Evidation Characterizing Life Events.pdf

Characterizing life events through self-report and wearable data

Traditional RWE leaves gaps in individuals’ health journeys

As the life sciences industry becomes more competitive and personalized, asset differentiation and market success often depend on identifying the factors impacting health beyond the confines of clinical settings. Real-world evidence (RWE) plays an important role in providing a holistic view of individual experiences, and healthcare industry stakeholders are increasingly requiring RWE of an intervention’s impact on health and quality of life to make approval, reimbursement, and prescribing decisions.

To meet these expectations, many life science companies rely on data from a snapshot in time such as electronic health records (EHRs) or infrequently collected patient-reported outcomes.

This information lacks the ability to show:

Table of Contents

Filling gaps with more complete, continuous data

Compelling RWE that more deeply captures what is happening in a person's life can be achieved by integrating conventional real-world data (RWD) sources (e.g., EHR, prescription records, claims) with data collected directly from people in their everyday lives, such as:

Traditional RWD paired with direct from patient data builds a more complete health journey

More comprehensive data collected directly from individuals, complemented by snapshot-in-time information, can inform evidence-based discussions around:

Pairing wearable data with surveys and PROs, for example, can uncover the effects of injury, disease, or stressful events on daily life that may not be evident even to the person itself.

Differences between self-reported sleep quality and wearable-reported sleep quality

Using an asthma survey paired with wearable sleep data from members on the Evidation platform, we detected discrepancies between the perceived effects and actual effects of asthma on sleep.

Survey: In the past 4 weeks how often did you wake up at night due to your asthma symptoms?

The respondents who reported they did not wake up at all due to asthma symptoms actually had the least amount of deep sleep.

Wearable: Minutes of deep sleep per night

This is a powerful example of how passively collected wearable data can supplement actively collected, self-reported feedback to:

In the following sections, we present examples of active and passive data collection from individuals that provides a more complete understanding of their health journeys.

Quantifying the impact of major life events on health and outcomes

Between June 10, 2023 and July 24, 2023, we offered Evidation Members the opportunity to complete a survey about major life events that had occurred over the previous five years. When members indicated they had experienced one of the events, we asked them to provide the date it occurred.

Survey information:

142,759

MEMBERS

Wearable data (steps, sleep, and heart rate):

109,554

OF THOSE MEMBERS

We explored 5-year trends in the occurrence of major life events for the entire sample to illustrate their prevalence. For those who experienced an event and had wearable data, we also compared the health activity metrics before and after each event to quantify the event’s impact.

Combination of data from the Evidation platform for this analysis:

Aggregated data from a large population such as this can be used not only to understand what events are happening in everyday life but also to establish a “normal” pattern, response to a treatment, or recovery from an illness or injury. Individual data can then be compared against the “normal” pattern to detect atypical responses that might benefit from additional support.

Examples of major life events from the survey and their impact on activity:

  1. Major illness or injury following hospitalization
  2. Being laid off or fired from a job

Mapping recovery after a major illness or injury requiring hospitalization

Challenge

Characterizing both collective and individual recovery journeys, including factors predictive of positive outcomes.

Commercial implication

Wearable data, combined with periodic self-reported information, provide an indication of a “normal” recovery and the ability to identify individual outliers.

Setting the scene

The trend analysis of the survey responses showed that the number of hospitalizations experienced by Evidation Members steadily increased from 2018 to 2022, which could have included COVID-19-related admissions.

Individuals who experienced a hospitalization

2023 only includes ~6 months of data

Insights from the wearable data

The wearable data show the changes in activity (steps), heart rate, and sleep duration before and after hospitalization (indicated by Day 0 on the x axis).

The post-hospitalization data provide insights into physiological responses and the time to resume baseline activity during recovery.

Other potential data sources that can be integrated in the Evidation platform for greater insights:

How these data could be used:

Real-world example of predicting post-surgical recovery using wearable data

Population:

1,324 individuals who underwent lower limb surgery

Challenge:

Lack of long-term individual baseline data for an accurate and objective assessment of functional post-surgical recovery

Approach:

Results

For more information, read the publication.

Identifying the health-related effects of losing one’s job

Challenge

Identifying individuals experiencing changes in SDOH, including income and access to healthcare, that could influence their health outcomes.

Commercial implication

Detecting changes to an individual’s social, behavioral, and environmental context facilitates help us understand what places someone at risk for treatment non-adherence and other barriers to positive outcomes.

Setting the scene

The trend analysis of the survey responses showed that a higher number of respondents reported being laid off or fired from their job in 2020. This could be related to the drastic effects of COVID-19 on both temporary and permanent business closures and the subsequent effects on the labor market.

Individuals who experienced a job loss

2023 only includes ~6 months of data

Insights from the wearable data

The wearable data show the changes in activity (steps), heart rate, and sleep duration before and after being fired or laid off (indicated by Day 0 on the x axis).

In addition to introducing chronic stress and disrupting health-promoting behaviors, losing one’s job could disrupt regular healthcare by impacting both income stability and availability of insurance benefits.

The impacts could be:

Other potential data sources that can be integrated in the Evidation platform for greater insights:

How these data could be used:

Gaining insights into health outside of clinic walls

Complementing traditional RWD sources, such as claims, EHRs, and patient registries, with objective wearable data, self-reported symptoms, and quality of life measures provides a novel view into individuals’ health outside of clinic walls. Coupling data continuously across the health journey (e.g., before a diagnosis, during an illness, post-treatment) with knowledge of major life events helps: