<?xml version="1.0" encoding="UTF-8"?>
<opml version="2.0">
  <head>
    <title>073-02 Incident investigation, diagnostic and medication error, quality measurement, and hazard analysis</title>
    <ownerName>Integrated Medical Foundations</ownerName>
  </head>
  <body>
    <outline text="Safety investigation and quality measurement">
      <outline text="Purpose of safety investigation">
        <outline text="Reconstruct why decisions seemed reasonable"/>
        <outline text="Find lost chances for detection or recovery"/>
        <outline text="Neither exoneration nor blame by default"/>
        <outline text="Evidence for fair accountability and redesign"/>
      </outline>
      <outline text="Immediate safety first">
        <outline text="Treat the patient, prevent further exposure">
          <outline text="Quarantine products, check similar patients"/>
        </outline>
        <outline text="Investigation must not delay treatment"/>
        <outline text="Never alter the original record">
          <outline text="Later notes dated and transparent"/>
        </outline>
        <outline text="Preserve time-sensitive evidence and logs"/>
        <outline text="Open disclosure: facts, regret, next steps">
          <outline text="Uncertainty is no reason for silence"/>
        </outline>
      </outline>
      <outline text="Reliable incident chronology">
        <outline text="Separate times of action, record, result, harm"/>
        <outline text="Electronic timestamps need interpretation">
          <outline text="Delayed entry, copied notes, clock differences"/>
        </outline>
        <outline text="Interview separately with psychological safety">
          <outline text="Keep hindsight out of earlier decisions"/>
        </outline>
        <outline text="Map care across settings and professions">
          <outline text="Transitions expose unowned tasks"/>
        </outline>
        <outline text="Barrier existed, functioned, or was bypassed"/>
      </outline>
      <outline text="Root-cause analysis without one root">
        <outline text="Factors from patient to external policy"/>
        <outline text="Asking why can stop at a convenient human"/>
        <outline text="Independent, multidisciplinary team"/>
        <outline text="Recommendations need owner, deadline, measures">
          <outline text="Otherwise documentation, not prevention"/>
        </outline>
        <outline text="Poor policy compliance is description, not cause"/>
        <outline text="Causal factor would plausibly change outcome">
          <outline text="Check whether another pathway still harms"/>
        </outline>
      </outline>
      <outline text="Diagnostic error across the pathway">
        <outline text="From access and symptoms to follow-up"/>
        <outline text="Ignored disconfirming evidence makes it unsafe">
          <outline text="Track first reasonable diagnostic moment"/>
        </outline>
        <outline text="Results must reach an explicit owner"/>
        <outline text="Distributed process, not one wrong thought"/>
        <outline text="Feedback from autopsy, returns, complaints"/>
      </outline>
      <outline text="Precise medication incident analysis">
        <outline text="Reconstruct indication, order, route, rate"/>
        <outline text="Units, decimals, defaults, pump libraries"/>
        <outline text="Ambiguous order versus prescribing error"/>
        <outline text="Double-checks weak when not truly independent">
          <outline text="Shared assumption or confirmed calculation"/>
        </outline>
        <outline text="Outcome severity does not decide importance">
          <outline text="No-harm large errors still deserve learning"/>
        </outline>
      </outline>
      <outline text="Measuring quality">
        <outline text="Structure, process, outcome, balancing measures">
          <outline text="Balancing detects unintended consequences"/>
        </outline>
        <outline text="Operational definitions fix numerator, denominator">
          <outline text="Percentages improve by excluding high risk"/>
        </outline>
        <outline text="Run charts show shifts and trends"/>
        <outline text="Control charts: common versus special cause">
          <outline text="Reacting to every fluctuation destabilises"/>
        </outline>
        <outline text="Risk adjustment can hide inequitable care">
          <outline text="Stratify and read adjusted and unadjusted"/>
        </outline>
      </outline>
      <outline text="Diagnose the problem before the solution">
        <outline text="Define the gap with baseline and observation"/>
        <outline text="Process maps, Pareto, cause-and-effect diagrams">
          <outline text="Tools generate questions, not causation"/>
        </outline>
        <outline text="Observe real work with patients and staff"/>
        <outline text="Prioritise severity, frequency, inequity">
          <outline text="Rare catastrophes justify strong controls"/>
        </outline>
      </outline>
      <outline text="Proactive hazard analysis">
        <outline text="Failure-mode and effects analysis, step by step">
          <outline text="Risk-priority scores give false precision"/>
        </outline>
        <outline text="Simulation tests teams and environments">
          <outline text="In situ exposes latent safety threats"/>
        </outline>
        <outline text="Safety cases for new technology">
          <outline text="Alert fatigue, automation bias, cybersecurity"/>
        </outline>
      </outline>
      <outline text="Learning from complaints and success">
        <outline text="Complaints reveal dignity, access, delay failures">
          <outline text="Few complaints may mean barriers to speaking"/>
        </outline>
        <outline text="Safety-II studies how work usually succeeds"/>
        <outline text="Positive deviance finds better-performing teams">
          <outline text="Workarounds must not excuse system defects"/>
        </outline>
        <outline text="Findings must lead to visible, monitored action"/>
      </outline>
    </outline>
  </body>
</opml>
