Impact of AllazoEngine-Targeted Interventions on Medication Adherence: Repeated Measures Difference-in-Differences Analysis
Bibliographic record
Abstract
Background: AllazoHealth utilizes predictive analytics to improve medication adherence by targeting patients whose behavior can be changed with intervention programs. The AllazoEngine utilizes Rx claims, previous intervention data, and demographics to predict future adherence, to prioritize the patients whose behaviors can be changed, and to select the intervention channel and messaging most effective for each individual patient. Blue Cross and Blue Shield of North Carolina commissioned AllazoHealth’s predictive analytics and separately commissioned medication intervention delivery services for this adherence program. Objective: This study aimed to evaluate the effectiveness of the AllazoEngine and targeted interventions to improve medication adherence. Methods: This was a double-blind, randomized controlled trial (RCT) focused on RAS antagonists, oral anti-diabetics, and Statins. Patients were randomized to receive no intervention, traditional non-Allazo-targeted interventions, or interventions targeted by the AllazoEngine. All interventions consisted of live calls, direct mail to patients, and faxes to prescribers. Patients were defined as adherent in accordance with Medicare Star ratings methodology if their proportion of days covered (PDC) was greater than 80%. Patients’ adherence status in 2015 was compared to their adherence status in 2016 after the intervention period. Difference-in-Differences (DiD) analysis was used to compare the effect of each intervention method. Statistical significance was set to 10%. Results: The primary study population consisted of 14,377 controls, 5,423 traditional non-Allazo targeted-intervention patients and 24,527 Allazo targeted intervention patients. Patients had comparable characteristics at baseline and comparable decrease in medication adherence in the pre-intervention observation period across the intervention groups. Non-Allazo-targeted interventions did not statistically improve the likelihood of adherence. Patients who received Allazo-targeted interventions performed statistically better than both the non-Allazo targeted group and control group (P=.06 and .03). Assuming net positive uplift from non-Allazo interventions, Allazo interventions accounted for 7.7 times the per-patient uplift in adherence compared to non-Allazo interventions. Conclusions: Due to the specific study design of not including new patients, adherence decreased in each intervention group over the intervention period. However, the decrease was significantly less for those in the Allazo group compared to both the non-Allazo and control groups.
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 | 0.014 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.004 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".