Skip to main content
PMC home page
PLOS ONE logoLink to PLOS ONE
. 2021 Jun 22;16(6):e0253640. doi: 10.1371/journal.pone.0253640

Body mass index and severity/fatality from coronavirus disease 2019: A nationwide epidemiological study in Korea

In Sook Kang 1,*, Kyoung Ae Kong 2
Editor: Jie V Zhao3
PMCID: PMC8219144  PMID: 34157043

Abstract

Objective

Obesity has been reported as a risk factor for severe coronavirus disease 2019 (COVID-19) in recent studies. However, the relationship between body mass index (BMI) and COVID-19 severity and fatality are unclear.

Research design and methods

This study included 4,141 COVID-19 patients who were released from isolation or had died as of April 30, 2020. This nationwide data was provided by the Korean Centers for Disease Control and Prevention Agency. BMI was categorized as follows; < 18.5 kg/m2, 18.5–22.9 kg/m2, 23.0–24.9 kg/m2, 25.0–29.9 kg/m2, and ≥ 30 kg/m2. We defined a fatal illness if the patient had died.

Results

Among participants, those with a BMI of 18.5–22.9 kg/m2 were the most common (42.0%), followed by 25.0–29.9 kg/m2 (24.4%), 23.0–24.9 kg/m2 (24.3%), ≥ 30 kg/m2 (4.7%), and < 18.5 kg/m2 (4.6%). In addition, 1,654 (41.2%) were men and 3.04% were fatalities. Multivariable analysis showed that age, male sex, BMI < 18.5 kg/m2, BMI ≥ 25 kg/m2, diabetes mellitus, chronic kidney disease, cancer, and dementia were independent risk factors for fatal illness. In particular, BMI < 18.5 kg/m2 (odds ratio [OR] 3.97, 95% CI 1.77–8.92), 25.0–29.9 kg/m2 (2.43, 1.32–4.47), and ≥ 30 kg/m2 (4.32, 1.37–13.61) were found to have higher ORs than the BMI of 23.0–24.9 kg/m2 (reference). There was no significant difference between those with a BMI of 18.5–22.9 kg/m2 (1.59, 0.88–2.89) and 23.0–24.9 kg/m2.

Conclusions

This study demonstrated a non-linear (U-shaped) relationship between BMI and fatal illness. Subjects with a BMI of < 18.5 kg/m2 and those with a BMI ≥ 25 kg/m2 had a high risk of fatal illness. Maintaining a healthy weight is important not only to prevent chronic cardiometabolic diseases, but also to improve the outcome of COVID-19.

Introduction

Since the first coronavirus disease 2019 (COVID-19) case was reported in December 2019 in China, the pandemic has been progressing worldwide. Nowadays, people are more likely to gain weight as a result of increasing social distancing, increasing intake of unhealthy food, and decreased physical activity [1]. Obesity and overweight are related to metabolic syndrome and cardiovascular events, and are important health issues that can cause various chronic illnesses [2]. Previous studies of body mass index (BMI) showed that mortality due to cardiovascular diseases and all other causes are increased in both underweight and obesity, which can be represented by a U-shaped curve [3–6].

Importantly, obesity adversely affects the immune system, and consequently increases the risk of infection [2]. Furthermore, obesity is a risk factor for severe COVID-19 [7]. In a French study, patients with BMI > 35 kg/m2 had a higher odds ratio of 7.36, than those with BMI < 25 kg/m2 for invasive mechanical ventilation among patients admitted to the intensive care unit (ICU) due to COVID-19 [8]. From a study in China, obese men (BMI ≥ 28 kg/m2) were associated with increased risk of severe COVID-19 than normal weight men (BMI < 24 kg/m2) with an odds ratio of 5.66 (95% CI; 1.8–17.75) [9]. Recent studies using Mendelian randomization showed a higher genetically proxied BMI-related increasing risk of severe respiratory COVID-19 and COVID-19 hospitalization [10, 11].

Previously, cardiac disease and obesity have also been reported as common comorbidities in the Middle East respiratory syndrome [12]. Similarly, in the 2009 influenza pandemic, obesity was reported as a common comorbidity in critically ill patients admitted to the ICU [13]. However, one study showed an apparent decrease in the rate of pneumonia with increasing BMI, and pneumonia was more common in underweight individuals during the same influenza pandemic [14]. Furthermore, the obesity survival paradox (the inverse relationship between obesity and mortality) has also been reported in patients with pneumonia in a meta-analysis [15]. Hence, this issue remained controversial, as other studies have showed that there is no association between BMI and survival during the influenza pandemic [16].

To date, there is scarcity of information regarding the obesity survival paradox or underweight risk associated with the outcomes of COVID-19. A systematic review found that the association of obesity with poor clinical presentation and the need for hospitalization due to COVID-19 was consistent, but the association of obesity mortality was not [17]. Hence, in this study, we aimed at evaluating the relationship between BMI and severe COVID-19, especially fatality.

Materials and methods

Data source and study population

This study was performed using nationwide data of COVID-19 confirmed patients who were released from isolation or who died from January 19 to April 30, 2020, in the Republic of Korea. The Korean government has centralized control systems for medical insurance and disease control. The Korea Disease Control and Prevention Agency (KDCA) has been actively tracking and managing almost all confirmed cases and contactors from the beginning of the COVID-19 outbreak. They constructed a nationwide registry of COVID-19 and provided the data (anonymized without personal information) to researchers with granted permission. We temporarily accessed the data from the KDCA through an encrypted remote system that contained anonymized data of 5,628 patients who were released from isolation or died during the aforementioned period. After excluding 1,487 patients (272 patients under the age of 20, 19 pregnant women, 26 without clinical severity information, 4 with no information regarding comorbidity, and 1,166 without available BMI value), 4,141 patients were included in the analysis for this study.

