Abstract
Background
It has been suggested that cardiorespiratory fitness (CRF) may be used to identify those at greatest risk for severe COVID-19 illness. However, no study to date has examined the association between CRF and COVID-19. The objectives of this study were to determine whether CRF is independently associated with testing positive with or dying from COVID-19.
Methods
This is a prospective cohort study of 2,690 adults from the UK Biobank Study that were followed from March 16th, 2020 to July 26th, 2020. Participants who were tested for COVID-19 and had undergone CRF assessment were examined. CRF was estimated (eCRF) and categorized as low (<20th percentile), moderate (20th to 80th percentile) and high (≥80th percentile) within sex and ten-year age groups (e.g. 50–60 years). Participants were classified as having COVID-19 if they tested positive (primarily PCR tests) at an in-patient or out-patient setting as of July 26, 2020. Participants were classified as having died from COVID-19 if the primary or underlying cause of death was listed ICD-10 codes U071 or U072 by June 30th, 2020. Adjusted risk ratios (aRR) and 95% confidence intervals (CI) were estimated and a forward model building approach used to identify covariates.
Findings
There was no significant association between eCRF and testing positive for COVID-19. Conversely, individuals with moderate (aRR = 0.43, 95% CI: 0.25, 0.75) and high fitness (aRR = 0.37, 95% CI: 0.16, 0.85) had a significantly lower risk of dying from COVID-19 than those with low fitness.
Conclusions
While eCRF was not significantly associated with testing positive for COVID-19, we observed a significant dose-response between having higher eCRF and a decreased risk of dying from COVID-19. This suggests that prior gains in CRF could be protective against dying from COVID-19 should someone develop the virus.
Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel virus that was first detected in December 2019 [1] and has rapidly evolved into a global pandemic. As of September 9, 2020, there were 27,628,190 reported cases of coronavirus-disease 19 (COVID-19) worldwide, with 898,757 reported deaths [2].
Physical inactivity is an area of public health concern. Research suggests that physical inactivity is associated with increased COVID-19 severity [3]. Regular physical activity can improve overall health by decreasing the incidence [4–9] and morbidity [3, 5, 10] of chronic and communicable diseases. However, because the greatest gains are from physical activity that was sustained over a period of time [6, 10], the assessment of episodic physical activity may not be the best indicator of heath outcomes.
Cardiorespiratory fitness (CRF) is the ability of the body to supply oxygen to skeletal muscles during sustained activity and increases with regular physical activity [11]. CRF is an objective, reproducible measure that captures the health benefits of sustained physical activity [12]. CRF is also an established predictor of cardiovascular disease [13–15] and all-cause mortality [13–15], independent of age and body mass index—risk factors associated with COVID-19 severity and mortality. In fact, CRF has been found to better predict heart disease than assessments of physical activity, suggesting that it may be the better assessment of being active [16]. As such, several groups have hypothesized that high CRF may reduce the risk, severity, and duration of viral infections, including COVID-19 [17, 18]. However, to date, no study has examined the association between CRF and COVID-19 infection or mortality.
The objective of this study was to examine the association between CRF and the risk of (1) testing positive for COVID-19, and (2) dying from COVID-19, among participants of the UK Biobank study (UKB).
Methods
This prospective cohort study has been conducted using the UK Biobank (UKB) Resource [19]. Research Ethics approval was received from the University of Toronto Research Ethics Board (REB) (protocol #39368). Individual participants provided written consent for their data to be shared with external researchers who have REB approval to access the UKB. Therefore, the specific research team for this project did not collect any consents. The UKB is a large population-based cohort that recruited approximately 500,000 adults aged 40 to 69 years from 2006 to 2010. In 2009, the UKB added a baseline CRF assessment, which 95,152 participants (18.9%) completed.
At baseline, questionnaires were used to capture demographic information (e.g., education, ethnicity) and medical history, while trained technicians measured height, weight, and conducted physical health assessments (e.g., CRF). Details of the recruitment process and data collection can be found elsewhere [19, 20].
