Top
Introduction
Results
Exclusions
Conclusion
 
Study Notes⏵
Methods and Data
Supplementary
References
 
All studies
Mortality
Ventilation
ICU admission
Hospitalization
Progression
Recovery
COVID‑19 cases
Peer reviewed
Exclusions
 
Feedback
Home
c19early.org COVID-19 treatment researchExerciseExercise (more..)
Melatonin Meta
Bromhexine Meta Metformin Meta
Budesonide Meta
Cannabidiol Meta Molnupiravir Meta
Colchicine Meta
Conv. Plasma Meta
Curcumin Meta Nigella Sativa Meta
Ensovibep Meta Nitazoxanide Meta
Famotidine Meta Paxlovid Meta
Favipiravir Meta Peg.. Lambda Meta
Fluvoxamine Meta Quercetin Meta
Hydroxychlor.. Meta Remdesivir Meta
Ivermectin Meta
Lactoferrin Meta
Home Adoption Other Treatments
Loading...
Analgesics⏵
Antiandrogens⏵
Bromhexine
Budesonide
Cannabidiol
Colchicine
Conv. Plasma
Curcumin
Ensovibep
Famotidine
Favipiravir
Fluvoxamine
Hydroxychlor..
Ivermectin
Lactoferrin
Lifestyle⏵
Melatonin
Metformin
Minerals⏵
Molnupiravir
Monoclonals⏵
Naso/orophar..⏵
Nigella Sativa
Nitazoxanide
Paxlovid
Peg.. Lambda
Quercetin
Remdesivir
Vitamins⏵

Exercise for COVID-19: real-time meta analysis of 61 studies

@CovidAnalysis, September 2023
Tweet
Share
PDF
Studies
Submit FeedbackFeedback
https://c19early.org/exmeta.html
 
0 0.5 1 1.5+ All studies 40% 61 1,801,085 Improvement, Studies, Patients Relative Risk Mortality 50% 18 1,483,571 Ventilation 46% 2 43,773 ICU admission 41% 4 708,149 Hospitalization 35% 18 897,490 Progression 75% 2 323 Cases 23% 22 214,381 Peer-reviewed 40% 59 1,794,406 Prophylaxis 40% 60 1,800,646 Late 47% 1 439 Exercise for COVID-19 c19early.org Sep 2023 after exclusions Favorsexercise Favorsinactivity
•Statistically significant lower risk is seen for mortality, ventilation, ICU admission, hospitalization, progression, recovery, and cases. 47 studies from 47 independent teams in 23 countries show statistically significant improvements.
•Meta analysis using the most serious outcome reported shows 40% [34‑46%] lower risk. Results are similar for higher quality and peer-reviewed studies.
•Results are robust — in exclusion sensitivity analysis 47 of 61 studies must be excluded to avoid finding statistically significant efficacy in pooled analysis.
•Results are consistent with the overall risk of all cause mortality based on cardiorespiratory fitness — Laukkanen show RR 0.55 [0.50-0.61] for the top vs. bottom tertiles.
•Most studies analyze exercise/physical activity levels before infection, comparing regular/moderate exercise and lower/no exercise. Risk may increase with more extreme activity levels.
•No treatment or intervention is 100% effective. All practical, effective, and safe means should be used based on risk/benefit analysis.
•All data to reproduce this paper and sources are in the appendix. Other meta analyses show significant improvements with exercise for mortality Ezzatvar, Halabchi, Liu, Rahmati, Sittichai, ICU admission Rahmati, hospitalization Ezzatvar, Halabchi, Li, Rahmati, severity Ezzatvar, Liu, Sittichai, and cases Ezzatvar.
Evolution of COVID-19 clinical evidence Exercise p<0.0000000001 Acetaminophen p=0.0000018 2020 2021 2022 2023 Effective Harmful c19early.org September 2023 meta analysis results (pooled effects) 100% 50% 0% -50%
Highlights
Exercise reduces risk for COVID-19 with very high confidence for mortality, ICU admission, hospitalization, recovery, cases, and in pooled analysis, and low confidence for ventilation and progression.
We show traditional outcome specific analyses and combined evidence from all studies.
Real-time updates and corrections, transparent analysis with all results in the same format, consistent protocol for 56 treatments.
A
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2+ Fernandez 47% 0.53 [0.29-0.90] death 16/201 62/238 Improvement, RR [CI] Treatment Control Tau​2​ = 0.00, I​2​ = 0.0%, p = 0.014 Late treatment 47% 0.53 [0.29-0.90] 16/201 62/238 47% lower risk Hamer 28% 0.72 [0.61-0.87] hosp. n/a n/a Improvement, RR [CI] Treatment Control Brawner 74% 0.26 [0.11-0.58] hosp. n/a n/a Tret'yakov 98% 0.02 [0.00-0.27] severe case 0/27 53/266 Gao -105% 2.05 [1.39-3.02] cases case control Ho 35% 0.65 [0.48-0.89] hosp. 213/123,588 59/14,887 Halabchi 89% 0.11 [0.01-1.46] death 0/249 79/4,445 Zhang 26% 0.74 [0.48-1.14] death n/a n/a Li 81% 0.19 [0.05-0.74] severe case n/a n/a per SD increase Tavakol 69% 0.31 [0.10-1.02] severe case 3/64 19/124 Yates 45% 0.55 [0.38-0.79] death 72/163,912 62/30,119 COVIDENCE UK Holt 17% 0.83 [0.63-1.09] cases 15,227 (all patients) Cho 53% 0.47 [0.26-0.87] death case control Sallis 59% 0.41 [0.22-0.76] death 11/3,118 170/6,984 Christensen 63% 0.37 [0.16-0.85] death 543 (n) 529 (n) Lobelo 20% 0.80 [0.66-0.97] hosp. 2,121 (n) 1,648 (n) Latorre-Román 76% 0.24 [0.05-1.04] hosp. n/a n/a Marcus 42% 0.58 [0.48-0.71] symp. case 240/10,627 134/3,708 Yuan 91% 0.09 [0.01-1.65] death 0/61 6/103 Brandenburg -6% 1.06 [0.23-4.79] hosp. 102 (n) 39 (n) Bielik 30% 0.70 [0.40-1.21] mod. case 775 (n) 365 (n) af Geijerstam 50% 0.50 [0.31-0.81] death 318,902 (n) 336,271 (n) Lee 74% 0.26 [0.07-0.99] death 2/11,072 32/41,293 Maltagliati 52% 0.48 [0.25-0.87] hosp. n/a n/a Baynouna AlKetbi 98% 0.01 [0.00-0.44] death n/a n/a Ahmadi 30% 0.70 [0.54-0.89] death 160/232,613 112/95,221 Nguyen 20% 0.80 [0.72-0.88] symp. case 904/2,836 483/1,111 Lin 47% 0.53 [0.12-2.33] cases n/a n/a de Souza 73% 0.27 [0.07-1.06] ventilation 3/611 6/327 Mohsin 19% 0.81 [0.66-0.99] severe case 86/258 224/544 Ekblom-Bak 48% 0.52 [0.30-0.92] severe case n/a n/a Lengelé 74% 0.26 [0.07-0.80] cases 23/229 4/12 Saadeh 9% 0.91 [0.56-1.43] symp. case 362 (n) 225 (n) Hamrouni 29% 0.71 [0.55-0.92] death 138/106,006 109/47,827 Huang 47% 0.53 [0.23-1.22] severe case 7/74 16/90 Hamdan 16% 0.84 [0.49-1.39] hosp. 22/128 37/172 Steenkamp 42% 0.58 [0.50-0.68] death 29,469 (n) 13,366 (n) Gilley -42% 1.42 [0.60-3.35] cases 172/1,917 5/79 Almansour 6% 0.94 [0.40-1.47] cases 35/71 38/71 Beydoun 43% 0.57 [0.32-1.01] cases 1,710 (n) 448 (n) Salgado-Aranda 83% 0.17 [0.05-0.56] death 4/223 41/297 Paul 38% 0.62 [0.30-1.20] PASC 1,811 (all patients) LONG COVID Kontopoulou 66% 0.34 [0.24-0.47] recov. time 42 (n) 24 (n) Malisoux 63% 0.37 [0.14-0.98] progression 115 (n) 108 (n) Antunes 80% 0.20 [0.03-1.41] ICU 1/14 9/25 Tsuzuki 56% 0.44 [0.36-0.52] severe case 3,340 (n) 1,528 (n) Reis 41% 0.59 [0.26-1.26] hosp. 9/241 29/305 Pływaczewska-J.. 11% 0.89 [0.72-1.11] m/s case 490 (n) 1,357 (n) Pitanga 33% 0.67 [0.45-1.01] cases 1,469 (n) 1,552 (n) Green 42% 0.58 [0.53-0.63] cases 1,267/11,144 16,198/101,931 Kapusta 71% 0.29 [0.20-0.43] severe case 181 (n) 387 (n) Young 74% 0.26 [0.20-0.33] death 11,279 (n) 29,099 (n) Wang 11% 0.89 [0.75-1.06] PASC 274/691 283/594 LONG COVID Park 26% 0.74 [0.53-1.04] death n/a n/a Sanchez 54% 0.46 [0.38-0.55] symp. case n/a n/a Cardoso 73% 0.27 [0.17-0.45] severe case case control Feter 26% 0.74 [0.58-0.95] PASC 52 (n) 95 (n) LONG COVID Sutkowska 62% 0.38 [0.08-1.72] death 71 (n) 60 (n) Frish 53% 0.47 [0.23-0.95] cases 212 (n) 1,202 (n) CoCo-Fakt Schmidt 31% 0.69 [0.50-0.94] cases 956 (n) 2,705 (n) Šebić 89% 0.11 [0.01-0.96] oxygen 0/53 4/47 Tau​2​ = 0.08, I​2​ = 82.8%, p < 0.0001 Prophylaxis 40% 0.60 [0.54-0.66] 3,646/1,042,018 18,212/741,590 40% lower risk All studies 40% 0.60 [0.54-0.66] 3,662/1,042,219 18,274/741,828 40% lower risk 61 exercise COVID-19 studies c19early.org Sep 2023 Tau​2​ = 0.08, I​2​ = 82.5%, p < 0.0001 Effect extraction pre-specified(most serious outcome, see appendix) Favors exercise Favors inactivity
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2+ Fernandez 47% death Improvement Relative Risk [CI] Tau​2​ = 0.00, I​2​ = 0.0%, p = 0.014 Late treatment 47% 47% lower risk Hamer 28% hospitalization Brawner 74% hospitalization Tret'yakov 98% severe case Gao -105% case Ho 35% hospitalization Halabchi 89% death Zhang 26% death Li 81% severe case per SD increase Tavakol 69% severe case Yates 45% death COVIDENCE UK Holt 17% case Cho 53% death Sallis 59% death Christensen 63% death Lobelo 20% hospitalization Latorre-Román 76% hospitalization Marcus 42% symp. case Yuan 91% death Brandenburg -6% hospitalization Bielik 30% mod. case af Geijerstam 50% death Lee 74% death Maltagliati 52% hospitalization Baynouna AlKetbi 98% death Ahmadi 30% death Nguyen 20% symp. case Lin 47% case de Souza 73% ventilation Mohsin 19% severe case Ekblom-Bak 48% severe case Lengelé 74% case Saadeh 9% symp. case Hamrouni 29% death Huang 47% severe case Hamdan 16% hospitalization Steenkamp 42% death Gilley -42% case Almansour 6% case Beydoun 43% case Salgado-Aranda 83% death Paul 38% PASC LONG COVID Kontopoulou 66% recovery Malisoux 63% progression Antunes 80% ICU admission Tsuzuki 56% severe case Reis 41% hospitalization Pływaczewska-.. 11% mod./sev. case Pitanga 33% case Green 42% case Kapusta 71% severe case Young 74% death Wang 11% PASC LONG COVID Park 26% death Sanchez 54% symp. case Cardoso 73% severe case Feter 26% PASC LONG COVID Sutkowska 62% death Frish 53% case CoCo-Fakt Schmidt 31% case Šebić 89% oxygen therapy Tau​2​ = 0.08, I​2​ = 82.8%, p < 0.0001 Prophylaxis 40% 40% lower risk All studies 40% 40% lower risk 61 exercise C19 studies c19early.org Sep 2023 Tau​2​ = 0.08, I​2​ = 82.5%, p < 0.0001 Effect extraction pre-specifiedRotate device for details Favors exercise Favors inactivity
B
Loading..
