Reassign pending screening work safely

Use the lock–preview–apply–unlock workflow

Do not treat reassignment as an ordinary team edit. It changes study-level ownership and therefore uses a controlled sequence.

1. Open Change Team as an administrator. Reassignment is allowed only for an active project.
2. Review active screening sessions. Ask analysts and reviewers to close their FPS/FTR pages when practical.
3. Lock Screening. A lock blocks new decision submissions while you prepare and apply the change.
4. Select the exact phase, arm, and reviewer targets. Preserved targets are not rewritten.
5. Choose one distribution strategy. Configure pools, weights, studies, or ranges as required.
6. Generate Preview. Resolve every blocker and confirm before/after workload.
7. Apply Reassignment. The server validates that the preview is still current and commits the operation atomically.
8. Review the refreshed workload and receipt. Confirm the intended owners and counts.
9. Finish and Unlock. Unlock only after active sessions have closed and the new allocation is accepted.

Understand who can own each arm

The available cards and locked controls depend on the project’s screening configuration.

FPS and FTR have independent analyst pools

An eligibility pool defines who may receive pending work for one phase and arm. FPS Arm A, FPS Arm B, FTR Arm A, and FTR Arm B are separate pools. The people available during FTR may differ from those who performed FPS. FPS Unsure and decision conflicts wait for reviewer resolution before ordinary FTR processing. If both FPS arms agree on Include but their criteria differ, the study may enter FTR while its FPS criteria review remains pending. An administrator can also explicitly approve an eligibility override.

• An arm with zero pending work is automatically preserved and its editing controls are disabled.
• Inactive historical users may remain visible so their existing work can be reviewed and moved safely.
• Legacy projects without saved arm pools require explicit pool selection; the system does not guess the split.
• FTR work exists only for eligible studies, such as studies included after FPS and not excluded by the FTR/PDF workflow.
• In double review statistics, Arm A and Arm B totals are assignment totals. A study screened twice contributes once to each arm; do not add the arms and interpret the sum as unique studies.
• Duplicate secondary records may be reported separately from active assignable screening workload.

Choose scope, targets, and one strategy

The Phase selector controls whether the operation can affect FPS, FTR, or both. Each visible arm then has its own target switch.

For analyst arms, equal distribution automatically calculates near-equal contiguous blocks of pending studies in study-ID order; there is no block-size input. For example, 300 studies with three analysts in each arm form three 100-study blocks, while two Arm A analysts and three Arm B analysts can share 240 studies in six 40-study blocks. Each block has one preferred analyst per selected arm, and the two people on a double review study must differ. Arm A stays fixed to AI Screener in AI+Humans projects. Completed decisions and unselected arms remain with their existing owners, so the preview may show exceptions to a block's preferred owner. Reviewer work still follows the independent, conflict-aware assignment rules when it becomes required.

Selected versus preserved

• Every editable analyst arm and Reviewer Assignment opens in Preserved mode. Turn on only the targets approved for the current operation.
• Will reassign means pending work in that target can change according to the chosen strategy.
• Preserved means the target’s assignments and stored pool are not changed by that operation.
• Locked identifies an AI-owned target that administrators cannot edit.
• No pending work disables that arm because there is nothing eligible to redistribute.

Build targeted rules carefully

Selected Studies and Study ID Ranges change only the studies named by their rules. They do not redistribute every pending study in the arm.

• Ranges for the same phase and target arm cannot overlap.
• Different arm rules may refer to the same study, but their assigned analysts must be different.
• Only pending work is changed. A completed target is preserved even if its study ID falls inside the range.
• If all analyst arms for the selected phase are preserved and Reviewer Assignment is enabled, Target automatically becomes Reviewer and the Arm field is hidden.
• A range check is an early warning. The full Preview remains the authoritative project-wide validation.

Reviewer pools are selected now; reviewer work is created only when needed

The project reviewer list is an eligibility pool, not an assignment to every study. AccuScreener waits for the required FPS or FTR screening decisions, then assigns one active independent reviewer only when the outcome is Unsure or conflicted. In sole AI Screener projects, definitive AI decisions remain final but are also available to selected reviewers as optional quality-assurance samples.

Read the reassignment preview before Apply

Preview does not write to the database. It calculates the proposed result against the current project revision.

Resolve validation errors instead of bypassing them

Blocking errors appear in the preview panel and, when phase/arm information is available, on the relevant assignment card.

Apply atomically, verify, then unlock

Apply rechecks the preview hash and project revision. Study assignments, selected project pools, assignment revision, FTR statistics where applicable, and the reassignment receipt are committed in one transaction.

Common reassignment workflows

Confirm these points before unlocking

• check_circle The project is active and the administrator has authority to change its team.
• check_circle Screening is locked and active users have been asked to close their screening pages.
• check_circle The selected phase and arm switches match the approved staffing change.
• check_circle Preserved arms, completed work, and AI-owned arms are not expected to change.
• check_circle FPS and FTR eligibility pools reflect who is actually available for each phase.
• check_circle Double review study assignments use different people in Arm A and Arm B.
• check_circle Every reviewer is independent from both human analysts on the same study.
• check_circle Range checks show the intended pending count and no opposite-arm conflicts.
• check_circle Preview shows no orphan blockers, unexpected removals, or unresolved anomalies.
• check_circle Before/after owner totals and selected targets match the intended allocation.
• check_circle The Apply receipt and refreshed workload were reviewed.
• check_circle Active sessions are closed before Finish and Unlock is selected.

Common questions