NICE, Vitamin D & Covid-19 – Pocket

Dear JULES,
NICE, Vitamin D & Covid-19

There’s a quick update to last week’s note before we move on to this week’s. Steve Root, an alert reader from New London, New Hampshire USA, noticed that something looked wrong with the odds ratios in the diagram. They didn’t match those in the table. A couple of exchanges between us later and we concluded that the legend in the diagram was wrong – the green dots should have been for hypertension and the purple dots for diabetes – not the other way round. We contacted the authors and they confirmed that this was wrong, and they are fixing it now. Spot of the year already goes to Steve – thank you!

Executive Summary

* The UK National Institute for Care & Health Excellence (NICE) undertook a rapid review of vitamin D for the treatment and prevention of Covid-19. The review was published in December 2020.

* Three research questions were asked about 1) vitamin D as a treatment for Covid-19, 2) vitamin D for prevention of Covid-19 and 3) any associations observed between vitamin D and Covid-19 outcomes.

* NICE found one randomised controlled trial (RCT) to consider for question 1, no RCTs for question 2 and 12 population/case control type studies for question 3.

* The one RCT examining vitamin D for treatment of Covid-19 found that administration of vitamin D at hospital admission significantly reduced the likelihood of being admitted to ICU and no one treated with vitamin D died. The odds ratios were striking.

* The 12 studies examining the association between vitamin D levels and incidence of Covid-19 and/or severity of Covid-19 (including survival from) were conducted in many different countries by many different research teams. Vitamin D levels were reviewed as absolute levels and as deficient or sufficient levels.

* Two UK studies (using Biobank data) found nothing statistically significant. The other 10 studies all concluded that higher/sufficient vitamin D levels were associated with significantly better outcomes from Covid-19. The risk ratios for lower/insufficient vitamin D levels were striking in many cases – up to 15-fold difference in one study and often beyond the 2-fold difference used as a starting point for causality with the Bradford Hill criteria.

* Every single study, every single piece of evidence, was dismissed in the NICE document as “very low” in quality and at “serious risk of bias” or “very serious risk of bias.”

* I saw a lot of evidence to support vitamin D being a valuable intervention to lessen the likelihood of bad outcomes (including death) from Covid-19. NICE essentially concluded “move along – there’s nothing to see here.”

Introduction

The UK National Institute for Care & Health Excellence (NICE) undertook a rapid review of vitamin D for the treatment and prevention of Covid-19. The review was published in December 2020 (Ref 1). The committee undertaking the work asked three research questions:

Research questions

1) What is the clinical effectiveness and safety of vitamin D supplementation for the treatment of Covid-19 in adults, young people, and children?

2) What is the clinical effectiveness and safety of vitamin D supplementation for the prevention of SARS CoV2 infection (and subsequent Covid-19) in adults, young people, and children?

3) Is vitamin D status independently associated with susceptibility to developing Covid-19, severity of Covid-19, and poorer outcomes from Covid-19 in adults, young people, and children?

The three key words in these research questions are 1) treatment 2) prevention and 3) association. The search for evidence was conducted on the period 2002 to October 27th, 2020. This allowed for other coronaviruses, such as SARS CoV1 (symptomatic) and MERS, to be included as indirect evidence. It transpired that only Covid-19 studies were included.

Outcomes of interest

The main outcomes of interest for research questions 1 and 2 were mortality (all-cause and Covid-19 related) and Intensive Care Unit (ICU) admissions. There were a number of secondary outcomes of interest including hospitalization, ventilation, time to clinical cure, complications, and others. I haven’t listed all the secondary outcomes of interest because, as you will see, only one study ended up being considered as evidence for Question 1 and no studies ended up being considered as evidence for question 2.

The outcomes of interest for research question 3 were:

– Incidence of Covid-19 (laboratory/virologically confirmed)

– Covid-19 rate (laboratory/virologically confirmed)

– Severity of confirmed Covid-19 (for example, but not limited to, Centers for Disease Control and Prevention definition, World Health Organization definition, proxy such as hospitalisation, ventilation, ICU admission, and others)

– Poor outcomes (for example, mortality, complications, morbidities post infection, long Covid).

Inclusion & exclusion criteria

The inclusion criteria for research questions 1 and 2 were: systematic review of randomised controlled trials (RCTs); RCTs; and controlled clinical trials. The exclusion criteria for research questions 1 and 2 were: case-control studies; cross section studies; case series and case reports; and qualitative studies (Ref 2).