C
Loading..
D
Loading..
Figure 1. A. Random effects meta-analysis. This plot shows pooled effects, see the specific outcome analyses for individual outcomes, and the heterogeneity section for discussion. Effect extraction is pre-specified, using the most serious outcome reported. For details of effect extraction see the appendix. B. Scatter plot showing the most serious outcome in all studies. The diamond shows the results of random effects meta-analysis. C. Results within the context of multiple COVID-19 treatments. 0.7% of 5,019 proposed treatments show efficacy c19early.org. D. Timeline of results in exercise studies. The marked dates indicate the time when efficacy was known with a statistically significant improvement of ≥10% from ≥3 studies for pooled outcomes and one or more specific outcome. Efficacy based on specific outcomes was delayed by 0.8 months, compared to using pooled outcomes.
Introduction
Insufficient physical activity is a risk factor for many diseases and is common around the world with prevalence increasing over time, and over two times greater in high-income countries Guthold. For upper respiratory tract infections, research shows lower risk for moderate activity vs. a sedentary lifestyle, however risk may increase with more extreme activity levels Nieman.
We analyze all significant studies reporting COVID-19 outcomes as a function of physical activity levels. Search methods, inclusion criteria, effect extraction criteria (more serious outcomes have priority), all individual study data, PRISMA answers, and statistical methods are detailed in Appendix 1. We present random effects meta-analysis results for all studies, studies within each treatment stage, individual outcomes, peer-reviewed studies, and after exclusions.
Results
Table 1 summarizes the results for all studies, with different exclusions, and for specific outcomes. Figure 2, 3, 4, 5, 6, 7, 8, 9, and 10 show forest plots for random effects meta-analysis of all studies with pooled effects, mortality results, ventilation, ICU admission, hospitalization, progression, recovery, cases, and peer reviewed studies.
Table 1. Random effects meta-analysis for all studies, with different exclusions, and for specific outcomes. Results show the percentage improvement with increased activity levels and the 95% confidence interval. * p<0.05  **** p<0.0001.
Improvement Studies Patients Authors
All studies40% [34‑46%]
****
61 1,801,085 553
After exclusions39% [33‑45%]
****
57 1,799,760 521
Peer-reviewed studiesPeer-reviewed40% [33‑45%]
****
59 1,794,406 547
Mortality50% [39‑60%]
****
18 1,483,571 175
VentilationVent.46% [32‑57%]
****
2 43,773 18
ICU admissionICU41% [35‑47%]
****
4 708,149 32
HospitalizationHosp.35% [27‑42%]
****
18 897,490 128
Recovery58% [41‑70%]
****
3 297 23
Cases23% [13‑32%]
****
22 214,381 245
Loading..
Loading..
Figure 2. Random effects meta-analysis for all studies with pooled effects. This plot shows pooled effects, see the specific outcome analyses for individual outcomes, and the heterogeneity section for discussion. Effect extraction is pre-specified, using the most serious outcome reported. For details of effect extraction see the appendix.
Loading..
Loading..
Figure 3. Random effects meta-analysis for mortality results.
Loading..
Figure 4. Random effects meta-analysis for ventilation.
Loading..
Figure 5. Random effects meta-analysis for ICU admission.
Loading..
Figure 6. Random effects meta-analysis for hospitalization.
Loading..
Figure 7. Random effects meta-analysis for progression.
Loading..
Figure 8. Random effects meta-analysis for recovery.
Loading..
Figure 9. Random effects meta-analysis for cases.
Loading..
Figure 10. Random effects meta-analysis for peer reviewed studies. Zeraatkar analyze 356 COVID-19 trials, finding no significant evidence that preprint results are inconsistent with peer-reviewed studies. They also show extremely long peer-review delays, with a median of 6 months to journal publication. A six month delay was equivalent to around 1.5 million deaths during the first two years of the pandemic. Authors recommend using preprint evidence, with appropriate checks for potential falsified data, which provides higher certainty much earlier. Effect extraction is pre-specified, using the most serious outcome reported, see the appendix for details.
Exclusions
To avoid bias in the selection of studies, we analyze all non-retracted studies. Here we show the results after excluding studies with major issues likely to alter results, non-standard studies, and studies where very minimal detail is currently available. Our bias evaluation is based on analysis of each study and identifying when there is a significant chance that limitations will substantially change the outcome of the study. We believe this can be more valuable than checklist-based approaches such as Cochrane GRADE, which may underemphasize serious issues not captured in the checklists, overemphasize issues unlikely to alter outcomes in specific cases (for example, lack of blinding for an objective mortality outcome, or certain specifics of randomization with a very large effect size), or be easily influenced by potential bias. However, they can also be very high quality.
The studies excluded are as below. Figure 11 shows a forest plot for random effects meta-analysis of all studies after exclusions.
Brawner, unadjusted results with no group details.
de Souza, unadjusted results with no group details. Excluded results: mechanical ventilation.
Huang, unadjusted results with no group details. Excluded results: severe case.
Kontopoulou, unadjusted results with no group details.
Mohsin, unadjusted results with no group details.
Tret'yakov, unadjusted results with no group details.
Yuan, excessive unadjusted differences between groups. Excluded results: death.
Loading..
Figure 11. Random effects meta-analysis for all studies after exclusions. This plot shows pooled effects, see the specific outcome analyses for individual outcomes, and the heterogeneity section for discussion. Effect extraction is pre-specified, using the most serious outcome reported. For details of effect extraction see the appendix.
Conclusion
More physically active people have reduced risk for COVID-19. Statistically significant lower risk is seen for mortality, ventilation, ICU admission, hospitalization, progression, recovery, and cases. 47 studies from 47 independent teams in 23 countries show statistically significant improvements. Meta analysis using the most serious outcome reported shows 40% [34‑46%] lower risk. Results are similar for higher quality and peer-reviewed studies. Results are robust — in exclusion sensitivity analysis 47 of 61 studies must be excluded to avoid finding statistically significant efficacy in pooled analysis. Results are consistent with the overall risk of all cause mortality based on cardiorespiratory fitness — Laukkanen show RR 0.55 [0.50-0.61] for the top vs. bottom tertiles.
Most studies analyze exercise/physical activity levels before infection, comparing regular/moderate exercise and lower/no exercise. Risk may increase with more extreme activity levels.
Other meta analyses show significant improvements with exercise for mortality Ezzatvar, Halabchi, Liu, Rahmati, Sittichai, ICU admission Rahmati, hospitalization Ezzatvar, Halabchi, Li, Rahmati, severity Ezzatvar, Liu, Sittichai, and cases Ezzatvar.
Study Notes
af Geijerstam
0 0.5 1 1.5 2+ Mortality 50% Improvement Relative Risk ICU admission 40% Hospitalization 27% Exercise  af Geijerstam et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study of 1,559,187 patients in Sweden (Mar - Sep 2020) Lower mortality (p=0.0047) and ICU admission (p=0.00032) c19early.org af Geijerstam et al., BMJ Open, July 2021 Favors exercise Favors inactivity
af Geijerstam: Prospective study of 1,559,187 men in Sweden with cardiorespiratory fitness levels measured on military conscription, showing high cardiorespiratory fitness associated with lower risk of COVID-19 hospitalization, ICU admission, and death.
Submit Corrections or Updates.
Ahmadi
0 0.5 1 1.5 2+ Mortality 30% Improvement Relative Risk Exercise for COVID-19  Ahmadi et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 327,834 patients in the United Kingdom Lower mortality with higher activity levels (p=0.0052) c19early.org Ahmadi et al., Brain, Behavior, and Im.., Aug 2021 Favors exercise Favors inactivity
Ahmadi: Retrospective 468,569 adults in the UK, showing no significant difference in COVID-19 mortality based on diet quality, however significantly lower mortality was seen with higher diet quality for pneumonia and infectious diseases.
Submit Corrections or Updates.
Almansour
0 0.5 1 1.5 2+ Case 6% Improvement Relative Risk Exercise for COVID-19  Almansour et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 142 patients in Saudi Arabia (April - June 2020) No significant difference in cases c19early.org Almansour et al., J. Multidisciplinary.., Feb 2022 Favors exercise Favors inactivity
Almansour: Retrospective 142 patients in Saudi Arabia, showing no significant difference in cases with physical activity.
Submit Corrections or Updates.
Antunes
0 0.5 1 1.5 2+ ICU admission 80% Improvement Relative Risk CT >50% 40% CT 25-50% 73% Hospitalization time 43% Disease duration 26% Exercise for COVID-19  Antunes et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 39 patients in Brazil (September - December 2020) Shorter hospitalization with higher activity levels (p=0.03) c19early.org Antunes et al., Sport Sciences for Hea.., Jun 2022 Favors exercise Favors inactivity
Antunes: Retrospective 39 hospitalized COVID-19 survivors >60 years old, showing shorter hospitalization for patients with active lifestyles before COVID-19 symptoms.
Submit Corrections or Updates.
Baynouna AlKetbi
0 0.5 1 1.5 2+ Mortality 98% Improvement Relative Risk Exercise  Baynouna AlKetbi et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in United Arab Emirates Lower mortality with higher activity levels (p=0.049) c19early.org Baynouna AlKetbi et al., J. Epidemiolo.., Aug 2021 Favors exercise Favors inactivity
Baynouna AlKetbi: Retrospective 234 COVID-19 cases in the United Arab Emirates, showing lower risk of mortality with increased physical activity.
Submit Corrections or Updates.
Beydoun
0 0.5 1 1.5 2+ Case, >1/week vs. none 43% Improvement Relative Risk Case, 1-4/mon vs. none 62% Exercise for COVID-19  Beydoun et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 2,158 patients in the USA Fewer cases with higher activity levels (not stat. sig., p=0.055) c19early.org Beydoun et al., American J. Infection .., Mar 2022 Favors exercise Favors inactivity
Beydoun: Retrospective 2,830 people in the USA, showing lower risk of COVID-19 with a history of moderate/vigorous exercise.
Submit Corrections or Updates.
Bielik
0 0.5 1 1.5 2+ Moderate case 30% Improvement Relative Risk Case -9% Exercise for COVID-19  Bielik et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 1,140 patients in Slovakia (December - December 2020) Fewer moderate/severe cases with higher activity levels (not stat. sig., p=0.1) c19early.org Bielik et al., Int. J. Environmental R.., Jul 2021 Favors exercise Favors inactivity
Bielik: Retrospective 1,544 participants in Slovakia, showing a lower risk of more severe COVID-19 for physically active participants, without statistical significance.
Submit Corrections or Updates.
Brandenburg
0 0.5 1 1.5 2+ Hospitalization, PA -6% Improvement Relative Risk Hospitalization, CRF 78% Hospitalization, CRF (b) 64% Severe case, PA 35% Severe case, CRF 24% Exercise for COVID-19  Brandenburg et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 211 patients in multiple countries Lower severe cases with higher activity levels (not stat. sig., p=0.3) c19early.org Brandenburg et al., J. Physical Activi.., Jul 2021 Favors exercise Favors inactivity
Brandenburg: Retrospective 263 COVID+ patients, showing lower hospitalization with higher self-reported cardiorespiratory fitness, but no significant differences for physical activity. Participants in the study were healthier and more fit than the general population.
Submit Corrections or Updates.
Brawner
0 0.5 1 1.5 2+ Hospitalization 74% unadjusted Improvement Relative Risk Exercise for COVID-19  Brawner et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in the USA (February - May 2020) Lower hospitalization with higher activity levels (p=0.0012) c19early.org Brawner et al., Mayo Clinic Proceedings, Oct 2020 Favors exercise Favors inactivity
Brawner: Retrospective 246 COVID-19 patients in the USA, showing the risk of hospitalization inversely associated with maximal exercise capacity. Adjusted results are only provided for MET as a continuous variable.
Submit Corrections or Updates.
Cardoso
0 0.5 1 1.5 2+ Severe case 73% Improvement Relative Risk Exercise for COVID-19  Cardoso et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 614 patients in Brazil (April 2020 - February 2022) Lower severe cases with higher activity levels (p<0.000001) c19early.org Cardoso et al., Medicina Clínica, May 2023 Favors exercise Favors inactivity
Cardoso: Case control study with 307 severe COVID-19 ICU patients and 307 matched COVID-19 outpatients in Brazil, showing significantly higher risk of severe cases with low physical activity.
Submit Corrections or Updates.
Cho
0 0.5 1 1.5 2+ Mortality 53% Improvement Relative Risk Case 10% Exercise for COVID-19  Cho et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 97,123 patients in South Korea Lower mortality (p=0.014) and fewer cases (p<0.0001) c19early.org Cho et al., J. Clinical Medicine, April 2021 Favors exercise Favors inactivity
Cho: Retrospective 6,288 COVID+ patients and 125,772 matched controls in South Korea, showing significantly lower risk of COVID-19 infection and mortality with higher physical activity.
Submit Corrections or Updates.
Christensen
0 0.5 1 1.5 2+ Mortality 63% Improvement Relative Risk Case 23% Exercise for COVID-19  Christensen et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study of 1,072 patients in the United Kingdom (Mar - Jul 2020) Lower mortality with higher activity levels (p=0.019) c19early.org Christensen et al., PLOS ONE, May 2021 Favors exercise Favors inactivity
Christensen: Prospective study of 2,690 adults in the UK Biobank showing lower cardiorespiritory fitness associated with COVID-19 mortality.
Submit Corrections or Updates.
de Souza
0 0.5 1 1.5 2+ Ventilation 73% unadjusted Improvement Relative Risk Hospitalization 34% Exercise for COVID-19  de Souza et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 938 patients in Brazil (June - August 2020) Lower hospitalization with higher activity levels (p=0.046) c19early.org de Souza et al., J. Science and Medici.., Sep 2021 Favors exercise Favors inactivity
de Souza: Retrospective survey of 938 COVID-19 recovered patients in Brazil, showing lower hospitalization with physical activity. NCT04396353.
Submit Corrections or Updates.
Ekblom-Bak
0 0.5 1 1.5 2+ Severe case 48% Improvement Relative Risk Exercise for COVID-19  Ekblom-Bak et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in Sweden Lower severe cases with higher activity levels (p=0.023) c19early.org Ekblom-Bak et al., Int. J. Behavioral .., Oct 2021 Favors exercise Favors inactivity
Ekblom-Bak: Retrospective 857 severe COVID-19 cases and matched controls in Sweden, showing lower risk of severe COVID-19 with higher cardiorespiratory fitness.
Submit Corrections or Updates.
Fernandez
0 0.5 1 1.5 2+ Mortality 47% Improvement Relative Risk Exercise  Fernandez et al.  LATE TREATMENT Is late treatment with exercise beneficial for COVID-19? Retrospective 439 patients in Chile Lower mortality with exercise (p=0.018) c19early.org Fernandez et al., J. Applied Physiology, Feb 2023 Favors exercise Favors inactivity
Fernandez: Retrospective 439 severe COVID-19 hospitalized patients with hypertension, 201 receiving a supervised exercise program, showing significantly lower mortality with exercise. Exercise included of aerobic, breathing, and musculoskeletal exercises, 3 to 4 times per week. There were significantly more control patients on beta-adrenergic blockers and thiazide diuretics.

There are many possible mechanisms of action, including improved circulation, stress reduction, hormone regulation, improved sleep, increased antioxidant levels, and increased nitric oxide levels in the respiratory system. Over-exercising may be detrimental and lead to impaired immune function.
Submit Corrections or Updates.
Feter
0 0.5 1 1.5 2+ PASC, before and during 26% Improvement Relative Risk PASC, during 17% Exercise  Feter et al.  Prophylaxis  LONG COVID Does physical activity reduce the risk of Long COVID (PASC)? Retrospective 237 patients in Brazil Lower PASC with higher activity levels (p=0.016) c19early.org Feter et al., Public Health, June 2023 Favors exercise Favors inactivity
Feter: Analysis of 237 COVID-19 patients in Brazil, showing lower risk of long COVID with physical activity.
Submit Corrections or Updates.
Frish
0 0.5 1 1.5 2+ Case 53% Improvement Relative Risk Exercise for COVID-19  Frish et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 3,038 patients in Israel (February - December 2020) Fewer cases with higher activity levels (p=0.037) c19early.org Frish et al., J. Clinical Medicine, Jun 2023 Favors exercise Favors inactivity
Frish: Retrospective 3,038 bariatric surgery patients in Israel, showing higher risk of SARS-CoV-2 infection with vitamin D deficiency, and lower risk with physical activity.
Submit Corrections or Updates.
Gao
0 0.5 1 1.5 2+ Case -105% Improvement Relative Risk Exercise for COVID-19  Gao et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 315 patients in China (February - March 2020) More cases with higher activity levels (p=0.00031) c19early.org Gao et al., PLOS ONE, November 2020 Favors exercise Favors inactivity
Gao: Case control study in China with 105 cases and 210 matched controls, showing COVID-19 cases associated with physical activity ≥5 times per week. Authors note that people may choose gyms for exercise in winter, leading to higher exposure risk.
Submit Corrections or Updates.
Gilley
0 0.5 1 1.5 2+ Case -42% unadjusted Improvement Relative Risk Exercise for COVID-19  Gilley et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 1,996 patients in the USA (September - December 2020) More cases with higher activity levels (not stat. sig., p=0.55) c19early.org Gilley et al., JMIR Mental Health, Feb 2022 Favors exercise Favors inactivity
Gilley: Retrospective survey of 1,997 college students in the USA, showing no significant difference in COVID-19 cases with exercise in unadjusted results.
Submit Corrections or Updates.
Green
0 0.5 1 1.5 2+ Case 42% Improvement Relative Risk Exercise for COVID-19  Green et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 113,075 patients in Israel (February - December 2020) Fewer cases with higher activity levels (p<0.000001) c19early.org Green et al., European J. General Prac.., Nov 2022 Favors exercise Favors inactivity
Green: Retrospective 113,075 people in Israel, showing lower risk of COVID-19 cases with physical activity and a dose dependent response.
Submit Corrections or Updates.
Halabchi
0 0.5 1 1.5 2+ Mortality 89% Improvement Relative Risk Hospitalization 28% Exercise for COVID-19  Halabchi et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 4,694 patients in Iran Lower hospitalization with higher activity levels (p=0.044) c19early.org Halabchi et al., J. Physical Activity .., Dec 2020 Favors exercise Favors inactivity
Halabchi (B): Retrospective 4,694 COVID-19 patients in Iran, showing lower risk of hospitalization and mortality with regular sports participation.
Submit Corrections or Updates.
Hamdan
0 0.5 1 1.5 2+ Hospitalization 16% Improvement Relative Risk Exercise for COVID-19  Hamdan et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 300 patients in Palestine Lower hospitalization with higher activity levels (not stat. sig., p=0.53) c19early.org Hamdan et al., J. Int. Medical Research, Dec 2021 Favors exercise Favors inactivity
Hamdan: Retrospective 300 participants in Palestine, showing lower risk of hospitalization with physical activity, without statistical significance.
Submit Corrections or Updates.
Hamer
0 0.5 1 1.5 2+ Hospitalization 28% Improvement Relative Risk Hospitalization (b) 34% Exercise for COVID-19  Hamer et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in the United Kingdom Lower hospitalization with higher activity levels (p=0.0004) c19early.org Hamer et al., Brain, Behavior, and Imm.., Jul 2020 Favors exercise Favors inactivity
Hamer: UK Biobank retrospective analysis of 387,109 people, showing lower risk of COVID-19 hospitalization with physical activity.
Submit Corrections or Updates.
Hamrouni
0 0.5 1 1.5 2+ Mortality 29% Improvement Relative Risk Exercise for COVID-19  Hamrouni et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study of 153,833 patients in the United Kingdom Lower mortality with higher activity levels (p=0.0093) c19early.org Hamrouni et al., BMJ Open, November 2021 Favors exercise Favors inactivity
Hamrouni: Prospective UK Biobank analysis, showing a history of low physical activity associated with COVID-19 mortality.
Submit Corrections or Updates.
Ho
0 0.5 1 1.5 2+ Hospitalization 35% Improvement Relative Risk Exercise for COVID-19  Ho et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 138,475 patients in the United Kingdom Lower hospitalization with higher activity levels (p=0.007) c19early.org Ho et al., BMJ Open, November 2020 Favors exercise Favors inactivity
Ho: UK Biobank retrospective 235,928 participants using walking pace as a proxy for physical fitness, showing lower risk of COVID-19 hospitalization with an average vs. slow walking pace.
Submit Corrections or Updates.
Holt
0 0.5 1 1.5 2+ Case 17% Improvement Relative Risk Exercise for COVID-19  COVIDENCE UK  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study of 15,227 patients in the United Kingdom (May 2020 - Feb 2021) Fewer cases with higher activity levels (not stat. sig., p=0.18) c19early.org Holt et al., Thorax, March 2021 Favors exercise Favors inactivity
Holt: Prospective survey-based study with 15,227 people in the UK, showing reduced risk of COVID-19 cases with lower impact physical activity. NCT04330599. COVIDENCE UK.
Submit Corrections or Updates.
Huang
0 0.5 1 1.5 2+ Severe case 47% unadjusted Improvement Relative Risk Severe case (b) 8% unadjusted Case 66% Exercise for COVID-19  Huang et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 164 patients in China (February - March 2020) Fewer cases with higher activity levels (p=0.0035) c19early.org Huang et al., Nature and Science of Sl.., Nov 2021 Favors exercise Favors inactivity
Huang: Retrospective 164 COVID-19 patients and 188 controls in China, showing lower risk of cases with regular exercise.
Submit Corrections or Updates.
Kapusta
0 0.5 1 1.5 2+ Severe case 71% Improvement Relative Risk Exercise for COVID-19  Kapusta et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 568 patients in Poland (March - August 2020) Lower severe cases with higher activity levels (p=0.001) c19early.org Kapusta et al., J. Infection and Publi.., Dec 2022 Favors exercise Favors inactivity
Kapusta: Retrospective 568 convalescent COVID-19 patients in Poland, showing lower risk of severe cases with regular physical activity in the 3 months before COVID-19.
Submit Corrections or Updates.
