Seasonality and population Vitamin D levels are parameters for COVID-19 models

In this article, Brent Walker brings to light seasonality and Vitamin D levels as factors to consider for COVID-19, particularly as Australia approaches its winter. 


According to the World Health Organisation on May 21, 2020 COVID-19 has infected nearly 5.5 million people worldwide. The true numbers of persons infected is much higher than this number due to no recording of asymptomatic and mild disease cases. Also, many countries (e.g. India, Pakistan, much of Africa, Indonesia and The Philippines) do not have the health systems to test much of their population. . The actual number of world-wide recorded deaths is almost 350,000 and again this is an understatement. It is likely to be a lot higher as many people die of this disease without being tested.

The official WHO figures on deaths by country is showing a pattern that, together with medical literature on what this virus does, leads to the conclusion that the timing of the pandemic in some countries and their population vitamin D levels will influence their infection levels and death rates. This is one reason for Australia and New Zealand having a good result so far. It may also be true for South Africa and possibly Argentina and other mid-latitude South American countries. Actuarial modelling of this disease should therefore include these factors.

WHO data at May 21

The WHO data for infection, death and testing rates for some continents and countries is provided.

Looking at this comparison it appears that data from China might not be correct[i]. China doesn’t provide testing numbers which would help verify its data. South Korea and Hong Kong appear to be stars in combatting this virus. Japan, Vietnam, Thailand and Taiwan may also be stars but their testing levels are way down on many OECD countries.  These Asian countries are superb at infection control. They learnt their lessons with SARS and bird flu (H5N1), both of which are primarily spread by airborne droplets, as does SARS-COV-2. In many Asian countries people automatically wear a face mask in public if they have any symptoms of a sore throat, head or chest cold. It is regarded as a form of politeness. Now in these countries almost all people wear masks in public. In mid-May 2020, the Department of Microbiology of University of Hong Kong released the results of an experimental study that proved conclusively that wearing face masks is an effective COVID-19 control measure[ii]. This study will be published in the Clinical Infectious Diseases Journal shortly. Clearly, widespread face mask use in a population would be a control factor when modelling infectious diseases.

 Russia’s infection numbers are increasing rapidly at present. This could be due to some catch up in reporting. Their deaths may not have hit their peak as their daily new cases appear to peak only on 11 May. Germany’s new cases hit their peak at the end of March and their deaths peaked around mid-April.

Australia and New Zealand have fantastic results and their testing levels are remarkably high. South Africa also has an amazing result although its testing level is only 19% of Australia’s. Argentina is also worth watching. So far, its infection levels are low, its death rate is relatively low, but its testing rate is also low at only 5.6% of Australia’s. Note that the country of Brazil is located near the equator and is doing poorly. Chile’s numbers are reasonable. The numbers for Uruguay and Paraguay are too.

Is Seasonality a Factor?

Why have Australia and New Zealand, probably South Africa and possibly Argentina got such low infection levels and deaths in comparison to many European countries and the US? Do these results have something to do with the season when COVID-19 hit these countries and if so, why?

Cold and Influenza viruses mainly hit mid-latitude countries in their winters and early springs. The reason for this is that viruses survive well in cold and/or humid conditions but do not survive for more than 30 minutes in bright sunlight. So, tropical countries tend to get influenza epidemics during their monsoon periods[iii]. Ultra-violet light kills many airborne viruses including corona viruses very quickly. They survive well indoors and particularly in the winter most mid-latitude countries. Because of this we in Australia get flu vaccination shots in autumn, not spring. So, as the SAR-COV-2 virus hit southern mid-latitude countries in their late summer, the virus did not have the same opportunity to infect their populations as it did in the late winter of northern mid-latitude countries. This does mean that a “second wave” during winter or early autumn is a distinct possibility in sunny, southern hemisphere mid-latitude countries[iv]. Actuarial models should allow for seasonality for COVID-19 in the same way as they do for influenza.

