What insights digital real-world data provide and where their limits lie
Digital therapy companions are increasingly being discussed in the pharmaceutical industry as a potential source of patient insights. At the same time, it often remains unclear what kind of insights these applications can realistically provide and what their added value is compared to established methods. There is a significant qualitative difference between simple usage statistics and robust statements about understanding of therapy or real-world care.
This article outlines under which conditions digital therapy companions become relevant for pharmaceutical questions. The focus is on real-world data from everyday medication use, the limits of their informative value, and the question of how qualitatively different solutions can be distinguished from one another.
What data actually arise in everyday life
Digital therapy companions are only meaningful if they are used regularly and have a clear connection to the therapy. Under these conditions, data are generated that reflect how patients handle their medication, initially without automatic interpretation.
For example, we can observe:
- which functions are used or ignored,
- when information is accessed,
- at which points users discontinue or return repeatedly,
- how usage patterns change over time.
These data are descriptive, not explanatory. They show that something happens, not why. Their benefit only arises when they are collected specifically along medical or care-related questions, for example regarding safe use, recurring uncertainties, or critical phases in the course of therapy.
Digital therapy companions therefore differ fundamentally in their insight value – depending on how they are designed, which data are systematically collected, and which post-processing and interpretation processes are applied to these data.
The quality spectrum of digital therapy companions
Digital therapy companions are not a homogeneous market. They differ significantly in their conceptual depth and therefore also in the quality of the insights that can be derived from them.
Simpler solutions often capture only activity data, such as log-ins, reminder confirmations, or general usage frequencies. This information allows conclusions about usage, but provides hardly any indication of causes, comprehension issues, or care-related questions.
In contrast, there are therapy companions that are developed along specific medical and content-related questions. In these applications, it is structurally defined which aspects of everyday therapy are to be observed – for example, how specific informational content is handled, recurring uncertainties, or changes in usage behavior across defined therapy phases.
The central difference lies less in the technology than in the professional concept. While simple applications generate data without making them interpretable, qualitatively sound approaches enable systematic analysis in a medical context.
This distinction is particularly important for pharmaceutical companies. It determines whether digital therapy companions are perceived as mere support tools or can serve as a reliable source of patient insights.
Real-world data as a complement, not as a substitute
Digital therapy companions are not primary instruments for generating statistically representative evidence in the sense of controlled studies or other investigations in artificial, highly regulated environments. Their contribution lies rather in the collection of real-world data: context-related, timely data from the actual course of therapy under everyday conditions.
In contrast to retrospective surveys, in which patients describe their behavior in hindsight, as well as to study settings in which processes are deliberately standardized and controlled, these data arise directly during use. This makes it possible to identify patterns that often remain hidden in traditional surveys: for example, recurring information gaps, systematic deviations from recommended processes, or typical declines in use over the course of therapy.
As part of a comprehensive RWE approach, digital therapy companions complement established methods rather than replace them. They do not provide causal evidence of effectiveness, but can offer valuable insights into real-world care and serve as a basis for further questions, hypotheses, or studies. Accordingly, EMA and FDA also primarily classify real-world data as a contextual basis for decision-making, not as a substitute for controlled evidence.
Relevance for pharmaceutical departments
If digital therapy companions are used along clearly defined questions, concrete, work-relevant insights can be derived from them:
| Department |
Concrete benefit |
|
Medical Affairs |
Indications of recurring comprehension issues; typical uncertainties in use; differences in understanding of therapy between patient groups |
|
Market Research & Analytics |
Reconstruction of actual usage paths, changes in behavior over time, comparison of different information formats |
|
Product strategy |
Identification of critical touchpoints in the course of therapy; indications of specific information needs; input for accompanying materials along the product life cycle |
These insights do not arise automatically, but from the targeted combination of data collection, professional questioning, and interpretation.
Digital education as part of the course of therapy
In addition to data collection, digital therapy companions often also take on an educational function. For assessing their benefit, it is less relevant that content is provided, but rather how patients engage with this content.
The following can be observed, among other things:
- whether certain content is accessed repeatedly,
- which formats are used or skipped,
- how usage changes after repeated exposure.
These patterns allow conclusions about where existing education is effective and where it reaches its limits. Digital therapy companions therefore do not provide a direct measurement of activation, but they do offer concrete indications of where support is actually used – or not. Particularly in the context of digital health literacy, it becomes clear that provision alone is not sufficient, but that use, understanding, and integration into everyday life are decisive.
👉 How digital education can actually contribute to the activation of patients in everyday life is explored in the article: “Promoting digital health literacy in everyday patient life.”
The value of the insights therefore lies in observing actual use, not in abstract activation concepts.
Why interpretation is more important than data depth
A common mistake in the evaluation of digital patient data is equating level of detail with insight. Even extensive datasets remain limited in their informative value if they are not professionally interpreted.
Interdisciplinary evaluation – incorporating medical, user-centered, and analytical perspectives – is necessary to meaningfully classify usage patterns and distinguish them from random effects or technical artifacts. For pharmaceutical companies, it is therefore less decisive how much data are collected, but rather how specifically and hypothesis-driven they are used.
Conclusion
Digital therapy companions do not deliver automatic patient insights. With clear objectives and methodological rigor, however, they can provide reliable indications of how patients actually manage their therapy.
The decisive factor is the quality of the concept. While simple solutions merely reflect activity, methodologically sound therapy companions enable a differentiated view of use, understanding, and information needs in everyday care. Digital therapy companions thus become a pragmatic instrument for pharmaceutical companies to review existing assumptions, refine communication strategies, and better support decisions along the product life cycle.
Accordingly, providers who understand digital therapy companions not as mere technology, but as a medically and methodologically sound instrument, and who consistently implement this perspective in design, evaluation, and application, gain particular relevance.
Sources
- Hibbard JH, Stockard J, Mahoney ER, Tusler M.
Development of the Patient Activation Measure (PAM): conceptualizing and measuring activation in patients and consumers.
Health Services Research. 2004;39(4 Pt 1):1005–1026. - Jordan S, Hoebel J.
Gesundheitskompetenz von Erwachsenen in Deutschland – Ergebnisse der HLS-EU-Studie.
Bundesgesundheitsblatt – Gesundheitsforschung – Gesundheitsschutz. 2015;58:942–950. - Chowdhury R, Khan H, Heydon E, et al.
Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences.
European Heart Journal. 2013;34(38):2940–2948. - Jankowska-Polańska B, Uchmanowicz I, Dudek K, Mazur G.
Relationship between patients’ knowledge and medication adherence among patients with hypertension.
Patient Preference and Adherence. 2016;10:2437–2447. - European Medicines Agency (EMA).
Real-world evidence framework to support EU regulatory decision-making. 2023.
https://www.ema.europa.eu/en/human-regulatory-overview/research-development/scientific-guidelines/clinical-efficacy-safety-guidelines/real-world-evidence - S. Food and Drug Administration (FDA).
Framework for FDA’s Real-World Evidence Program. 2021.
https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence