Evidence Supports a Causal Model for Vitamin D in COVID-19 Outcomes

  1. Gareth Davies
  2. Attila R Garami
  3. Joanna Byers

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Background

The COVID-19 pandemic caused by the coronavirus SARS-CoV-2 seemed to affect locations in the northern hemisphere most severely appearing to overlap with the pattern of seasonal vitamin D deficiency. Integrating available knowledge, we hypothesised that vitamin D could play a causal role in COVID-19 outcomes.

Objective

We set out to analyse the relationship between COVID-19 severity and latitude, and construct a causal inference framework to validate this hypothesis.

Methods

We analysed global daily reports of fatalities and recoveries from 239 locations from 22nd Jan 2020 to 9th April 2020. We quantified local COVID-19 outbreak severity to clearly distinguish the latitude relationship and identify any outliers breaking this pattern, and analysed the timeline of spread. We then used a causal inference framework to distinguish correlation from cause using observational data with a hypothetico-deductive method of proof. We constructed two contrasting directed acyclic graph (DAG) models, one causal and one acausal with respect to vitamin D and COVID-19 severity, allowing us to make 19 verifiable and falsifiable predictions for each.

Results

Our analysis confirmed a striking correlation between COVID-19 severity and latitude, and ruled out the temporal spread of infection as an explanation. We compared observed severity for 239 locations with our contrasting model. In the causal model, 16 predictions matched observed data and 3 predictions were untestable; in the acausal model, 14 predictions strongly contradicted observed data, 2 appeared to contradict data, and 3 were untestable.

Discussion

We show in advance of RCTs that observed data strongly match predictions made by the causal model but contradict those of the acausal model. We present historic evidence that vitamin D supplementation prevented past respiratory virus pandemics. We discuss how molecular mechanisms of vitamin D action can prevent respiratory viral infections and protect against ARDS. We highlight vitamin D’s direct effect on the renin-angiotensin-system (RAS), which in concert with its other effects, can modify host responses thus preventing a cytokine storm and SARS-CoV-2-induced pathological changes. Emerging clinical research confirms striking correlations between hypovitaminosis D and COVID-19 severity, in full alignment with our study.

Conclusion

Our novel causal inference analysis of global data supports the hypothesis that vitamin D levels play a key role in COVID-19 outcomes. The data set size, supporting historical, biomolecular, and emerging clinical research evidence altogether suggest that a very high level of confidence is justified. Vitamin D prophylaxis offers a widely available, low-risk, highly-scalable, and cost-effective pandemic management strategy including the mitigation of local outbreaks and a second wave. Timely implementation of vitamin D supplementation programmes worldwide is critical; initial priority should be given to those who are at the highest risk, including the elderly, immobile, homebound, BAME and healthcare professionals. Population-wide vitamin D sufficiency could prevent seasonal respiratory epidemics, decrease our dependence on pharmaceutical solutions, reduce hospitalisations, and thus greatly lower healthcare costs while significantly increasing quality of life.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.05.01.20087965: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Some CSV files were converted to Excel format for previsualisation checks before being imported into Matlab.
    Matlab
    suggested: (MATLAB, RRID:SCR_001622)

    Results from OddPub: Thank you for sharing your code and data.

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Einstein famously expressed this limitation when he said, “No amount of experimentation can ever prove me right; a single experiment can prove me wrong”. Our proof considers two contrasting models, the verification of the causal model can be considered proof since the contrasting acausal model was falsified thus removing any remaining doubt about the result. Historical Evidence Suggests Causal Role of Vitamin D in Pandemic Prevention: Vitamin D was discovered five years after the 20th century’s deadliest pandemic, Spanish Influenza, caused by the avian virus H1N1, which infected one third of the world’s population (500M/1,8B) with a mortality rate of 10% to 20%. Vitamin D deficiency was rife: rickets was present in 80% to 90% of children living in northern Europe and north eastern US [7]. Lacking antibiotics, many died of secondary bacterial infections [47]. Sanatoria promoted fresh air and sunshine as a treatment that worked, but no-one knew why. Between 1930 and 1950, Vitamin D fortification was commonplace and additives eradicated hypovitaminosis D. The practice was banned in 1950 on the grounds that uncontrolled dosing had led to instances of hypercalcaemia and hypovitaminosis D returned. The introduction of sunscreen creams containing UVB blockers in the 1970s further exacerbated this [7]. We note with interest that on the timeline of major influenza epidemics [48] there is a 37 year period from 1920 to 1957 where no new flu strains seem to appear and no new pandemics oc...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2020.05.01.20087965v3 on medRxiv
  3. ScreenIT

    SciScore for 10.1101/2020.05.01.20087965: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Some CSV files were converted to Excel format for previsualisation checks before being imported into Matlab.
    Matlab
    suggested: (MATLAB, SCR_001622)

    Results from OddPub: Thank you for sharing your code.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore is not a substitute for expert review. SciScore checks for the presence and correctness of RRIDs (research resource identifiers) in the manuscript, and detects sentences that appear to be missing RRIDs. SciScore also checks to make sure that rigor criteria are addressed by authors. It does this by detecting sentences that discuss criteria such as blinding or power analysis. SciScore does not guarantee that the rigor criteria that it detects are appropriate for the particular study. Instead it assists authors, editors, and reviewers by drawing attention to sections of the manuscript that contain or should contain various rigor criteria and key resources. For details on the results shown here, including references cited, please follow this link.

  4. Version published to 10.1101/2020.05.01.20087965v2 on medRxiv
  5. Version published to 10.1101/2020.05.01.20087965v1 on medRxiv