Going to find the patient who has not come

There is a number I am more attached to than any other on this site, and it looks like an administrative metric: five out of five obstetric patients on my panel enrolled in prenatal care before twenty weeks' gestation.

Here is why it is not administrative.

You cannot achieve it by waiting

Reliable pre-twenty-week enrolment requires the physician to know which of her patients are pregnant before those patients come to tell her. There is no version of that which happens from behind a desk. It is a direct measurement of whether the doctor is actually present in the community, dressed up as a coverage statistic.

At CMF-4 I held fourteen families — five obstetric patients, eight paediatric, ten geriatric — and every fortnight I reviewed the whole panel house to house. Not as a courtesy. As the mechanism.

What the visit finds that the clinic cannot

Being in someone's home changes what you are able to say to them. Advice stops being general and starts being specific to that household:

• An elderly patient at risk of falling. In clinic I would have said “consider a grab rail”. In her bathroom I could say where, and see why the obvious wall was the wrong one.

• A house with young children. Where the medication actually lives, as opposed to where it is supposed to live.

• An obstetric patient with a new cough. The ventilation of the room she sleeps in, which was the answer.

None of that is available in a consulting room, and the patients trusted the advice because they could see it had been made for them.

The people who stop appearing

The distinguishing skill in primary care is not managing the patient in front of you. Any competent intern does that adequately. It is going to find the patient who has not come.

The elderly woman who has quietly stopped attending. The pregnant patient whose partner has moved her out of the district. The child whose vaccination has slipped by six weeks and then by six months.

None of these people will ever appear as a problem in a clinic's throughput figures. They are the reason the throughput figures look fine. A service that only measures the patients who arrive will always report that it is doing well, right up until one of the absent ones arrives at an emergency department.

The list is the whole intervention

What made this possible was not skill. It was a list, and the discipline of going through it on a fixed cycle whether or not anything appeared to be wrong.

I have since found myself doing the same thing in a completely different context — keeping the record of which students in a scholarship cohort had not been heard from in a term. It is the same failure mode. People in difficulty stop showing up to things, and if nobody holds the list, nobody notices for months.

Why this is the rotation I would defend hardest

Family medicine has no theatre and no resuscitation bay, and students arriving from hospital rotations often find it beneath them. I think they have it exactly backwards.

Hospital medicine is what happens after the front door has already failed. My conclusion from my own family's experience was never that we needed better hospitals. It was that we needed a working front door to the hospitals we already had — someone who knows the population, who is reachable, and who knows where a patient should go.

Fourteen families is a very small demonstration of that. It is also exactly the right shape.