Saturday, 19 December 2020

Covid–19 and Vitamin D: Efficacy of vaccines

Efficacy of Covid-19 vaccines

This Blog post might be a bit dull, but it contains information about the performance of the three new vaccines, which are about to be introduced.

First a reminder of Cacifediol

Calcifediol / vitamin D 

We have seen that vitamin D in its rapid-acting part-activated form Calcifediol is remarkably effective the treatment of Covid-19 pneumonia. We have seen that its efficacy is 96%, and also how this is worked out. 

Proportionate benefit = efficacy = 96%

Absolute benefit = 48%

NNT = 2, meaning that 2 treatment are necessary to prevent 1 ICU admission/death.

In the control group 2 deaths occurred among the 13 patients needing admission to the ICU.

We have also seen that the cost of preventing an ICU admission or death would be approximately £10. 

We have also seen that a detailed report from Massachusetts Institute of Technology indicated that the  Córdoba RCT of Calcifediol was conducted properly, and that the probability of the positive result being by chance is less than one in a million.

We have seen that the study was rubbished by NICE and as a result it has not been used in the UK. It is likely that since the trial's results were published on September 3rd, 25,500 UK citizens have died from Covid-19. This would have been a much smaller number if the results of the Córdoba study had been put into action, and hospital patients with Covid-19 pneumonia treated with Calcifediol.

Physical or immunological protection?

The government and its advisors have done a great deal in what can only be described as a futile attempt to conquer the air-borne respiratory virus by physical means. It was clear from the outset to people with a knowledge of biological science that the most important approach would be to protect our citizens by mechanisms of immunity.

There would be two approaches to this. One would be to develop a vaccine but that would obviously take time. The vaccines are becoming ready for use at the end of 2020, by when more than 65,000 deaths from Covid-19 will have occurred in the UK. There are still some concerns about vaccine safety.

The other approach would be to optimise innate defensive immunity, and this is best undertaken by correcting widespread vitamin D deficiency. Even before the pandemic developed, it was or should have been well-known in medical–scientific circles that vitamin D is the vital key that unlocks the enhancement of the defensive immune process. 

On March 23rd 2020 there had been 285 deaths from Covid-19 in the UK. Lockdown was introduced, with huge social and economic cost. Vaccines were well in the distance, but vitamin D could have been and should have been used to help defend the people while waiting for a safe vaccine to be produced.

The vaccines are now available or are being tested. It is known that in the presence of vitamin D deficiency, vaccines are less likely to produce an immune response.

I have used prices of vaccines that I have seen. More accurate or dependable prices might become available, and can then be substituted.

Pfizer vaccine

The Pfizer trial, 1st release “90% effective”

Pfizer released this data to the press.

36,995 subjects recruited into the trial, divided into two equal sized groups 18,497 in each (forget the one extra for now as we were not told which group this person was in).

Placebo 85 cases, no deaths

Vaccine 9 cases, no deaths

This is all the data that we have been given. 

What it means is this:

Reduction of cases in the vaccination group is 85 (control) minus 9 (vaccine), equals 76.

Efficacy method 

Expected number of deaths in vaccine group =

number of deaths in placebo group = 85

Observed number in vaccine group = 9

Cases prevented by vaccine = 85 - 9 = 76 

76 prevented out of 85 = (76/85)x100 = 89.411 %

Proportionate reduction of cases = Efficacy = 90%

Absolute benefit

Control group : 85 cases out of 18477 = 0.46%

Vaccine group:   9 cases out of 18477 = 0.049%

Gain from vaccine = 0.46 - 0.049 = 0.41%

Absolute risk reduction of cases = 0.41%


Number needed to treat

18,497 vaccinations are necessary to prevent (85 - 9) = 76 “cases” (whatever a case might be). 

18497 ÷ 76 

= 243 vaccines are necessary to prevent one case

243 is the NNT, the number needed to treat to prevent one case.

From this we can calculate the cost of one case prevented, on the basis that the cost of one success must be the sum of the NNT.


We are told that the price of one vaccination (assume a pair of vaccinations in one person) is $20. We assume that this includes distribution and storage costs.

We are told that in the UK there will be a £12.85 ($17.19) doctor administration fee. I will ignore this in the calculations but it will almost double the costs of vaccination.

Cost = $20

NNT 243 x $20 = 

$4,860  the cost per case prevented. 

Some people would rather have the money.

Deaths prevented (prediction only)

World wide there are 50 times more cases than deaths, ratio of cases to deaths is 50:1. 

