By: Josh Herigon, MS 2, MPH

Smartphones represent the latest evolution in wireless communications technologies. These devices have the potential to profoundly impact clinical medicine. Not only are they able to speed communication through voice and two-way text communication (texting and email), they provide handheld reference tools for clinical decision-making, point-of-care access to imaging studies and laboratory data, as well as resources for patient education.

Despite these advantages, many physicians and hospitals are still relying on relatively antiquated paging systems. For those interested in capitalizing on this tool as a means of improving patient care and efficiency, the biggest question is—exactly how many physicians are using smartphones?

Physicians are peculiar creatures when it comes to technology. Medicine is a field consistently on the cutting edge of technology. Advanced imaging with CTs and MRIs, robotic surgery, and diagnostics with flow cytometry and PCR are just a few advanced technologies in routine clinical use. At the same time, doctors rely on technology everyday that is nearly 200 years old–the stethoscope–and even older (i.e.–paper charts, microscopes). Communications technologies exhibit this same dimorphism—although many physicians appear to be using smartphones more and more, they also rely heavily on older technologies such as pagers.

According to marketing research conducted by independent research firms and technology companies, smartphone use among physicians is wholly on the rise. Manhattan Research, a pharmaceutical and healthcare market research firm, conducted a study last year and showed 72% of physicians are using smartphones. Their research also showed this will reach 82% in 2012. Another marketing research company, Spyglass Consulting Group, showed 94% of physicians using some sort of smartphone.

Epocrates, an app maker, released a study last year showing 96% of surveyed medical students are using some sort of smartphone (although this is not explicitly stated in their presentation of the results; they focused more on what kind of smartphone med students are using).

Sampling bias leading to inflated numbers

Although these studies show a high degree of smartphone adoption among physicians, these results should be interpreted cautiously. These firms provide few details on how they actually conducted these studies. A major hurdle to conducting such research is sampling bias. This can occur in survey research when researchers get a low response rate (i.e.—researchers approach a large number of individuals to fill out a survey but few actually fill it out).

This biases the result in that the group responding to the survey is likely different than those who declined to respond.  More specifically, those who responded to requests by these marketing researchers may have been more likely to be avid technology enthusiasts and thus, more willing to talk about their use of technology with a surveyor. This biases the sample towards giving a higher estimate of technology use than there actually may be.

Certainly the Epocrates study suffers from selection bias because the medical students participating in the study were already Epocrates users (i.e.—if you are aware of Epocrates and use it, then you are inherently more technologically inclined than someone who relies on a PDR for drug reference). Epocrates may have been aware of this bias and this may be why they chose not to explicitly state the rate of smartphone use in their results.

Manhattan Research seems to have employed several strategies to avoid these biases. In their summary materials, they indicate they employed random digit dialing (a system commonly employed in telephone surveying to generate a random sample) and online surveys. Unfortunately, they don’t provide proportions for how many of their 2,033 surveys were random digit dialing or online nor do they provide response rates in each category. They also indicate they attempted to obtain a representative sample by filling quotas for 25 specialties and weight their data according to key demographics. All of these strategies help to reduce sampling biases. Interestingly, the Manhattan Research figure (72%) is significantly lower than their less methodologically strenuous counterparts’ estimates.

Current state of literature on smart phone adoption rates

Formal, peer-reviewed studies generally work very hard to avoid such biases and could provide a much clearer picture of actual smartphone use among a representative sample of physicians. Unfortunately, I could not find any such articles. Searches in PubMed and Google Scholar yielded no studies that specifically examined the prevalence of smartphone use among physicians. I found several studies examining the use of older PDAs in clinical medicine, but no recent work on smartphone use. Smartphones represent a very recent technological evolution. Given that peer-reviewed literature generally lags at least one year behind current trends, it is very possible such studies are in the works and have yet to appear in the journals.

Despite a lack of peer-reviewed research on the subject, the available evidence does suggest most physicians are using some sort of smartphone. My own experience in hospitals and clinics points towards rapid adoption by clinicians. Although I do see most physicians using a smartphone, they also have a pager clipped to their white coat. Comprehensive communications systems with smartphones as their backbone seem to be scarce. Hospitals and clinics are currently unwilling to foot the bill for fully integrating smartphones into their communications systems and phasing out pagers. Texas Children’s Hospital recently announced a new initiative to install Voalté, a comprehensive, smartphone-based system.

In the future, I hope to see more companies providing such integrated solutions as well as hard research examining such systems’ impact on patient care and clinical efficiency. Hopefully, when I start my internship in two years, pagers will be a laughable anachronism.

Editors note: Further reading on Sampling Bias & Image Credit: Scholarpedia.org

Josh Herigon, MPH is a second year medical student and regularly blogs about medical education, medical research, health policy, and patient care at Number Needed to Treat.