Vitamin D

One reason why colds, and SARS type corona viruses are more prevalent in winter and early spring is low population vitamin D levels. Population vitamin D levels vary significantly by country. For example, a 2019 study found that: “Vitamin D deficiency (serum 25-hydroxyvitamin D (25(OH)D) <50 nmol/L or 20 ng/mL) is common in Europe and the Middle East. It occurs in <20% of the population in Northern Europe, in 30–60% in Western, Southern and Eastern Europe and up to 80% in Middle East countries”. Severe deficiency (serum 25(OH)D <30 nmol/L or 12 ng/mL) is found in >10% of Europeans[v]”. The same study found that vitamin D deficiency was high in non-western immigrants in European countries. So, this 2019 study suggests that Italy and Spain, for example, would get a higher infection and death rates than Germany and Denmark. Correct! The study also found that UK population has a high level of Vitamin D deficiency. The UK also has a higher infection and death rate than Germany and Denmark. Sweden has opted not to have the controls of its neighbouring countries and this shows in its figures.

Vitamin D levels also vary by season with the lowest levels normally encountered in winter and early spring. Natural sunlight is not sufficient for the skin to produce vitamin D at all during the months of May to August in Melbourne and Hobart and during June and July in Sydney, Adelaide and Canberra and perhaps during June in Perth. Of course, if there are long cold, cloudy and rainy periods, people will not be exposing their skin to sunlight anyway. Besides these factors, the darker the skin pigmentation the less vitamin D is produced, and older people’s skin is up to 50% less efficient in producing vitamin D than young people’s skin. 

Vitamin D is produced by cholesterol in the surface of the skin and stored initially in fat cells, where it can last for months. The formation of vitamin D3, the active form of vitamin D in the body requires both the liver and the kidneys to be functioning well and this may not be the case in elderly people. Vitamin D3 has a half-life of about 8 hours for adults and about a day for children. Some medications also inhibit the formation of Vitamin D3. So, during winter, older, dark-skinned people who are on certain medications are most unlikely to have adequate vitamin D3 levels. People who are institutionalised or living in residential care often get little sun exposure and hence do not have adequate vitamin D3 levels either. Vitamin D3 levels are also generally lower in obese people and diabetics. So, the vitamin D dice is really loaded against the elderly and particularly dark-skinned, elderly, fat people.

Shift workers and those who work long daylight hours indoors often do not get any late morning or early afternoon sun exposure on enough skin for it to produce much vitamin D. Usually early morning or late afternoon sun exposure does not have sufficient ultra-violet light to produce vitamin D in the skin. So those taking outside exercise in the early morning and late afternoon may not be getting their vitamin D hit from that exposure. Even golfers are sometimes found to have vitamin D deficiency due their use of sunscreens and their covering of skin that they do not want to be exposed to the sun.

Therefore, the proportion of the population with inadequate vitamin D3 levels is going to be particularly high in late winter and early spring but select groups of the population such as the elderly, hospital doctors and nurses working long hours in hospitals, shift workers, people in institutions and work-a-holics could be vitamin D deficient at any time of the year. Whereas people who are most likely to have good vitamin D3 levels are those people who are young, fit and play outdoor sport during the day, children who play outside regularly, beach lovers and regular gardeners.

Medical Literature

There is a growing list of medical literature that indicate why good vitamin D levels are somewhat protective against many respiratory viruses. Recent medical literature is also specific to the SARS-COV-2 virus. A sample of this literature is summarised.

A meta-analysis concluded that vitamin D supplementation was safe and protective against acute respiratory tract infections. It suggested that patients who were severe vitamin D deficient experienced the most benefit[vi].

Vitamin D has multiple roles in the immune system that can modulate the body reaction to an infection. Abu-Amer et al. have described that vitamin D deficiency impairs the ability of macrophages to mature, to produce macrophage-specific surface antigens that produce the lysosomal enzyme acid phosphatase, and to secrete H2O2, a function integral to their antimicrobial function[vii]. Crucial to the immune response are toll-like receptors which recognise molecules related to pathogens and, when activated, release cytokines and induce reactive oxygen species and antimicrobial peptides, cathelicidins and defensins. Several of the toll-like receptors affect or are affected by vitamin D receptor induction[viii] .

