Focus on Research January 2015

Integrating Patient-Generated Data into Primary Care

By Sara Keller, MPH, MSW
Research Associate

Sara Keller, MPH, MSW

There are many new and innovative ways that individuals can track their personal health data, including through third-party applications. These data are considered patient-generated data (PGD) and some patients are tracking information about their health on an ongoing basis. These data may present opportunities for clinicians to have a deeper understanding of the health of their patients between visits, since most patients only see their doctors a few times a year.

One type of PGD is observations of daily living (ODL). ODLs are people’s observations, patterns and experiences of daily life such as diet, physical activity, sleep, pain episodes and mood. ODLs are unique in that they are patient-generated and patient-informed, meaning that they are not predetermined by others, but provide the opportunity for clinicians and patients to work together to choose the cue, behaviors and experiences to track and record.

Deborah Cohen, PhD, and Sara Keller, MPH, MSW, conducted an evaluation of Project HealthDesign (PHD), which was a national program of the Robert Wood Johnson Foundation. PHD grantees developed and tested innovative health information technology (health IT) tools to help people capture, sort and use ODLs with the aim of using the information patients collect to foster patient engagement and inform personal health decision making and clinical care.

PHD consisted of five grantees that looked at a variety of health experiences including poorly controlled asthma; overweight young adults; caregivers of premature, high-risk infants; elders at risk for cognitive decline; and patients with Crohn’s disease. Each grantee team worked with both patients and clinicians, during design phase focus groups to identify the ODLs to be collected. This required finding the balance between ODLS that patients wanted to track and ODLs that were clinically relevant for physicians.

The evaluation team conducted and analyzed qualitative data from each of the grantees, and conducted qualitative interviews with participating clinicians and study team members. This led to cross-project finding including a theoretical model identifying factors that influence ODL tracking behavior and use. This model was the basis for a manuscript in Personal and Ubiquitous Computing, Developing a model for understanding patient collection of observations of daily living: A qualitative meta-synthesis of the Project HealthDesign Program (this article comes out in print in January). Upon analysis of clinician and study team interviews, the evaluation team identified three aspects to integrating ODL data into primary care practices.

First, clinicians found potential value in patient generated ODL data. These data provided the opportunity for clinicians to take a deeper look into the lives of their patients outside of visits. A nurse practitioner for patients with Crohn’s disease, for example, found that it gave her a better sense of the day-to-day life of her patients. It allowed her to recognize that a day in the clinic might be a good day in a bad month. This is especially relevant for patients with chronic diseases where symptoms are ever changing.

The second key feature is that these data need to be accessible to the clinic and synthesized in a meaningful way. Clinicians have no need for many individual data points, but do have use for a relevant summary that can lead to quick clinical decision-making. In some PHD projects, patients took their data to other clinicians. This was particularly true for caregivers of premature infants. These caregivers found that when ODL data were presented to clinicians who were not participating in the study, they weren’t interested in looking at the data. In part, this was because clinicians were not willing to look at the data in a more “raw” or unsynthesized form, since they did not have the web-based decisions support tools participating clinicians did have. In part, study participation may also explain motivation to use these data, and that is something that requires further exploration.

Finally, protocols to integrate the new tools and data that patients collect into the daily workflow need to be developed by clinics. Protocols can include: who is responsible for reviewing the data? (is it a clinician or an ancillary staff member?), how often will these data be reviewed?, and, finally, what steps will be needed to address alerts or “red flags?” (for instance, when data suggests a patient’s asthma is poorly controlled, or when a caregiver’s data shows a lack of wet diapers in a premature infant).

Project HealthDesign has done a great job opening the door to thinking about how smart technology and health tracking can impact the relationship between provider and patient and the potential for these data to be integrated into clinics. More work is yet to be done on the type of health experiences that would most benefit from collecting ODLs and data ownership in regards to clinical access to patient-generated data.