The red cross claims that donating blood can 'save up to three lives' That seems a highly exaggerated best case scenario. I'm curious, how many lives are really saved from a unit of blood? Specifically, what is the odds of one pint of whole blood producing a product which will be used to successfully treat an otherwise fatal injury/illness etc, and that if the donor had not donated that pint a death would have occurred due to lack of sufficient supplies, or being forced to use fewer resources of less computable ones?

What I would love is the ability to do get as nuanced and exact numbers as possible, as part of the motivation of this question is to have the numbers needed to address another question on Skeptics. If an answer doesn't want to do the exact math pointing me to the resources I need for me to do the math would be fine as well; but I don't even know what products are produced from a whole blood donation much less how to calculate the benefit of any one of those products.

I know that blood type of the donor could play a factor here, but to address the question I'm interested in I'd prefer to know about the 'average' pint of blood, so in essence if all the donor of all the varous blood types in the US donated and you averaged all those various pings of different types together what would the average pint do?

Of course being B+ myself I would be personally curious to know what the B+ blood type does, it would be cool to calculate a 'statistical lives saved' ratio for myself :)

  • 4
    I will be very surprised if credible data exists to answer this question. Even defining "lives saved" is difficult, bordering on impossible.
    – Carey Gregory
    Commented Jun 18, 2015 at 14:33

3 Answers 3


I think your skepticism may come from not understanding the process behind the claim.

A single unit of blood is separated into 4 main "blood products": red blood cells, plasma, platelets, and white blood cells. Another product called cryoprecipitate can be produced from frozen then thawed plasma and is used in special circumstances. Whole blood is rarely used for transfusions anymore because of problems with transfusion reactions and, quite frankly, except in the case of massive hemorrhage, a single person rarely needs all these components all at once (and even then they probably won't need white blood cells). Red blood cells (or packed red blood cells) are what most people think of when they get a "blood transfusion." Plasma is given to people who do not have enough clotting factors in their blood to stop bleeding that is currently occurring or if it should occur. Platelets are given to people who are not producing enough platelets to keep them from bleeding to death. White blood cells are rarely given anymore, but there may be occasion to use them in specific cases.

Now did that one unit save up to 3 lives? Well, to answer that the question of whether their lives were in danger to begin with has to be answered. The answer is, "Yes!" When will they die without the product? For some it's in the next several minutes, for others it may be hours, days or weeks, but in every case these peoples lives are in danger without the products derived from human blood.

But did that unit all by itself achieve this? No, but in concert with the donations of others it contributed to saving "up to 3 lives." Because doctors try avoid using precious blood products until it is absolutely necessary by treating patients with other methods if possible, so by the time a person definitely requires blood products, they will get more than one unit, but had they been given products earlier without giving the 'other methods' a chance to work they would have needed just as much, maybe more, over time.

The American Cancer Society has an excellent summary of types of blood transfusions and what they are used for here.

I don't think the claim is derived from any particular set of statistics or the hard and fast numbers that you are seeking, but from the process involved.


The best collection of statistical information online may be Blood Transfusion and Donation published by the National Institutes of Health and The 2011 National Blood Collection and Utilization Survey Report (PDF), published by the US Department of Health and Human Services (and several other departments and agencies). These 2 documents contain the most current comprehensive data available on blood collection and transfusion.

  • I understand all of the information you provided, and appreciate the effort, but I'm not sure this addresses my actual concern. I would like to get a statistical lives saved per pint of blood average, factoring in all the ways the blood product is used, the odds of a treatment not occurring due to lack of a product, odds of a treatment successfully saving a life that would have otherwise died within the next few weeks due to lack of treatment etc. I know that a doctor will never say "this treatment saved this life", but a general statistical number can be produced without doing so.
    – dsollen
    Commented Jun 22, 2015 at 19:19
  • I do realize I'm asking allot in the question, but if I can get information on where to find statistics, such as what the most common treatments are, what there NNT is, and statistical odds of a treatment not being performed due to lack of product I could do the math. Even if you could point me to a lower bound by suggesting the few most important treatments (ones more likely to save lives) that would help. By defining a number I'm better able to make further calculations on viability of blood collection options I'm interested in.
    – dsollen
    Commented Jun 22, 2015 at 19:22

I'm answering my own question, yay! This is my attempt to answer my question despite my lack of familiarity with much of the medicine. I therefore stress that I am not guaranteeing this answer, it's meant to be a very rough estimate and shouldn't be considered exact. Perhaps others will comment on things I missed and allow me to make it better...

First, Here is a quick link that discusses the concept: http://blog.inceptsaves.com/blog/2010/10/27/donor-recruitment-how-can-one-pint-of-blood-save-three-lives/

So there are three types of products that can be produced from whole blood, and each is divided into a nice 'unit', a base number used to define how much of a product is ordered by a hospital. However, one whole blood donation does not make up a full unit in any of these three. So what we get is:

  1. Red Blood Cells(RDC): little less then 2 donations per unit
  2. Platelets: 5-6 (I've seen both numbers, 6 seems more common) donations per unit
  3. plasma: less certain, I think about 2 donations per unit?

Thus for 10 donations of red blood cells you will have produced 2 unit if platelets and 5 units of RDC and Plasma. If one unit was required to save a life then you will save 12 lives with those 10 donations, or 1.2 lives per donation.

However, it looks like many units are required per operation/transfusion/emergency.

