GalenOps prod · 0.1.0

Agent

Dropout risk scored

Horizon 2 retention_agent

Active patients scored; at-risk cohort proposed for outreach.

Escalates: Participants above the dropout-risk threshold, with proposed outreach a coordinator approves before anyone is contacted — to the study coordinator.

Agents

Runs

3

Escalations

1

Failures

0

Avg duration

4.4 s

Literature & benchmarks

  • Dropout-risk model — missed visits, overdue check-ins and AE burden weighted into a per-participant score
  • Human-approved outreach only; the agent never contacts a participant

Run history (3 across all studies)

Newest first, 50 per page. Expand a row for the recorded reasoning and evidence.

Expand When Study Outcome Model Duration Reasoning
2026-09-29 23:42 UTC A Multicenter, Randomized, Open-Label Phase 2b Study to Investigate the Preliminary Efficacy and Safety of INNO-206 (Doxorubicin-EMCH) Compared to Doxorubicin in Subjects With Metastatic, Locally Advanced, or Unresectable Soft Tissue Sarcoma completed claude-sonnet-5 4.2 s This is a retrospective review of a completed study (INNO-206 vs Doxorubicin, soft-tissue sarcoma, Phase II), with 0 active patients and 0 patients scored, so no live dropout risk scoring applies. Among 126 ever-consented subjects, 20 withdrew, giving a withdrawal density of 15.9%, which is below the 25% red-flag threshold cited in docs/intent/lit_review_synthesis.md (heuristic proxy, no validated cutoff); accordingly withdrawal_density_flag is false. Median time to withdrawal was 60.5 days. Top disposition reasons were WITHDRAWAL BY SUBJECT (9), PROTOCOL VIOLATION (6), OTHER (3), and PHYSICIAN DECISION (2). Cited literature (Dombernowsky et al. 2019, Trials 20:708; Zhang & DuBois 2023, Cancer Med 12:5517) notes retention/staffing issues can compound accrual-driven termination risk, but no such escalation is indicated here. No review was opened (review_opened: false), so no human decision is required at this time.
2026-09-29 21:13 UTC A Phase III, Randomized, Double-Blind Study of Torivumab Plus Chemotherapy Versus Placebo Plus Chemotherapy in Previously Untreated Advanced Non-Small Cell Lung Cancer completed claude-sonnet-5 4.6 s This run scored 269 active patients with a heuristic logistic model (coefficients are priors, not fitted to this study). Since fewer than 5 patients reached p≥0.5, the selection rule fell back to the top decile, flagging 27 patients as at elevated dropout risk. The highest-risk patient, SITE013-0226, has p=0.472, driven by 1 grade ≥3 AE, 1 serious AE in the last 60 days, and treatment stopped 40 days ago. Cohort-wide withdrawal density is 1.6% (5/317), well below the 0.25 heuristic red-flag threshold from docs/intent/lit_review_synthesis.md, so the withdrawal_density_flag is false. Per Zhang & DuBois 2023 (Cancer Med 12:5517), retention failure contributes to 34.5% of terminations, underscoring why this monitoring matters. The review task is already pending (review_opened: false, status "already_pending"). The human reviewer is asked to decide whether outreach should be initiated for the 27 flagged patients, particularly the top-ranked SITE013 cases.
2026-09-19 08:59 UTC A Phase III, Randomized, Double-Blind Study of Torivumab Plus Chemotherapy Versus Placebo Plus Chemotherapy in Previously Untreated Advanced Non-Small Cell Lung Cancer escalated human claude-sonnet-5 4.3 s The retention model scored 269 active patients using a heuristic logistic function (coefficients described as "heuristic priors from the retention literature, not fitted to this study's outcomes"). Since no patient reached the 0.5 threshold, the top decile was used per the selection rule "p>=0.5 or top decile (with at least one patient-level signal)," flagging 27 patients as at elevated dropout risk. Highest risk is SITE013-0226 at p=0.472, driven by 1 grade ≥3 AE, 1 serious AE in the last 60 days, and treatment stopped 40 days ago. Cohort-level withdrawal density is 1.6% (5/317), well below the 0.25 red-flag threshold from the lit-review heuristic, so withdrawal_density_flag is false. Citations reference Dombernowsky et al. 2019 and Zhang & DuBois 2023 on retention/accrual risk. A review task was opened; the human is asked to decide whether to approve proactive outreach to the flagged patients.