Skip to main content
. 2021 May 10;13(5):1589. doi: 10.3390/nu13051589

Table 5.

COVID-19 Severity indicated by outcomes with study-specific definitions.

Author [Ref] Micronutrient in Question Outcome Definition (Study-Specific) Reported Summary Risk Estimate: Odds Ratio (OR) (95% Confidence Interval); % Population Infected (Infected/Population Size) Key Findings
Supplementation Deficiency
Supplemented Non-supplemented Quintile 1 Quintile 2 Quintile 3
Outcome: Clinical severity
Hastie [17] Vitamin D Hospitalisation rate Reference; NR 1.06 (0.89–1.26) a; NR 1.1 (0.88–1.37) a; NR

  • No significant difference in hospitalisation rates between those with insufficient or deficient Vitamin D, compared to those with sufficient Vitamin D.

  • No significant difference in hospitalisation rates for every 10 nmol/L increase in Vitamin D levels (Incidence rate ratio: 1.00 (0.96–1.05) a)
Merzon [33] Vitamin D Hospitalisation rate

  • Significantly lower mean plasma 25(OH)D levels in hospitalised patients (18.38 ng/mL (95% CI: 16.79–19.96)), compared to non-hospitalised patients. 20.45 ng/mL (95% CI: 20.22–20.68), p < 0.001).
Radujkovic [37] Vitamin D Hospitalisation rate NR; 44.4% (64/144) * NR; 70.7% (29/41) *

  • Significantly higher proportion of hospitalised patients in those with deficiency than those with no deficiency (p = 0.004)

  • Significantly different median (IQR) Vitamin D levels between inpatients (14.6 ng/mL (10.7–21.0)) and outpatients (18.6 ng/mL (14.2–26.0), p = 0.001)
Karahan [27] Vitamin D3 Chinese Clinical Guideline (Moderate disease) NR; 100% (12/12) * NR; 79.4% (27/34) * NR; 7.8% (8/103) *

  • Significantly different proportions of patients with moderate and severe/critical disease onset across those with normal, insufficient and deficient Vitamin D3 levels (p < 0.001)
Chinese Clinical Guideline (Severe/critical disease) NR; 0% (0/12) * NR; 18.9% (7/37)* NR; 92.2% (95/103)
Maghbooli [30] Vitamin D3 CDC criteria (Severe/critical disease) Reference; 63.6% (49/77) * 1.59 (1.05–2.41) *,b; 77.2% (122/158) *

  • 59% b higher odds of severe/critical disease amongst patients with deficiency or insufficiency, compared to those without deficiency

  • Significantly higher proportion of severe/critical onset amongst deficient or insufficient patients compared non-deficient patients (p = 0.02)
Ye [43] Vitamin D Chinese National Health Commission Guidelines (6th edition) (Mild/moderate disease) Reference; 88.9% (32/36) # NR; 69.2% (18/26) #

  • 14.18 c times higher odds for severe/critical COVID-19 amongst patients with vitamin D deficiency compared to those without deficiency (p = 0.034)

  • Significantly higher median (IQR) 25(OH)D (nmol/L) levels in mild/moderate patients (56.6 nmol/L (44.6–66.4); n = 50 patients), compared to severe/critical patients (38.2 nmol/L (33.2–50.5); n = 10 patients, p < 0.05)
Chinese National Health Commission Guidelines (6th edition) (Severe/critical disease) Reference; 5.6% (2/36) # 15.18 (1.23–187.45) c; 30.8% (8/26) #
Number of symptomatic patients NR; 80% (8/10) NR; 100% (26/26) NR; 100% (26/26)

  • Significantly different proportions of symptomatic patients across those with sufficient, insufficient and deficient Vitamin D levels (p = 0.004)
Pizzini [36] Vitamin D Requiring hospitalisation, respiratory support or intensive care treatment

  • No significant difference in mean ± SD 25(OH)D levels, 8 weeks after disease onset, across patients with varying COVID-19 severity: overall 54 ± 25 nmol/L, mild 64 ± 31 nmol/L, moderate 54 ± 19 nmol/, and severe 50 ± 24 nmol/L (p > 0.05)
Calcium (total, ionised)

