January 17, 2011

Healthcare in a Petri Dish
By Elizabeth S. Roop
For The Record
Vol. 23 No. 1 P. 20

The Healthcare Innovation and Technology Lab examines the culture of medical text messaging and condition-specific EHRs and their effects on health disparities.

More than 1.5 trillion text messages—nearly 5 billion per day—were sent and received on U.S. carrier networks during 2009, according to CTIA-The Wireless Association. As texting has become ubiquitous in American society, it was a natural target for forward-thinking healthcare organizations that have begun utilizing texting for purposes ranging from patient communications and public health education initiatives to consumer and provider marketing.

“Text messaging can provide a form of communication and a way to access healthcare or health information that many couldn’t access [otherwise],” says Ilene Hollin, MPH, research manager at the Healthcare Innovation and Technology (HIT) Lab. “Millions have access to mobile [technology] who never had access to landlines or computers. [Texting] is leapfrogging over other technologies and can reach into developing populations and countries.”

Though healthcare texting campaigns are still in the early adoption stage, the technology has already raised questions about ethical implications and overall effectiveness. In search of answers, the HIT Lab undertook a comprehensive literature review of academic and popular press coverage to differentiate between health interventions and marketing campaigns, evaluate their success, and explore the issues that arise when recipients are unable to differentiate between the two.

The findings of the study, “Differentiation of Text Messaging Meta-Data for Critical Clinical Intervention Purposes From Marketing Purposes,” were presented in November at the American Public Health Association’s (APHA) Annual Meeting & Exposition. It was one of four studies focused on advanced technologies in healthcare that were presented at the conference by the HIT Lab. A second, “Assessing EHRs and Health IT in Outpatient HIV/AIDS Clinics Across Four Continents,” evaluated the ways in which HIV-specialty EHR functionality contributes to the elimination of health disparities.

Texting Shows Promise, Raises Concerns
In its text messaging study, the HIT Lab sought to differentiate health campaigns from marketing campaigns and explored the ethical implications of instances where recipients would likely be unable to do so. Among the “intervention” campaigns were those focused on treatment compliance, clinical care management, and preventive behaviors such as breast self-examination, weight management, smoking cessation, and vaccinations.

Among the “marketing” campaigns were those that directly promoted a product or a service offered by a commercial vendor. These included the promotion of free anticongestion products and specific drugs to consumers and healthcare providers to raise top-of-mind awareness.

“Especially from our literature review of formal academic studies, we definitely saw positive results. Half of the studies show statistically significant differences in favor of text messaging and a positive trend. A few didn’t, but even those that showed a positive trend that was not statistically significant, it was because of small sample sizes. We could still see them moving in the right direction,” says Hollin.

The challenge in extrapolating success rates lies in the fact that text messaging is still too young to measure any long-term effectiveness. Expectations are that once the novelty wears off, the effectiveness of certain types of campaigns, such as smoking cessation or weight loss, will go down.

There is also the risk that text messaging, given its rapid rate of growth, will reach a saturation point long before it can make a real impact on healthcare and public health. Hollin points to the recent mHealth Summit, where the majority of attendees shared text messaging success stories.

“They said everyone is sick of e-mail because it’s a source of spam, but text is above the fray. That’s great, but it was hundreds of people doing the same thing,” she says. “There is the possibility that if it all happens in large quantities, it becomes something that is easily ignored. We’ve seen it with behavioral change websites. People stop logging in and stop using them. That might happen to text messaging eventually. … The key is how we use it and how it’s tailored to people’s individual needs.”

While the effectiveness of text messaging in the long and short term was an important component of the HIT Lab’s research, the primary focus was on the ethical implications involved in utilizing the technology for marketing purposes. To evaluate potential ethical pitfalls, researchers evaluated campaigns based on the following qualitative parameters:

• Claim/purpose: What is the purpose of the campaign, as specified by the sponsor?

• Product mention or recommendation: Is a product named or recommended by the sponsor?

• Follow-up events: Does the campaign follow a related product-specific commercial event?

• Contact information: Does the sponsor ask participants to visit a branded website?

• Target population: Is the campaign neutral in the selection of the target population?

• Departments involved: Is the marketing department involved in conducting the study?

• Conditions applied: Are there any conditions for participating in the campaign?

Based on their evaluation, researchers identified a number of ethical implications surrounding the use of text messaging in healthcare. These included marketing campaigns masquerading as health interventions, such as a campaign recommending mammograms sponsored by a screening provider and a pollen alert text message program that followed the launch of a new allergy product.

“I don’t think [marketing and health intervention campaigns] can ever be completely separate. Even if they are as neutral as they possibly can be, if a brand is associated, at minimum it’s building brand loyalty,” says Hollin. “It’s great we have industry partners that can sponsor these interventions, but keep in mind that they’re doing so with the intention of driving the bottom line.”