On March 16th, 2020, the UKB was linked to COVID-19 testing data and results [21]. As of July 26, 2020, a total of 20,554 tests had been performed on 13,502 UKB participants; our sample represents the number of individuals tested who also completed the CRF assessment (n = 2,722). Participants were excluded if they were tested for COVID-19 posthumously (n = 3), if they were missing data on body mass index (BMI) (n = 4), alcohol use frequency (n = 12), or smoking status (n = 13). This left a total of 2,690 participants for analysis. Analyses examining risk of COVID-19 mortality included a sub-set of participants who tested positive for COVID-19 before the end of mortality data follow-up (July 1, 2020) (n = 346).
CRF assessment
CRF was assessed by trained technicians using a submaximal bicycle test [22]. Each CRF assessment followed a similar pattern: (1) a pretest phase of 15 seconds; (2) a constant phase of 2 minutes where women received a workload of 30 W and men of 40 W; (3) the incremental phase which was 4 minutes. The incremental phase varied slightly among participants based on their responses to an initial questionnaire, which assigned them individuals to different protocols based on perceived risk Heart rate was captured throughout the test using a 4-lead ECG device (CAM-USB 6.5 with Cardiosoft v6.51 software). We used linear regression to estimate workload at the predicted maximum heartrate. Maximum oxygen consumption (VO2 max) was estimated using the formula: 7ml/min/kg + ((10.8ml/min/watt x predicted maximum workload in watts)/weight in kilograms) [23]. Estimated CRF (eCRF) was operationalized in two ways: first, participants were classified as having low eCRF (yes/no) if their estimated VO2 max was in the lowest 20th percentile within their sex and 10-year age bands [14, 24]. Second, eCRF was classified as a three-category variable with participants being classified as having low (<20th percentile), moderate (20th to <80th percentile), and high eCRF (≥ 80th percentile) by sex and within 10-year age bands [24]. We also conducted a sensitivity analysis for the three-category eCRF variable where we classified individuals as having low (<20th percentile), moderate (20th to <60th percentile), and high CRF (≥60th percentile) by sex and within 10-year age bands [25].
Outcomes
Individuals were classified as having COVID-19 if they tested positive at an out-patient or in-hospital laboratory between March 16, 2020 and July 26, 2020 by PCR test. The majority of specimen were taken in the nose or upper respiratory tract (94% or 2524 of 2690). Primary and underlying cause of death was obtained from the National Health Service Information Centre for participants in England and Wales, and the National Health Service Central Register for participants in Scotland through June 30th, 2020. Individuals were classified as having died from COVID-19 if their primary or underlying cause of death were ICD-10 codes U071 [COVID-19, virus identified] or U072 [COVID-19, virus not identified].
Covariates
Age at testing was treated as a continuous variable, and calculated using age at UKB assessment, month of COVID-19 testing, and month of birth. Other demographic and lifestyle variables were self-reported at baseline and assumed to remain constant over the follow-up period. The following variables were considered as covariates: sex (male/female); race (white/Asian/Black/Other); education (secondary/post-secondary/missing); smoking (never/current/previous); alcohol use frequency (never/special occasions only/one to three times a month/once to twice a week/three or four times a week)/body mass index (underweight/normal weight[<25 kg/m2]/overweight [≥ 25 kg/m2 and <30 kg/m2]/obesity[≥30 kg/m2]).
Chronic conditions were assessed at baseline and updated over the follow-up period using in-patient hospital administrative databases and cancer registries. Evaluated chronic conditions included: immune disorders (e.g., whole organ transplant), cardiovascular disorders (e.g., high blood pressure), respiratory disease (e.g., chronic obstructive pulmonary disorders), liver disease, kidney failure, cancer, and diabetes (Table 1).
