One drop, ten seconds, and a number that decides three months of syrup.

"Anaemia hai, doctor. The school camp report says so."

She unfolded a small printed slip and pushed it across the desk, smoothing it twice, the way people do with paper they have been carrying around and worrying about. Her daughter, six, was spinning slowly on the stool, entirely uninterested in her own haemoglobin.

I asked her two things. Did anyone squeeze the child's finger to get the drop out? And what else was written on the report?

She did not remember about the squeezing. And there was nothing else on the report. One number, with three months of iron syrup written underneath it.

Both answers matter. The second one matters more.

Two surveys, one country, very different answers

India has counted anaemia in its children twice over and got two different countries back.

The National Family Health Survey, carried out at enormous scale by teams taking a drop from the fingertip, reported that about 67 in every 100 children under five were anaemic. The Comprehensive National Nutrition Survey, run between 2016 and 2018, drew blood from a vein and sent it to a laboratory. Among children aged one to four it found 40 in 100 (Yadav and colleagues, PLOS Global Public Health, 2024). Across the wider one-to-nineteen group it found 30.7 in 100, less than half the survey figure (Kurpad and Sachdev, Indian Pediatrics, 2022).

Two explanations are on the table for that gap. One is the cut-off itself. Researchers went back to the healthy children inside the venous survey and worked out what haemoglobin actually looks like in a child with no deficiency and no inflammation. Those levels sat one to two points below the numbers we have been using. Applied to all 41,210 children and teenagers with a valid reading, anaemia fell from 30.0 in 100 to 10.8 (Sachdev and colleagues, Lancet Global Health, 2021). The other explanation is what happens to a sample collected in a hurry, in the field, by a team getting through hundreds of children before lunch (Rai and colleagues, British Journal of Nutrition, 2022).

Let me be honest rather than dramatic, because you deserve the honest version. Not everyone agrees about the cut-off. A WHO-funded analysis pooling healthy children from eight countries found that for two- to five-year-olds the line sits almost exactly where it already is, and found no evidence it should differ between ancestries at all (Braat and colleagues, Lancet Haematology, 2024). Below two years, it did come out lower.

"So the finger-prick machine is useless?"

Not at all. I use one in my own clinic, most days of the week, and I trust it.

A capillary sample is only as good as the drop it is taken from. Milk or squeeze a fingertip and tissue fluid comes out along with the blood, thinning the sample and pushing the reading down. Take it from a cold, tightly clenched hand and you can get a falsely high one. But wipe away the first drop, let the next one well up on its own from a warm relaxed hand, and the number tracks a laboratory result closely enough to act on.

So it is not the finger that is the problem. It is the hurry.

And there is a second problem, bigger than the first one. Anaemia is a number, not a diagnosis. In that national survey, only about a third of the anaemia found in children was explained by iron deficiency. Three out of four anaemic toddlers were short of iron, B12 or folate, so for a quarter of them none of the three was the answer. And in this part of the country, a stubbornly low haemoglobin with small red cells is sometimes thalassaemia trait, which no amount of iron will fix.

Which is why the reflex of low number, start iron, repeat in three months quietly misses most of the children it is meant to help, and gives iron to some who never needed it at all.

Lemon on the dal, guava after lunch. Much of the answer is already on the table.

What I actually ask parents to do

Do not start three months of anything on one number. A careful repeat, or a complete blood count from a laboratory, costs little and settles whether there is anything to treat. This is my own practice rather than a national rule.

Ask for the whole report, not just the haemoglobin. The size of the red cells sits in the same printout and is the first clue to what is actually going on.

If iron is started, fix a date to check it. A genuinely iron-deficient child's haemoglobin climbs within four to six weeks. If it has not moved, the answer is not a bigger dose. It is a different question.

Put a sour thing next to the iron-rich thing. Lemon on the dal, guava or amla after lunch. And keep tea and a large glass of milk away from meals, because both block iron absorption at exactly the wrong moment.

The mother that morning had done nothing wrong. She had taken a piece of paper seriously, which is what a good parent does.

A single number is a question, not an answer. Ask the next one before you open the bottle.

Sourced from PubMed; all studies linked above. This is general guidance and not a substitute for examining your own child.