Detection of Behavioral Anomalies in Medication Adherence Patterns Among Patients With Serious Mental Illness Engaged With a Digital Medicine System
Bibliographic record
Abstract
BACKGROUND: Adherence to medication is often represented in the form of a success percentage over a period of time. Although noticeable changes to aggregate adherence levels may be indicative of unstable medication behavior, a lack of noticeable changes in aggregate levels over time does not necessarily indicate stability. The ability to detect developing changes in medication-taking behavior under such conditions in real time would allow patients and care teams to make more timely and informed decisions. OBJECTIVE: This study aims to develop a method capable of identifying shifts in behavioral (medication) patterns at the individual level and subsequently assess the presence of such shifts in retrospective clinical trial data from patients with serious mental illness. METHODS: We defined the term adherence volatility as "the degree to which medication ingestion behavior fits expected behavior based on historically observed data" and defined a contextual anomaly system around this concept, leveraging the empirical entropy rate of a stochastic process as the basis for formulating anomaly detection. For the presented methodology, each patient's evolving behavior is used to dynamically construct the expectation bounds for each future interval, eliminating the need to rely on model training or a static reference sequence. RESULTS: Simulations demonstrated that the presented methodology identifies anomalous behavior patterns even when aggregate adherence levels remain constant and highlight the temporal dependence inherent in these anomalies. Although a given sequence of events may present as anomalous during one period, that sequence should subsequently contribute to future expectations and may not be considered anomalous at a later period-this feature was demonstrated in retrospective clinical trial data. In the same clinical trial data, anomalous behavioral shifts were identified at both high- and low-adherence levels and were spread across the whole treatment regimen, with 77.1% (81/105) of the population demonstrating at least one behavioral anomaly at some point in their treatment. CONCLUSIONS: Digital medicine systems offer new opportunities to inform treatment decisions and provide complementary information about medication adherence. This paper introduces the concept of adherence volatility and develops a new type of contextual anomaly detection, which does not require an a priori definition of normal and allows expectations to evolve with shifting behavior, removing the need to rely on training data or static reference sequences. Retrospective analysis from clinical trial data highlights that such an approach could provide new opportunities to meaningfully engage patients about potential shifts in their ingestion behavior; however, this framework is not intended to replace clinical judgment, rather to highlight elements of data that warrant attention. The evidence provided here identifies new areas for research and seems to justify additional explorations in this area.
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 | 0.001 | 0.010 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".