2026 ECNP Congress

PS04-4096

Mapping dimensional symptom trajectories in bipolar disorder via digital phenotyping to enable targeted early interventions

Fabian Huth — C. Bittendorf, E.M. Langner, E. Mühlbauer, M. Bauer, A. Pfennig, V. Ludwig, U. Ebner-Priemer · Carl Gustav Carus University Hospital, Department of Psychiatry and Psychotherapy, Dresden, Germany

2026 ECNP Congress

Introduction

Bipolar disorder involves symptom fluctuations across euthymic, prodromal, and episodic phases [1]. Digital phenotyping often focuses on binary episode onset, overlooking potential dynamic shifts that can help define the transition from stability to relapse [2]. Consequently, subtle trajectories during euthymic and prodromal phases remain under-detected. Seeking to refine early detection, the BipoSense study tracked bipolar patients with biweekly interviews [3,4]. The aim of the current study was to investigate which dimensional affective symptoms can be captured by digital phenotyping across illness phases.

Methods

We analyzed 12-month longitudinal data from the BipoSense study (N=29 individuals with bipolar disorder). Clinical status and dimensional symptom severity were assessed biweekly via semi-structured interviews. To minimize noise from floor effects, only items with a standard deviation ≤5% of the scale range were included (YMRS items 1-7; MADRS items 1-9; and total scores). Continuous behavioral data (GPS-based fast mobility, self-reported sleep, outgoing calls) were collected via the movisensXS app. Associations were tested using multilevel models. P-values were adjusted via the Holm-Bonferroni method, applied independently for each sensing predictor per clinical outcome.

Results

Regarding manic psychopathology (YMRS), shorter sleep duration predicted increased mood, activity, sexuality and speech during acute mania (p <.001), while its association with thought disorder emerged exclusively during the prodrome (p < .001), and the link to increased sexual interest was also significant in euthymia (p < .05). Increased fast mobility predicted elevated mood and decreased sleep need only during acute mania (p < .001). Regarding communication, increased outgoing calls predicted less elevated motor activity/mood in the manic prodrome (p < .01). During acute mania, increased calls were linked to elevated mood, but with less pressured speech and thought disorder (p < .001). Regarding depressive symptoms (MADRS), significant associations emerged only during acute episodes. Reduced calls predicted anhedonia and pessimism (p < .01), reduced fast mobility linked to appetite loss (p < .05), and hypersomnia correlated with emotional blunting (p < .01).

Discussion

Our analyses indicate that symptom trajectories and their predictability by digital phenotyping may vary across euthymic, prodromal and illness phases only in mania. Here, sleep loss was associated with different symptoms across all phases, while mobility predicted symptoms only in acute weeks. Further, we observed a “behavioral switch” regarding social communication: during the prodrome, increased outgoing calls were associated with less elevated mood, whereas the acute episode showed a synchronization of outgoing calls and euphoria. Thus, manic mood and social behavior might be decoupled before episode onset. Conversely, in depression, significant associations only emerged during the acute state. These findings suggest that sensing markers capture stage-dependent symptom clusters. This granularity could help develop individualized precision interventions, such as social rhythm stabilization, to target specific early risk warnings and interventions at the symptom level and at right time. Future research should prioritize subject-level modeling in larger longitudinal cohorts to validate these symptom-level trajectories.

References

  1. Van Meter, A. R., Burke, C., Youngstrom, E. A., Faedda, G. L., & Correll, C. U. (2016). The Bipolar Prodrome: Meta-Analysis of symptom prevalence prior to initial or recurrent mood episodes. Journal of the American Academy of Child & Adolescent Psychiatry, 55(7), 543–555.
  2. Faurholt-Jepsen, M., Busk, J., Bardram, J. E., Stanislaus, S., Frost, M., Christensen, E. M., Vinberg, M., & Kessing, L. V. (2023). Mood instability and activity/energy instability in patients with bipolar disorder according to day-to-day smartphone-based data – An exploratory post hoc study. Journal of Affective Disorders, 334, 83–91.
  3. Ebner-Priemer, U. W., Mühlbauer, E., Neubauer, A. B., Hill, H., Beier, F., Santangelo, P. S., Ritter, P., Kleindienst, N., Bauer, M., Schmiedek, F., & Severus, E. (2020). Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders. International Journal of Bipolar Disorders, 8(1), 35.
  4. Ludwig, V. M., Bittendorf, C. A., Reinhard, I., Neubauer, A. B., Mennigen, E., Mühlbauer, E., Severus, W. E., Bauer, M., & Ebner-Priemer, U. W. (2025). Early warning signals of bipolar relapse: Investigating critical slowing down in smartphone data. Journal of Affective Disorders, 391, 119972.