Table 1. Characteristics of the whole study population and those who tested positive for COVID-19.
| Variable | Tested for COVID-19 | Tested Positive for COVID-19 |
|---|---|---|
| Sample size (n) | 2,690 | 346 |
| DEMOGRAPHICS | ||
| Age (median (IQR)) | 70 (61, 75) | 67 (57, 74) |
| Male (n (%)) | 1,314 (49.0) | 180 (52) |
| Race, n (%) | ||
| White | 2,416 (90) | 289 (84) |
| Asian | 100 (4) | 17 (5) |
| Black | 104 (4) | 25 (7) |
| Mixed/Other/Missing | 70 (3) | 15 (4) |
| Education, n (%) | ||
| Secondary | 535 (20) | 65 (19) |
| Post-secondary | 1,657 (62) | 227 (66) |
| Missing | 498 (19) | 54 (16) |
| ANTHROPOMETRIC MEASURES | ||
| Body Mass Index in kg/m2 (mean ± SD) | 28.2 ± 5.2 | 28.9 ± 5.5 |
| Body Mass Index Category, n (%) | ||
| Underweight/Normal weight | 758 (28) | 83 (24) |
| Overweight | 1,099 (41) | 148 (43) |
| Obesity | 833 (31) | 115 (33) |
| LIFESTYLE RISK FACTORS | ||
| Smoke Status, n (%) | ||
| Never | 1,048 (39) | 151 (44) |
| Current | 308 (11) | 35 (10) |
| Previous | 1,334 (50) | 160 (46) |
| Alcohol Use Frequency, n (%) | ||
| Daily | 547 (20) | 60 (17) |
| Three or four times a week | 551 (20) | 56 (16) |
| Once or twice a week | 672 (25) | 86 (25) |
| One to three times a month | 280 (10) | 43 (12) |
| Special occasions only | 365 (14) | 48 (14) |
| Never (n (%)) | 275 (10) | 53 (15) |
| Estimated Cardiorespiratory Fitness | ||
| Estimated VO2 max, mean ± SD | 27.3 ± 5.5 | 27.3 ± 5.4 |
| Three category, n (%) | ||
| Low Fitness | 529 (20) | 77 (22) |
| Moderate Fitness | 1,618 (60) | 214 (63) |
| High Fitness | 543 (20) | 55 (15) |
| CHRONIC CONDITIONS | ||
| Immune System Disorders, n % | ||
| Whole Organ Transplant | 170 (6) | 22 (6) |
| HIV/AIDs | 195 (7) | 30 (9) |
| Inflammatory Disease of the CNS | 27 (1) | 8 (2) |
| Other immune system disorders | 694 (26) | 92 (27) |
| Cardiovascular Disease, n (%) | ||
| Hypertension | 1,244 (46) | 169 (49) |
| Cholesterol Disease | 1,131 (42) | 147 (42) |
| Ischaemic Heart Disease | 1,061 (39) | 140 (40) |
| Pulmonary Heart Disease | 1,254 (47) | 169 (49) |
| Other forms of Heart Disease | 1,116 (41) | 151 (44) |
| Cerebrovascular Disease | 1,129 (42) | 155 (45) |
| Respiratory Disorders, n (%) | ||
| Bronchitis/Emphysema | 709 (26) | 85 (25) |
| COPD | 681 (25) | 85 (25) |
| Asthma | 807 (30) | 105 (30) |
| Other Respiratory Disorder | 794 (30) | 98 (28) |
| Liver Disease, n (%) | 1,387 (52) | 173 (50) |
| Kidney Failure, n (%) | 560 (21) | 72 (21) |
| Diabetes, n (%) | 498 (19) | 76 (22) |
| Cancer, n (%) | 554 (21) | 69 (20) |
| Chronic Condition, n (%) | 2,363 (88) | 304 (88) |
| Number of Chronic Conditions, median (IQR) | 5 (1, 8) | 5 (1, 8) |
| COVID-19 Specific Death, n (%) | -- | 59 (17) |
Abbreviations: IQR = interquartile range, % = percent, SD = standard deviation, CNS = central nervous system, COPD = chronic obstructive pulmonary disorder, COVID-19 = coronavirus-disease 19.