The Ewha Womans University Mokdong Hospital Institutional Review Board deemed this study exempt from ethical review and waived the requirement for informed consent (EUMC2020-07-002), since it was completely anonymized and without personal information.

Study definitions and outcome assessment

COVID-19 was diagnosed based on the detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) by real-time polymerase chain reaction test with respiratory specimens [18]. The investigation of comorbid diseases was performed depending on whether the patients were previously diagnosed with specific diseases or not. Body temperature and BMI were the initial findings of hospital admission.

BMI was calculated as body weight in kilograms divided by height in meter squared and categorized as followings [19–21]: < 18.5 kg/m2, 18.5–22.9 kg/m2, 23.0–24.9 kg/m2, 25.0–29.9 kg/m2, and ≥ 30 kg/m2.

Disease severity was defined according to the guidelines of the KDCA and World Health Organization (WHO) [22, 23]. Briefly, we defined a patient as having a “critical illness” if they required more than invasive mechanical ventilation and “fatal illness” if they died. Critical illnesses included patients requiring invasive mechanical ventilation, those with multi-organ failure, those requiring extracorporeal membrane oxygenation therapy, and/or those who died. The severity evaluation was based on the most severe condition during the hospital stay. Critical illness was defined as severe COVID-19.

Statistical analysis

Descriptive statistics were described using frequencies and proportions for categorical variables. The chi-square test was used to compare the variables. The values of continuous variables were expressed as mean and standard deviation. Analysis of variance was performed for the body temperature.

Multivariable logistic regression was used to analyze the independent risk for COVID-19 critical illness and COVID-19 fatal illness after adjusting for age, sex, BMI, and five comorbid diseases: diabetes mellitus (DM), hypertension, chronic kidney disease (CKD), cancer (excluding cured cases) and dementia. The age variable was given as a categorical variable in units of 10 years, but we calculated the odds ratio of every 10 years as a continuous variable. Since dementia had missing data (n = 314) that were not replaced, 3827 participants were included in the multivariable analysis. Although the BMI of 23.0–24.9 kg/m2 was classified as overweight rather than normal-weight according to the Asia-Pacific cutoff values [21], we used it as a reference. Since this group is of normal weight according to the WHO classification [24], and a previous study reported a BMI cutoff value of Koreans as 24.2 kg/m2 [25].

To analyze the relationship between BMI and fatality by sex, we performed a multivariable analysis after adjusting for age and the five comorbid conditions. In addition to sex, we analyzed the relationship between BMI and fatality according to whether they had DM or HTN. We performed multivariable analysis after adjusting for age, sex, and four comorbidities other than the target disease (DM or HTN). Then, the relationships between BMI and fatality in the groups with and without the diseases (DM or HTN) were evaluated separately.

The threshold for statistical significance was set at P <0.05. Statistical analyses were performed using SAS software (version 9.4, SAS Institute, Cary, NC, USA).

Results

Baseline characteristics according to BMI are presented in Table 1. A BMI of 18.5–22.9 kg/m2 was the most common, with 1,741 (42.0%) patients, followed by BMI of 25.0–29.9, 23.0–24.9 kg/m2 as 1,011 (24.4%), and 1,005 (24.3%), respectively. Moreover, BMI of < 18.5, and ≥ 30 kg/m2 were less than 5%. In the BMI < 18.5, 18.5–22.9, and ≥ 30 kg/m2 groups, the most patients were in their 20s, whereas in the BMI of 23.0–24.9, and 25.0–29.9 kg/m2 groups, most patients were in their 50s. The BMI of 18.5–22.9 kg/m2 was the most common both in male and female, whereas the proportions of the BMI ranging 23.0–24.9 and 25.0–29.9 kg/m2 were higher in male than in female (raw %, male: 27.3, 32.0 vs. female: 22.1, 19.0%, respectively). Those with BMI ≥ 25.0 kg/m2 had higher body temperature than others (P < 0.0001). Dyspnea was a more frequent complaint in patients with a BMI of < 18.0 and ≥ 25.0 kg/m2 than in others (P = 0.0135). Hypertension and DM were more common in the BMI of 23.0–24.9 and ≥ 25.0 kg/m2 than in others (P < 0.0001).

Table 1. Baseline characteristics of the study population according to the body mass index.

