Feasibility of Collecting and Linking Digital Phenotyping, Clinical, and Genetics Data for Mental Health Research: Pilot Observational Study
Bibliographic record
Abstract
Background: Digital phenotyping-the use of digital data to measure and understand behavior and internal states-shows promise for advancing predictive analytics in mental health, particularly when combined with other data sources. However, linking digital phenotyping data with sources of highly sensitive clinical or genetic data remains rare, primarily due to technical, ethical, and procedural challenges. Understanding the feasibility of collecting and linking these data types is a critical first step toward developing novel multimodal datasets. Objective: The Mobigene Pilot Study examines the feasibility of collecting smartphone-based digital phenotyping and mental health data and linking it to genetic data from an existing cohort of adults with a history of depression (ie, the Australian Genetics of Depression Study). This paper aims to describe (1) rates of study uptake and adherence; (2) levels of adherence and engagement with daily mood assessments; (3) willingness to take part in similar research; and (4) whether feasibility indicators varied according to mental health symptoms. Methods: Participants aged 18-30 years with genetic data from the Australian Genetics of Depression Study were invited to participate in a two-week digital phenotyping study. They completed a baseline mental health survey and then downloaded the MindGRID digital phenotyping app. Active data from cognitive, voice, and typing tasks were collected once per day on days 1 and 11. Daily momentary assessments of self-reported mood were collected on days 2-10 (once per day for 9 days). Passive data (eg, from GPS, accelerometers) were collected throughout the two-week period. A second mental health survey was then completed after two weeks. To measure feasibility, we examined metrics of study uptake (eg, consent) and adherence (eg, proportion of completed momentary assessments), and willingness to participate in similar future research. Pearson correlations and t tests explored the relationship between feasibility indicators and mental health symptoms. Results: Of 174 consenting and eligible participants, 153 (87.9%) completed the baseline mental health survey and 126 (72.4%) provided data enabling linkage of genetic, self-report, and digital data. After removal of duplicates, we found that 100 (57.5%) of these identified as unique participants and 69 (39.7%) provided complete post-study data. A small proportion of participants dropped out prior to completing the baseline survey (21/174, 12.1%) or during app-based data collection (31/174, 17.8%). Participants completed an average of 5.30 (SD 2.76) daily mood assessments. All 69 (100%) participants who completed the post-study surveys expressed willingness to participate in similar studies in the future. There was no significant association between feasibility indicators and current mental health symptoms. Conclusions: It is feasible to collect and link multimodal datasets involving digital phenotyping, clinical, and genetic data, although there are some methodological and technical challenges. We provide recommendations for future research related to data collection platforms and compliance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.042 | 0.067 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".