Understanding the Long Term Effectiveness of Varying Doses of a New Therapy for Major Depression


A large biopharmaceutical client conducted a Phase II, 6-week, randomized, placebo-controlled, dose-finding trial to evaluate the efficacy, tolerability and safety of a new oral medication in patients with major depressive disorder who were not responding to antidepressant therapy. The trial plans to enroll 400 patients. About 160 are enrolled currently and 125 have consented to tokenizing their data. The study has six arms, four of which included different doses of the new therapy, one arm that included an adjunctive therapy and one placebo-controlled population.


The client wanted to understand the long-term impact of the various doses of therapy on the rate of depression episodes to identify if any of the doses offered longer-term benefits. The trial population was relatively small so they questioned whether they could find enough patients in a single data set that would overlap with their trial population.


Datavant helped the client tokenize the trial and run overlaps with ten different claims and personally identifiable information collected and used to generate tokens included first name, last name, date of birth, gender, address, and zip code.

  • Linking claims enabled the Sponsor to capture healthcare utilization trends, such as doctor visits, not captured in the trial. Using Datavant, the client overlapped patient-level trials data with claims data sets from different data partners in our ecosystem. The client can combine these different claims sources while also de-duplicating them to ensure unique patient records are linked.
  • Linking consumer data created opportunities to examine how differences in income, employment status, family composition, race, ethnicity, and other social determinants correlate to depressive episodes and response to therapy.


The client found that for consumer data partners, almost 70% of their trial patients could be found in partner datasets. For claims data partners, the client found that between 70-90% of their trial patients were found in claims data sets. The client will prioritize partnerships with those partners that had the highest number of overlapping patients and analyze the linked data to look at long-term effectiveness and safety, and to conduct exploratory analyses on patient sub-cohorts using consumer data variables.

Why Datavant

50+ trials already use Datavant tokens to enable linking to real-world data

Datavant connects the largest health data ecosystem with more than 500 real-world data partners

Datavant tokenization can be seamlessly embedded into trial operation workflows for easy implementation by site administrators

To view this case study, please fill out the form below.