A completed health check is not a completed health outcome

The health-check dashboard arrives. It shows how many employees were eligible, how many booked, how many attended and how many reports were delivered. The programme has a completion rate. The invoice can be reconciled. The campaign is over.

Now look at the same programme from the employee’s side. Their report says a result needs attention. They do not know how concerned to be. The recommendation says “consult a physician”, but not which kind, how soon or whether the consultation is covered. They put the report aside and return to work.

Both accounts can be true. The health check was delivered exactly as promised. The employee’s care still went nowhere.

A programme can be operationally complete and clinically unfinished.

The dashboard stops where the uncertainty begins

Invitations, bookings, attendance and report delivery are worth measuring. Without them, there is no programme. But they are measures of delivery. They tell an employer whether the service reached people and whether the provider did the work it was paid to do.

They do not tell us whether a finding was understood, whether the next step was realistic or whether the employee reached the right care. Yet these numbers are often made to carry more meaning than they can. “Reports delivered” begins to sound like “employees helped”. “Consultations booked” begins to sound like “care completed”.

That is not just loose language. It changes what the programme pays attention to. If success ends at attendance, the system gets very good at filling slots. If success ends at report delivery, it gets very good at producing PDFs. Nobody is responsible for the confusion left behind.

Completion means different things to different people

For the diagnostic provider, complete may mean the sample was processed and the report released. For the benefits team, it may mean the eligible employee used the service. For the employee, complete is more personal: I understand what this means, I know what to do, and I have been able to do it—or a doctor has told me that no further action is needed.

Those definitions should not be collapsed into one percentage. Nor should every completed next step be called a health outcome. Reaching a doctor is progress. Completing a follow-up test is progress. Neither, by itself, proves that someone’s health improved.

Healthcare quality has long separated what a system has, what it does and what happens to health afterwards. The distinction matters here because real health outcomes take time and are influenced by far more than one employer programme. The honest middle ground is to show whether care moved forward without pretending that movement is the final outcome.

Four questions are enough to expose the gap

After a finding that needs attention, ask four plain questions. Did someone explain it? Was the next step clear? Could the employee actually get that care? Did anyone check what happened next?

The answers do not have to produce another appointment. A normal result may need nothing more. A clinician may recommend waiting. An employee may understand the options and choose not to continue. The point is not to push everyone through a funnel. It is to make sure uncertainty is not mistaken for a decision and abandonment is not mistaken for completion.

This also changes how a programme interprets “low engagement”. If people repeatedly stop at the same point, the problem may not be motivation. Perhaps the recommendation is vague. Perhaps the relevant service is not covered. Perhaps employees are asked to repeat their history to a provider who cannot see the original finding. Another reminder will not repair any of those breaks.

Do not grade the employee for failing to navigate a pathway the programme never designed.

HR needs a programme view, not a medical chart

Better follow-up can sound uncomfortably close to employee surveillance. It should not be. An employer does not need to know who had an abnormal result, what was said in a consultation or why a particular person chose not to continue.

The employer’s questions are about the service it bought. Could people book? Were results that needed explanation reviewed? Where did access repeatedly fail? Was follow-up available when it was supposed to be? Those answers can be reported as anonymised patterns, with groups large enough that no individual can be worked backwards from the data.

The employee’s questions stay on the other side of that boundary: What does my result mean? What should I do? What have I already tried? What did my doctor advise? That is their care, not management information.

A better review meeting sounds different

Instead of asking only, “What was our participation rate?”, an HR team can ask: At what point were employees most often left without a clear next step? Which recommendations could not be acted on because the right service was unavailable? How long did it take to get an explanation? Did anyone remain stuck after asking for help?

These questions are less flattering than a single completion number. They are also more useful. They reveal work the programme can improve without making claims about health that the data cannot support.

What we are building

Oxology is being built around the part that begins after the report. The employee gets a continuous record, an explanation they can use, a visible next step and a physician when clinical judgment is needed. The programme can see where its handovers work and where they fail, without receiving the individual’s results or private conversation.

We are early, and the measures will change as we learn. The standard is simpler: say what was delivered, show where care moved forward, and do not call either of those a health outcome until there is evidence of one.