Red Blood Cells

in 2010 on average 2.75 units were used per patient. This is not anywhere near an accurate estimate of units needed per 'life saved', but I'm trying for a very very rough estimate. So for now lets say that each transfusion saved a life, and thus it took roughly 5.25 whole blood donations per 'life saved' via RBC. In actuality the odds are not every patient was in a life critical situation, so we should look at the number of units used per patient who was in a life critical situation, but I don't have that number. Since it seems likely that those in non-life critical situations would likely require less RBC then those in a life critical situation I would assume the units used per life-critical operation are higher, meaning the overall lives saved per whole blood donation is lower, but I don't have statistics on this.


I know it can take up to 30 units for a single organ donation; but that's probably a high estimate, to stress the importance of donations.

Bone marrrow transplants used 54 PC in the 100 days post transfer, where a PC is defined to be the amount of platelets from a single donation of whole blood, but was also defined as requiring 9-6 PC to make a full unit; so a slight discrepancy from other numbers. I would put that at an average of 5 units per bone marrow transfer?

The average pre-transfusion platelet count was 32,055 with a healthy platelet count being 150,000 to 400,000. One unit should raise platelet count by 30,000-50,000. If we assume they would transfere enough platelets to get someone up to close to the lower end of 'healthy', 150,000, then they would need another 3-4 units per transfusion. This seems in keeping with other math I saw, with all surgeries looking to require 5+ units. Going with a favorable comparison lets say an average of 3 units per life-saving transfusion. I'm not at all confident with this number, but it's the best I've managed so far.

With that number were looking at 15-18 whole blood donations per one life saved with platelets.


I give up and throw up my hands here. Plasma is broken down into many different products, and trying to get averages for all those products to expand on the average units used is just too much.

for now, until I get around to doing better research, lets be extra generous and assume every unit saves one life. I highly doubt this, I would say it's likely that it's at least a minimum of 2 units per life saved, but I'm trying to stay on the generous side; and lacking any statistics I can at least say that they define units as a minimal quantity likely suggests they don't expect adults to ever need less then 1 unit per transfer which implies at least one is needed per life saving intervention.

Giving that likely generous presumption were looking at 1 life saved per two whole blood donations via plasma.

Other products

There are other products that can be made out of whole blood. None of them are nearly as useful as the above, and rarely are collected, but at least some of them are utilized and would add to overall lives saved by a very small amount. I don't feel like adding them up since it sounds like they are such a minimal affect, and won't. That's because I think any lives preserved via these methods is counteracted by...

Waste and outdating

Just under 5% of all blood products go unused due to waste or being too old. I figure this waste more then counteracts the above other products, so I'm going to ignore both and just pretend they two affects perfectly countered each other.

Final assessment

This is all very very rough math by a layman, so I am not swearing to any of it. I was trying to error on the side of presuming maximum number of lives saved per donation; though it's entirely possible I missed an important factor which caused me to underestimate these numbers; feedback is welcome.

However, as the math works out we have:

  • 5.25 Whole Blood Donations per life saved (HBD/L) for RBC
  • 15 HBD/L for platelets
  • 2 HBD/L as a very generous presumption for plasma.

This works out to .757 lives saved per donation.

so a little less then 1 life saved per donation at my most generous; though I fear my being lazy and not doing a plasma calculation likely raised that number higher then it really should be. Assuming I haven't made some obvious miscalculation; which is quite possible!

So every 2 whole blood donations you make will save a life! I've been donating for about 10 years now at about 1 donation per 2 months, so that still means more then 30 lives saved. Still not bad considering how little it really costs me.

I welcome feedback from anyone who has more knowledge about this then...well a geek googling random statistics who may be able to point out factors I missed. This is definitely not definitive; but it's nice to have at least a rough rule of thumb.


Here are some statistics: Each year, an estimated 6.8 million people in the U.S. donate blood. 13.6 million whole blood and red blood cells are collected in the U.S. in a year. Giving blood saves 4.5 million lives each year in the U.S.

The Jehovah's Witness patients' decision to forego transfusions for major surgical procedures appears to add 0.5% to 1.5% mortality to the overall operative risk

13.6 million blood donations is used for 4.5 million people. This is .33 people per donation. Using the Jehovah's Witness information then about .5 to 1.5 % of these would actually die without blood. This is .0016 to .0050 people per blood transfusion. So this implies something like 200 to 600 blood transfusions to save a life. At first glance this seems small which is why the number is not readily available. However, you can look at it another way. People who need blood are often of middle age where they have a life expectancy of 30 years or about 8,000 days. So you are saving somewhere between 13 and 40 days of life for every blood donation on the average. One hour or so and 13-40 days of life are added. This seems wonderful.

  • Interesting approach to the question. Welcome to the site. Could you please make links when you post instead of inserting URLs as plain text?
    – Carey Gregory
    Commented Aug 9, 2020 at 20:52
  • I appreciate your analysis, but I think you missed quite a few lives saved by blood donation. Each donation is split into three separate units and each unit can be used to save lives. Even whole blood is most often used not for the sort of elective surgeries refereed to in your article, but for situations where blood loss was already severe enough to mandate transfusion to save a life. In other words your focusing on one of the least important ways a donation is utilized to save lives and ignoring a multitude of other ways it's utilized.
    – dsollen
    Commented Aug 10, 2020 at 14:55

Not the answer you're looking for? Browse other questions tagged or ask your own question.