The inclusion criteria for research question 3 were: systematic review of non-randomised studies; prospective and retrospective cohort studies; case-control studies; cross-sectional studies; and case series. The exclusion criteria for research question 3 were case reports.

Studies found meeting criteria

For research question 1, one study was included (Entrenas Castillio et al 2020 RCT Spain) (Ref 3).

For research question 2, no studies were included.

For research question 3, 12 studies were included (see below).

I checked the studies found with a vitamin D expert, who knows the literature far better than I do, and he thought that the committee had found the studies that met their criteria. I am proceeding on the basis that the studies available for each of these research questions are as presented in the report.

The results

Research question 1

The one study included to answer this research question was conducted in a hospital in Spain. This study involved 76 patients who were hospitalised with lab-confirmed Covid-19. They were randomised in a 2:1 ratio to the treatment arm or the control arm; so 50 people ended up in the treatment arm and 26 in the control arm. The treatment was administration of oral calcifediol (vitamin D3) (0.532 mg) on the day of admission. Patients in the calcifediol treatment group continued with oral calcifediol (0.266 mg) on days 3 and 7, and then weekly until discharge or ICU admission. People in the control arm were not given any vitamin D. Other treatment between the patients remained the same.

Of the 50 patients treated with calcifediol, 1 required admission to ICU (2%) while, of 26 untreated patients, 13 required admission (50%). That was a statistically significant and clinically significant difference. Of the patients treated with calcifediol, none died, and all were discharged without complications. The 13 patients not treated with calcifediol, who were not admitted to the ICU, were discharged. Of the 13 patients admitted to the ICU, 2 died and the remaining 11 were discharged.

The odds ratio (OR) for ending up in ICU was 0.02 (95 %CI 0.002−0.17) for the treatment group vs the no-treatment group. That was almost a 0% chance of the vitamin D group ending up in ICU. When the results were adjusted for differences between the two groups for hypertension and type 2 diabetes, the odds ratio was still 0.03 (95 %CI: 0.003-0.25).

The NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias” for ICU admission and as “very low” in quality and at “very serious risk of bias” for mortality.

Research question 2

NICE found nothing to examine for the research question about prevention.

Research question 3

NICE found 12 studies to help with the research question – is there any evidence for vitamin D being associated with developing Covid-19 or the severity of Covid-19 if diagnosed? The 12 studies were all from 2020. Two studies each came from France, the UK, the US, and from Spain and one each came from China, Germany, Israel, and Turkey. The studies were mainly of population studies and case-control studies.

We’ll whizz through them – extracting the salient point from the original papers. (I’m going to keep numbers and confidence intervals to a minimum for ease of reading. All of these can be found in the paper abstracts, which the references can point you to. Many of the findings are so significant in the normal sense of the word, that confirmation of the statistical significance will just clutter the narrative).

1) Annweiler et al. Retrospective quasi-experimental study France (Ref 4).

The first Annweiler et al study involved 66 frail elderly residents in a French nursing home with an average age of 88. The intervention group was defined as all residents diagnosed with Covid-19 who received an oral dose of 80,000 IU vitamin D3 either in the week following the suspicion or diagnosis of Covid-19, or during the previous month. This definition placed 57 people in the intervention group and 9 in the control. 82.5% of participants in the Intervention group survived Covid-19, compared to 44.4% in the Comparator group.

The researchers concluded that vitamin D supplementation was independently associated with less severe Covid-19 and better survival rate.

The NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias.”

2) Annweiler et al. Retrospective cohort study France (Ref 5).

The second Annweiler et al study involved 77 patients hospitalised for Covid-19 in a geriatric unit with an average age of 88. Group 1 (29 people) regularly supplemented with vitamin D over the previous year. Group 2 (16 people) were supplemented with vitamin D after Covid-19 diagnosis and the comparator group (Group 3 – 32 people) received no vitamin D supplementation. In Group 1, 93% of Covid-19 participants survived at day 14, compared to 81% survivors in Group 2 and 69% survivors in Group 3.

The researchers concluded that vitamin D supplementation was independently associated with less severe Covid-19 and better survival rate.

The NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias.”

3) Hastie et al. Retrospective cohort study UK (Ref 6).