Body mass index (kg/m2), N (column %)
<18.5 18.5–22.9 23.0–24.9 25.0–29.9 ≥30 p
N (row %) 191 (4.6) 1741 (42.0) 1005 (24.3) 1011 (24.4) 193 (4.7)
Age (years) <0.0001
20–29 68 (35.6) 455 (26.1) 172 (17.1) 193 (19.1) 54 (28.0)
30–39 24 (12.6) 183 (10.5) 102 (10.1) 117 (11.6) 43 (22.3)
40–49 13 (6.8) 270 (15.5) 133 (13.2) 145 (14.3) 34 (17.6)
50–59 20 (10.5) 347 (19.9) 241 (24.0) 234 (23.1) 31 (16.1)
60–69 16 (8.4) 249 (14.3) 204 (20.3) 186 (18.4) 19 (9.8)
70–79 28 (14.7) 136 (7.8) 111 (11.0) 95 (9.4) 8 (4.1)
≥ 80 22 (11.5) 101 (5.8) 42 (4.2) 41 (4.1) 4 (2.1)
Female 135 (70.7) 1182 (67.9) 534 (53.1) 459 (45.4) 105 (54.4) <0.0001
Male 56 (29.3) 559 (32.1) 471 (46.9) 552 (54.6) 88 (45.6)
BT (°C) 36.97 ± 0.5 36.91 ± 0.5 36.92 ± 0.5 37.02 ± 0.6 37.05 ± 0.6 <0.0001
Dyspnea 0.0135
Yes 26 (13.6) 176 (10.1) 118 (11.7) 146 (14.4) 26 (13.5)
No 165 (86.4) 1565 (89.9) 886 (88.2) 865 (85.6) 167 (86.5)
SBP* (mmHg) <0.0001
<120 81 (42.4) 529 (30.4) 183 (18.2) 152 (15.0) 19 (9.8)
120–129 36 (18.8) 387 (22.2) 219 (21.8) 207 (20.5) 31 (16.1)
130–139 27 (14.1) 344 (19.8) 205 (20.4) 209 (20.7) 55 (28.5)
140–159 30 (15.7) 363 (20.9) 291 (29.0) 329 (32.5) 65 (33.7)
≥ 160 16 (8.4) 115 (6.6) 107 (10.6) 113 (11.2) 23 (11.9)
Diabetes mellitus <0.0001
Yes 18 (9.4) 157 (9.0) 137 (13.6) 160 (15.8) 29 (15.0)
No 173 (90.6) 1584 (91.0) 868 (86.4) 851 (84.2) 164 (85.0)
Hypertension <0.0001
Yes 28 (14.7) 234 (13.4) 247 (24.6) 301 (29.8) 52 (26.9)
No 163 (85.3) 1507 (86.6) 758 (75.4) 710 (70.2) 141 (73.1)
Heart failure 0.0132
Yes 6 (3.1) 12 (0.7) 8 (0.8) 13 (1.3) 1 (0.5)
No 185 (96.9) 1729 (99.3) 997 (99.2) 998 (98.7) 192 (99.5)
Chronic heart disease** 0.0686
Yes 7 (3.7) 42 (2.4) 40 (4.0) 41 (4.1) 4 (2.1)
No 183 (95.8) 1693 (97.2) 961 (95.6) 965 (95.5) 189 (97.9)
Asthma 0.0333
Yes 3 (1.6) 31 (1.8) 32 (3.2) 23 (2.3) 9 (4.7)
No 188 (98.4) 1710 (98.2) 973 (96.8) 988 (97.7) 184 (95.3)
Chronic obstructive lung disease 0.0006
Yes 6 (3.1) 14 (0.8) 5 (0.5) 3 (0.3) 1 (0.5)
No 185 (96.9) 1727 (99.2) 1000 (99.5) 1008 (99.7) 192 (99.5)
Chronic kidney disease 0.7807
Yes 3 (1.6) 18 (1.0) 8 (0.8) 12(1.2) 3 (1.6)
No 188 (88.4) 1723 (99.0) 997 (99.2) 999 (98.8) 190 (98.4)
Malignancy 0.2349
Yes 7 (3.7) 48 (2.8) 26 (2.6) 18 (1.8) 8 (4.1)
No 184 (98.4) 1693 (97.2) 979 (99.2) 993 (98.2) 185 (95.9)
Dementia <0.0001
Yes 20 (10.5) 65 (3.7) 20 (2.0) 16 (1.6) 0
No 162 (84.8) 1528 (87.8) 911 (90.6) 924 (91.4) 181 (93.8)
Disease severity 0.0003
Non critical 175 (91.6) 1686 (96.8) 980 (97.5) 962 (95.2) 186 (96.4)
Critical 16 (8.4) 55 (3.2) 25 (2.5) 49 (4.8) 7 (3.6)
Discharge <0.0001
Live 175 (91.6) 1695 (97.4) 985 (98.0) 972 (96.1) 188 (97.4)
Death 16 (8.4) 46 (2.6) 20 (2.0) 39 (3.9) 5 (2.6)
Open in a new tab

BT, body temperature, mean ± standard deviation; SBP, systolic blood pressure; DM, diabetes mellitus; CHD, chronic heart disease; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease.

Missing data

* n = 5

** n = 16

¶ n = 314.

S1 Table shows the patient characteristics according to the fatal illness. We noted that 41.7% were male. There were no deaths in patients aged < 40 years. The overall proportions of patients in their 20s and 50s were 942 (22.7%) and 873 (21.1%), respectively. Systolic blood pressures of < 120 mmHg and ≥ 140 mmHg were more related to fatal illness than those between 120 mmHg and 140 mmHg (P < 0.0001).

Independent risk factors associated with severe COVID-19 are shown in Table 2. With every 10 years of increase in age, the odds ratio of multivariable analyses for critical and fatal illness increased to 2.64 (95% CI: 2.20–3.15) and 3.04 (2.45–3.78), respectively. Male sex was also found to be an independent risk factor. When the BMI of 23.0–24.9 kg/m2 group was used as a reference, the risk of BMI ranging 18.5–22.9 kg/m2 was as high as 1.59 (0.88–2.89) for fatal illness without statistical significance. All five comorbidities (DM, hypertension, CKD, cancer, and dementia) were related to a higher critical illness rate compared to the absence of comorbidities. Hypertension had an odds ratio of 1.34 in fatal illness, but had no statistical power (95% CI: 0.96–2.30) in multivariable analysis.

Table 2. Logistic analyses of in critical and fatal illness patients.