Kontopoulou
0 0.5 1 1.5 2+ Recovery time 66% Improvement Relative Risk Dyspnea after hospitalizat.. 67% Exercise for COVID-19  Kontopoulou et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 66 patients in Greece (November - December 2020) Faster recovery with higher activity levels (p<0.000001) c19early.org Kontopoulou et al., J. Personalized Me.., Apr 2022 Favors exercise Favors inactivity
Kontopoulou: Retrospective 66 hospitalized COVID-19 patients in Greece, showing significantly improved recovery with a history of exercise in unadjusted results. Exercise after hospitalization was also associated with lower levels of dyspnea one month post hospitalization.
Submit Corrections or Updates.
Latorre-Román
0 0.5 1 1.5 2+ Hospitalization 76% Improvement Relative Risk Hospitalization (b) 87% Exercise  Latorre-Román et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in Spain Lower hospitalization with higher activity levels (not stat. sig., p=0.05) c19early.org Latorre-Román et al., Research in Spor.., Jun 2021 Favors exercise Favors inactivity
Latorre-Román: Retrospective 420 people in Spain, showing lower risk of COVID-19 hospitalization with a history of physical activity.
Submit Corrections or Updates.
Lee
0 0.5 1 1.5 2+ Mortality 74% Improvement Relative Risk Severe case 58% Case 16% Exercise for COVID-19  Lee et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 52,365 patients in South Korea (Jan - Jul 2020) Lower mortality (p=0.046) and severe cases (p=0.03) c19early.org Lee et al., British J. Sports Medicine, Jul 2021 Favors exercise Favors inactivity
Lee: Retrospective 212,768 adults in South Korea, showing lower risk of COVID-19 cases, severity, and mortality with physical activity. Notably, results for aerobic and muscle strengthening activities combined were much better than results for either one in isolation.
Submit Corrections or Updates.
Lengelé
0 0.5 1 1.5 2+ Case 74% Improvement Relative Risk Exercise for COVID-19  Lengelé et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study of 241 patients in Belgium (Mar 2020 - Apr 2021) Fewer cases with higher activity levels (p=0.028) c19early.org Lengelé et al., Aging Clinical and Exp.., Oct 2021 Favors exercise Favors inactivity
Lengelé: Analysis of 241 adults >65yo in Belgium, showing lower risk of COVID-19 with a history of physical activity.
Submit Corrections or Updates.
Li
0 0.5 1 1.5 2+ Severe case 81% Improvement Relative Risk Hospitalization 56% Exercise for COVID-19  Li et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in the United Kingdom Lower severe cases with higher activity levels (p=0.02) c19early.org Li et al., BMC Medical Genomics, February 2021 Favors exercise Favors inactivity
Li (B): Mendelian randomization study showing lower risk of severe COVID-19 with physical activity.
Submit Corrections or Updates.
Lin
0 0.5 1 1.5 2+ Case 47% Improvement Relative Risk Exercise for COVID-19  Lin et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study in multiple countries (March - October 2020) Fewer cases with higher activity levels (not stat. sig., p=0.4) c19early.org Lin et al., BMJ Open, September 2021 Favors exercise Favors inactivity
Lin: Prospective survey analysis of 28,575 people in 99 countries, showing a lower risk of COVID-19 with a exercise, without statistical significance.
Submit Corrections or Updates.
Lobelo
0 0.5 1 1.5 2+ Hospitalization 20% Improvement Relative Risk Exercise for COVID-19  Lobelo et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 5,712 patients in Georgia (March - October 2020) Lower hospitalization with higher activity levels (p=0.022) c19early.org Lobelo et al., BMJ Open, May 2021 Favors exercise Favors inactivity
Lobelo: Retrospective 5,712 COVID-19 patients in the USA, showing higher risk of COVID-19 hospitalization with a history of physical inactivity.
Submit Corrections or Updates.
Malisoux
0 0.5 1 1.5 2+ Progression 63% Improvement Relative Risk Progression (b) 52% Progression (c) 43% Exercise for COVID-19  Malisoux et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 224 patients in Luxembourg (May 2020 - June 2021) Lower progression with higher activity levels (p=0.045) c19early.org Malisoux et al., BMJ Open, April 2022 Favors exercise Favors inactivity
Malisoux: Retrospective 452 participants in Luxembourg, showing lower risk of moderate cases with higher physical activity.
Submit Corrections or Updates.
Maltagliati
0 0.5 1 1.5 2+ Hospitalization 52% Improvement Relative Risk Exercise for COVID-19  Maltagliati et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in multiple countries Lower hospitalization with higher activity levels (p=0.02) c19early.org Maltagliati et al., J. Sports Sciences, Aug 2021 Favors exercise Favors inactivity
Maltagliati: Retrospective 3,139 adults >50 in Europe, with 66 COVID-19 hospitalizations, showing lower risk of hospitalization with higher physical activity and with higher muscle strength. Note that model 2 includes muscle strength which is correlated with physical activity eurapa.biomedcentral.com.
Submit Corrections or Updates.
Marcus
0 0.5 1 1.5 2+ Symp. case 42% Improvement Relative Risk Exercise for COVID-19  Marcus et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study of 14,335 patients in multiple countries (Mar - May 2020) Fewer symptomatic cases with higher activity levels (p<0.000001) c19early.org Marcus et al., PLOS ONE, June 2021 Favors exercise Favors inactivity
Marcus: Prospective survey based study with 14,335 participants, showing lower risk of viral symptoms with regular exercise.
Submit Corrections or Updates.
Mohsin
0 0.5 1 1.5 2+ Severe case 19% Improvement Relative Risk Severe case (b) -1% Exercise for COVID-19  Mohsin et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 1,500 patients in Bangladesh (November 2020 - April 2021) Lower severe cases with higher activity levels (p=0.036) c19early.org Mohsin et al., Infection and Drug Resi.., Sep 2021 Favors exercise Favors inactivity
Mohsin: Retrospective 1,500 COVID+ patients in Bangladesh, showing lower risk of severe cases with regular exercise in unadjusted results.
Submit Corrections or Updates.
Nguyen
0 0.5 1 1.5 2+ Symp. case 20% Improvement Relative Risk Symp. case (b) 42% Exercise for COVID-19  Nguyen et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 3,947 patients in Vietnam (February - March 2020) Fewer symptomatic cases with higher activity levels (p=0.000011) c19early.org Nguyen et al., Nutrients, September 2021 Favors exercise Favors inactivity
Nguyen: Analysis of 3,947 participants in Vietnam, showing significantly lower risk of COVID-19-like symptoms with physical activity and with a healthy diet. The combination of being physically active and eating healthy reduced risk further compared to either alone. The analyzed period was Feb 14 to Mar 2, 2020, which may have been before testing was widely available.
Submit Corrections or Updates.
Park
0 0.5 1 1.5 2+ Mortality 26% Improvement Relative Risk Mortality (b) 38% Case 7% Case (b) 10% Exercise for COVID-19  Park et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in South Korea (January - August 2020) Fewer cases with higher activity levels (p=0.016) c19early.org Park et al., Frontiers in Public Health, Feb 2023 Favors exercise Favors inactivity
Park: Retrospective 4,363 COVID-19 patients and 67,125 controls in South Korea, showing higher risk of mortality and cases with insufficient physical activity.
Submit Corrections or Updates.
Paul
0 0.5 1 1.5 2+ Long COVID 38% Improvement Relative Risk Long COVID (b) 4% Exercise  Paul et al.  Prophylaxis  LONG COVID Does physical activity reduce the risk of Long COVID (PASC)? Retrospective 1,811 patients in the United Kingdom Lower PASC with higher activity levels (not stat. sig., p=0.16) c19early.org Paul et al., medRxiv, April 2022 Favors exercise Favors inactivity
Paul: Retrospective 1,811 COVID-19 patients in the UK, showing lower risk of self-reported long COVID with 3+ hours of exercise per week in the month before infection, without statistical significance (p=0.16).
Submit Corrections or Updates.
Pitanga
0 0.5 1 1.5 2+ Case 33% Improvement Relative Risk Case (b) 43% Case (c) 20% Exercise for COVID-19  Pitanga et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 4,476 patients in Brazil Fewer cases with higher activity levels (not stat. sig., p=0.052) c19early.org Pitanga et al., Int. J. Environmental .., Oct 2022 Favors exercise Favors inactivity
Pitanga: Retrospective 4,476 participants in Brazil, showing lower risk of COVID-19 cases with a history of physical activity, statistically significant only for those following specific practices to protect against COVID-19.
Submit Corrections or Updates.
Pływaczewska-Jakubowska
0 0.5 1 1.5 2+ Moderate/severe case 11% Improvement Relative Risk PASC 14% Exercise  Pływaczewska-Jakubowska et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 1,847 patients in Poland (May 2020 - January 2022) Fewer moderate/severe cases (p=0.3) and lower PASC (p=0.24), not sig. c19early.org Pływaczewska-Jakubowska et al., Fronti.., Oct 2022 Favors exercise Favors inactivity
Pływaczewska-Jakubowska: Retrospective 1,847 COVID+ patients in Poland, showing no significant difference in moderate/severe cases with physical activity. Hospitalized patients were excluded.
Submit Corrections or Updates.
Reis
0 0.5 1 1.5 2+ Hospitalization 41% Improvement Relative Risk Exercise for COVID-19  Reis et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 546 patients in the USA (December 2020 - February 2021) Lower hospitalization with higher activity levels (not stat. sig., p=0.18) c19early.org Reis et al., American J. Lifestyle Med.., Oct 2022 Favors exercise Favors inactivity
Reis: Retrospective 546 COVID+ patients in the USA, showing lower risk of hospitalization with higher frequency of strength training, without statistical significance.
Submit Corrections or Updates.
Saadeh
0 0.5 1 1.5 2+ Symp. case 9% Improvement Relative Risk Symp. case (b) 4% Exercise for COVID-19  Saadeh et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 587 patients in Sweden (March - June 2020) No significant difference in symptomatic cases c19early.org Saadeh et al., Aging Clinical and Expe.., Oct 2021 Favors exercise Favors inactivity
Saadeh: Retrospective 904 patients in Sweden, showing higher risk of COVID-19-like symptoms with poor muscle strength. Risk was slightly higher for physical inactivity, without statistical significance.
Submit Corrections or Updates.
Salgado-Aranda
0 0.5 1 1.5 2+ Mortality 83% Improvement Relative Risk Exercise  Salgado-Aranda et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 520 patients in Spain (February - April 2020) Lower mortality with higher activity levels (p=0.003) c19early.org Salgado-Aranda et al., Infectious Dise.., Mar 2022 Favors exercise Favors inactivity
Salgado-Aranda: Retrospective 520 COVID-19 patients in Spain, showing significantly lower mortality with a history of physical activity.
Submit Corrections or Updates.
Sallis
0 0.5 1 1.5 2+ Mortality 59% Improvement Relative Risk ICU admission 42% Hospitalization 53% Exercise for COVID-19  Sallis et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 10,102 patients in the USA Lower mortality (p=0.0047) and ICU admission (p=0.0056) c19early.org Sallis et al., British J. Sports Medic.., Apr 2021 Favors exercise Favors inactivity
Sallis: Retrospective 48,440 COVID-19 patients in the USA, showing significantly lower mortality, ICU admission, and hospitalization with exercise.
Submit Corrections or Updates.
Sanchez
0 0.5 1 1.5 2+ Symp. case 54% Improvement Relative Risk Exercise for COVID-19  Sanchez et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in Spain Fewer symptomatic cases with higher activity levels (p<0.000001) c19early.org Sanchez et al., Fisioterapia, April 2023 Favors exercise Favors inactivity
Sanchez: Retrospective 29,875 university staff and students in Spain, 3,662 with data, showing lower risk of COVID-19 symptoms for people that exercise. Exercise more than 5 days/week was the most protective, and intense exercise was more effective than moderate exercise.
Submit Corrections or Updates.