The Hastie et al study examined the UK Biobank data between 16th March and 14th April 2020 for lab confirmed Covid-19 and patient vitamin D levels. Complete data were available for 348,598 participants. Of these, 449 had confirmed Covid-19. After adjustments, vitamin D was not associated with infection rates. The most interesting finding is surely how low the incidence of Covid-19 was given that the peak of the UK Covid-19 curve was 8th April, 2020.

The researchers’ findings did not support a potential link between vitamin D concentrations and risk of Covid-19 infection.

The NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias.”

4) Hernández et al. Case-control study Spain (Ref 7).

The Hernández et al study examined 216 Covid-19 patients and 197 population-based controls (average age 60-61) to see if vitamin D levels were associated with the incidence of Covid-19. Vitamin D deficiency was found in 82% of Covid-19 cases and 47% of population-based controls (Ref 8).

The researchers concluded that vitamin D levels were lower in hospitalised Covid-19 patients than in population-based controls and that these patients had a higher prevalence of deficiency.

The NICE document dismissed this evidence as “very low” in quality and at “serious risk of bias.”

5) Karahan and Katkat. Case-control study Turkey (Ref 9).

The Karahan and Katkat study examined 149 patients (average age 64) hospitalised with Covid-19 for severity and/or mortality. 47 patients had moderate Covid-19; 102 patients were severe-critical. Average blood vitamin D levels were significantly lower in patients with severe-critical Covid-19, compared with moderate Covid-19 (Ref 10). Vitamin D insufficiency was present in 93% of the patients with severe-critical Covid-19.

The researchers concluded that vitamin D was independently associated with mortality in Covid-19 patients.

The NICE document dismissed this evidence as “very low” in quality and at “serious risk of bias.”

6) Kaufman et al. Retrospective cohort study US (Ref 11).

The Kaufman et al study examined over 190,000 patients (average age 54) from the 50 US states and matched them with vitamin D level results (25(OH)D) from the preceding 12 months. The SARS-CoV-2 positivity rate was higher (at 12.5%) in the 39,190 patients with “deficient” vitamin D (<20 ng/mL) than in the 27,870 patients with “adequate” vitamin D (30–34 ng/mL) (8.1%) and the 12,321 patients with values ≥55 ng/mL (5.9%). The association between 25(OH)D levels and SARS-CoV-2 positivity produced an R2 value of 0.96, which indicated strong correlation in the total population (strong is an understatement with an R2 of 0.96).

The researchers concluded that SARS-CoV-2 positivity was strongly and inversely associated with circulating vitamin D levels, a relationship that persisted across latitudes, races/ethnicities, both sexes, and age ranges.

The NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias.”

7) Macaya et al. Case series Spain (Ref 12).

The Macaya et al study examined 80 patients with lab confirmed Covid-19 seen at the emergency department of a hospital where recent vitamin D (25(OH)D) levels were known. Vitamin D deficiency was associated with an increased risk of developing severe Covid-19 (after adjusting for all relevant factors.) The odds ratio was 3.2 – which is beyond a level at which causation can start to be assessed.

The researchers concluded that vitamin D deficiency did show an association with severe Covid-19 infection.

The NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias.”

8) Meltzer et al. Retrospective cohort study US (Ref 13).