This means that to prevent one death will require:

 243 x 50 = 12,156 vaccinations. 

This will result in a cost of 

50 x $4,860  

= $243,000 to prevent one death. 

My family would perhaps rather have the money!

Remember that most deaths are after the age of 80, with life expectancy about ten years =

$24,300 per year gained (quality not considered). 

In the UK the ratio of cases to deaths is 25:1

This means that cost per death prevented is half the worldwide average

243 x 25 = 6981 vaccinations to prevent one death = $139,620.

Pfizer trial, 2nd release “95% effective”

43,000 subjects, assumed equally divided.

21,500 placebo, 162 cases

21,500 vaccine, 8 cases

Vaccine group

Expected cases = 162

Observed cases = 8

benefit: (162 - 8) = 154

Cases prevented = 154 out of 162

= (154 ÷ 162) x 100 = 95%

Efficacy = 95%

21,500 vaccinations prevented 154 cases

(21,500 ÷ 154) = 140 = vaccinations to prevent one case

140 x $20 per vaccination 

= $ 2,800 per case prevented

= $ 140,000 per death prevented (world average)

Absolute benefit

Control group : 162 cases out of 21500 = 0.753%

Vaccine group:    8 cases out of 21500 = 0.037%

Gain from vaccine = 0.753 - 0.037 = 0.716%

Absolute risk reduction of cases = 0.72%

This does not sound quite so good, and so proportionate reduction of cases in expressed rather than absolute.

Moderna vaccine

The publication process is very slow. 

On November 12th a paper was published in the New England Journal of Medicine with he results of a phase 1 study involving 45 subjects receiving the mRNA-1273 Covid-19 vaccine. It had appeared as a pre-print on July 14th. The purpose was to define optimal dose.

The results of the clinical trial were released to the press in advance of peer-reviewed publication, and appeared in New Scientist on November 16th.

The Moderna mRNA-1273 vaccine seems to be 95 per cent effective and to work in those who need protecting the most – people aged over 65 

This is an "interim analysis", but the result is to bring the Moderna vaccine into use.

More than 30,000 people in the US aged 18 and over are taking part in the phase III trial of the Moderna vaccine.
Assume 15,000 in each group, placebo and active vaccination.
Placebo 90 cases (15 severe)
Vaccine group 5 cases (none severe)

Vaccine group, 90 expected, 5 observed.
Reduction of cases = 90-5 = 85
Reduction is 85 out of 90
= (85 ÷ 90) x 100 = 94.44 = 95% approx.

Absolute risk reduction
Risk of infection:
Placebo group:  90 out of 15,000 = 0.6%
Vaccine group :  5  out of 15,000 = 0.33%
Risk reduction = 0.6 - 0.33 = 0.47%

Number needed to Treat
15,000 ÷ 85 = 176
NNT = 176. 
176 vaccinations needed to prevent one case.

Price of Moderna vaccine: $10 to $50 per dose.
If $10 per dose, $1760 to prevent one case
If $50 per dose, $ 8750 to prevent one case. 

Cost of one death prevented (case x 50) = 
$88,000 or $437,500

Prices of vaccines

AstraZeneca vaccine

Different dosage groups, in UK and Brazil
One example:

Low Dose + Standard Dose
Vaccine: 1367 subjects, 3 cases
Placebo : 1374 subjects, 30 cases
Cases prevented = 30 - 3 = 27
27 cases prevented out of 30 = 90% efficacy

Number Needed to Treat
1367 vaccinations to prevent 27 cases
= 51 vaccinations to prevent 1 case
NNT = 51

Price $4 per dose
Cost per case prevented = $4 x 51= $204
Cost per death prevented = $204 x 50 = $10,200

Standard Dose + Standard Dose
Vaccine: 1879 subjects, 14 cases
Placebo : 1922 subjects, 35 cases
Cases prevented = 35-14 = 21
21 cases prevented out of 35 = 60%

Number Needed to Treat
1879 vaccinations to prevent 21 cases
= 89 vaccinations to prevent 1 case
NNT = 89

Price $4 per dose
Cost per case prevented = $4 x 89 = $356

Cost per death prevented = $356 x 50 = $17,800

No conclusions, just observations.

Vitamin D and vaccines are complementary.

Giving vitamin D now would reduce the impact of the Covid-19 pandemic before the vaccines are widely used. Also vitamin D is necessary to activate the immune system so as to allow the vaccine to induce a good response. 

Also vitamin D is available now and distribution costs are minimal. There are no costs of storage or administration. Safety is not a concern in the doses recommended (up to 4,000 units daily or equivalent).