  • No significant difference in mean ± SD total calcium levels, 8 weeks after disease onset, across patients with varying COVID-19 severity: overall 2.37 ± 0.09 mmol/L, mild 2.37 ± 0.09 mmol/L, moderate: 2.36 ± 0.09 mmol/L, and severe: 2.39 ± 0.08 mmol/L (p > 0.05)

  • No significant difference in mean ± SD ionised calcium levels, 8 weeks after disease onset, across patients with varying COVID-19 severity: overall 1.22 ± 0.04 mmol/L, mild: 1.24 ± 0.03 mmol/L, moderate: 1.22 ± 0.04 mmol/L, and severe: 1.22 ± 0.04 mmol/L (p > 0.05)
Iron (Serum Ferritin)

  • Significantly different mean ± SD serum ferritin levels, 8 weeks after disease onset, across patients with varying COVID-19 severity: overall 263 ± 230 µg/L, mild: 139 ± 118 µg/L, moderate: 260 ± 183 µg/L, and severe: 317 ± 271 µg/L (p = 0.001)
Dahan [24] Iron (Serum Ferritin) Report of the WHO-China Joint Mission

  • Significant difference in mean serum ferritin levels across disease severity groups: mild 327.27 ng/mL, moderate 1555 ng/mL, severe 2817.6 ng/mL (p = 0.003).

  • Significantly higher mean serum ferritin levels, in the moderate and severe disease groups, compared to the mildly ill group (p = 0.006 and 0.005, respectively).

  • No significant difference between the moderate and severe disease groups ferritin levels (p > 0.05) after excluding extremely deviant cases.

  • Significantly higher mean serum ferritin levels in severe patients (2817.6 ng/mL), than in non-severe patients (708.6 ng/mL, p = 0.02).
Sun [19] Iron (Serum Ferritin) New Coronavirus Pneumonia Prevention and Control Program, 7th edition

  • Significant difference in serum ferritin levels (times the upper limit of the normal) between the mild and critically ill groups, and between the moderate disease and critically ill groups (p < 0.01);

  • Serum ferritin levels, mean ± SD (times the upper limit of the normal)—mild: 0.55 ± 0.5, moderate: 2.00 ± 2.20, severe: 3.20 ± 1.47, critically ill: 5.08 ± 3.29
Zhao [44] Iron (Serum) Chinese National Health Commission Guidelines (7th edition)

  • Significantly lower median pre-treatment serum iron levels in patients with severe COVID-19 compared to those with mild COVID-19 (p < 0.05).

  • No significant difference in median(IQR) pre-treatment serum iron levels across all groups—mild: 6.6 μmol/L (5.4–10.9), severe: 4.9 μmol/L (4.0–8.1), critical: 5.2 μmol/L (2.6–7.3) (p > 0.05).
Smith [39] Iron (Ferritin) Hospitalisation and/or ICU admission, requiring mechanical ventilation and/or death

  • Significantly higher median (IQR) ferritin levels in severe patients (1163 ng/mL (640.0–1967.0)) compared to those with moderate condition (624.0 ng/mL (269.7–954.0), p < 0.01)
Sonnweber [40] Iron/Ferritin ICU admission, requiring oxygen therapy or respiratory support

  • No significant difference in mean ± SD iron levels between severity groups: mild 18 ± 6 µmol/L, moderate:16 ± 6 µmol/L, and severe/critical 15 ± 6 µmol/L (p > 0.05).