Schmidt
0 0.5 1 1.5 2+ Case, above vs. below gui.. 31% Improvement Relative Risk Case, above vs. meeting.. 34% Case, any vs. none 23% Case, moderate/vigorou.. 22% Exercise for COVID-19  CoCo-Fakt  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 5,338 patients in Germany Fewer cases with higher activity levels (p=0.02) c19early.org Schmidt et al., Sports Medicine - Open, Jun 2023 Favors exercise Favors inactivity
Schmidt: Retrospective 5,338 individuals with confirmed contact with a COVID-19 patient, showing lower risk of COVID-19 with exercise.
Submit Corrections or Updates.
Steenkamp
0 0.5 1 1.5 2+ Mortality 42% Improvement Relative Risk Ventilation 45% ICU admission 41% Hospitalization 34% Exercise for COVID-19  Steenkamp et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 42,835 patients in South Africa (Mar 2020 - Jun 2021) Lower mortality (p<0.0001) and ventilation (p<0.0001) c19early.org Steenkamp et al., British J. Sports Me.., Feb 2022 Favors exercise Favors inactivity
Steenkamp: Retrospective 65,361 COVID-19 patients in South Africa, showing significantly lower hospitalization, ICU admission, ventilation, and mortality with exercise.
Submit Corrections or Updates.
Sutkowska
0 0.5 1 1.5 2+ Mortality 62% Improvement Relative Risk Recovery 61% Exercise for COVID-19  Sutkowska et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Prospective study of 131 patients in Poland (Jan - Feb 2022) Lower mortality (p=0.21) and improved recovery (p=0.19), not sig. c19early.org Sutkowska et al., J. Clinical Medicine, Jun 2023 Favors exercise Favors inactivity
Sutkowska: Prospective study of 131 hospitalized patients in Poland, showing lower mortality and improved recovery with a history of higher physical activity.
Submit Corrections or Updates.
Tavakol
0 0.5 1 1.5 2+ Severe case 69% Improvement Relative Risk Exercise for COVID-19  Tavakol et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 188 patients in Iran (March - April 2020) Lower severe cases with higher activity levels (not stat. sig., p=0.05) c19early.org Tavakol et al., J. Public Health, February 2021 Favors exercise Favors inactivity
Tavakol: Retrospective 206 patients in Iran, showing COVID-19 disease severity associated with lower physical activity.
Submit Corrections or Updates.
Tret'yakov
0 0.5 1 1.5 2+ Severe case 98% Improvement Relative Risk Exercise for COVID-19  Tret'yakov et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 293 patients in Russia Lower severe cases with higher activity levels (p=0.0067) c19early.org Tret'yakov et al., Pulmonologiya, October 2020 Favors exercise Favors inactivity
Tret'yakov: Retrospective 293 COVID+ patients in Russia, showing lower risk of severe COVID-19 for individuals who regularly practice aerobic training in unadjusted results.
Submit Corrections or Updates.
Tsuzuki
0 0.5 1 1.5 2+ Severe case 56% Improvement Relative Risk Exercise for COVID-19  Tsuzuki et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 4,868 patients in Japan (January - May 2022) Lower severe cases with higher activity levels (p<0.000001) c19early.org Tsuzuki et al., medRxiv, July 2022 Favors exercise Favors inactivity
Tsuzuki: Retrospective 4,868 elderly COVID-19 patients in Japan, showing higher risk of severe cases with poor physical activity status.
Submit Corrections or Updates.
Wang
0 0.5 1 1.5 2+ PASC, exercise 11% Improvement Relative Risk PASC, healthy lifestyle 49% Hospitalization, healthy lif.. 78% Exercise  Wang et al.  Prophylaxis  LONG COVID Does physical activity reduce the risk of Long COVID (PASC)? Prospective study of 1,285 patients in the USA (Apr 2020 - Nov 2021) Lower PASC with higher activity levels (not stat. sig., p=0.2) c19early.org Wang et al., JAMA Internal Medicine, Feb 2023 Favors exercise Favors inactivity
Wang: Prospective analysis of 32,249 women, showing lower risk of PASC with a healthy lifestyle, in a dose-dependent manner. Participants with 5 or 6 healthy lifestyle factors had significantly lower COVID-19 hospitalization and PASC. BMI and sleep were independently associated with risk of PASC.
Submit Corrections or Updates.
Yates
0 0.5 1 1.5 2+ Mortality 45% Improvement Relative Risk Severe case 47% Exercise for COVID-19  Yates et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 194,031 patients in the United Kingdom Lower mortality (p=0.0015) and severe cases (p<0.0001) c19early.org Yates et al., Int. J. Obesity, February 2021 Favors exercise Favors inactivity
Yates: UK Biobank retrospective 412,596 people, showing severe COVID-19 and COVID-19 mortality inversely associated with self-reported walking pace.
Submit Corrections or Updates.
Young
0 0.5 1 1.5 2+ Mortality 74% Improvement Relative Risk Mortality (b) 65% Mortality (c) 48% Mortality (d) 35% Hospitalization 48% Hospitalization (b) 42% Hospitalization (c) 30% Hospitalization (d) 20% Exercise for COVID-19  Young et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 94,731 patients in the USA (January 2020 - May 2021) Lower mortality (p<0.0001) and hospitalization (p<0.0001) c19early.org Young et al., American J. Preventive M.., Dec 2022 Favors exercise Favors inactivity
Young: Retrospective 194,191 COVID-19 patients in the USA, showing lower risk of hospitalization and mortality with physical activity, with a dose response relationship.
Submit Corrections or Updates.
Yuan
0 0.5 1 1.5 2+ Mortality 91% Improvement Relative Risk Severe case 70% Exercise for COVID-19  Yuan et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 164 patients in China (February - March 2020) Lower severe cases with higher activity levels (p=0.033) c19early.org Yuan et al., Therapeutic Advances in R.., Jun 2021 Favors exercise Favors inactivity
Yuan: Retrospective 164 COVID-19 patients in China, showing physical inactivity associated with an increased risk of severe COVID-19.
Submit Corrections or Updates.
Zhang
0 0.5 1 1.5 2+ Mortality 26% Improvement Relative Risk Case 18% Exercise for COVID-19  Zhang et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective study in the United Kingdom Fewer cases with higher activity levels (p=0.012) c19early.org Zhang et al., J. Glob. Health, December 2020 Favors exercise Favors inactivity
Zhang (B): UK Biobank retrospective showing significantly lower COVID-19 cases with objectively measured physical activity.
Submit Corrections or Updates.
Šebić
0 0.5 1 1.5 2+ Oxygen therapy 89% Improvement Relative Risk Hospitalization 91% Progression 84% Recovery 47% Exercise for COVID-19  Šebić et al.  Prophylaxis Does physical activity reduce risk for COVID-19? Retrospective 100 patients in Bosnia and Herzegovina Lower need for oxygen therapy (p=0.045) and lower hospitalization (p=0.02) c19early.org Šebić et al., Sports Science and Health, Jul 2023 Favors exercise Favors inactivity
Šebić: Retrospective 100 COVID-19 patients in Bosnia and Herzegovina, showing lower symptom severity and faster recovery with a history of regular physical activity.
Submit Corrections or Updates.
Appendix 1. Methods and Data
We performed ongoing searches of PubMed, medRxiv, ClinicalTrials.gov, The Cochrane Library, Google Scholar, Collabovid, Research Square, ScienceDirect, Oxford University Press, the reference lists of other studies and meta-analyses, and submissions to the site c19early.org. Search terms were (exercise OR "physical activity") AND COVID-19. Automated searches are performed every few hours with notification of new matches. All studies regarding the use of exercise for COVID-19 that report a comparison with a control group are included in the main analysis. Sensitivity analysis is performed, excluding studies with major issues, epidemiological studies, and studies with minimal available information. This is a living analysis and is updated regularly.
We extracted effect sizes and associated data from all studies. If studies report multiple kinds of effects then the most serious outcome is used in pooled analysis, while other outcomes are included in the outcome specific analyses. For example, if effects for mortality and cases are both reported, the effect for mortality is used, this may be different to the effect that a study focused on. If symptomatic results are reported at multiple times, we used the latest time, for example if mortality results are provided at 14 days and 28 days, the results at 28 days are used. Mortality alone is preferred over combined outcomes. Outcomes with zero events in both arms were not used (the next most serious outcome is used — no studies were excluded). For example, in low-risk populations with no mortality, a reduction in mortality with treatment is not possible, however a reduction in hospitalization, for example, is still valuable. Clinical outcome is considered more important than PCR testing status. When basically all patients recover in both treatment and control groups, preference for viral clearance and recovery is given to results mid-recovery where available (after most or all patients have recovered there is no room for an effective treatment to do better). If only individual symptom data is available, the most serious symptom has priority, for example difficulty breathing or low SpO2 is more important than cough. When results provide an odds ratio, we computed the relative risk when possible, or converted to a relative risk according to Zhang. Reported confidence intervals and p-values were used when available, using adjusted values when provided. If multiple types of adjustments are reported including propensity score matching (PSM), the PSM results are used. Adjusted primary outcome results have preference over unadjusted results for a more serious outcome when the adjustments significantly alter results. When needed, conversion between reported p-values and confidence intervals followed Altman, Altman (B), and Fisher's exact test was used to calculate p-values for event data. If continuity correction for zero values is required, we use the reciprocal of the opposite arm with the sum of the correction factors equal to 1 Sweeting. Results are expressed with RR < 1.0 favoring treatment, and using the risk of a negative outcome when applicable (for example, the risk of death rather than the risk of survival). If studies only report relative continuous values such as relative times, the ratio of the time for the treatment group versus the time for the control group is used. Calculations are done in Python (3.11.4) with scipy (1.11.1), pythonmeta (1.26), numpy (1.25.0), statsmodels (0.14.0), and plotly (5.15.0).
Forest plots are computed using PythonMeta Deng with the DerSimonian and Laird random effects model (the fixed effect assumption is not plausible in this case) and inverse variance weighting. Mixed-effects meta-regression results are computed with R (4.1.2) using the metafor (3.0-2) and rms (6.2-0) packages, and using the most serious sufficiently powered outcome.
We received no funding, this research is done in our spare time. We have no affiliations with any pharmaceutical companies or political parties.
We have classified studies as early treatment if most patients are not already at a severe stage at the time of treatment (for example based on oxygen status or lung involvement), and treatment started within 5 days of the onset of symptoms. If studies contain a mix of early treatment and late treatment patients, we consider the treatment time of patients contributing most to the events (for example, consider a study where most patients are treated early but late treatment patients are included, and all mortality events were observed with late treatment patients). We note that a shorter time may be preferable. Antivirals are typically only considered effective when used within a shorter timeframe, for example 0-36 or 0-48 hours for oseltamivir, with longer delays not being effective McLean, Treanor.
A summary of study results is below. Please submit updates and corrections at the bottom of this page.
A summary of study results is below. Please submit updates and corrections at https://c19early.org/exmeta.html.
Late treatment
Effect extraction follows pre-specified rules as detailed above and gives priority to more serious outcomes. For pooled analyses, the first (most serious) outcome is used, which may differ from the effect a paper focuses on. Other outcomes are used in outcome specific analyses.
Fernandez, 2/2/2023, retrospective, Chile, peer-reviewed, 10 authors.
Submit Corrections or Updates.
 risk of death, 47.5% lower, RR 0.53, p = 0.02, high activity levels 16 of 201 (8.0%), low activity levels 62 of 238 (26.1%), NNT 5.5, adjusted per study, odds ratio converted to relative risk, multivariable.
Prophylaxis
Effect extraction follows pre-specified rules as detailed above and gives priority to more serious outcomes. For pooled analyses, the first (most serious) outcome is used, which may differ from the effect a paper focuses on. Other outcomes are used in outcome specific analyses.
af Geijerstam, 7/5/2021, prospective, Sweden, peer-reviewed, 9 authors, study period March 2020 - September 2020.