Critical illness Fatal illness
Odds ratio (95% CI) Odds ratio (95% CI)
Univariable Multivariable* Univariable Multivariable*
Age** 3.09 (2.65–3.60) 2.64 (2.20–3.15) 3.65 (3.03–4.39) 3.04 (2.45–3.78)
Male (vs. female) 1.97 (1.42–2.74) 2.85 (1.94–4.19) 1.90 (1.33–2.72) 2.92 (1.90–4.48)
BMI (kg/m2)
<18.5 3.58 (1.88–6.85) 3.29 (1.54–7.04) 4.50 (2.29–8.86) 3.97 (1.77–8.92)
18.5–22.9 1.28 (0.79–2.07) 1.59 (0.94–2.71) 1.34 (0.79–2.27) 1.59 (0.88–2.89)
23–24.9 1 (Reference) 1 (Reference) 1 (Reference) 1 (Reference)
25.0–29.9 2.00 (1.22–3.26) 2.37 (1.38–4.07) 1.98 (1.14–3.41) 2.43 (1.32–4.47)
≥30 1.48 (0.63–3.46) 4.40 (1.66–11.66) 1.31 (0.49–3.53) 4.32 (1.37–13.61)
Diabetes mellitus 5.74 (4.10–8.04) 1.95 (1.32–2.86) 6.42 (4.46–9.25) 2.38 (1.50–3.51)
Hypertension 6.60 (4.72–9.23) 1.64 (1.11–2.43) 6.47 (4.48–9.33) 1.48 (0.96–2.30)
CKD 10.6 (5.35–21.02) 3.12 (1.36–7.19) 10.1 (4.87–20.92) 2.74 (1.11–6.76)
Cancer 3.88 (2.12–7.09) 2.28 (1.12–4.62) 4.80 (2.61–8.83) 3.01 (1.44–6.28)
Dementia 12.48 (8.09–19.26) 2.13(1.25–3.63) 16.17 (10.36–25.22) 2.50 (1.43–4.36)
Open in a new tab

*n = 3827, The ORs of multivariable analyses were adjusted for age, sex, and five comorbidities (diabetes mellitus, hypertension, chronic kidney disease, cancer, and dementia).

**Age was given as a categorical variable in units of 10 years, but we calculated the odds ratio of every 10 years as a continuous variable. BMI, body mass index; CKD, chronic kidney disease.

S1 Fig shows the number of critical and fatal illnesses according to the BMI group and the fatality rate of the study population. The BMI of 23.0–24.9 kg/m2 group was found to have the lowest fatality. Fatality rate increased in both BMI of < 18.5 and 25.0–29.9 kg/m2 groups, but slightly decreased in the BMI of ≥ 30.0 kg/m2 group. As shown in S1 Fig, patient fatality was the highest in the BMI of < 18.5 kg/m2, followed by the BMI of 25.0–29.9, and ≥ 30.0 kg/m2. After multivariable analysis with adjusting covariates, a similar result was shown in the group with BMI < 18.5 kg/m2 (Fig 1). However, the result according to the degree of obesity showed that the odds ratio of BMI ≥ 30.0 kg/m2 was higher in patients with critical and fatal illness than in those with a BMI of 25.0–29.9 kg/m2. Hence, the risk was increased in the BMI < 18.5 kg/m2 and BMI of ≥ 25.0 kg/m2 groups, resulting in a U-shaped curve for critical and fatal illness (Fig 1).

Fig 1. Odds ratios for critical and fatal illness according to body mass index.

Fig 1

Open in a new tab

Odds ratios were adjusted for age, gender, and five comorbidities (diabetes mellitus, hypertension, chronic kidney disease, cancer, and dementia).

Fig 2 shows the association between BMI and sex for critical and fatal illnesses. This was expressed as an odds ratio after adjusting for age and comorbidities. Fig 2 shows similar U-shaped curves in both males and females, although it seems to be slightly different in detail depending on sex. Women showed a higher risk in their BMI < 18.5 kg/m2 population than men. While, men showed a higher risk in their BMI ≥ 25.0 kg/m2 population than women (Fig 2A and 2B). However, the Pinteraction for among males and females were not significant.

Fig 2.

Fig 2

Open in a new tab

Odds ratios for critical and fatal illness of male and female according to body mass index (A) critical illness, (B) fatal illness Odds ratios were adjusted for age and five comorbidities (diabetes mellitus, hypertension, chronic kidney disease, cancer, and dementia).

Similar non-linear relationships are shown in Fig 3. This figure shows subgroup analyses of subjects with (A) DM or (B) hypertension. Each group showed a non-linear pattern with increasing BMI. In addition, there was no significant difference according to the presence or absence of the diseases (Pinteraction = 0.62 for DM, Pinteraction = 0.19 for hypertension)

Fig 3.

Fig 3

Open in a new tab

Subgroup analyses for fatal illness according to the BMI in patients with/without diabetes mellitus or hypertension (A) Diabetes Mellitus*, (B) Hypertension**. *Odds ratios were adjusted for age, gender, and four comorbidities (hypertension, chronic kidney disease, cancer, and dementia). **Odds ratios were adjusted for age, gender, and four comorbidities (diabetes mellitus, chronic kidney disease, cancer, and dementia).

Discussion

In this nationwide study, we showed that BMI ≥ 25.0 kg/m2 was associated with an increased risk for fatal illness as well as critical illness from COVID-19. Furthermore, BMI < 18.5 kg/m2 also increased the risk of critical and fatal illness caused by COVID-19, similar to that seen with BMI ≥ 25.0 kg/m2. The frequency data of S1 Fig shows that fatality decreased in those with BMI ≥ 30.0 kg/m2, and it appeared as if the obesity paradox existed. However, the odds ratio after adjusting for covariates showed a U-shaped curve, as shown in Fig 1. Hence, BMI ≥ 25.0 kg/m2 and < 18.5 kg/m2 were found to have an association with increased fatality from COVID-19 independent of combined comorbidities. Although the male sex showed a higher risk of fatal illness from COVID-19, there was no definite evidence that BMI had a more detrimental effect on men than on women in this study.

