Engagement With Digital Adherence Technologies as Measures of Intervention Fidelity Among Adults With Drug-Susceptible Tuberculosis and Health Care Providers: Descriptive Analysis Using Data From Cluster-Randomized Trials in Five Countries
Bibliographic record
Abstract
Background: Digital adherence technologies (DATs) are promising tools for supporting tuberculosis (TB) treatment. DATs can serve as reminders for people with TB to take their medication and act as proxies for adherence monitoring. Strong engagement with DATs, from both the person with TB and health care provider (HCP) perspectives, is essential for ensuring intervention fidelity. The Adherence Support Coalition to End TB (ASCENT) project evaluated 2 types of DATs, pillboxes and medication labels (99DOTS), in cluster-randomized trials across 5 countries. Objective: This study aims to investigate participant and HCP engagement with DATs for TB treatment, stratified by DAT type and country. Methods: This study is a subanalysis of data generated through the ASCENT trials, which enrolled adults with drug-susceptible TB. A digital dose was defined as either a pillbox opening (for pillbox users) or a dosing confirmation SMS text message sent by the participant (for label users), both of which were recorded on the adherence platform. Descriptive analysis was used to provide an overview of dose-day outcomes. DAT engagement was assessed from both participant and HCP perspectives. To enhance participant engagement, we summarized the frequency of digital engagement overall and by treatment phase, as well as the frequency of consecutive days without engagement. For HCP engagement, we summarized the frequency of doses added manually, the number of days between the actual dose day and when a manual dose was added, and instances of consecutive manual dosing lasting more than 3 and more than 7 days, where doses were added more than 1 week after the dose day. Results: Of the 9511 participants included, 6719 (70.64%) were using the pillbox, 3544 (37.26%) were female, and the median age was 40 years. Across DAT types, there were 1,384,879 dose days, with 973,876 (70.32%) contributed by pillbox users. Of all dose days, 1,165,195 (84.14%) were recorded as digital, 156,664 (11.31%) as manual, 59,045 (4.26%) had no information, and 3975 (0.29%) were confirmed as missed. Digital dosing decreased slightly from the intensive to the continuation phase. The percentage of digital dose days was higher among pillbox users (851,496/973,876, 87.43%) compared with label users (313,699/411,003, 76.33%). Among label users, manual dosing was most common in the Philippines (37,919/171,786, 22.07%) and least common in Tanzania (11,108/76,231, 14.57%). Among pillbox users, manual dosing was most common in the Philippines (24,015/208,130, 11.54%) and Ukraine (13,209/111,901, 11.80%). Overall, 512 out of 2792 (18.34%) label users and 588 out of 6719 (8.75%) pillbox users experienced a run of more than 7 consecutive nondigital dose days that were resolved more than 1 week after the dose day. The highest occurrence was observed in the Philippines (368/1142, 32.22%, for label users and 224/1351, 16.58%, for pillbox users). Conclusions: There was considerable variation in DAT engagement across countries and DAT types, reflecting differences in how the intervention was implemented. Further refinement of the intervention and improvements in its delivery may be necessary to enhance outcomes.
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 | 0.145 | 0.187 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.010 |
| Bibliometrics | 0.007 | 0.007 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 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".