  • Significantly different mean ± SD ferritin levels between severity groups: mild 139 ± 118 µmol/L, moderate 260 ± 183 µmol/L, and severe/critical 317 ± 271 µmol/L (p = 0.001).
Yasui [42] Iron (Ferritin) ICU admission, requiring oxygen therapy or respiratory support All patients:
NR; 36.7% (11/30) *
Subset of inpatient: NR; 28.6% (6/21) 		All patients:
NR; 6.3% (2/32) *
Subset of inpatient: NR; 12.5% (1/8)

  • Significantly higher proportion of severe patients among those with ≥300 ng/mL (males)/≥200 ng/mL (females) ferritin levels, compared to those with ferritin levels lower than that, amongst all patients (p = 0.008)

  • Significantly higher mean ± SD ferritin levels in severe patients (1117 ± 654 ng/mL), compared to the mild/moderate patients (386 ± 393 ng/mL, p = 0.00002)
Zinc Subset of inpatient: NR; 5% (1/20) * Subset of inpatient: NR; 66.7% (6/9) *

  • Significantly higher proportion of severe patients among those with≥70µg/dL zinc levels, compared to those with zinc levels lower than that, amongst all inpatients (p = 0.0003)

  • Significantly lower mean ± SD zinc levels in severe patients (62.4 ± 19.2 μg/dL), compared to mild/moderate patients (87.7 ± 19.1 μg/dL, p = 0.005)
Jothimani [26] Zinc Number of symptomatic patients Reference; 90% (18/20) 3.15 (0.58–17.67); 96.3% (26/27)

  • No association between zinc deficiency and odds of symptoms onset

  • No significant difference in proportion of symptomatic patients between patients with and without zinc deficiency (p > 0.05)
Outcome: Progression to respiratory-related complication
Jothimani [26] Zinc ARDS development NR; 0% (0/20) NR; 18.5% (5/27)

  • No significant difference proportion of patients developing ARDS between those with and without zinc deficiency (p > 0.05)
Wu [41] Iron (Serum Ferritin) ARDS development 3.53 (1.52-8-16); NR Reference; NR

  • 253% higher odds of ARDS development in those with serum ferritin levels >300 ng/mL, compared to those with levels ≤300 ng/mL (p = 0.003)

  • Significantly higher median serum ferritin levels in patients with ARDS, compared to those without ARDS (difference: 545.5 ng/mL (IQR 332.15–754.44), p < 0.001)
Maghbooli [30] Vitamin D3 ARDS development NR; 11.7% (9/77) NR; 17.1% (27/158)

  • No significant difference in patients developing ARDS between those with and without Vitamin D deficiency (p > 0.05)
Im [25] Vitamin D3 Pneumonia incidence, or requiring high-flow nasal cannula, mechanical ventilator, and extracorporeal membrane oxygenation or death NR; 50% (6/12) # NR; 68.4% (26/38) #

  • Similar proportions of patients with outcomes between those with deficiency and no deficiency (p = not reported)

  • About half as many patients with Vitamin B6 deficiency developing respiratory complications compared to patients without deficiency (p = not reported)
Vitamin B6 NR; 66% (31/47) # NR; 33.3% (1/3) #
Vitamin B9 NR; 64.6% (31/48) # NR; 50% (1/2) #
Selenium NR; 65.5% (19/29) # NR 61.9% (13/21) #
≥1 deficiency NR; 44.4% (4/9) # NR; 68.3% (28/41) #
Capone [46] Vitamin C & Zinc Requiring invasive mechanical ventilation

  • Requiring invasive mechanical ventilation inversely correlated with zinc and vitamin C supplementation amongst 73 hospitalised patients (Pearson correlation coefficient = −0.20)
Radujkovic [37] Vitamin D Requiring any form of oxygen therapy NR; 35.4% (54/144) * NR;63.4% (26/41) *

  • Significantly higher proportion of deficient patients requiring any form of oxygen therapy compared to non-deficient patients (p < 0.001)
Tan [49] Vitamin D, Magnesium & Vitamin B12 Requiring oxygen therapy 0.195 (0.041–0.926) d,
0.182 (0.038–0.859) e;
11.7% (2/17) *		 Reference; 30.8% (8/26) *

  • 80.5% d to 81.8% e lowered odds of requiring oxygen therapy amongst patients provided with micronutrient blend, compared to those not provided with it.