Vaccines will not be of help to those who are ill from Covid-19. For this, Calcifediol the part-activated  form of vitamin D is essential.


  1. David Grimes, you are a bloody genius! I'll take the D3 (cost £8 for one year) and the Pfizer money (£100,000, say) for preventing my own death for a year (profit £99,992)!

  2. This is an excellent analysis! I would disagree with one small part, 4000 units of D is not enough to get to the desired level of 50ng/ml. You would need twice that amount and there is no safety issue here also. It would be a good idea to take some magnesium and K2 to offset and direct the rise is calcium, which you will get by taking additional D.

  3. After speaking to David a few years ago and seeing my gp. I had a level of 10ng/ml so took a 3,000 iu for apox 3 months along with K2 rich foods such as curly kale.. Nuts to aid absorbtion Gouda cheese throughout the week and my level went up to 40ng/ml ... I dropped down the 100iu after 12vweeks. I continue to take 1000 iu a day.. Go in the sun in summer for approx 15... 2o mins as I'm fair to allow sun exposure then put cream on. My levels have not gone below 40ng/ml since. I am slim so aids absorbtion and I eat salmon and mushrooms regularly. Everyone will be different depending on body weight, muscle mass.. Fat in the body so best to get a good calculation of approx iu to take depending on what's base level and loading dose needed. My gp gave me 20,000 loading when I was deficient but I took 5,000.. In doses over 4 weeks and then 100iu. I also contacted the vitamin D Council as they provide useful information.

  4. Doctors have used one off doses of 400,000iu to treat Covid ICU patients successfully; and authorities suggest 400iu is sufficient when we can make 20,000iu in an hour of sun bathing. Is nature right or medical authorities?

    1. A single large bolus dose failed in one trial and should not be used. In the event of acute infection active analog needs to be used to ramp up levels immediately followed by daily doses along with high dose melatonin. together the two hormones ramp up the innate and adaptive immune system and are able to tamp down significant mitochondrial oxidative stress while also adding a supporting role to endothelium.

  5. This needs to be shared with the practising medical and scientific community. Thank you David.

  6. I live in Texas and I have been on 5000 iu of vitamin D daily since April. In late March we (fellow lab techs) analyzed our 25(oh)D levels and found them to be between 12-15 ng/mL. We started taking large doses 10,000 daily then we were potentially exposed and ramped up much higher 80,000 daily for a few days then back down to 20,000 then 10,000 then 5000 daily. After two weeks our levels were in the 30's. By end if April we were in the high 40's. Still taking 5000 iu in summer we topped out in the high 50's lower 60's. Two weeks ago reanalyzed and was surprised that my 25 (oh) D level fell 17% (48 ng/mL) just by going from summer to winter while maintaining 5000 iu daily. That did not put me in a dangerous level, but what about people who live at higher latitudes whose Drs think 20-30 ng/mL is sufficient and it's summer. My fellow lab tech just analyzed a younger family member. she gives him a multi vitamin daily and 1,000 iu tab every other day and this young man's level was 19 ng/mL. RDA's are way too low potentially by the power of Ten and the pump needs to be primed day

  7. I tested 66.6 ng/mL in late August 2020 after being on 7000IU over the summer. I spent every day outdoors between 9am to 3pm in the sun dressed in shorts, a short sleeve T-shirt and baseball hat. I live 48 deg latitude and are overweight, bmi just below 30 and I am 67 years old. Now after 1st September I increased my daily dose to 12000IU due to less or no sunshine here in the "rainy" PNW. She maintains level 40+ ng/mL. We will test again in early January. Feels good to be right on track
    My wife takes 10000IU per day year around

    1. For the past 20 years I have taken 3,000 units daily, all the year round. My blood level is stable at 60ng/ml, 150nmol/L, which I regard as ideal.

  8. Dr Grimes, thank you for what you're doing to bring D to the public attention. My wife and I have been taking 2000IU every other day for almost 5 months now.

    I think this one article may have a problem with the cost analysis.

    Both the NTT and "cost per case prevented" numbers are actually only valid for the snapshot of data (meaning: for the few months during which trial data was captured).

    What you're calculating is actually "cost per case prevented in N months" (N being the duration of the vaccine's trial). That cost will only go lower as cases increase over time. You'd need to project expected cases over the pandaemic -say, three years?- in order to derive a "cost per case prevented, period".

    It may not make a huge numerical difference, but rigour always helps make a better case.