Statistical analysis
Modified Poisson regression with log link was used to estimate adjusted risk ratios (aRR) and 95% confidence intervals (CI) for the association of eCRF with testing positive for COVID-19 and COVID-19 specific mortality. A forward model building approach [26] was undertaken separately for each outcome. Variables included as potential covariates are those described above. Given the well documented association between CRF and BMI [27–30], BMI was forced into all models. The final models for both testing positive for COVID-19 and COVID-19-related mortality included age at testing and BMI category, regardless of the eCRF variable used (i.e., binary low eCRF vs. three-level categorical variable). The models for testing positive for COVID-19 also included race, and the mortality models included sex. All analyses were conducted using SAS version 9.4. All tests were two-sided, and findings were considered statistically significant at an alpha of 0.05.
Results
Characteristics of all participants are presented in Table 1. Approximately 13% of the sample (n = 346) tested positive for COVID-19, and there was a high case fatality rate of 17% (n = 59). Individuals who tested positive for COVID-19 were slightly younger than all those tested (median (IQR) = 67 (57, 74) versus 70 (61, 75)). All participants who were tested for COVID-19 had a high prevalence of comorbidities (median of 5 conditions) and approximately 88% had at least one chronic condition other than obesity. Approximately 31% of participants had obesity. However, participants were in the overweight BMI category on average (tested for COVID-19: 28.2 ± 5.2 kg/m2; tested positive: 28.9 ± 5.5 kg/m2).
There was no difference in the mean estimated VO2 max for those who were tested for COVID-19 and those who tested positive for COVID-19 (27.3 ± 5.5 ml/kg/min and 27.3 ± 5.4 ml/kg/min respectively). Most of the participants who were tested for COVID-19 had moderate fitness (n = 1,618, 60%), while approximately 20% had low (n = 529, 20%) and high fitness (n = 543, 20.2), respectively.
Compared to individuals with low eCRF, those with moderate (aRR = 0.93, 95% CI: 0.72, 1.21) or high (aRR = 0.77, 95% CI: 0.52, 1.15) eCRF were not at an increased risk of testing positive for COVID-19 (Table 2). Conversely, individuals with low eCRF had more than 2 times the risk of dying from COVID-19 compared to those with moderate or high fitness (aRR = 2.34, 95% CI: 1.35, 4.05). Further, when eCRF was categorized as high, moderate, and low, compared to individuals with low fitness, those with moderate fitness had a 57% (aRR = 0.43, 95% CI: 0.25, 0.75) lower risk and those with high fitness had a 63% (aRR = 0.37, 95% CI: 0.16, 0.85) lower risk of dying from COVID-19 (Table 2).
Table 2. Association between estimated cardiorespiratory fitness (eCRF) and testing positive for COVID-19 and COVID-19 mortality.
| Risk of Testing Positive | Mortality | |
|---|---|---|
| aRR (95% CI)a | aRR (95% CI)b | |
| ESTIMATED CARDIORESPIRATORY FITNESS | ||
| Low Fitness | 1.08 (0.83, 1.41) | 2.34 (1.35, 4.05) |
| Three Category Variable | ||
| Low Fitness | Ref | Ref |
| Moderate Fitness | 0.93 (0.72, 1.21) | 0.43 (0.25, 0.75) |
| High Fitness | 0.77 (0.52, 1.15) | 0.37 (0.16, 0.85) |
Abbreviations: aRR = adjusted relative risk, CI = confidence interval.
aEstimates are adjusted for age at testing, race (white, Asian, Black, Other), and BMI category (underweight/normal weight, overweight, obesity).
bEstimates are adjusted for age at testing, sex (male, female), BMI category (underweight/normal weight, overweight, obesity), and alcohol use frequency.
As a sensitivity analysis, eCRF was reclassified as low (<20th percentile), moderate (20th to <60th percentile) and high (≥ 60th percentile). Using this alternative classification, eCRF was still not a predictor of testing positive for COVID-19 (p = 0.41). The findings were also similar with regards to dying from COVID-19 in that individuals with moderate (RR = 0.42, 95% CI: 0.24–0.75) and high (RR = 0.44, 95% CI: 0.23–0.85) were at a significantly lower risk of dying from COVID-19 than individuals with low fitness.