Submit Corrections or Updates.
 risk of death, 50.0% lower, OR 0.50, p = 0.005, high vs. low fitness, model 7, RR approximated with OR.
risk of ICU admission, 40.0% lower, OR 0.60, p < 0.001, high vs. low fitness, model 7, RR approximated with OR.
risk of hospitalization, 27.0% lower, OR 0.73, p < 0.001, high vs. low fitness, model 7, RR approximated with OR.
Ahmadi, 8/31/2021, retrospective, United Kingdom, peer-reviewed, 5 authors.
Submit Corrections or Updates.
 risk of death, 30.0% lower, RR 0.70, p = 0.005, adjusted per study, sufficient vs. inactive, model 2, multivariable.
Almansour, 2/17/2022, retrospective, Saudi Arabia, peer-reviewed, 12 authors, study period April 2020 - June 2020.
Submit Corrections or Updates.
 risk of case, 5.7% lower, RR 0.94, p = 0.85, high activity levels 35 of 71 (49.3%), low activity levels 38 of 71 (53.5%), NNT 24, adjusted per study, odds ratio converted to relative risk, multivariable.
Antunes, 6/11/2022, retrospective, Brazil, peer-reviewed, survey, 5 authors, study period September 2020 - December 2020.
Submit Corrections or Updates.
 risk of ICU admission, 80.2% lower, RR 0.20, p = 0.06, high activity levels 1 of 14 (7.1%), low activity levels 9 of 25 (36.0%), NNT 3.5.
risk of miscellaneous, 40.5% lower, RR 0.60, p = 0.48, high activity levels 3 of 14 (21.4%), low activity levels 9 of 25 (36.0%), NNT 6.9, CT abnormalities >50%.
risk of miscellaneous, 72.5% lower, RR 0.27, p = 0.04, high activity levels 2 of 14 (14.3%), low activity levels 13 of 25 (52.0%), NNT 2.7, CT abnormalities 25-50%.
hospitalization time, 43.4% lower, relative time 0.57, p = 0.03, high activity levels 14, low activity levels 25.
miscellaneous, 25.5% lower, relative time 0.74, p = 0.02, high activity levels 14, low activity levels 25.
Baynouna AlKetbi, 8/23/2021, retrospective, United Arab Emirates, peer-reviewed, 16 authors.
Submit Corrections or Updates.
 risk of death, 98.5% lower, OR 0.01, p = 0.049, adjusted per study, multivariable, RR approximated with OR.
Beydoun, 3/12/2022, retrospective, USA, peer-reviewed, survey, 7 authors.
Submit Corrections or Updates.
 risk of case, 43.0% lower, OR 0.57, p = 0.05, high activity levels 1,710, low activity levels 448, adjusted per study, multivariable, >1/week vs. none, model 2, RR approximated with OR.
risk of case, 62.0% lower, OR 0.38, p = 0.010, high activity levels 672, low activity levels 448, adjusted per study, multivariable, 1-4/mon vs. none, model 2, RR approximated with OR.
Bielik, 7/4/2021, retrospective, Slovakia, peer-reviewed, survey, 3 authors, study period 7 December, 2020 - 12 December, 2020.
Submit Corrections or Updates.
 risk of moderate case, 30.4% lower, RR 0.70, p = 0.10, high activity levels 775, low activity levels 365, adjusted per study, physically active group.
risk of case, 9.1% higher, RR 1.09, p = 0.36, high activity levels 775, low activity levels 365, adjusted per study, physically active group.
Brandenburg, 7/1/2021, retrospective, multiple countries, peer-reviewed, survey, 4 authors.
Submit Corrections or Updates.
 risk of hospitalization, 6.0% higher, OR 1.06, p = 0.60, high activity levels 102, low activity levels 39, adjusted per study, multivariable, PA, >1h vigorous vs. no/low, RR approximated with OR.
risk of hospitalization, 78.0% lower, OR 0.22, p = 0.05, high activity levels 177, low activity levels 34, adjusted per study, multivariable, CRF, 6.2-8.7 vs. >10, RR approximated with OR.
risk of hospitalization, 64.0% lower, OR 0.36, p = 0.04, high activity levels 97, low activity levels 34, adjusted per study, multivariable, CRF, 8.7-10 vs. >10, RR approximated with OR.
risk of severe case, 35.0% lower, OR 0.65, p = 0.30, high activity levels 102, low activity levels 39, adjusted per study, multivariable, PA, >1h vigorous vs. no/low, RR approximated with OR.
risk of severe case, 24.0% lower, OR 0.76, p = 0.60, high activity levels 52, low activity levels 34, adjusted per study, multivariable, CRF, 4.4-6.2 vs. >10, RR approximated with OR.
Brawner, 10/10/2020, retrospective, USA, peer-reviewed, 10 authors, study period 29 February, 2020 - 30 May, 2020, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of hospitalization, 74.2% lower, OR 0.26, p = 0.001, unadjusted, inverted to make OR<1 favor high activity levels, highest fitness quartile vs. lowest fitness quartile, RR approximated with OR.
Cardoso, 5/9/2023, retrospective, Brazil, peer-reviewed, 6 authors, study period April 2020 - February 2022.
Submit Corrections or Updates.
 risk of severe case, 73.0% lower, OR 0.27, p < 0.001, high activity levels 307, low activity levels 307, adjusted per study, inverted to make OR<1 favor high activity levels, case control OR, moderate/high vs. low physical activity, multivariable.
Cho, 4/6/2021, retrospective, South Korea, peer-reviewed, 9 authors.
Submit Corrections or Updates.
 risk of death, 53.0% lower, OR 0.47, p = 0.01, high activity levels 17 of 48 (35.4%) cases, 3,223 of 4,536 (71.1%) controls, case control OR, moderate to vigorous vs. inactive.
risk of case, 10.0% lower, OR 0.90, p < 0.001, high activity levels 3,223 of 4,536 (71.1%) cases, 68,609 of 92,587 (74.1%) controls, NNT 142, case control OR, moderate to vigorous vs. inactive.
Christensen, 5/5/2021, prospective, United Kingdom, peer-reviewed, 5 authors, study period 16 March, 2020 - 26 July, 2020.
Submit Corrections or Updates.
 risk of death, 63.0% lower, RR 0.37, p = 0.02, high activity levels 543, low activity levels 529, adjusted per study, high fitness vs. low fitness, multivariable.
risk of case, 23.0% lower, RR 0.77, p = 0.20, high activity levels 55 of 543 (10.1%), low activity levels 77 of 529 (14.6%), NNT 23, adjusted per study, high fitness vs. low fitness, multivariable.
de Souza, 9/30/2021, retrospective, Brazil, peer-reviewed, 8 authors, study period June 2020 - August 2020, trial NCT04396353 (history).
Submit Corrections or Updates.
 risk of mechanical ventilation, 73.2% lower, RR 0.27, p = 0.07, high activity levels 3 of 611 (0.5%), low activity levels 6 of 327 (1.8%), NNT 74, unadjusted, excluded in exclusion analyses: unadjusted results with no group details.
risk of hospitalization, 34.3% lower, RR 0.66, p = 0.046, high activity levels 49 of 611 (8.0%), low activity levels 42 of 327 (12.8%), NNT 21, adjusted per study, sufficient vs. insufficient, model 3, multivariable.
Ekblom-Bak, 10/19/2021, retrospective, Sweden, peer-reviewed, 13 authors.
Submit Corrections or Updates.
 risk of severe case, 47.6% lower, OR 0.52, p = 0.02, inverted to make OR<1 favor high activity levels, case control OR, model 3, high vs. very low CRF.
Feter, 6/13/2023, retrospective, Brazil, peer-reviewed, survey, mean age 37.1, 17 authors.
Submit Corrections or Updates.
 risk of PASC, 26.0% lower, RR 0.74, p = 0.02, high activity levels 52, low activity levels 95, adjusted per study, before and during pandemic, multivariable.
risk of PASC, 17.0% lower, RR 0.83, p = 0.04, high activity levels 67, low activity levels 170, adjusted per study, during pandemic, multivariable.
Frish, 6/15/2023, retrospective, Israel, peer-reviewed, 7 authors, study period 1 February, 2020 - 31 December, 2020.
Submit Corrections or Updates.
 risk of case, 53.0% lower, OR 0.47, p = 0.04, high activity levels 212, low activity levels 1,202, adjusted per study, >3 times per week vs. none, multivariable, RR approximated with OR.
Gao, 11/5/2020, retrospective, China, peer-reviewed, survey, median age 55.0, 11 authors, study period 10 February, 2020 - 1 March, 2020.
Submit Corrections or Updates.
 risk of case, 105.0% higher, HR 2.05, p < 0.001, high activity levels 59 of 105 (56.2%) cases, 69 of 210 (32.9%) controls, case control OR, Cox proportional hazards.
Gilley, 2/10/2022, retrospective, USA, peer-reviewed, survey, 21 authors, study period September 2020 - December 2020, trial NCT04766788 (history).
Submit Corrections or Updates.
 risk of case, 41.8% higher, RR 1.42, p = 0.55, high activity levels 172 of 1,917 (9.0%), low activity levels 5 of 79 (6.3%), unadjusted.
Green, 11/7/2022, retrospective, Israel, peer-reviewed, 9 authors, study period 1 February, 2020 - 31 December, 2020.
Submit Corrections or Updates.
 risk of case, 41.7% lower, RR 0.58, p < 0.001, high activity levels 1,267 of 11,144 (11.4%), low activity levels 16,198 of 101,931 (15.9%), adjusted per study, odds ratio converted to relative risk, >3 times per week vs. none, multivariable.
Halabchi (B), 12/1/2020, retrospective, Iran, peer-reviewed, 8 authors.
Submit Corrections or Updates.
 risk of death, 88.8% lower, RR 0.11, p = 0.08, high activity levels 0 of 249 (0.0%), low activity levels 79 of 4,445 (1.8%), NNT 56, adjusted per study, odds ratio converted to relative risk, multivariable.
risk of hospitalization, 28.3% lower, RR 0.72, p = 0.04, high activity levels 30 of 249 (12.0%), low activity levels 878 of 4,445 (19.8%), adjusted per study, odds ratio converted to relative risk, multivariable.
Hamdan, 12/23/2021, retrospective, Palestine, peer-reviewed, survey, mean age 30.5, 7 authors.
Submit Corrections or Updates.
 risk of hospitalization, 16.4% lower, RR 0.84, p = 0.53, high activity levels 22 of 128 (17.2%), low activity levels 37 of 172 (21.5%), NNT 23, adjusted per study, odds ratio converted to relative risk, multivariable.
Hamer, 7/31/2020, retrospective, United Kingdom, peer-reviewed, 4 authors.
Submit Corrections or Updates.
 risk of hospitalization, 27.5% lower, RR 0.72, p < 0.001, adjusted per study, inverted to make RR<1 favor high activity levels, model 2, sufficient vs. no activity, multivariable.
risk of hospitalization, 33.8% lower, RR 0.66, p < 0.001, adjusted per study, inverted to make RR<1 favor high activity levels, model 1, sufficient vs. no activity, multivariable.
Hamrouni, 11/3/2021, prospective, United Kingdom, peer-reviewed, 5 authors.
Submit Corrections or Updates.
 risk of death, 29.0% lower, RR 0.71, p = 0.009, high activity levels 138 of 106,006 (0.1%), low activity levels 109 of 47,827 (0.2%), adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, high vs. low physical activity, multivariable.
Ho, 11/19/2020, retrospective, United Kingdom, peer-reviewed, survey, 13 authors.