A second area of concern was the exposure of recipient metadata, particularly behavioral data, which are extremely valuable as a marketing tool. In many instances, researchers identified vulnerable areas where key metadata, including demographics, health conditions and treatments, and behavioral patterns such as changes in adherence to treatment or preferences (eg, pills vs. liquid medications) and the amount of mobile usage, could potentially be exposed.

Not only do these metadata expose participants to unwanted marketing, but such exposure could take place at multiple, unrelated points. For example, a consumer group reaches out to its database of potential participants, who in turn share personal information as part of the campaign. That information is then shared with a pharmaceutical company, which uses it for future drug promotions or with a solution provider that uses it for a different pharmaceutical campaign or campaigns for entirely separate industries.

This potential for exposure was particularly troublesome for Hollin, who notes that “people don’t always realize the information that’s being collected when they sign up for these things. The personal information is obvious, but what might not be so obvious is the wealth of information they’re providing in terms of behaviors. I see that as a little more concerning because it’s so hard to make people see [the dangers].”

Two additional ethical implications identified by researchers were the “halo effect” enjoyed by companies that sponsor true health interventions and the emergence of text campaigns targeting providers and pharmacists.

Ultimately, Hollin notes, the power of text messaging may require regulatory intervention as has taken place with other communications among vendors, consumers, and providers.

“The fact that it opens up a whole new channel for communicating with people [raises the question of] will it need to be regulated the way physician detailing is and the way direct-to-consumer advertising is. Who is looking at these text messaging programs?” she asks.

In Praise of Condition-Specific EHRs
While the HIT Lab’s text messaging research revealed several of the ethical pitfalls of next-generation communication technologies, its examination of condition-specific EHRs showed how another technology can positively impact the serious problem of health disparities.

In this study, researchers conducted site visits at HIV clinics across the United States, using surveys and observational methods to collect data on the use of EHRs designed specifically for the HIV/AIDS population. Secondary research examined the HIT infrastructure and EHR use in Europe, Africa, and Asia to determine HIV/AIDS healthcare needs and how EHRs can contribute to the elimination of health disparity.

“One of the major impacts we saw, which was also the thesis of this research, is that specialty EHRs can help eliminate health disparities. That’s very important when you’re looking at this population because it disproportionately affects developing countries. AIDS is among the top five causes of death in the 27 poorest counties. In the U.S., it disproportionately impacts minority groups,” says Margaret Griffin, a research coordinator at the HIT Lab.

In her APHA presentation, Griffin noted that this disparity exists despite substantial progress in the prevention and treatment of HIV/AIDS. In the United States, blacks, Hispanics, and men who have sex with men have much higher rates of HIV/AIDS. Globally, in 2008, adult prevalence of HIV/AIDS in sub-Saharan Africa was 5.2% compared with 0.4% in North America.

Disparities also extend to insufficient treatment options. In developing and transitional economies, 9.5 million people living with HIV/AIDS are in immediate need of life-saving antiretroviral drugs, but only 4 million (42%) receive them.

“From this research, we saw that having [HIV/AIDS-specific] systems in place can highlight these disparities but also create a way for providers to intervene,” Griffin says. “The first step in eliminating disparities is acknowledging that they exist, being able to help track race and ethnicity with quality indicators, and looking at how to move forward from there.

“Another important point is that HIV has been transitioning from an acute disease to a manageable chronic disease. Having a system in place that can help [manage] HIV will have a huge impact on care,” she adds. “As medical research keeps adding years on to life after diagnosis, it’s important to have a system in place [for long-term management].”

Specifically, researchers found that EHRs designed for HIV/AIDS populations help eliminate disparities by standardizing treatment, improving patient education, and streamlining reporting capabilities. They also help providers efficiently gauge patients’ progress by tracking disease-specific health indicators such as CD4 and white blood cell counts and viral load levels. Further, these health indicators can be tracked according to race and ethnicity, which allows medical histories to be evaluated to determine the causes of disparities.

Other key areas where condition-specific EHRs had an impact include the following:

• patient education tools such as visit summaries tailored to the concerns of HIV/AIDS patients and providers;

• medication reconciliation tools customized to support complicated antiviral regimens;

• HIV-specific provider reminders for health items such as vaccinations and hepatitis testing; and

• reporting functions that ease the burden on providers to comply with federal funding requirements.

“These are all very critical in having a comprehensive HIV treatment plan,” says Griffin. “Having an EHR that is specific to a disease, especially one as complicated as HIV, is really important.”

When it comes to drawbacks, Griffin says there are few. In fact, the overriding concern is the loss of the systems in the wake of accelerated EHR adoption under HITECH.
“One of the concerns we were hearing from some of the clinics in the U.S. is that many are attached to a hospital or other parent organization. They were getting pressured to use the system in place within the hospital and to move away from the specialty-specific EHR,” says Griffin. “Staying interoperable and being able to create interfaces will be very important for keeping these systems [in place].”

— Elizabeth S. Roop is a Tampa, Fla.-based freelance writer specializing in healthcare and HIT.