    1. Thanks Vladimir – you are correct. The data presented applies only to the trial. As experience progresses the data will change. However further data might not be released as once the show is on the road there is nothing to be gained by the manufacturer, and there could be bad news that it might be better to suppress. However if further data is released I will update the analysis.

  9. Para manter níveis otimizados (entre 80ng e 100ng) de vitamina D, a suplementação deve ser diária e continua, de acordo com o peso. 200UI por cada kg de peso.

  10. It's no wonder the drug manufacturers were collectively scrambling to produce a viable product in short order. Producing a vaccine is a licence to print money, as long as your ethics don't get in the way.

    1. Not just ethics, but recent history to be forgotten.
      The vaccine Pandemrix was manufactured in the UK by GSK and was introduced in 2009 as vaccine to prevent swine flu, a pandemic that did not materialise. It was given an accelerated licence by the EU but not in the USA. It was given to 6 million people in Europe, many of them young people. In 2010 there were reports in Finland and Sweden of an unusual increase in the frequency of narcolepsy in young people. This is a brain disorder characterised by uncontrollable attacks of sudden deep sleep. It became a small but widespread epidemic of about 700, mainly young adults, and it was determined, by the UK Patient Safety Agency (now part of PHE) to be due to Pandemrix. Its licence was revoked in 2014 and it was withdrawn from use. Sir Patrick Vallance was head of R&D at GSK at that time, but he does not mention Pandemrix today. This shows how difficult it is to determine safety in a short time, especially when an uncommon but serious side-effect might not appear immediately and when the vaccine is new.

  11. Very interesting information. Not heard about that on the BBC! How is it possible that Vallance can hold his current position after that happening? Absolutely scandalous!

    1. What about the trail of inaccurate predictions behind Neil Ferguson? They are still listening to him and have put him back on SAGE!

  12. Your analysis of the Pfizer data shows that the vaccine prevented 154 cases. Based on this number you calculate the cost to treat and save a life in a pool of 21,500 subjects. Your calculation is faulty because these are only partial results from 2 months. Theoretically, all 21,500 subjects will eventually be exposed to SARS-Cov2. If they are all susceptible to SARS-Cov2 infection, then a vaccine with 95% effectiveness would prevent over 20,000 infections- not 154.

    1. The calculation was based on the results of the clinical trial. That is all that I can do. Extension beyond that is guesswork until the company releases long-term results.

  13. If the purpose of vaccination is to provide immunity why are we not testing to see if people already have immunity?

    Some studies show that the UK may already have around 50% immunity so it would seem pointless to vaccinate without ascertaining whether or not the person is already immune!

    I tried to start a petition on the UK government site to this effect but it was rejected because: "It included confidential, libellous, false, unproven or defamatory information, or a reference to a case where there are active legal proceedings."

    I asked them to substantiate that statement and they came back with this:
    "I appreciate that this is a complex area, but we couldn’t accept a petition that implied the immunity of certain people to Covid-19, when there does not appear to be any independent research demonstrating that T-cell tests can establish that a person is immune to – as opposed to having an effective immune response – to Covid-19. Similarly, we couldn’t accept a petition that implied that certain categories of people are at particular risk of certain reactions to any Covid-19 vaccines, when there does not appear to be any evidence that this is the case."

    Perhaps someone could explain what the difference is between being "immune to" and "having an immune response to" as in their statement above? Clearly they think we are all stupid - hence the "complex area" comment!

    They also didn't know the difference between antibodies and T-cells!

  14. But NICE is full of experts, The government has access to the best experts in the world. (Oh well in Britain since this year) And is David Davis a scientist? His degree was in computing and chemistry, he has not worked as a professional scientist. Why should we trust his expertise?

    Perhaps this quote from Prof Jose Lopez Miranda the academic lead on the Cordoba study may shed some light

    "The British parliamentarian David Davis has been in contact with us for several months and after his parliamentary interpellation he wants to follow the nutritional recommendations that we have made regarding the administration of vitamin D to the elderly, since many of them have deficiencies, already which reduces the admissions in the ucis of these hospitalized patients, "Dr. López Miranda told Diario CÓRDOBA after hearing the news.

  15. Ferguson predicted 500 000 deaths if the epidemic was allowed to run without mitigation. To date there have been 100 000 deaths on 14% of the population affected. If the epidemic would be allowed to run to herd immunity -80%, Then the death would be 5.7 times the current. ie 570 000. A bit higher than Ferguson's calculation.

  16. COVID Vaccination near me Really I enjoy your site with effective and useful information. It is included very nice post with a lot of our resources.thanks for share. i enjoy this post.