How do Project Baseline researchers use participant survey data?
Health isn't all about the numbers in your chart when you visit the doctor. One of the ways researchers can get a holistic view of health, across different populations, is by designing scientifically valid surveys that can help answer research questions. We sat down with biostatistician Sarah Short to learn more.
Why is it important and valuable to get self-reported data from participants for research?
Historically, researchers focused more on clinical, objective data because there was less trust in the quality and usefulness of subjective, self-reported information. As research has matured, we've taken a more patient-centered focus, which means taking people's perceptions and experiences into account when evaluating the effectiveness of treatments. There are important aspects of patient experience that cannot be captured through clinical measurements like temperature, blood pressure, or lab results.
For example, when we're studying a new treatment, we can measure how effective it is clinically, but that doesn't tell us about patient satisfaction with the treatment. If we discount that, we run the risk of creating something that doesn't fit into people's lifestyles. Also, with remote research increasing, it's much easier to collect such data than it was when everything had to be collected at a physical study site.
What were some of the challenges associated with surveying people before?
Having a human interviewer or doctor in the room (or even on the phone), asking the questions, can introduce bias. Electronic surveys can lead to more honest results. They can also help us collect information about people who may not be as easy to include in typical trials, including people who face challenges with mobility issues or access to care.
One way we're trying to use more inclusive, digital means of conducting research to reach underrepresented groups is through Research Goes Red, our collaboration with the American Heart Association. Heart disease is thought to be more of a problem for men, and women may not be aware of their risks. That's an important gap to fill. We're focusing on learning about the community's motivations and health concerns so researchers can get more insight on how to help.
One challenge we've always had, even with more remote research opportunities, is the digital divide. Many groups don't have reliable access to the internet. While research is growing more inclusive, there's still a lot of work to be done to engage traditionally underserved groups.
What's the process for creating a scientific survey? How do you ensure the results are valid enough to draw conclusions for research?
It's definitely not as simple as putting a few interesting questions together and sending out a poll! Some of the surveys we use are validated instruments – surveys that have been developed to serve a specific purpose and which have been rigorously evaluated for reliability and validity. However, to answer more specialized research questions, we often create our own surveys. This involves bringing together clinicians, researchers, survey design experts, statisticians, and user experience teams to collaborate on clear, unbiased questions that will lead to the right kind of data to answer your research question. This can be inherently challenging when you're conducting experimental research and trying to uncover new markers of health and disease.
It's also important to get surveys right the first time. Changing survey questions makes it more difficult to analyze them over time. Some surveys might measure more clinical information – for example, changes in symptoms – while others can be more qualitative. Both types of data are valuable to researchers.
Can you give an example of scientifically interesting insights or research we've gleaned from scientific surveys?
As part of the Baseline COVID-19 Research Project, survey data collection serves multiple purposes. With the intense pace of research to find vaccines and treatments – and save lives – it's important to find the right people to join studies. We're able to reach a broad sample of people and learn about their symptoms, which can help with pre-screening for research opportunities. The COVID-19 Research Project can give us a lot of interesting longitudinal, or long-term, population health data. For example, our surveys around social distancing and behaviors reveal that mask use is increasing in our cohort over time. This can help us understand how public health information is understood on an aggregate level.
We can also see patterns over time around people's emotional experiences and coping mechanisms. We found an association between hours of exercise and improvements in mood. This isn't necessarily a surprising finding, but it shows that even in this time, there are activities that can help reduce stress. Access to data can help us pinpoint strategies for addressing the virus' impact and provide help where people need it. Based on the findings from our wellbeing surveys, we recently hosted a webinar specifically focused on coping strategies for the pandemic drawing from research data.
Why is it important to return data from research?
It can be tricky to return results when we're asking people subjective questions about their feelings and experiences, particularly when asked repeatedly over time. We have to consider whether sharing results will influence further responses down the road. This is one reason why results usually are not shared until the end of a study. However, it's important for us to prioritize navigating those challenges to identify data we can share without introducing bias. We value our participants, not just the data they provide. This is one way we show participants that they are valued, and that their contributions have a meaningful impact.
To learn more about participating in research through Project Baseline – including our COVID-19 focused efforts – visit projectbaseline.com.