The results of the current study showed that maintaining a healthy weight in the general population, regardless of underlying comorbidities, is still important in the COVID-19 pandemic era. In this study, normal weight (18.5 < BMI <25 kg/m2) according to the WHO criteria, especially 23–25 kg/m2 in the South Korean population, is considered a healthy weight in the COVID-19 pandemic era.

Obesity, defined using BMI, has different cutoff values between the global and Asia-Pacific standards. The WHO announced the Asia-Pacific perspective: redefining obesity and its treatment’ in 2000 on the evidence that the Asian population has an increased risk of diseases at BMI > 23 kg/m2 [21]. The Asia-Pacific BMI was categorized as followings; underweight (< 18.5 kg/m2), normal-weight (18.5–22.9 kg/m2), overweight (23.0–24.9 kg/m2), obesity-1 (25.0–29.9 kg/m2), and obesity-2 (≥ 30 kg/m2). Proposed mechanisms of this inter-population discrepancy of defined obesity by BMI include differences in percentage and distribution of body fat across populations, the interplay of genetic susceptibility, and environmental factors related to diet and sedentary lifestyle [19, 20].

However, the debate on whether racial differences exist in defining obesity based on BMI is still ongoing. Despite the differences in BMI cutoff points for obesity in these populations, a meta-analysis of four continents in 239 prospective studies showed that BMI and all-cause mortality were broadly consistent across the four continents [3]. In 2004, WHO expert consultation recommended that the use of global standards is appropriate because the BMI standards for obesity do not differ greatly by race, and it is not appropriate to stipulate different standards for the Asia-Pacific region due to small differences [19]. In a study comparing the BMI cutoff value of the South Koreans with the global standard in 2015, the cutoff point of Korean BMI was 24.2 kg/m2 (sensitivity 78%, specificity 71%, as the result of receiver operating characteristic cure, based on the body fat percentage) and it was only 1.3 kg/m2 lower than the global standard [25]. The result supports the recommendation of the WHO expert consultation in 2004. Nonetheless, the Korean government continues to follow the 2000 Asian-Pacific cut-off values. Hence, to avoid confusion, we did not express obesity or normal for the BMI group, but only used numeric values in this study.

There are limited studies on the clinical implications of BMI, from underweight to obesity, in viral pandemics. In the case of influenza-associated pneumonia, an increase in risk has been observed in underweight and obesity [4]. However, regarding obesity, contradictory results–obesity paradox-were reported in previous studies [15]. Similarly, there was a recent study on ‘an obesity survival paradox’ that authors showed the projected rates of COVID-19 infection and mortality drop with elevated prevalence of obesity in United States [26]. However, in the current COVID-19 pandemic, the majority of studies reported obesity as a risk factor for progressing to severe COVID-19 and were associated with the need for hospitalization and admission to the critical care unite due to COVID-19 [7, 17, 27, 28]. Importantly, our study also showed increased BMI as a risk factor for critical and fatal illness of COVID-19 and cannot prove the ‘obesity survival paradox.’

In contrast to obesity, the relationship between underweight and COVID-19 is unknown. A recent study showed a J-shaped (non-linear) relationship between BMI and risk for COVID-19 related hospitalization, ICU admission, and death [28]. In our study, subjects with a BMI ≥ 30 kg/m2 presented higher fatality with an odds ratio of 4.32, compared to subjects of a 23 ≤BMI < 25 kg/m2. In addition, BMI < 18.5 kg/m2 also showed an odds ratio of 3.97 (1.77–8.92) for fatal illness.

One recent remarkable study is the analysis of the UK biobank, even though preliminary data are available [29, 30]. They showed that BMI was strongly associated with positive results in the COVID-19 test and the risk of death related to COVID-19 [29]. Interestingly, both BMI and waist circumference were associated with testing positive for COVID-19 in a dose-response fashion [30]. However, another study using a two-sample multivariable Mendelian randomization failed to show the impact of body composition (waist circumference, trunk fat ratio) on COVID-19 susceptibility and severity; only BMI was significant [31]. Our study has a limitation in that it did not include body composition.

In infectious diseases, research on whether there are gender differences in the effect of BMI on mortality is limited. In a UK study [32], with the national mortality data of 3.6 million adults, the mortality specific outcome related to respiratory infection, the hazard ratio (HR) of underweight increased similarly to obesity, showing a U-shaped curve, which is consistent with our study results. Further, the UK study showed the difference between men and women in all-cause mortality among never-smokers. Underweight women were associated with a greater increase in HR than men, while obese men were associated with a greater increase in HR than women (Pinteraction < 0.0001). In other words, underweight and obesity were associated with an increase in the HR of all-cause mortality in both men and women, but the increased risk was greater in underweight women and obese men. This is consistent with our results shown in Fig 2. However, there are some differences between the study populations and the target outcomes, since our study aimed at evaluating the association between the fatality of COVID-19 and BMI. Additionally, we did not have information regarding smoking history. Furthermore, the sample size in our study was relatively small, and when divided by subgroup analysis, a similar pattern was shown as the UK study between men and women, although the Pinteraction did not show a significant value. Therefore, it is necessary to verify this through large-scale research. A Chinese study of 383 patients found that obesity, especially in men, significantly increased the risk of developing severe COVID-19 [9]. In this study, we found that there was a slight difference in the odds ratio and increasing pattern in those with BMI of < 18.5 and ≥ 25.0 kg/m2 according to sex. Male sex was associated with an increased odds ratio of fatal illness (2.92 [1.9–4.48]) than the female sex. However, there was no definite evidence that men had a higher risk of severe COVID-19 due to BMI than women.

