What This Case Supports, and What It Doesn't

What the case supports: the baseline data show a real, specific bottleneck, accessioning delay concentrated at two predictable arrival peaks, not a vague sense that the laboratory is slow. The two-week PDSA test moved the target metric in the predicted direction while holding the balancing measures steady, and it surfaced one real, unpredicted complaint about handoff timing. Act, in this case, was adapt: keep the countermeasure, adjust the specific break time, and extend the test to a full month with weekend coverage before adopting it laboratory-wide.

What the case does not support: two weeks of data from one countermeasure at one laboratory do not prove that staggered breaks fix turnaround time everywhere. Outside results do not set this laboratory’s target or determine its next step; use the local baseline, the peak-accession constraint, and the balancing limits defined for the pilot. A favorable indicator alone is not sufficient without reviewing its unintended effects.

What depends on method and local policy: the exact 45-minute stat BMP target, which tests carry a stat designation, the quality-indicator threshold that triggers mandatory root-cause investigation, the staffing model and how peak coverage is staggered, and the balancing-measure thresholds that define how much overtime or how many missed breaks are acceptable are all decisions the laboratory director and local quality program set, not values fixed by any Lean or Six Sigma method itself. A DPMO or Sigma figure is only meaningful once the laboratory has stated its own defect and opportunity definitions, and a QC plan is still built from the measuring system, environment, clinical application, regulation, and patient risk, not from a Sigma value alone.

Hold every improvement to the same test: did the target measure move as predicted, and did anything else get worse? If either answer is unclear, the cycle is not finished.