Harvard Medical School derives richer health research through patient app

Critical data on hypoglycemia not collected before

In a mobile, texting, socially engaged society, one would expect medical researchers to move beyond clipboards and phone surveys to make the most of technology. Harvard Medical School and Boston Children’s Hospital are starting to do that with a series of studies they’ve done querying people with diabetes, a growing population with multiple, severe health needs. Elissa Weitzman, an assistant professor at HMS and Children’s Hospital, discusses a recent study that she and her team published, using a social network called TuDiabetes, the open source patient data repository Indivo (developed at HMS), and an app they developed called TuAnalyze.

The team succeeded in gathering a huge amount of information at a depth that researchers could not traditionally capture through standard research methods such as phone surveys and the examination of emergency room visits. What has been missing from organized research efforts in diabetes is the many small incidents that take place, where hypoglycemia hampers or disables the victim but does is not bad enough to send him to the doctor. Understanding these incidents is important to protect patients and improve their quality of life.

The TuAnalyze app collected reports of these adverse effects. It captured incidents of both severe and moderate hypoglycemia that had not been previously reported. Furthermore, patients volunteered in huge numbers for follow-up research.

Weitzman’s research suggests the value of creating a two-way street between research participants and researchers, who usually have obtained their data by “sucking” it away from patients. Toward the end of our interview, Weitzman explores the importance of signing up large numbers of research subjects, which requires researchers to respect their concerns about being exploited and having commercial interests use their data in ways that make them unsafe or uncomfortable.

Major topics of the interview follow.

  • Goals of this research at HMS and Children’s Hospital [Discussed at the 0:16 mark]
  • The TuDiabetes online community and the advantages that engaging with that community offers to researchers. [Discussed at the 1:34 mark]
  • The value of adding structure to the information that individuals exchange about their medical conditions. This give patients feedback on other people as well as providing researchers with information that’s hard to get: a closed-loop approach. [Discussed at the 4:05 mark]
  • The TuAnalyze app. The importance of a personal health record for privacy and patient control. Displays that offer the users information on people in their community, which can motivate better life choices. [Discussed at the 5:29 mark]
  • Tracking adverse events: hypoglycemic. Got more information than researchers get from telephone interviews (shallow but broad) and clinical events such as emergency room admissions. Both data sources miss a huge range of experiences of people on insulin. [Discussed at the 10:00 mark]
  • Participants using TuAnalyze showed they were willing to be contacted to receive results and for further research. [Discussed at the 14:20 mark]
  • How can we use the information gathered by this kind of app? Understanding hypoglycemia. Feeding back data can also be used by patients to share more of their problems with their doctors. [Discussed at the 15:33 mark]
  • This research is a small step toward creating citizen health. We have to “keep talking” to patients. Some conditions, such as asthma, continue over a life. [Discussed at the 17:27 mark]
  • Discussing the current mania for patient engagement. Patients are already engaged in online communities. But who else could be listening? Clinicians and researchers are not involved in the social networks. Data collection through an app can make a bridge where both the researchers and the patients benefit. [Discussed at the 19:45 mark]
  • Alignment of goals between researchers and their subjects. Patients bring conditions to research: they want to be in control, and prefer giving their data to non-profit organizations. [Discussed at the 22:40 mark]
  • How can efforts like those of Elissa’s research team expand? Businesses are heavily engaged in this area, with the goal of data mining. Patients are engaged for different reasons, and could perhaps accept the involvement of businesses, but only under conditions of transparency. Patients will leave environments that they feel are predatory. For difficult, complex conditions, the participation of large populations is crucial, so their interests must be respected. [Discussed at the 23:36 mark]

View the entire conversation in the following video:

tags: , , , ,

Get the O’Reilly Data Newsletter

Stay informed. Receive weekly insight from industry insiders.