Discussion
This is the first study to examine the relationship between eCRF and COVID-19 infection and mortality. While we found that eCRF was not associated with risk of testing positive for COVID-19, we found evidence of a dose-response relationship wherein people with higher eCRF have a lower risk of dying from COVID-19.
Few population-based assessments of CRF exist, likely due to the higher technical and financial requirements of these tests compared to measured and self-reported physical activity27, limiting research into the impact of CRF on developing or dying from a communicable disease. Though partly heritable [31], measured and eCRF is also predicted by physical activity. It has been hypothesized that high CRF resulting from regular physical activity, especially exercise training, confers innate immune protection, attenuating the risk of infectious diseases including COVID-19 [18, 32]. However, the benefits of CRF in preventing COVID-19 may be complicated by the fact that participating in activities that promote CRF (e.g. physical or exercise in groups) could increase exposure to and spread of the virus. This paradoxical relationship may explain, in part, why we did not observe a significant association between CRF and testing positive for COVID-19.
Patients with severe COVID-19 may experience significant decreases in lung function, potentially requiring mechanical ventilation [33], and respiratory and circulatory failure are common causes of death among COVID-19 patients [34]. Since CRF measures the ability of these systems to supply oxygen to skeletal muscles during sustained activity, CRF may help identify individuals at the greatest risk for severe COVID-19 outcomes [17]. Consistent with this hypothesis, we found that low CRF more than doubled individuals’ risk of dying from COVID-19. In addition, having moderate to high CRF significantly decreased the risk of COVID-19 mortality, with high CRF reducing the risk of mortality even more than moderate CRF, suggestive of dose-response.
UKB participants are known to differ from the broader UK population. Previous research has suggested that the UKB is impacted by healthy volunteer bias, wherein UKB participants have better health and are less likely to live in socially deprived areas than the general UK population [20]. In the current study, approximately 88% of participants tested for COVID-19 in the UKB had at least one chronic condition other than obesity. This is consistent with an American study where 89% of patients hospitalized for COVID-19 had at least one underlying condition [35].
In addition, participants appeared to have a lower CRF than similar populations tested for COVID-19. Women in the current study had a comparable CRF to women in some [36], but not all studies. For example, CRF values for women in the current study were much lower than three cohorts from the United States and Norway (range: 30.4 to 34.4 ml/kg/min) [36–38]. Men had a lower CRF on average than men in all four of the above-mentioned studies (current study: 30.5 ± 5.2 ml/kg/min versus range: 33.8 to 42.6 ml/kg/min [36–38]). One possible explanation for these differences could be that while the cohorts directly measured CRF, in the current study CRF was estimated from a submaximal fitness assessment. Another possible explanation could be that testing was initially restricted to in-hospital symptomatic patients, such that those tested for COVID-19 would tend to have poorer health than the general public. As the risk of dying from COVID-19 was inversely associated with CRF, this could mean that the protective effects of CRF against COVID-19 mortality may be more pronounced in the general population.
This study has numerous strengths. First, exercise-based CRF assessments are rare, and the UKB uses trained technicians and a validated protocol to conduct the assessment. The comprehensive data available through the UKB ensured we had access to several important confounders, either self-reported or via linkage to administrative databases over the follow-up period. While previous studies of COVID-19 mortality have examined all cause-mortality, we had a sufficiently large sample to examine COVID-19-specific mortality, which is another important strength. Limitations of the current study include the approximate 10-year lag between baseline measurements (demographics, lifestyle variables, CRF assessment) and COVID-19 testing. We assumed that these variables remained constant over the follow-up period. As BMI and CRF are known to increase and decrease respectively with age, our results may be biased to the null. We attempted to mitigate this potential bias by categorizing BMI and CRF, as while these factors may change with age, they would be less likely to change so severely as to change categorizes. We also likely underestimated the true burden of comorbid conditions in this population as we only had access to hospital admissions and cancer registry data to capture diseases over the follow-up period.
