Health Sciences and Technology Division
Harvard Medical School
Institute of Technology
Medical Informatics Laboratory
Dept of Anesthesia and Critical Care
Archetype Technologies, Inc
Unlike computing in most professions, clinical computing is primarily about communications. Clinicians and pharmacists are concerned with accessing patient, formulary and other clinically relevant data – as well as updating data – in a timely manner. Similarly, just as the cellphone and pocket pager free clinicians from the confines of their desk, mobility and wireless connectivityare essential to providing clinicians with immediate access to relevant data. As such, a powerful, number-crunching behemoth is of little value compared with a personal digital assistant (PDA), or tablet PC, especially when one of these mobile devices is equipped with wireless connectivity to the hospital network and to other devices. This article provides an overview of tablet PC and PDA technologies in terms of the practical considerations relevant to creating department- or hospital-wide systems capable of supporting daily practice.
Size matters only up to a point
Given the marketing hype behind the various makes and models of PDAs and tablet PCs, it is often difficult to discern which models would be most appropriate in a given clinical setting. As noted in the article on handheld computers by Scot Pegler (HPE 2003;7:49–50), although handheld computers are readily available in Europe, very few hospital pharmacists currently use them to enhance everyday practice. The reasons for this state of affairs are numerous. There are the more obvious issues of cost, software availability, size, features and availability of add-ons. However, aside from cost and availability of software, many of the issues touted by the hardware manufacturers should be of secondary consideration. The reason that many clinicians – and many department heads – embrace handheld computing is because the technology has the potential to save them time.
In my experience with handheld computing at Harvard Medical School and the affiliated hospitals, the first-generation handheld computing devices of the 1990s were cumbersome and difficult to learn to use. As a result, even when the handheld hardware and custom software were provided free of charge, most clinicians and even medical students refused to use them. It was quicker to look up a diagnosis or formulary from a pocket notebook than to fumble with an electronic device.
Times have changed, and the latest generation of PDAs is fast, affordable and, most importantly, when properly configured and used, can save clinicians time. At Massachusetts General Hospital, for example, residents and staff routinely “beam” patient data from one PDA to the next through built-in infrared links. Medical students and residents can access clinical reference material anywhere on the medical campus and in most clinical locations through wireless WiFi (802.11b) LAN technology. Formulary data, differential diagnosis lists, email referrals and other “just-intime” data are available at the touch of a stylus or thumbwheel. This is possible because of a robust underlying information system, including the servers where the data are stored and the hospital network that transports the data when and where it is needed. A wireless tablet PC or PDA is simply an interface or a means of interacting with the data. A lone PDA or tablet PC is of relatively little value, especially when time-saving is the metric of success.
Selection criteria – the big picture
Paradoxically, what really matters most in selecting a diminutive handheld computing solution is not as much the individual device as consideration of the big picture. Taking the perspective of a hospital chief information officer, the key selection criteria are seemingly mundane: vendor assessment, operating system, battery life, warranty, cost of maintenance, insurability, compatibility with third-party software and other systems in the hospital or clinic, and the migration path to the next-generation devices.
The most successful handheld computing solutions satisfy departmental and hospital criteria for cost, operability and demands on the support infrastructure, and this normally translates to establishing a standard device for the organisation. Not only is it more cost-effective to buy a single model of a tablet PC or PDA in bulk, but there is only one type of replacement battery or stylus to keep on hand, and one style of synching cradle to install and support. Similarly, there is only one operating system and one suite of software applications to maintain. Because most tablet PCs on the market run a version of Windows optimised for the tablet, the major operating system issues revolve around PDAs, which operate either PalmOS or Microsoft’s PocketPC system. Unfortunately, the two operating systems are incompatible, so software written for one operating system won’t work with the other.
When it comes to computer technology, especially leading-edge technology like tablet PCs, it is prudent to assume that the lifespan of a typical model is at most a year. As such, it is important to select a vendor that plans to be in the market for the long haul, as opposed to one that offers an inexpensive, one-time alternative to mainstream models.
A problem unique to handheld computing is theft. It is easy for a wandering patient to slip a tablet PC into a backpack or purse, and it is even easier to pocket a PDA. In a busy clinic it is not always practical to keep an eye on a tablet PC all day. Another issue is catastrophic failure due to accidental damage. A heavy tablet PC sandwiched between papers can be difficult to manage. As someone who has dropped a tablet PC onto a hard clinic floor, I can attest that they are built for lightness and size, not for durability. A drop of about a metre resulted in a shattered LCD screen and a cracked motherboard. PDAs generally fare better. However, the fragility of these devices points to the necessity of insuring against theft or accidental damage.
One reason that the early PalmOSbased PDAs have been so successful in the medical area is extended battery life. A single set of batteries may last several weeks on a charge under heavy use. In contrast, a problem with many of the latest-generation PDAs is their relatively short battery life. For example, although the colour screen of a high-end Casio PDA is beautiful, and there are numerous add-ons to increase its functionality, battery life is typically measured in hours or, at most, days. This can be a severe limitation in a clinical environment. However, if a high-quality colour screen is a necessity – say, to distribute pathology images – then a Casio or other highend PDA with a colour screen may be the perfect choice. The best technology for a particular job depends on the nature of the work, the work environment and the way things are done without the technology.
Evaluating the two general classes of technologiesavailable for handheld computing, tablet PCs and PDAs, is facilitated by exploring the return on investment associated with each technology. Initial investments for the more expensive tablet devices are different compared with the more affordable PDAs (see Table 1). However, the return on investment over the life of a typical device can favour the tablet PC if using PDAs will require significant custom software development or if it fails to support the needs of clinicians and pharmacists. Whereas tablet PCs can run most standard Windows applications with only modes modification, developing custom software for PDAs requires more work. However, development costs for the often simpler PDA applications are typically much less than those for tablet PCs.
Handheld computing, like every other information technology available to pharmacists, makes sense only if it truly saves time. As such, a holistic view of the clinical practice, not a narrow view of a particular handheld technology, provides the best metrics for how technology may be able to save time, cut costs and provide enhanced patient care. A PDA or other handheld technology provides value when it’s used appropriately as part of the pharmacist’s or clinician’s arsenal of data gathering and manipulating procedures.