  • Significantly higher proportion of patients not provided with micronutrient blend requiring oxygen therapy, compared to those provided with micronutrient blend (p = 0.006)
Outcome: Composite outcome (with multiple outcomes)
Liu [28] Calcium (Serum calcium) Need for mechanical ventilation, ICU admission due to COVID-19 episode, or all-cause mortality during admission Reference; 25% (10/40) * 2.962 (1.085–8.090) f; 47.8% (32/67) *

  • 196% f higher odds for composite outcome in those with hypocalcemia compared to those with normal serum calcium levels.

  • Significantly lower median (IQR) serum calcium levels in patients with composite outcome (2.01 mmol/L (1.97–2.05)) compared to those without the outcome (2.10 mmol/L (2.03–2.20), p < 0.001).
Macaya [29] Vitamin D3 α Death, ICU admission or requiring high flow oxygen (greater than nasal cannula) Reference; 31.4% (11/35) 3.2 (0.9–11.4) g; 44.4% (20/45)

  • 220% g higher odds for death, ICU admission or requiring oxygen in those with deficiency, compared to those without deficiency

  • Patient age significantly modified the association (pinteraction = 0.03)

  • Age over 75 years (3rd tertile) and male gender were significantly associated with the composite outcome [OR 10.4 (95% CI: 2.0–54.8) vs. the first tertile, p = 0.006; OR 6.2 (95% CI: 2.0–19.5), p = 0.002, respectively].

  • No significant difference in the median (IQR) levels of 25(OH)D between patients with outcome (13 ng/mL (8–25)) and without the outcome (19 ng/mL (9–30), p > 0.05)
Vitamin D3 β NR; 45.5% (20/44) NR; 20.6% (11/36)

  • No significant difference in proportion of patients with composite outcome, between those supplemented or not supplemented (p > 0.05)
Radujkovic [37] Vitamin D3 ᶧ Mechanical invasive ventilation and/or death from COVID-19 episode For all subjects: Reference; NRFor inpatients only: Reference; 87.5% (56/64) # For all subjects: 6.12(2.79–13.42) h; NR
For inpatients only: 4.65 (2.11–10.25) h; 89.7% (26/29) #

  • 365% h and 512% h higher hazards for composite outcome in those with vitamin D deficiency, compared to those without deficiency in the inpatient and all patients respectively.

  • Significantly different proportion of patients receiving different types of maximum oxygen therapy (including none) differed significantly between the deficiency and non-deficiency group, for entire cohort (p < 0.001) and inpatient subgroup (p = 0.004)
Vitamin D3 ᶲ Reference; NR For all subjects: 5.75 (1.73–19.09) h; NR
For inpatients only: 3.99 (1.2–13.28) h; NR

  • 299% h and 475% h higher hazards for composite outcome in those with vitamin D insufficiency, compared to those without insufficiency in the inpatient and all patients respectively.
Tan [49] Vitamin D, Magnesium & Vitamin B12 Requiring oxygen therapy or ICU admission due to COVID-19 episode NR; 5.9% (1/17) * NR; 30.8% (8/26) *

  • Significantly higher proportion of composite outcome in those not provided with the micronutrient blend, compared to those provided with the blend (p = 0.006)

* between-group p < 0.05; # between group p-value not reported; α Exposure was deficiency assessed as an ordinal variable (Yes/No); β Exposure was supplementation status assessed as an ordinal variable (Yes/No); ᶧ quintiles classification: no deficiency and deficiency; ᶲ quintiles classification: no insufficiency and insufficiency; a Incidence rate ratio adjusted for age, sex, ethnicity, month of assessment, Townsend deprivation quintile, household income, BMI category, smoking status, diabetes, systolic blood pressure, diastolic blood pressure, self-reported health rating, and long-standing illness, disability or infirmity; b Adjusted for age, sex, BMI, smoking and history of a chronic medical disorder; c Adjusted for age (every 10 years), gender, renal failure, diabetes, and hypertension; d Adjusted for age; e Adjusted for hypertension; f Adjusted for age, and c-reactive protein, procalcitonin, interleukin-6 and D-dimer levels; g Adjusted for age, gender, obesity and severe chronic kidney disease; h Hazard ratio adjusted for age (≥60 years), gender and comorbidity status (Yes/No).