In summary, this is the first study to examine the association between CRF and testing positive for or dying from COVID-19. Importantly, CRF was not significantly associated with testing positive for COVID-19 but having moderate or high CRF was associated with a significant decrease in the risk of dying from COVID-19. Physical activity is a modifiable behaviour that positively influences CRF [39] and has also been identified as a way to mitigate some of the potential negative health impacts from COVID-19 lockdowns [40–42]. This study provides additional support for these recommendations and suggests that prior physical activity could be protective against dying from COVID-19 among those that test positive.
Acknowledgments
This research has been conducted using the UK Biobank Resource under Application Number 52609.
Data Availability
Data cannot be shared publicly because this is third party data which also contains identifiable information. Data are available from the UK Biobank for researchers who meet the criteria for access to the data. Researchers interested in accessing the data can contact access@ukbiobank.ac.uk.
Funding Statement
The authors received no specific funding for this work.
References
- 1.The World Health Organization, “Coronavirus disease 2019 (COVID-19) Situation Report-94 HIGHLIGHTS,” Geneva, 2020. Accessed: Nov. 16, 2020. [Online]. Available: https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200423-sitrep-94-covid-19.pdf#:~:text=The first human cases of,%2C in December 2019.
- 2.Dong E., Du H., and Gardner L., “An interactive web-based dashboard to track COVID-19 in real time,” Lancet Infect. Dis., 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hamer M., Kivimäki M., Gale C. R., and David Batty G., “Lifestyle risk factors, inflammatory mechanisms, and COVID-19 hospitalization: A community-based cohort study of 387,109 adults in UK,” Brain. Behav. Immun., 2020, 10.1016/j.bbi.2020.05.059 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Laaksonen D. E. et al. , “Physical activity in the prevention of type 2 diabetes: The finnish diabetes prevention study,” Diabetes, 2005, 10.2337/diabetes.54.1.158 [DOI] [PubMed] [Google Scholar]
- 5.Romeo J., Wärnberg J., Pozo T., and Marcos A., “Physical activity, immunity and infection,” 2010, 10.1017/S0029665110001795 [DOI] [PubMed] [Google Scholar]
- 6.Florido R. et al. , “Six-year changes in physical activity and the risk of incident heart failure ARIC study,” Circulation, 2018, 10.1161/CIRCULATIONAHA.117.030226 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Nieman D. C., Henson D. A., Austin M. D., and Sha W., “Upper respiratory tract infection is reduced in physically fit and active adults,” Br. J. Sports Med., 2011, 10.1136/bjsm.2010.077875 [DOI] [PubMed] [Google Scholar]
- 8.Matthews C. E., Ockene I. S., Freedson P. S., Rosal M. C., Merriam P. A., and Hebert J. R., “Moderate to vigorous physical activity and risk of upper-respiratory tract infection,” Med. Sci. Sports Exerc., 2002, 10.1097/00005768-200208000-00003 [DOI] [PubMed] [Google Scholar]
- 9.Fondell E. et al. , “Physical activity, stress, and self-reported upper respiratory tract infection,” Med. Sci. Sports Exerc., 2011, 10.1249/MSS.0b013e3181edf108 [DOI] [PubMed] [Google Scholar]
- 10.Moholdt T., Lavie C. J., and Nauman J., “Sustained Physical Activity, Not Weight Loss, Associated With Improved Survival in Coronary Heart Disease,” J. Am. Coll. Cardiol., 2018, 10.1016/j.jacc.2018.01.011 [DOI] [PubMed] [Google Scholar]
- 11.Duscha B. D. et al. , “Effects of exercise training amount and intensity on peak oxygen consumption in middle-age men and women at risk for cardiovascular disease,” Chest, vol. 128, no. 4, pp. 2788–2793, 2005, 10.1378/chest.128.4.2788 [DOI] [PubMed] [Google Scholar]
- 12.Ingelsson E. et al. , “Heritability, linkage, and genetic associations of exercise treadmill test responses,” Circulation, 2007, 10.1161/CIRCULATIONAHA.106.683821 [DOI] [PubMed] [Google Scholar]
