Germ theory and antisepsis

The intervention here is a change in what people do, and people change what they do when they have looked down the tube themselves.

Grind simple lenses from glass: a small bead of high curvature gives 100–300×, which is enough to see bacteria as moving rods and dots, and protozoa clearly. A compound instrument with an objective and an eyepiece gets you to 400×. Mount it rigidly with a mirror beneath the stage for light, a fine screw for focus, and use thin slips of glass. Then show everyone: pond water, a tooth scraping, water from the sick house.

Wiping and rinsing do not sterilise. Heat does.

• Boiling for 20 minutes kills all vegetative bacteria and most viruses. It does not kill spores, including tetanus.
• Steam under pressure — 121 °C for 15 minutes at 15 psi — kills spores, and is what makes surgery and later bacteriology possible. A strong sealed vessel with a safety valve and a gauge is the tool.
• Dry heat, 160 °C for 2 hours, for glass and metal that steam would spoil.
• Flame for an instant need: red heat, then cooled without touching anything.

Anything sterilised stays wrapped until the moment of use. An instrument laid on a bench is no longer sterile.

This single behaviour, adopted, is worth more than every drug at this tier. It was resisted by trained physicians for decades, and the resistance is the thing to plan for: make washing structurally unavoidable rather than a matter of conscience.

A basin and soap at the door of every sick room, refilled by a named person. 40–60 seconds, nails short, forearms included, before and after every patient and after every removal of a dressing. Post the five moments where it matters: before touching a patient, before a clean procedure, after body fluids, after touching a patient, after touching their surroundings.

Make cross-contamination difficult by arrangement, so it does not depend on anyone remembering.

One direction of flow: clean store → preparation → patient → dirty collection → laundry, never backwards. Two separate sets of buckets, cloths, bowls and tables, marked and coloured differently. Dirty linen goes straight into a lidded container, is boiled 20 minutes with soap, and dried in sun. Wash water from dressings goes to a soakaway, not to the yard where children play or where it can reach a well — keep 30 m and downhill from any water source.

Isolation without dedication is theatre: the carer who moves between the sick house and the kitchen is the route of transmission.

So: a separate room or hut, dedicated carers who do no food preparation and no attendance at births, dedicated bedding, bowls and utensils that never leave, an overshirt left at the door, and a written list of who has entered and when. Keep a 1-metre distance from coughing patients and cover the face. For gut illness, the carer's hands after any contact with stool are the whole problem — wash before touching anything else.

A rolling boil for 1 minute, 3 minutes above 2,000 m, for everything anyone in the sick house drinks — including the carers.

Store it in a narrow-necked covered vessel with a tap or a dedicated dipper, never a cup dipped in by hand, and treat a stored batch as good for 24 hours. Sick people drink far more than usual and someone gets tired and takes a shortcut with a bucket — which is how a treatable outbreak becomes a sustained one. Name the person responsible per shift. See making water safe.

Culture is what turns "an infection" into a named organism with known behaviour.

You need: nutrient broth (meat extract, salt, a little peptone from digested protein), a gelling agent — agar from seaweed, which holds at 37 °C where gelatine melts — poured into flat covered dishes; a warm box holding 35–37 °C; a wire loop, flamed; and stains. Gram staining — crystal violet, iodine, alcohol, then a red counter-stain — splits bacteria into two great classes in five minutes and predicts much of how they behave. Keep a written record of what grew from what, with drawings.

Antisepsis is judged by a rate, not by how clean the room looks. Record every operation and every birth, and against each whether infection followed within 30 days — fever, pus, a wound that opened, puerperal fever. Total monthly, as a percentage.

The historical figures show what the method can reveal. Semmelweis's First Clinic in Vienna lost 11.4% of mothers in 1846; after chlorinated-lime handwashing was enforced, 1.27% in 1848. Nobody at the bedside could see that difference; the register could. Keep counting when the trained person is away — the rate on their days off is the real test of whether the practice has taken.