive.

Security will continue to be a major concern. While WAP and other wireless Web technologies have been slower to catch on in the United States and Europe, Japans i-Mode has over 25 million subscribers and is growing at 1.3 million new users per month. This makes i-Mode the worlds most successful wireless data service by a wide margin. Interestingly, viruses, spam, and Trojan horses, security issues that target only wired systems in the United States have hit the i-Mode network. Consider this foreshadowing.The future might even find a home for wireless PKI (public key infrastructure). Security experts widely consider PKI critical to deploying secure mobile applications. However, a quick survey finds many PKI companies faltering. The adoption, even in the wired world, just isnt there.

Bottom line
Although wireless computing may spawn the new new economy, Ive never been completely convinced that any technology thats considered new new will succeed in the clinical trial technology market. Maybe not-too-old technology is as leading-edge as we can hope for. Newness aside, wireless computing appears to represent the next great technology frontier for clinical trials.

Think of wireless data today as the Web in 1994. Although our low expectations of the Web allowed the technology to evolve satisfactorily, our expectations for over-hyped wireless computing outstrip reality by such a large measure that frustration is inevitable. On the other hand, those of you running clinical trials who crave the latest technology will find wireless applications deployed today.

Proceed with caution, though, and acknowledge the limitations of todays solutions. WLANs offer mobility within the clinic, but additional security measures must be applied. Cellular data networks offer true mobility for electronic diaries and Web portalbased information access, but beware of gaps in coverage and limitations in bandwidth and processing power.