The Meltzer et al study examined 489 patients (average age 49) at an urban medical centre for whom vitamin D levels had been measured within the previous year. Between March 3rd, and April 10th, 2020, 71 of these patients tested positive for Covid-19. Vitamin D deficiency was defined as 25(OH)D < 20 ng/mL. Testing positive for Covid-19 was associated with vitamin D deficiency (relative risk 1.77) when compared with vitamin D sufficiency. Predicted Covid-19 rates in the deficient group were 21.6% vs 12.2% in the sufficient group. The researchers concluded that vitamin D deficiency was associated with increased Covid-19 risk. The NICE document dismissed this evidence as “very low” in quality and at “serious risk of bias.” 9) Merzon et al. Case-control study Israel (Ref 14). The Merzon et al Israeli study examined 14,000 members of a health service who were tested for Covid-19 between Feb 1st, and April 30th, 2020, who had at least one previous blood test for their 25(OH)D level. Suboptimal vitamin D levels in this study were defined as < 30 ng/mL. The 782 people who tested positive for Covid-19 had significantly lower average vitamin D levels. People with low vitamin D levels had an increased likelihood of Covid-19 infection (1.58 times) and increased likelihood of hospitalization (2.09 times). The researchers concluded that “low plasma 25(OH)D levels appear to be an independent risk factor for COVID-19 infection and hospitalization.” The NICE document dismissed this evidence as “very low” in quality and at “serious risk of bias.” 10) Radujkovic et al. Retrospective cohort study Germany (Ref 15). The Radujkovic et al study examined 185 patients diagnosed with Covid-19 who presented at a German health centre. Vitamin D status was assessed at the time of admission. Vitamin D (25(OH)D) deficiency was set at a low bar of < 12 ng/mL. During an average period of 66 days, 93 patients required hospitalisation; 28 patients had a severe outcome (IMV/D – invasive mechanical ventilation and/or death). Among these 28 severe outcomes, there were 16 deaths. A total of 41 patients were vitamin D deficient. When adjusted for age, gender, and comorbidities, vitamin D deficiency was associated with higher risk of IMV/D and death. The risk ratios were 6.12 for IMV/D and 14.73 for death. The researchers concluded that “our study demonstrates an association between vitamin D deficiency and severity/mortality of COVID-19.” The NICE document dismissed this evidence as “very low” in quality and at “serious risk of bias.” 11) Raisi-Estabragh et al. Nested case-control study UK (Ref 16). The Raisi-Estabragh et al studywas the second study to use the UK Biobank data. This study specifically examined men and Black, Asian, and Minority Ethnic (BAME) individuals, as these were patients identified as at higher risk of a poor outcome from Covid-19 than women and non-BAME individuals. This study asked a different research question and I’m not sure that it should have been included. It sought to see if any factors (of which vitamin D was just one) could explain why men and BAME individuals were having worse outcomes. A total of 4,510 people were assessed. The study confirmed "there was over-representation of men and BAME ethnicities in the COVID-19 positive group." However, the study couldn’t explain why. The conclusion was: “In this study, sex and ethnicity differential pattern of COVID-19 was not adequately explained by variations in cardiometabolic factors, 25(OH)-vitamin D levels or socio-economic factors.” This mattered little anyway as the NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias.” 12) Ye et al. Case-control study China (Ref 17). The Ye et al study compared the vitamin D (25(OH)D) levels and rates of vitamin D deficiency between 80 healthy controls and 62 patients diagnosed with Covid-18 and admitted to a hospital in China. Cases were categorized into asymptomatic, mild/moderate, and severe/critical disease. The vitamin D concentration in Covid-19 patients was much lower than that in healthy controls. Vitamin D levels were the lowest in severe/critical cases, compared with mild cases. Significantly higher rates of vitamin D deficiency were found in Covid-19 cases (42%) compared to healthy controls (11%). Vitamin D deficiency was the greatest in severe/critical cases (80%), compared with mild cases (36%). The researchers concluded that vitamin D was associated with Covid-19 severity and added "a potential threshold of 25(OH)D (41.19nmol/L) to protect against COVID-19 was identified.” The NICE document dismissed this evidence as “very low” in quality and at “very serious risk of bias.” Summary The one RCT examining vitamin D for treatment of Covid-19 found that administration of oral calcifediol (vitamin D) at hospital admission significantly reduced the likelihood of being admitted to ICU and no one treated with vitamin D died. The odds ratios were striking. There were 12 studies (largely population and case control studies) examining the association between vitamin D levels and incidence of Covid-19 and/or severity of Covid-19 (including survival from). The studies were conducted in