All-cause mortality due to obesity is attenuated by aging. Large-scale epidemiological studies have reported that heathy BMI shifts to the right in old age compared to young age [3, 32, 33]. Among the those with COVID-19, young obese patients (< 60 years old and BMI > 35 kg/m2) were reported to be 3.6 times more likely to be admitted to the critical care unit than those with BMI < 30 kg/m2 [34]. However, in our study, the incidence of severe COVID-19 in patients aged < 60 years was low (fatal illness was only 8 cases and critical illness was as small as 15 cases), so there was a limit to whether obesity was related to critical illness at a young age.

Obesity is regarded as an important public health issue, mainly related to chronic diseases. Obesity potentiates multiple cardiovascular risk factors and causes atherosclerosis. It also causes insulin resistance and reduces beta cell function [35]. It can cause functional immunological deficits through dysregulation of the immune system, resulting in diseases mainly related to cardiometabolic problems and chronic inflammation [2]. However, in the recent SARS-CoV2 pandemic, obesity was reported to be associated with severe COVID-19, and its clinical significance in acute infections may be receiving new attention from researchers. In addition to chronic metabolic conditions, proposed mechanisms for the detrimental effect of obesity in severe COVID-19 are related to the cardiorespiratory system, increased thrombogenicity, and immune hyper-reactivity [7, 36, 37]. Beyond these proposed mechanisms, COVID-19 can induce direct target organ damage, regardless of the underlying disease. Direct invasion or inflammation of cardiomyocyte is related to poor disease outcomes [18, 38, 39]. In the case of SARS, induced by SARS-CoV, the possibility of disruption of pancreatic beta cell function through pancreatic invasion through binding to the cellular entry, ACE-2, was confirmed in an autopsy report [40]. However, there is paucity of evidence in COVID-19 pancreatitis through direct beta cell involvement of SARS-CoV-2 [41]. Furthermore, there is no available data regarding whether obesity or underweight can augment the virus infection related to direct target organ damage by altering the susceptibility or ACE-2.

The present study has some limitations. Despite the advantage of nationwide data compared to previous studies related to COVID-19 and BMI, this study was limited to Koreans and did not include a diverse population. Although the fatality of BMI and COVID-19 was evaluated by adjusting for sex, age, and comorbid conditions, there was an issue in subgroup analysis because the sample was insufficient in terms of specific diseases. Furthermore, detailed information about comorbidities was not provided. The number of patients with BMI < 18.5, or ≥ 30 kg/m2, was small; 191 (4.6%) or 193 (4.7%), respectively. Therefore, there is a chance of being statistically underpowered or having an increased margin of error. Furthermore, we could not check the distribution of continuous BMI according to fatal illness status. The KCDA provided BMI data in a categorized form. Finally, it was an observational study and could not clearly explain causal relationships. Although the characteristics of acute infectious disease act as a limitation in the study of COVID-19, a large-scale prospective cohort study of various populations is of utmost important.

Conclusions

We analyzed the relationship between BMI and criticality/fatality from COVID-19. In particular, those with a BMI of < 18.5 kg/m2 and ≥ 25.0 kg/m2 were found to have a higher association with critical and fatal illness compared to those with a BMI of 23.0–24.9 kg/m2. Maintaining a healthy weight is important not only to prevent chronic cardiometabolic diseases, but also to improve the outcome of COVID-19.

Supporting information

S1 Table. Baseline characteristics of the study population according to disease severity.

(DOCX)

Click here for additional data file. (18.8KB, docx)
S1 Fig. Number of critical and fatal illnesses, and fatalities according to body mass index.

The left y-axis presents bar graphs showing the number of critical and fatal illnesses, and the right y-axis presents as a line graph showing fatality.

(TIF)

Click here for additional data file. (793.7KB, tif)

Acknowledgments

We acknowledge all health-care workers involved in the diagnosis and treatment of patients with COVID-19 in South Korea. We thank the Korea Disease Control & Prevention Agency, National Medical Center and the Health Information Manager in hospitals for their efforts in collecting the medical records.

Data Availability

Data cannot be shared publicly for protecting personal information by the Korea Disease Control and Prevention Agency (KDCA). Broad information regarding Korean COVID-19 statistics are released daily on a public web site (http://ncov.mohw.go.kr/en/) and through the media by KCDA. You may contact KDCA for the detailed data in current study.

Funding Statement

The authors received no specific funding for this work.

