Three small incidents, one real finding: themes and trends in practice

The first error is a missed morning dose, caught at the next round. The second, a fortnight later, is a dose given late. The third, a couple of weeks after that, is a signature missing from the record. Each is logged, investigated, found to be low harm, and closed with a note. By any reasonable standard, the home is doing the right thing with each one.

And that is exactly how the real problem stays hidden. When every incident is handled well in isolation, there is no moment where anyone is forced to look at them together. Three tidy closed records feel like three problems solved. They are actually one problem solved zero times.

The thing that changes everything is looking at the three together rather than one at a time. Same category, medication. Same shift, the early one. Same handover window, the half hour where the night and day teams overlap and the medication round is being picked up mid-handover. Individually, three unrelated slips. Together, a clear signal: something about that handover is making medication errors more likely, and it will keep doing so until it is addressed at the source.

This is what trend analysis actually is. Not a chart for its own sake, but the discipline of asking, across a set of small events, whether they share a cause. A home that never does this will keep closing medication errors one by one, indefinitely, and never wonder why there are always more.

Out of three small incidents comes one finding worth the name: the early-shift handover does not protect the medication round. That is a different kind of statement from any of the three incident notes. It is not about a missed dose, it is about the condition that keeps producing missed doses. It is the thing you can actually fix.

The finding is recorded as a theme that links back to all three incidents, so the evidence for it is the incidents themselves, not an opinion. Anyone can follow it from the conclusion to the three events that support it.

Because the finding is about the cause, the action is too. Not three separate fixes for three separate slips, but one thematic action: protect the medication round during the early handover, by changing when the round happens or who owns it across the overlap. The action is owned, dated, and linked to the theme, and through it to all three incidents.

And because this is a recurring weakness rather than a one-off, it earns a place on the risk register: medication safety at shift handover, scored, with the new handover arrangement named as its control. The risk links back to the theme and the incidents. The home is no longer managing three closed records. It is managing one named risk, with one control, supported by the evidence that justified it.

The step that proves whether any of this mattered is the one most often skipped. After the handover change is delivered, the question is simple: in the weeks that follow, do medication errors on the early shift recur? If they stop, the finding was right and the action worked, and there is a record to show it. If they continue, the cause was somewhere else, and the home knows to look again rather than assume the problem is solved.

An action closed against a theme that then recurs is not a success, it is a signal that the real cause has not been reached. Checking for that, rather than ticking the action as done, is the difference between a home that processes medication errors and one that learns its way out of them.

The failures here are quiet, because every individual record looks fine.

The first is that the three incidents are never looked at together, so the theme is never seen and the home keeps closing the same kind of error forever. The second is that a theme is spotted but produces only a general reminder to be careful, which changes nothing, rather than a change to the handover itself. The third is that the action is taken but never checked, so a fix that did not work is mistaken for one that did, and the errors quietly continue under a closed record.

Read together, the three incidents, the theme, the action, the risk and the recurrence check tell one story: a home noticed a pattern in small events, found the cause, changed the thing that caused it, and confirmed the change held. That is the most valuable kind of governance a registered manager can show, because it is governance that made care safer, not governance that documented it. A reviewer reads it as a service that learns from trends rather than reacting to events. The residents on the early shift simply get their medication on time.

This is the evidence loop applied to the hardest case, the one where no single event is big enough to force attention. It is the same idea we set out in how the evidence loop works in practice, worked through themes and trends rather than a single chain.

Verivius surfaces these themes from your own records, links the incidents to the finding, and flags when an action closed against a theme recurs. See how it works for adult social care, or talk to us about a design-partner engagement.