- 13.Kodama S. et al. , “Cardiorespiratory fitness as a quantitative predictor of all-cause mortality and cardiovascular events in healthy men and women: A meta-analysis,” JAMA—Journal of the American Medical Association. 2009, 10.1001/jama.2009.681 [DOI] [PubMed] [Google Scholar]
- 14.Wei M. et al. , “Relationship between low cardiorespiratory fitness and mortality in normal-weight, overweight, and obese men,” J. Am. Med. Assoc., 1999, 10.1001/jama.282.16.1547 [DOI] [PubMed] [Google Scholar]
- 15.Harber M. P. et al. , “Impact of Cardiorespiratory Fitness on All-Cause and Disease-Specific Mortality: Advances Since 2009,” Progress in Cardiovascular Diseases. 2017, 10.1016/j.pcad.2017.03.001 [DOI] [PubMed] [Google Scholar]
- 16.Williams P. T., “Physical fitness and activity as separate heart disease risk factors: A meta-analysis,” Med. Sci. Sports Exerc., 2001, 10.1097/00005768-200105000-00012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ahmed I., “COVID-19 –does exercise prescription and maximal oxygen uptake (VO2 max) have a role in risk-stratifying patients?,” Clin. Med. J. R. Coll. Physicians London, 2020, 10.7861/clinmed.2020-0111 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zbinden-Foncea H., Francaux M., Deldicque L., and Hawley J. A., “Does High Cardiorespiratory Fitness Confer Some Protection Against Proinflammatory Responses After Infection by SARS-CoV-2?,” Obesity. 2020, 10.1002/oby.22849 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.UK Biobank Coordinating Centre, “UK Biobank: Protocol for a large-scale prospective epidemiological resource,” 2007. Accessed: Mar. 20, 2019. [Online]. Available: https://www.ukbiobank.ac.uk/wp-content/uploads/2011/11/UK-Biobank-Protocol.pdf. [Google Scholar]
- 20.Fry A. et al. , “Comparison of Sociodemographic and Health-Related Characteristics of UK Biobank Participants with Those of the General Population,” Am. J. Epidemiol., 2017, 10.1093/aje/kwx246 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.United Kingdom Biobank Study, “UKB: External Info: COVID19_tests,” 2020. http://biobank.ndph.ox.ac.uk/ukb/exinfo.cgi?src=COVID19_tests (accessed May 12, 2020). [Google Scholar]
- 22.UK Biobank, “Cardio Assessment Version 1.0,” 2011. Accessed: Mar. 17, 2019. [Online]. Available: http://www.ukbiobank.ac.uk/. [Google Scholar]
- 23.Swain D. P., “Energy cost calculations for exercise prescription: An update,” Sports Medicine, 2000. 10.2165/00007256-200030010-00002 [DOI] [PubMed] [Google Scholar]
- 24.Blair S. N., Kohl H. W., Paffenbarger R. S., Clark D. G., Cooper K. H., and Gibbons L. W., “Physical Fitness and All-Cause Mortality: A Prospective Study of Healthy Men and Women,” JAMA J. Am. Med. Assoc., 1989, 10.1001/jama.262.17.2395 [DOI] [PubMed] [Google Scholar]
- 25.Peel J. B., Sui X., Adams S. A., HIbert J. R., Hardin J. W., and Blair S. N., “A prospective study of cardiorespiratory fitness and breast cancer mortality,” Med. Sci. Sports Exerc., 2009, 10.1249/MSS.0b013e31818edac7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gortmaker S. L., Hosmer D. W., and Lemeshow S., “Applied Logistic Regression.,” Contemp. Sociol., 1994, 10.2307/2074954 [DOI] [Google Scholar]
- 27.Hung T. H., Liao P. A., Chang H. H., Wang J. H., and Wu M. C., “Examining the relationship between cardiorespiratory fitness and body weight status: Empirical evidence from a population-based survey of adults in Taiwan,” Sci. World J., 2014, 10.1155/2014/463736 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.DiPietro L., Kohl H. W., Barlow C. E., and Blair S. N., “Improvements in cardiorespiratory fitness attenuate age-related weight gain in healthy men and women: The aerobics center longitudinal study,” Int. J. Obes., 1998, 10.1038/sj.ijo.0800543 [DOI] [PubMed] [Google Scholar]