References

  • 1.Lim S, Shin SM, Nam GE, Jung CH, Koo BK. Proper Management of People with Obesity during the COVID-19 Pandemic. J Obes Metab Syndr. 2020;29(2):84–98. doi: 10.7570/jomes20056 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lavie CJ, Laddu D, Arena R, Ortega FB, Alpert MA, Kushner RF. Healthy Weight and Obesity Prevention: JACC Health Promotion Series. Journal of the American College of Cardiology. 2018;72(13):1506–31. doi: 10.1016/j.jacc.2018.08.1037 [DOI] [PubMed] [Google Scholar]
  • 3.Di Angelantonio E, Bhupathiraju SN, Wormser D, Gao P, Kaptoge S, de Gonzalez AB, et al. Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. The Lancet. 2016;388(10046):776–86. doi: 10.1016/S0140-6736(16)30175-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Dobner J, Kaser S. Body mass index and the risk of infection—from underweight to obesity. Clin Microbiol Infect. 2018;24(1):24–8. Epub 2017/02/25. doi: 10.1016/j.cmi.2017.02.013 . [DOI] [PubMed] [Google Scholar]
  • 5.Kong KA, Park J, Hong S-h, Hong YS, Sung Y-A, Lee H. Associations between body mass index and mortality or cardiovascular events in a general Korean population. PLOS ONE. 2017;12(9):e0185024. doi: 10.1371/journal.pone.0185024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wu C-Y, Chou Y-C, Huang N, Chou Y-J, Hu H-Y, Li C-P. Association of Body Mass Index with All-Cause and Cardiovascular Disease Mortality in the Elderly. PLOS ONE. 2014;9(7):e102589. doi: 10.1371/journal.pone.0102589 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Sattar N, McInnes IB, McMurray JJV. Obesity Is a Risk Factor for Severe COVID-19 Infection: Multiple Potential Mechanisms. Circulation. 2020;142(1):4–6. Epub 2020/04/23. doi: 10.1161/CIRCULATIONAHA.120.047659 . [DOI] [PubMed] [Google Scholar]
  • 8.Simonnet A, Chetboun M, Poissy J, Raverdy V, Noulette J, Duhamel A, et al. High Prevalence of Obesity in Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) Requiring Invasive Mechanical Ventilation. Obesity. 2020;28(7):1195–9. doi: 10.1002/oby.22831 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Cai Q, Chen F, Wang T, Luo F, Liu X, Wu Q, et al. Obesity and COVID-19 Severity in a Designated Hospital in Shenzhen, China. Diabetes Care. 2020:dc200576. doi: 10.2337/dc20-0576 [DOI] [PubMed] [Google Scholar]
  • 10.Li S, Hua X. Modifiable lifestyle factors and severe COVID-19 risk: a Mendelian randomisation study. BMC Medical Genomics. 2021;14(1):38. doi: 10.1186/s12920-021-00887-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Ponsford MJ, Gkatzionis A, Walker VM, Grant AJ, Wootton RE, Moore LSP, et al. Cardiometabolic Traits, Sepsis, and Severe COVID-19. Circulation. 2020;142(18):1791–3. doi: 10.1161/CIRCULATIONAHA.120.050753 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Badawi A, Ryoo SG. Prevalence of comorbidities in the Middle East respiratory syndrome coronavirus (MERS-CoV): a systematic review and meta-analysis. International Journal of Infectious Diseases. 2016;49:129–33. doi: 10.1016/j.ijid.2016.06.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Bangalore S, Sharma A, Slotwiner A, Yatskar L, Harari R, Shah B, et al. ST-Segment Elevation in Patients with Covid-19—A Case Series. New England Journal of Medicine. 2020;382(25):2478–80. doi: 10.1056/NEJMc2009020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Blumentals WA, Nevitt A, Peng MM, Toovey S. Body mass index and the incidence of influenza-associated pneumonia in a UK primary care cohort. Influenza Other Respir Viruses. 2012;6(1):28–36. Epub 2011/05/19. doi: 10.1111/j.1750-2659.2011.00262.x . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Nie W, Zhang Y, Jee SH, Jung KJ, Li B, Xiu Q. Obesity survival paradox in pneumonia: a meta-analysis. BMC Medicine. 2014;12(1):61. doi: 10.1186/1741-7015-12-61 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Domínguez-Cherit G, Lapinsky SE, Macias AE, Pinto R, Espinosa-Perez L, de la Torre A, et al. Critically Ill Patients With 2009 Influenza A(H1N1) in Mexico. JAMA. 2009;302(17):1880–7. doi: 10.1001/jama.2009.1536 [DOI] [PubMed] [Google Scholar]
  • 17.Peres KC, Riera R, Martimbianco ALC, Ward LS, Cunha LL. Body Mass Index and Prognosis of COVID-19 Infection. A Systematic Review. Frontiers in Endocrinology. 2020;11(562). doi: 10.3389/fendo.2020.00562 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Duan J, Wu Y, Liu C, Yang C, Yang L. Deleterious effects of viral pneumonia on cardiovascular system. European Heart Journal. 2020;41(19):1833–8. doi: 10.1093/eurheartj/ehaa325 [DOI] [PubMed] [Google Scholar]
  • 19.consultation* We. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. The Lancet. 2004;363(9403):157–63. doi: 10.1016/S0140-6736(03)15268-3 [DOI] [PubMed] [Google Scholar]
  • 20.Pan WH, Yeh WT. How to define obesity? Evidence-based multiple action points for public awareness, screening, and treatment: an extension of Asian-Pacific recommendations. Asia Pac J Clin Nutr. 2008;17(3):370–4. Epub 2008/09/27. . [PubMed] [Google Scholar]
  • 21.World Health Organization. Regional Office for the Western P. The Asia-Pacific perspective: redefining obesity and its treatment: Sydney: Health Communications Australia; 2000. [Google Scholar]
  • 22.Henry BM, Lippi G. Chronic kidney disease is associated with severe coronavirus disease 2019 (COVID-19) infection. International Urology and Nephrology. 2020;52(6):1193–4. doi: 10.1007/s11255-020-02451-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Marshall JC, Murthy S, Diaz J, Adhikari N, Angus DC, Arabi YM, et al. A minimal common outcome measure set for COVID-19 clinical research. The Lancet Infectious Diseases. 2020;20(8):e192–e7. doi: 10.1016/S1473-3099(20)30483-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Obesity WHOCo, World Health Organization. Division of Noncommunicable D, World Health Organization. Programme of Nutrition F, Reproductive H. Obesity: preventing and managing the global epidemic: report of a WHO Consultation on Obesity, Geneva, 3–5 June 1997. Geneva: World Health Organization; 1998. [PubMed]
