Why we started counting repeat donors

A child with thalassemia major will receive somewhere between 500 and 700 blood transfusions in their lifetime. Eighteen to twenty a year, every two to three weeks, from early childhood onwards. We have written that number down before, in a post last year about whether India can afford nucleic acid testing, and I have said it in enough rooms that I had stopped hearing it properly.

It started meaning something different to me when I looked at it from the donor's side. Five hundred to seven hundred transfusions is also five hundred to seven hundred separate encounters with somebody else's blood. And the question that follows from that is not how much blood we collect. It is how many different people that blood comes from, and how well we know them.

That is not the question the sector is set up to ask. We count units. Units collected at a camp, units issued, units short this month. I have spent years counting them myself. Units are what run out, so units are what get counted, and there is nothing wrong with the instinct. But for thalassemia it turns out to be an incomplete measure, and I want to explain why.

What the Indian evidence actually says

Earlier this year a systematic review and meta-analysis of transfusion-transmitted infections in Indian blood donors was published in Vox Sanguinis, by Golia and colleagues. It pooled 41 studies covering 1,860,594 donors. The headline numbers are the pooled prevalence rates — HIV at 0.12 per cent, hepatitis B at 0.91 per cent, hepatitis C at 0.28 per cent, syphilis at 0.14 per cent and malaria at 0.01 per cent.

The finding I keep returning to is not in the headline. The authors report that replacement donors and first-time donors had a higher prevalence of infection than voluntary donors and repeat donors respectively. Not dramatically framed, not hedged, simply stated as one of the patterns the data shows.

Single studies have found the same thing for years. A five-year study of 192,348 units at a centre in north India, published in The Scientific World Journal in 2014, reported hepatitis B seropositivity of 1.67 per cent among replacement donors against 0.24 per cent among voluntary donors, with HIV at 0.15 against 0.08 per cent. That particular study was very heavily weighted towards replacement donors, so I would not lean on its exact ratios, and I would not want anyone else to. But the direction has been consistent across a lot of Indian data for a long time.

There is a straightforward reason for it, and it is not that one kind of person is more careful than another. A replacement donor is usually somebody's relative, giving under pressure, on the worst day of that family's month, because a hospital has asked for a unit before it will release one. A repeat voluntary donor has been screened before, has been deferred before if something was found, and has no reason to be anything other than honest in the questionnaire. The system is asking two very different things of them.

Why this matters more for thalassemia than for almost anyone else

Most people who receive blood receive it once or twice, in a surgery or after an accident. For them the relevant question is whether that unit is safe, and the answer in India is now usually yes, because every unit is tested.

A thalassemia patient is in a different position arithmetically. Whatever the residual risk on a single unit — the small window before an infection becomes detectable, the rare test failure — a child meeting it five hundred to seven hundred times is in a different relationship with that risk than someone meeting it once. Cumulative exposure is the thing, and cumulative exposure is a function of how many separate donations a patient receives and who they come from.

I want to be careful here, because this is the point where an argument like this can do harm. Nothing about this replaces testing. Every unit still has to be screened, and the case for better testing does not weaken because donors are better chosen. Nor am I saying that replacement donation is bad, or that families who arrange it are doing something wrong. They are doing what the system requires of them at the moment they are least able to argue with it. The problem is the requirement, not the people meeting it.

What we do about it

The Blood Bridge is our attempt at the practical version of this. For each thalassemia fighter we map a small group of blood-group-matched voluntary donors, usually eight to ten, who take turns so that the next transfusion is covered before it is due and no individual donor is asked more often than they should be. We built the matching, the rotation and the reminders into software, and we run most of it over WhatsApp, because that is where people already are.

We started this to solve a scheduling problem. The need arrives every two to three weeks forever, and a camp produces a great deal of blood on a single day, and those two shapes do not fit, which is why families end up ringing around at night looking for one unit. The safety argument came later, and honestly it came from reading rather than from our own data. But it follows from the same design. A bridge, if it works, is a machine for turning first-time and replacement donors into repeat voluntary ones for a specific child.

I would not claim more than that. I do not have our own longitudinal infection data to put next to the published studies, and I am not going to imply that blood arranged through us is safe in some way that other blood is not. What I can say is that the mechanism the literature keeps pointing at — donor repetition — is the exact thing the model produces, and that seems worth saying out loud because I have not seen anyone else in India say it.

The hard part is not finding donors. People genuinely want to help. The hard part is keeping a commitment alive across years, through job changes and moves and the ordinary way that good intentions decay, and we are still learning what that takes. Some bridges hold and some thin out, and we do not yet fully understand the difference between them.

The thing I would like to see change

If donor repetition matters as much as the evidence suggests, then it should show up somewhere in how the system measures itself. At the moment a blood drive is judged by units collected on the day. Nothing in that number distinguishes three hundred people who will never return from a hundred and fifty who will come back twice a year for a decade, and for a patient population like this one, the second is worth considerably more than the first.

I do not think that is anyone's fault. It is just what happens when a measure is easy to collect and the alternative is not. But it is measurable now in a way it was not twenty years ago, and I would like to see return rate sit alongside units collected in how camps and blood banks report themselves.

If you work in transfusion medicine, at a blood bank, or at another organisation doing this — particularly if you have data on donor return rates and what moves them — I would like to hear from you, including if you think I have this wrong. We are a small organisation and this is an argument built mostly on other people's research. It would be better for having been argued with.