Talking Touchscreen: Reaching the Hard-to-Reach Patient

EVANSTON, Illinois—Patients with low literacy have often been excluded from assessments of self-reported health outcomes, but that may change thanks to a new multimedia tool called the Talking Touchscreen.

The Touchscreen, which is being tested in three Chicago cancer centers, may finally allow researchers to identify the factors influencing health outcomes in this understudied and vulnerable population, Elizabeth A. Hahn, MA, told ONI in an interview. Ms. Hahn, one of the Touchscreen’s developers, is a research assistant professor at the Institute for Health Services Research and Policy Studies, Northwestern University, and director of biostatistics at the Center on Outcomes, Research and Education (CORE), Evanston Northwestern Healthcare, both in Evanston, Illinois.

Few studies have assessed the prevalence or impact of low literacy in the health care setting, according to Ms. Hahn. But the available evidence suggests that about half of all patients have impaired literacy, and that low literacy has tangible health consequences.

"Patients with low literacy are considered to be vulnerable and at risk of having poorer health outcomes," she said. Studies have found, for example, that individuals with low literacy are diagnosed with cancer at a later stage, are more likely to rate their health as poor, and have higher health care costs.

However, the reasons for these associations remain uncertain because this population has generally been left out of studies assessing self-reported outcomes, Ms. Hahn said. "In fact, in most research studies, literacy is explicitly listed as an inclusion criterion," she said.

Some researchers have used face-to-face interviews to work around literacy barriers. But this approach has problems of its own, she observed, most notably the need for resources and the potential for introducing bias.

"We don’t really know if the problems facing patients with low literacy are the same as those in patients who can read because we don’t have enough information about them," Ms. Hahn said. So the primary motivation in developing the Talking Touchscreen, she said, was to find a way to include these patients in future research or "any kind of health outcomes assessment in hopes of learning more about these patients and ultimately being able to do something to improve their outcomes."

How It Works

The Talking Touchscreen is a multimedia tool that presents a patient with quality-of-life questions and answers on a touch-sensitive screen while it simultaneously reads them to the patient through a headset, Ms. Hahn explained.

Patients are presented with only a single question and set of answers per screen, displayed in large type and offset with color to promote readability (see Figure). Each answer appears on a gray "button" on the screen, and patients choose an answer by simply touching one of these buttons. Some questions are accompanied by graphics to illustrate difficult concepts, such as probability, she said.

The interface gives patients a variety of options for tailoring a session: They can play the sound again by pressing the ear icon next to the question and each answer; they can go back to previous questions and change their answers; and they can take the headset off and advance at a faster pace if they read quickly.

The physical features of the Talking Touchscreen strike a balance between sturdiness and portability, Ms. Hahn noted. The tool is heavier than most laptops, weighing in at about 12 pounds, but it withstands use and travels well in a carrying case. It has a 12-inch diagonal screen, an audio jack for plugging in the headset, and a keyboard for programming that is detached during patient use. It can be set up on any flat surface.

To begin a Talking Touchscreen session, a researcher starts the program and enters a patient’s ID number, used to preserve confidentiality, Ms. Hahn said. The researcher orients the patient to the Talking Touchscreen’s features, helps the patient work through a few practice questions, and then steps away and allows the patient to continue alone.

The questions presented by the Talking Touchscreen come from three sources: the Functional Assessment of Cancer Therapy-General (FACT-G), the Short Form-36 Health Survey (SF-36), and the Standard Gamble Utility Questionnaire (SGUQ). Patients can work through each of these in about 10 minutes, Ms. Hahn said.

Flexibility

Part of the beauty of the Talking Touchscreen is its flexibility, Ms. Hahn noted. For example, the Talking Touch-screen is currently programmed with a woman’s voice in English and Spanish. But future versions could offer patients a choice between a man’s and a woman’s voice, and could present questions and answers in virtually any language.

The Talking Touchscreen has other merits as well. "People feel that it gives them more privacy," Ms. Hahn said. "It feels more confidential to them than sitting down and filling out a paper-and-pencil questionnaire." Patients have also reported that the Talking Touchscreen’s headset blocks out background noise, making it ideally suited to a noisy clinic setting, she said. And anecdotal evidence from other studies suggests that adding sound enhances concentration, even if an individual has good reading skills.

Although the Talking Touchscreen was designed with low-literacy patients in mind, a key consideration was ensuring that it would be acceptable to all patients, regardless of their literacy level and computer skills, Ms. Hahn said.

Singling out low-literacy patients for Talking Touchscreen assessment might reinforce the stigma of low literacy, she explained. Furthermore, some patients with higher literacy might nonetheless have no or poor computer skills.

"We wanted to design something that was flexible and that could be used by everyone," Ms. Hahn commented.

Two Parallel Studies

Ms. Hahn and her colleagues are currently testing the Talking Touchscreen in two parallel studies, one each in English-speaking and Spanish-speaking outpatients with cancer. The studies are taking place at three cancer centers in Chicago’s inner city, and each will enroll up to 400 adult patients, approximately half of whom will have low literacy (a reading ability below the seventh grade level). The patient populations are intentionally diverse with respect to sex, age, ethnicity, type of cancer, and disease stage. "We are looking for maximum generalizability of the findings," she said.

The goals of the studies are threefold, Ms. Hahn said. The first goal is to establish the feasibility of using the Talking Touchscreen in the cancer outpatient setting. The second is to check for the presence of bias in measurement across different patient subgroups, such as the possibility that patients with low and higher literacy might interpret questions differently. And the third is to assess potential differences in self-reported health outcomes between patients with low and higher literacy.

The study in Spanish-speaking patients will also look at the feasibility of using the data collected with the Talking Touchscreen in real time, Ms. Hahn said. In this scenario, a patient would take a Talking Touchscreen assessment while waiting to see his or her physician, and the Touchscreen would generate a quality-of-life score on the spot.

"We can print that out and give it to the physician and the patient, and they can decide how best to use that information to make decisions about treatment or the next step in their care," she said.

Possible future applications of the Talking Touchscreen, she said, include talking patients through the informed consent process, delivering education and preventive interventions, and giving instructions on tasks such as preparing for medical tests and taking medication.