Submit Corrections or Updates.
 risk of hospitalization, 34.6% lower, RR 0.65, p = 0.007, high activity levels 213 of 123,588 (0.2%), low activity levels 59 of 14,887 (0.4%), adjusted per study, inverted to make RR<1 favor high activity levels, model 2, average vs. slow walking pace, multivariable.
Holt, 3/30/2021, prospective, United Kingdom, peer-reviewed, 34 authors, study period 1 May, 2020 - 5 February, 2021, trial NCT04330599 (history) (COVIDENCE UK).
Submit Corrections or Updates.
 risk of case, 17.0% lower, OR 0.83, p = 0.18, adjusted per study, fully adjusted, ≥2 hours lower impact physical activity vs. 0 hours, RR approximated with OR.
Huang, 11/30/2021, retrospective, China, peer-reviewed, survey, 5 authors, study period 10 February, 2020 - 28 March, 2020.
Submit Corrections or Updates.
 risk of severe case, 46.8% lower, RR 0.53, p = 0.18, high activity levels 7 of 74 (9.5%), low activity levels 16 of 90 (17.8%), NNT 12, unadjusted, exercise habit, ≥1 time per week, excluded in exclusion analyses: unadjusted results with no group details.
risk of severe case, 8.0% lower, RR 0.92, p = 1.00, high activity levels 3 of 23 (13.0%), low activity levels 20 of 141 (14.2%), NNT 88, unadjusted, ≥30 minutes ≥3 times per week, excluded in exclusion analyses: unadjusted results with no group details.
risk of case, 65.9% lower, OR 0.34, p = 0.004, adjusted per study, inverted to make OR<1 favor high activity levels, case control OR, regular exercise, multivariable.
Kapusta, 12/12/2022, retrospective, Poland, peer-reviewed, survey, mean age 70.4, 7 authors, study period 1 March, 2020 - 30 August, 2020, trial NCT05018052 (history).
Submit Corrections or Updates.
 risk of severe case, 70.9% lower, OR 0.29, p = 0.001, high activity levels 181, low activity levels 387, inverted to make OR<1 favor high activity levels, RR approximated with OR.
Kontopoulou, 4/17/2022, retrospective, Greece, peer-reviewed, survey, 4 authors, study period November 2020 - December 2020, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 recovery time, 66.2% lower, relative time 0.34, p < 0.001, high activity levels mean 22.0 (±14.0) n=42, low activity levels mean 65.0 (±32.0) n=24.
relative dyspnea after hospitalization, 66.7% better, RR 0.33, p < 0.001, high activity levels mean 1.0 (±1.0) n=42, low activity levels mean 3.0 (±1.0) n=24, inverted to make RR<1 favor high activity levels.
Latorre-Román, 6/15/2021, retrospective, Spain, peer-reviewed, survey, 7 authors.
Submit Corrections or Updates.
 risk of hospitalization, 76.0% lower, OR 0.24, p = 0.05, moderate physical activity, >150 min per week, RR approximated with OR.
risk of hospitalization, 87.0% lower, OR 0.13, p = 0.07, moderate physical activity, 30-150 min per week, RR approximated with OR.
Lee, 7/22/2021, retrospective, South Korea, peer-reviewed, 25 authors, study period 1 January, 2020 - 31 July, 2020.
Submit Corrections or Updates.
 risk of death, 74.0% lower, RR 0.26, p = 0.046, high activity levels 2 of 11,072 (0.0%), low activity levels 32 of 41,293 (0.1%), NNT 1683, adjusted per study, odds ratio converted to relative risk, model 2,aerobic and muscle strengthening vs. insufficient aerobic and muscle strengthening, multivariable.
risk of severe case, 57.8% lower, RR 0.42, p = 0.03, high activity levels 39 of 11,072 (0.4%), low activity levels 273 of 41,293 (0.7%), adjusted per study, odds ratio converted to relative risk, model 2,aerobic and muscle strengthening vs. insufficient aerobic and muscle strengthening, multivariable.
risk of case, 15.6% lower, RR 0.84, p = 0.03, high activity levels 291 of 11,072 (2.6%), low activity levels 1,293 of 41,293 (3.1%), NNT 199, adjusted per study, odds ratio converted to relative risk, model 2,aerobic and muscle strengthening vs. insufficient aerobic and muscle strengthening, multivariable.
Lengelé, 10/23/2021, prospective, Belgium, peer-reviewed, median age 75.6, 8 authors, study period March 2020 - April 2021.
Submit Corrections or Updates.
 risk of case, 73.6% lower, RR 0.26, p = 0.03, high activity levels 23 of 229 (10.0%), low activity levels 4 of 12 (33.3%), NNT 4.3, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk.
Li (B), 2/3/2021, retrospective, United Kingdom, peer-reviewed, 2 authors, per SD increase.
Submit Corrections or Updates.
 risk of severe case, 81.0% lower, OR 0.19, p = 0.02, RR approximated with OR.
risk of hospitalization, 56.0% lower, OR 0.44, p = 0.07, RR approximated with OR.
Lin, 9/21/2021, prospective, multiple countries, peer-reviewed, survey, 19 authors, study period 26 March, 2020 - 8 October, 2020.
Submit Corrections or Updates.
 risk of case, 47.4% lower, OR 0.53, p = 0.40, inverted to make OR<1 favor high activity levels, exercise ≥1/month vs. exercise <1/month, RR approximated with OR.
Lobelo, 5/19/2021, retrospective, Georgia, peer-reviewed, 7 authors, study period 3 March, 2020 - 29 October, 2020.
Submit Corrections or Updates.
 risk of hospitalization, 20.0% lower, OR 0.80, p = 0.02, high activity levels 2,121, low activity levels 1,648, adjusted per study, inverted to make OR<1 favor high activity levels, active vs. inactive, multivariable, RR approximated with OR.
Malisoux, 4/29/2022, retrospective, Luxembourg, peer-reviewed, survey, median age 42.0, 6 authors, study period May 2020 - June 2021, trial NCT04380987 (history).
Submit Corrections or Updates.
 risk of progression, 63.0% lower, OR 0.37, p = 0.045, high activity levels 115, low activity levels 108, moderate case, >82 vs. <30 MET-hour/week, RR approximated with OR.
risk of progression, 52.0% lower, OR 0.48, p = 0.14, high activity levels 116, low activity levels 108, moderate case, >52-82 vs. <30 MET-hour/week, RR approximated with OR.
risk of progression, 43.0% lower, OR 0.57, p = 0.28, high activity levels 113, low activity levels 108, moderate case, 30-52 vs. <30 MET-hour/week, RR approximated with OR.
Maltagliati, 8/11/2021, retrospective, multiple countries, peer-reviewed, survey, 8 authors.
Submit Corrections or Updates.
 risk of hospitalization, 52.0% lower, OR 0.48, p = 0.02, adjusted per study, model 1, more than once a week vs. hardly ever or never, multivariable, RR approximated with OR.
Marcus, 6/17/2021, prospective, multiple countries, peer-reviewed, survey, 12 authors, study period 26 March, 2020 - 3 May, 2020.
Submit Corrections or Updates.
 risk of symptomatic case, 42.1% lower, RR 0.58, p < 0.001, high activity levels 240 of 10,627 (2.3%), low activity levels 134 of 3,708 (3.6%), NNT 74, adjusted per study, odds ratio converted to relative risk, multivariable.
Mohsin, 9/30/2021, retrospective, Bangladesh, peer-reviewed, survey, 10 authors, study period November 2020 - April 2021, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of severe case, 19.0% lower, RR 0.81, p = 0.04, high activity levels 86 of 258 (33.3%), low activity levels 224 of 544 (41.2%), NNT 13, exercise >30 minutes.
risk of severe case, 0.9% higher, RR 1.01, p = 0.91, high activity levels 290 of 698 (41.5%), low activity levels 224 of 544 (41.2%), exercise <30 minutes.
Nguyen, 9/18/2021, retrospective, Vietnam, peer-reviewed, survey, 17 authors, study period 14 February, 2020 - 2 March, 2020.
Submit Corrections or Updates.
 risk of symptomatic case, 20.3% lower, RR 0.80, p < 0.001, high activity levels 904 of 2,836 (31.9%), low activity levels 483 of 1,111 (43.5%), NNT 8.6, adjusted per study, odds ratio converted to relative risk, active vs. inactive, COVID-19-like symptoms, multivariable.
Park, 2/14/2023, retrospective, South Korea, peer-reviewed, survey, 4 authors, study period 1 January, 2020 - 14 August, 2020.
Submit Corrections or Updates.
 risk of death, 25.6% lower, OR 0.74, p = 0.08, inverted to make OR<1 favor high activity levels, sufficient vs. insufficient PA, model 3, RR approximated with OR.
risk of death, 38.4% lower, OR 0.62, p = 0.02, inverted to make OR<1 favor high activity levels, sufficient vs. insufficient PA, model 2, RR approximated with OR.
risk of case, 7.2% lower, OR 0.93, p = 0.02, inverted to make OR<1 favor high activity levels, sufficient vs. insufficient PA, model 3, RR approximated with OR.
risk of case, 10.4% lower, OR 0.90, p < 0.001, inverted to make OR<1 favor high activity levels, sufficient vs. insufficient PA, model 2, RR approximated with OR.
Paul, 4/13/2022, retrospective, United Kingdom, preprint, survey, 2 authors.
Submit Corrections or Updates.
 risk of long COVID, 38.1% lower, RR 0.62, p = 0.16, adjusted per study, odds ratio converted to relative risk, 3+ hours per week vs. none, multivariable, model 4, control prevalance approximated with overall prevalence.
risk of long COVID, 4.1% lower, RR 0.96, p = 0.89, adjusted per study, odds ratio converted to relative risk, ≤2 hours per week vs. none, multivariable, model 4, control prevalance approximated with overall prevalence.
Pitanga, 10/29/2022, retrospective, Brazil, peer-reviewed, survey, 11 authors.
Submit Corrections or Updates.
 risk of case, 33.0% lower, OR 0.67, p = 0.05, high activity levels 1,469, low activity levels 1,552, combined results with and without protection practices, RR approximated with OR.
Pływaczewska-Jakubowska, 10/24/2022, retrospective, Poland, peer-reviewed, median age 51.0, 5 authors, study period May 2020 - January 2022.
Submit Corrections or Updates.
 risk of moderate/severe case, 11.0% lower, OR 0.89, p = 0.30, high activity levels 490, low activity levels 1,357, adjusted per study, multivariable, model 3, RR approximated with OR.
risk of PASC, 14.0% lower, OR 0.86, p = 0.24, high activity levels 389, low activity levels 1,128, adjusted per study, multivariable, model 3, RR approximated with OR.
Reis, 10/24/2022, retrospective, USA, peer-reviewed, survey, 6 authors, study period December 2020 - February 2021.
Submit Corrections or Updates.
 risk of hospitalization, 40.7% lower, RR 0.59, p = 0.18, high activity levels 9 of 241 (3.7%), low activity levels 29 of 305 (9.5%), adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, strength training 2+/week vs. <2, multivariable.
Saadeh, 10/30/2021, retrospective, Sweden, peer-reviewed, 6 authors, study period March 2020 - June 2020.
Submit Corrections or Updates.
 risk of symptomatic case, 9.1% lower, OR 0.91, p = 0.71, high activity levels 362, low activity levels 225, adjusted per study, inverted to make OR<1 favor high activity levels, 2+ symptoms, Table 8, physically active vs. inactive, multivariable, RR approximated with OR.
risk of symptomatic case, 3.8% lower, OR 0.96, p = 0.85, high activity levels 362, low activity levels 225, adjusted per study, inverted to make OR<1 favor high activity levels, 1+ symptoms, Table 2, model 2, physically active vs. inactive, multivariable, RR approximated with OR.