The link between hypertension and other cardiovascular diseases and the increased rate of COVID-19 infection is due to the ACE2 receptors being increased by medication so effectively giving the SARS-COV-2 virus more entry points. However, higher levels of ACE2 are associated with better outcomes for coronavirus disease and it has been shown that in the lung, ACE2 protects against acute lung injury[ix]. Also, vitamin D3 supplementation adds to this positive effect[x].

The extent of ACE2 expression in lungs is affected by ageing and gender. In a rodent model study, the decrease of ACE2 was relatively small between the young-adult and the middle-aged groups (25% for male and 18% for female, respectively), but the ACE2 decrease was 67% in older female rats and 78% in older male rats as compared to younger groups.[xi] This decrease of ACE2 with age and gender seems to parallel the increase in COVID-19 mortality as well as the higher mortality in the human males.


  • The outstanding results in Australia, New Zealand, probably in South Africa and possibly Argentina and other mid-latitude countries in South America are partly due to the timing of the primary infection of SARS-COV-2 hitting these sunny mid-latitude countries in late summer. This is not the peak time for respiratory infections in those countries. Also, this timing was co-incident with peak levels of vitamin D levels in their populations. The combination of these factors appears to have conferred some level of immunity from the severe form of this disease.
  • The danger to these countries is that a “second wave” will occur during winter and early spring.
  • Actuarial models of COVID-19 should include parameters for seasonality and population vitamin D levels.
  • Medical exams for large life insurance policies should include Vitamin D level tests.

[i]Brent Walker



[iv] Grant WB, Lahore H, McDonnell SL, et al. Evidence that Vitamin D Supplementation Could Reduce Risk of Influenza and COVID-19 Infections and Deaths. Nutrients. 2020;12(4):E988. Published 2020 Apr 2. doi:10.3390/nu12040988.


[vi] Martineau Adrian R, Jolliffe David A, Richard HL et al (2017) Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data. BMJ 356:i6583

[vii] Abu-Amer Y, Bar-Shavit Z (1993) Impaired bone marrow-derived macrophage differentiation in vitamin D deficiency. Cell Immunol.

[viii] Gruber-Bzura BM (2018) Vitamin D and influenza-prevention or therapy? Int J Mol Sci 19:2419.

[ix] Kuka K, Imai Y, Penninger JM (2006) Angiotensin-converting enzyme 2 in lung diseases. Curr Opin Pharmacol 6:271–276

[x] Martineau Adrian R, Jolliffe David A, Hooper Richard L, Greenberg Lauren, Aloia John F, Bergman Peter et al. Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data BMJ 2017; 356 :i6583

[xi] Xie X, Chen J, Wang X et al (2006) Age- and gender-related difference of ACE2 expression in rat lung [published correction appears in Life Sci 79:2499. Xudong, Xie [corrected to Xie, Xudong]; Junzhu, Chen [corrected to Chen, Junzhu]; Xingxiang, Wang [corrected to Wang, Xingxiang]; Furong, Zhang [corrected to Zhang, Furong]; Yanrong, Liu [corrected to Liu, Yanrong]]. Life Sci 78:2166–2171.

CPD: Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.


Image of Andrew Gower
Andrew Gower says

31 May 2020

Brett, I don’t disagree with your conclusions in general and actuaries should be thinking through how this may be impacted by weather.

The statement the UV light kills viruses is only partly true. UVC does kill viruses (and other pathogens), however whatever UVC is in the suns rays is captured by the ozone layer. UVA and UVB, which are in the suns rays, don’t kill viruses.

Image of Jon Harwood
Jon Harwood says

7 June 2020

Thank you for this thought provoking article Brent. There has been a lot of 'the virus won't affect us because it is hot here' mentally in various countries in tropical environments and whilst the jury is still out, there may be some truth in it. India and Bangladesh's experience with covid has worsened recently, co-inciding with the monsoon season and we are observing a similar deterioration here where I am currently in Nicaragua. Also an interesting inference that Australia was the 'lucky country' in being exposed to the virus over summer.

Comment on the article (Be kind)

Your comment will be revised by the site if needed.