many different countries by many different research teams. Vitamin D levels were reviewed as absolute levels and as deficient or sufficient levels. The two UK studies using Biobank found nothing statistically significant. The other 10 studies all concluded that higher/sufficient vitamin D levels were associated with significantly better outcomes from Covid-19. The risk ratios for lower/insufficient vitamin D levels were striking in many cases – up to 15-fold difference in one study and often beyond the 2-fold difference used as a starting point for causality with the Bradford Hill criteria (Ref 18). Every single study, every single piece of evidence, was dismissed in the NICE document as “very low” in quality and at “serious risk of bias” or “very serious risk of bias.” Look at the evidence for yourself and decide if NICE were right to dismiss it all as they did. Look at the evidence for yourself and ask – can I (or my loved ones) really risk having low or insufficient vitamin D levels right now? Ask yourself – is there anything to lose by taking a vitamin D supplement for the foreseeable future and for my loved ones to do the same? (Ref 19). You may also like to ponder, as I found myself doing, why NICE would dismiss such a cheap, safe, and effective nutrient, which seems highly likely to help and does no harm. Until the next time All the best - Zoë References Ref 1: https://www.nice.org.uk/guidance/ng187/evidence/evidence-reviews-for-the-use-of-vitamin-d-supplementation-as-prevention-and-treatment-of-covid19-pdf-8957587789 Ref 2: https://www.zoeharcombe.com/2020/08/the-hierarchy-of-evidence/ Ref 3: Entrenas Castillo et al. Effect of calcifediol treatment and best available therapy versus best available therapy on intensive care unit admission and mortality among patients hospitalized for COVID-19: A pilot randomized clinical study.; J Steroid Biochem Mol Biol. 2020. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7456194/ Ref 4: Annweiler et al. Vitamin D and survival in COVID-19 patients: A quasi-experimental study. 2020. The Journal of steroid biochemistry and molecular biology. https://pubmed.ncbi.nlm.nih.gov/33065275/ Ref 5: Annweiler et al. Vitamin D Supplementation Associated to Better Survival in Hospitalized Frail Elderly COVID-19 Patients: The GERIA-COVID Quasi-Experimental Study. Nutrients. 2020. https://pubmed.ncbi.nlm.nih.gov/33147894/ Ref 6: Hastie et al. Vitamin D concentrations and COVID-19 infection in UK Biobank. Diabetes & metabolic syndrome. 2020. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7204679/ Ref 7: Hernandez et al. Vitamin D Status in Hospitalized Patients With SARS-CoV-2 Infection. The Journal of Clinical Endocrinology & Metabolism. 2020. https://academic.oup.com/jcem/advance-article/doi/10.1210/clinem/dgaa733/5934827 Ref 8: In COVID-19 patients, mean ± standard deviation 25OHD levels were 13.8 ± 7.2 ng/mL, compared with 20.9 ± 7.4 ng/mL in controls (P < .0001). Ref 9: Karahan, S. and Katkat, F. Impact of Serum 25(OH) Vitamin D Level on Mortality in Patients with COVID-19 in Turkey. Journal of Nutrition, Health and Aging. 2020. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7533663/ Ref 10: The differences were quite striking 10.1 ± 6.2 vs. 26.3 ± 8.4 ng/mL, respectively, p<0.001. Ref 11: Kaufman et al. SARS-CoV-2 positivity rates associated with circulating 25-hydroxyvitamin D levels. PloS one. 2020. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0239252 Ref 12: Macaya et al. Interaction between age and vitamin D deficiency in severe COVID-19 infection. Nutricion hospitalaria. 2020. https://pubmed.ncbi.nlm.nih.gov/32960622/ Ref 13: Meltzer et al. Association of Vitamin D Status and Other Clinical Characteristics With COVID-19 Test Results. JAMA network open. 2020. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2770157 Ref 14: Merzon et al. Low plasma 25(OH) vitamin D level is associated with increased risk of COVID-19 infection: an Israeli population-based study. The FEBS journal. 2020. https://pubmed.ncbi.nlm.nih.gov/32700398/ Ref 15: Radujkovic et al. Vitamin D Deficiency and Outcome of COVID-19 Patients. Nutrients. 2020. https://pubmed.ncbi.nlm.nih.gov/32927735/ Ref 16: Raisi-Estabragh et al. Greater risk of severe COVID-19 in Black, Asian and Minority Ethnic populations is not explained by cardiometabolic, socioeconomic or behavioural factors, or by 25(OH)-vitamin D status: study of 1326 cases from the UK Biobank. Journal of public health (Oxford, England). 2020. https://academic.oup.com/jpubhealth/article/42/3/451/5859581 Ref 17: Ye et al. Does Serum Vitamin D Level Affect COVID-19 Infection and Its Severity? A Case-Control Study. Journal of the American College of Nutrition. 2020. https://www.tandfonline.com/doi/full/10.1080/07315724.2020.1826005 Ref 18: https://www.zoeharcombe.com/2016/09/the-bradford-hill-criteria/ Ref 19: Eating a 200g tin of oily fish with the bones and skin will provide approximately 15mcg of vitamin D, which is a basic intake. Who consumes this daily? Sunbathing is a natural way to acquire vitamin D, but not possible for most people for many months of the year.

Print Friendly, PDF & Email