- 29.Sui X., “Longitudinal analyses of physical activity and cardiorespiratory fitness on adiposity and glucose levels,” 2012. [Google Scholar]
- 30.Williams P. T., “Self-selection Accounts for inverse association between weight and cardiorespiratory fitness,” Obesity, 2008, 10.1038/oby.2007.5 [DOI] [PubMed] [Google Scholar]
- 31.Bouchard C. et al. , “Familial aggregation of VO2 max response to exercise training: Results from the HERITAGE family study and elite athletic performance,” J. Appl. Physiol., vol. 87, no. 3, pp. 1003–1008, 1999. 10.1152/jappl.1999.87.3.1003 [DOI] [PubMed] [Google Scholar]
- 32.Laddu D. R., Lavie C. J., Phillips S. A., and Arena R., “Physical activity for immunity protection: Inoculating populations with healthy living medicine in preparation for the next pandemic,” Progress in Cardiovascular Diseases. 2020, 10.1016/j.pcad.2020.04.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Wang D. et al. , “Clinical Characteristics of 138 Hospitalized Patients with 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China,” JAMA—J. Am. Med. Assoc., vol. 323, no. 11, pp. 1061–1069, March. 2020, 10.1001/jama.2020.1585 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Rajagopal K. et al. , “Advanced Pulmonary and Cardiac Support of COVID-19 Patients: Emerging Recommendations from ASAIO—A ‘living Working Document,’” ASAIO J., 2020, 10.1097/MAT.0000000000001180 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Garg S. et al. , “Hospitalization Rates and Characteristics of Patients Hospitalized with Laboratory-Confirmed Coronavirus Disease 2019—COVID-NET, 14 States, March 1–30, 2020,” MMWR. Morb. Mortal. Wkly. Rep., vol. 69, no. 15, pp. 458–464, April. 2020, 10.15585/mmwr.mm6915e3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Kaminsky L. A., Imboden M. T., Arena R., and Myers J., “Reference Standards for Cardiorespiratory Fitness Measured With Cardiopulmonary Exercise Testing Using Cycle Ergometry: Data From the Fitness Registry and the Importance of Exercise National Database (FRIEND) Registry,” Mayo Clin. Proc., 2017, 10.1016/j.mayocp.2016.10.003 [DOI] [PubMed] [Google Scholar]
- 37.Loe H., Rognmo Ø., Saltin B., and Wisløff U., “Aerobic Capacity Reference Data in 3816 Healthy Men and Women 20–90 Years,” PLoS One, 2013, 10.1371/journal.pone.0064319 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Edvardsen E., Hansen B. H., Holme I. M., Dyrstad S. M., and Anderssen S. A., “Reference values for cardiorespiratory response and fitness on the treadmill in a 20- to 85-year-old population,” Chest, 2013, 10.1378/chest.12-1458 [DOI] [PubMed] [Google Scholar]
- 39.Jackson A. S., Sui X., Hébert J. R., Church T. S., and Blair S. N., “Role of lifestyle and aging on the longitudinal change in cardiorespiratory fitness,” Arch. Intern. Med., 2009, 10.1001/archinternmed.2009.312 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Jurak G. et al. , “Physical activity recommendations during the coronavirus disease-2019 virus outbreak,” Journal of Sport and Health Science. 2020, 10.1016/j.jshs.2020.05.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Hall G., Laddu D. R., Phillips S. A., Lavie C. J., and Arena R., “A tale of two pandemics: How will COVID-19 and global trends in physical inactivity and sedentary behavior affect one another?,” Progress in Cardiovascular Diseases. 2020, 10.1016/j.pcad.2020.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Schwendinger F. and Pocecco E., “Counteracting physical inactivity during the COVID-19 pandemic: Evidence-based recommendations for home-based exercise,” Int. J. Environ. Res. Public Health, 2020, 10.3390/ijerph17113909 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data cannot be shared publicly because this is third party data which also contains identifiable information. Data are available from the UK Biobank for researchers who meet the criteria for access to the data. Researchers interested in accessing the data can contact access@ukbiobank.ac.uk.