Salgado-Aranda, 3/14/2022, retrospective, Spain, peer-reviewed, 15 authors, study period 15 February, 2020 - 15 April, 2020.
Submit Corrections or Updates.
 risk of death, 83.1% lower, HR 0.17, p = 0.003, high activity levels 4 of 223 (1.8%), low activity levels 41 of 297 (13.8%), NNT 8.3, inverted to make HR<1 favor high activity levels, active vs. sedentary, Cox proportional hazards.
Sallis, 4/13/2021, retrospective, USA, peer-reviewed, 8 authors.
Submit Corrections or Updates.
 risk of death, 59.2% lower, RR 0.41, p = 0.005, high activity levels 11 of 3,118 (0.4%), low activity levels 170 of 6,984 (2.4%), adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, consistently active vs. consistently inactive, multivariable.
risk of ICU admission, 41.5% lower, RR 0.58, p = 0.006, high activity levels 32 of 3,118 (1.0%), low activity levels 195 of 6,984 (2.8%), adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, consistently active vs. consistently inactive, multivariable.
risk of hospitalization, 53.0% lower, RR 0.47, p < 0.001, high activity levels 99 of 3,118 (3.2%), low activity levels 732 of 6,984 (10.5%), adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, consistently active vs. consistently inactive, multivariable.
Sanchez, 4/25/2023, retrospective, Spain, peer-reviewed, 3 authors, trial NCT04624048 (history).
Submit Corrections or Updates.
 risk of symptomatic case, 54.1% lower, OR 0.46, p < 0.001, inverted to make OR<1 favor high activity levels, exercise vs. no exercise before COVID-19, RR approximated with OR.
Schmidt, 6/21/2023, retrospective, Germany, peer-reviewed, 8 authors, CoCo-Fakt trial.
Submit Corrections or Updates.
 risk of case, 31.1% lower, OR 0.69, p = 0.02, high activity levels 956, low activity levels 2,705, adjusted per study, inverted to make OR<1 favor high activity levels, above guidelines vs. below guidelines, multivariable, RR approximated with OR.
risk of case, 34.5% lower, OR 0.66, p = 0.02, high activity levels 956, low activity levels 1,113, adjusted per study, inverted to make OR<1 favor high activity levels, above guidelines vs. meeting guidelines, multivariable, RR approximated with OR.
risk of case, 22.7% lower, OR 0.77, p = 0.02, high activity levels 3,658, low activity levels 1,680, adjusted per study, inverted to make OR<1 favor high activity levels, exercise vs. no exercise, multivariable, RR approximated with OR.
risk of case, 21.6% lower, OR 0.78, p = 0.03, high activity levels 3,371, low activity levels 1,716, adjusted per study, inverted to make OR<1 favor high activity levels, moderate-to-vigorous vs. low intensity, multivariable, RR approximated with OR.
Steenkamp, 2/9/2022, retrospective, South Africa, peer-reviewed, 10 authors, study period 19 March, 2020 - 30 June, 2021.
Submit Corrections or Updates.
 risk of death, 42.0% lower, RR 0.58, p < 0.001, high activity levels 29,469, low activity levels 13,366, adjusted per study, high activity vs. low activity, poisson regression, multivariable.
risk of mechanical ventilation, 45.0% lower, RR 0.55, p < 0.001, high activity levels 29,469, low activity levels 13,366, adjusted per study, high activity vs. low activity, poisson regression, multivariable.
risk of ICU admission, 41.0% lower, RR 0.59, p < 0.001, high activity levels 29,469, low activity levels 13,366, adjusted per study, high activity vs. low activity, poisson regression, multivariable.
risk of hospitalization, 34.0% lower, RR 0.66, p < 0.001, high activity levels 29,469, low activity levels 13,366, adjusted per study, high activity vs. low activity, poisson regression, multivariable.
Sutkowska, 6/14/2023, prospective, Poland, peer-reviewed, 14 authors, study period 31 January, 2022 - 11 February, 2022, trial NCT05200767 (history).
Submit Corrections or Updates.
 risk of death, 62.0% lower, HR 0.38, p = 0.21, high activity levels 71, low activity levels 60, inverted to make HR<1 favor high activity levels, IPAQ 1/2 vs. IPAQ 0, Cox proportional hazards.
risk of no recovery, 61.0% lower, HR 0.39, p = 0.19, high activity levels 71, low activity levels 60, IPAQ 1/2 vs. IPAQ 0, Cox proportional hazards.
Tavakol, 2/4/2021, retrospective, Iran, peer-reviewed, 9 authors, study period 20 March, 2020 - 24 April, 2020.
Submit Corrections or Updates.
 risk of severe case, 68.5% lower, RR 0.31, p = 0.05, high activity levels 3 of 64 (4.7%), low activity levels 19 of 124 (15.3%), NNT 9.4, adjusted per study, odds ratio converted to relative risk, moderate to high activity versus low activity, multivariable.
Tret'yakov, 10/26/2020, retrospective, Russia, peer-reviewed, 8 authors, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of severe case, 98.3% lower, RR 0.02, p = 0.007, high activity levels 0 of 27 (0.0%), low activity levels 53 of 266 (19.9%), NNT 5.0, relative risk is not 0 because of continuity correction due to zero events (with reciprocal of the contrasting arm).
Tsuzuki, 7/5/2022, retrospective, Japan, preprint, 4 authors, study period 1 January, 2022 - 16 May, 2022.
Submit Corrections or Updates.
 risk of severe case, 56.3% lower, OR 0.44, p < 0.001, high activity levels 3,340, low activity levels 1,528, adjusted per study, inverted to make OR<1 favor high activity levels, good vs. poor physical activity status, multivariable, RR approximated with OR.
Wang, 2/6/2023, prospective, USA, peer-reviewed, survey, mean age 64.7, 8 authors, study period April 2020 - November 2021.
Submit Corrections or Updates.
 risk of PASC, 10.7% lower, RR 0.89, p = 0.20, high activity levels 274 of 691 (39.7%), low activity levels 283 of 594 (47.6%), NNT 13, adjusted per study, inverted to make RR<1 favor high activity levels, ≥210 vs. 0-30, multivariable, model 2.
risk of PASC, 49.0% lower, RR 0.51, p = 0.002, high activity levels 188, low activity levels 66, 5 or 6 healthy lifestyle factors vs. 0.
Yates, 2/26/2021, retrospective, United Kingdom, peer-reviewed, 7 authors.
Submit Corrections or Updates.
 risk of death, 45.3% lower, RR 0.55, p = 0.001, high activity levels 72 of 163,912 (0.0%), low activity levels 62 of 30,119 (0.2%), adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, multivariable.
risk of severe case, 46.7% lower, RR 0.53, p < 0.001, high activity levels 291 of 163,912 (0.2%), low activity levels 180 of 30,119 (0.6%), adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, multivariable.
Young, 12/14/2022, retrospective, USA, peer-reviewed, 7 authors, study period 1 January, 2020 - 31 May, 2021.
Submit Corrections or Updates.
 risk of death, 74.4% lower, OR 0.26, p < 0.001, high activity levels 11,279, low activity levels 29,099, inverted to make OR<1 favor high activity levels, always active vs. always inactive, RR approximated with OR.
risk of death, 65.3% lower, OR 0.35, p < 0.001, high activity levels 11,279, low activity levels 83,452, inverted to make OR<1 favor high activity levels, always active vs. mostly inactive, RR approximated with OR.
risk of death, 47.9% lower, OR 0.52, p < 0.001, high activity levels 11,279, low activity levels 42,490, inverted to make OR<1 favor high activity levels, always active vs. some activity, RR approximated with OR.
risk of death, 35.5% lower, OR 0.65, p = 0.002, high activity levels 11,279, low activity levels 27,871, inverted to make OR<1 favor high activity levels, always active vs. consistently active, RR approximated with OR.
risk of hospitalization, 47.6% lower, OR 0.52, p < 0.001, high activity levels 11,279, low activity levels 29,099, inverted to make OR<1 favor high activity levels, always active vs. always inactive, RR approximated with OR.
risk of hospitalization, 41.9% lower, OR 0.58, p < 0.001, high activity levels 11,279, low activity levels 83,452, inverted to make OR<1 favor high activity levels, always active vs. mostly inactive, RR approximated with OR.
risk of hospitalization, 30.1% lower, OR 0.70, p < 0.001, high activity levels 11,279, low activity levels 42,490, inverted to make OR<1 favor high activity levels, always active vs. some activity, RR approximated with OR.
risk of hospitalization, 20.0% lower, OR 0.80, p < 0.001, high activity levels 11,279, low activity levels 27,871, inverted to make OR<1 favor high activity levels, always active vs. consistently active, RR approximated with OR.
Yuan, 6/20/2021, retrospective, China, peer-reviewed, 9 authors, study period 15 February, 2020 - 14 March, 2020.
Submit Corrections or Updates.
 risk of death, 90.5% lower, RR 0.09, p = 0.09, high activity levels 0 of 61 (0.0%), low activity levels 6 of 103 (5.8%), NNT 17, relative risk is not 0 because of continuity correction due to zero events (with reciprocal of the contrasting arm), excluded in exclusion analyses: excessive unadjusted differences between groups.
risk of severe case, 70.0% lower, RR 0.30, p = 0.03, high activity levels 3 of 61 (4.9%), low activity levels 26 of 103 (25.2%), NNT 4.9, adjusted per study, inverted to make RR<1 favor high activity levels, odds ratio converted to relative risk, multivariable.
Zhang (B), 12/6/2020, retrospective, United Kingdom, peer-reviewed, 9 authors.
Submit Corrections or Updates.
 risk of death, 26.0% lower, OR 0.74, p = 0.17, adjusted per study, AMPA, per SD increase, multivariable, RR approximated with OR.
risk of case, 18.0% lower, OR 0.82, p = 0.01, adjusted per study, AMPA, per SD increase, multivariable, RR approximated with OR.
Šebić, 7/15/2023, retrospective, Bosnia and Herzegovina, peer-reviewed, 5 authors.
Submit Corrections or Updates.
 risk of oxygen therapy, 89.5% lower, RR 0.11, p = 0.045, high activity levels 0 of 53 (0.0%), low activity levels 4 of 47 (8.5%), NNT 12, relative risk is not 0 because of continuity correction due to zero events (with reciprocal of the contrasting arm).
risk of hospitalization, 91.4% lower, RR 0.09, p = 0.02, high activity levels 0 of 53 (0.0%), low activity levels 5 of 47 (10.6%), NNT 9.4, relative risk is not 0 because of continuity correction due to zero events (with reciprocal of the contrasting arm).
risk of progression, 83.9% lower, RR 0.16, p < 0.001, high activity levels 4 of 53 (7.5%), low activity levels 22 of 47 (46.8%), NNT 2.5, pneumonia.
no recovery, 47.3% lower, RR 0.53, p < 0.001, high activity levels 22 of 53 (41.5%), low activity levels 37 of 47 (78.7%), NNT 2.7, day 14.
Supplementary Data
Supplementary Data
References
Please send us corrections, updates, or comments. Vaccines and treatments are complementary. All practical, effective, and safe means should be used based on risk/benefit analysis. No treatment, vaccine, or intervention is 100% available and effective for all current and future variants. We do not provide medical advice. Before taking any medication, consult a qualified physician who can provide personalized advice and details of risks and benefits based on your medical history and situation. FLCCC and WCH provide treatment protocols.
  or use drag and drop   
Submit    
Tweet
Share
@CovidAnalysis
FAQ
Public domain CC0 1.0