  • 25.Yoon JL, Cho JJ, Park KM, Noh HM, Park YS. Diagnostic performance of body mass index using the Western Pacific Regional Office of World Health Organization reference standards for body fat percentage. J Korean Med Sci. 2015;30(2):162–6. Epub 2015/02/06. doi: 10.3346/jkms.2015.30.2.162 ; PubMed Central PMCID: PMC4310942. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Arbel Y, Fialkoff C, Kerner A, Kerner M. Can reduction in infection and mortality rates from coronavirus be explained by an obesity survival paradox? An analysis at the US statewide level. International Journal of Obesity. 2020;44(11):2339–42. doi: 10.1038/s41366-020-00680-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Anderson MR, Geleris J, Anderson DR, Zucker J, Nobel YR, Freedberg D, et al. Body Mass Index and Risk for Intubation or Death in SARS-CoV-2 Infection: A Retrospective Cohort Study. Ann Intern Med. 2020:M20–3214. doi: 10.7326/M20-3214 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Kompaniyets L, Goodman AB, Belay B, Freedman DS, Sucosky MS, Lange SJ, et al. Body Mass Index and Risk for COVID-19–Related Hospitalization, Intensive Care Unit Admission, Invasive Mechanical Ventilation, and Death—United States, March–December 2020. MMWR Morb Mortal Wkly Rep. 2021;70:355–61. doi: 10.15585/mmwr.mm7010e4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Sattar N, Ho FK, Gill JM, Ghouri N, Gray SR, Celis-Morales CA, et al. BMI and future risk for COVID-19 infection and death across sex, age and ethnicity: Preliminary findings from UK biobank. Diabetes Metab Syndr. 2020;14(5):1149–51. Epub 2020/07/16. doi: 10.1016/j.dsx.2020.06.060 ; PubMed Central PMCID: PMC7326434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Yates T, Razieh C, Zaccardi F, Davies MJ, Khunti K. Obesity and risk of COVID-19: analysis of UK biobank. Prim Care Diabetes. 2020;14(5):566–7. Epub 2020/06/05. doi: 10.1016/j.pcd.2020.05.011 ; PubMed Central PMCID: PMC7254007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Freuer D, Linseisen J, Meisinger C. Impact of body composition on COVID-19 susceptibility and severity: A two-sample multivariable Mendelian randomization study. Metabolism. 2021;118:154732. doi: 10.1016/j.metabol.2021.154732 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Bhaskaran K, dos-Santos-Silva I, Leon DA, Douglas IJ, Smeeth L. Association of BMI with overall and cause-specific mortality: a population-based cohort study of 3·6 million adults in the UK. The Lancet Diabetes & Endocrinology. 2018;6(12):944–53. doi: 10.1016/s2213-8587(18)30288-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Yi SW, Ohrr H, Shin SA, Yi JJ. Sex-age-specific association of body mass index with all-cause mortality among 12.8 million Korean adults: a prospective cohort study. Int J Epidemiol. 2015;44(5):1696–705. Epub 2015/07/26. doi: 10.1093/ije/dyv138 ; PubMed Central PMCID: PMC4681110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Lighter J, Phillips M, Hochman S, Sterling S, Johnson D, Francois F, et al. Obesity in Patients Younger Than 60 Years Is a Risk Factor for COVID-19 Hospital Admission. Clinical Infectious Diseases. 2020;71(15):896–7. doi: 10.1093/cid/ciaa415 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Cerf ME. Beta cell dysfunction and insulin resistance. Frontiers in endocrinology. 2013;4:37–. doi: 10.3389/fendo.2013.00037 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Kim I-C, Han S. Epicardial adipose tissue: fuel for COVID-19-induced cardiac injury? European Heart Journal. 2020;41(24):2334–5. doi: 10.1093/eurheartj/ehaa474 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Rebello CJ, Kirwan JP, Greenway FL. Obesity, the most common comorbidity in SARS-CoV-2: is leptin the link? International Journal of Obesity. 2020;44(9):1810–7. doi: 10.1038/s41366-020-0640-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Tao G, Guo T, Fan Y, Chen M, Wu X, Zhang L, et al. Cardiovascular Implications of Fatal Outcomes of Patients With Coronavirus Disease 2019 (COVID-19). JAMA Cardiology. 2020;5(7):811–8. doi: 10.1001/jamacardio.2020.1017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Tavazzi G, Pellegrini C, Maurelli M, Belliato M, Sciutti F, Bottazzi A, et al. Myocardial localization of coronavirus in COVID-19 cardiogenic shock. European Journal of Heart Failure. 2020;22(5):911–5. doi: 10.1002/ejhf.1828 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Ding Y, He L, Zhang Q, Huang Z, Che X, Hou J, et al. Organ distribution of severe acute respiratory syndrome (SARS) associated coronavirus (SARS-CoV) in SARS patients: implications for pathogenesis and virus transmission pathways. The Journal of Pathology. 2004;203(2):622–30. doi: 10.1002/path.1560 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Thaweerat W. Current evidence on pancreatic involvement in SARS-CoV-2 infection. Pancreatology. 2020;20(5):1013–4. Epub 2020/05/27. doi: 10.1016/j.pan.2020.05.015 . [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision Letter 0

Jie V Zhao

12 Mar 2021

PONE-D-21-03568

Body mass index and fatality from coronavirus disease 2019: A nationwide epidemiological study in Korea

PLOS ONE

Dear Dr. Kang,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by 11 May, 2021. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Jie V Zhao

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study investigated the relationship between BMI and COVID-19 fatal illness in a Korean sample. While this study is of quality, it can be improved in the following parts.

1. The introduction cited some studies about the relationship between BMI and severe COVID-19 illness. However, these studies are observational studies only and cannot provide high level evidence about the causation. There are several Mendelian randomisation studies on this relationship, such as Ponsford et al Circulation 2020 & Li et al BMC Medical Genomics 2021. These studies should be cited to give a more comprehensive review.

2. BMI is a continuous variable, but this study interestingly analysed it as a categorical variable. Transforming continuous variable to categorical would lose information. The study should provide the results for BMI as a continuous variable as well, and more importantly, to investigate if the U-shape still exists after using continuous BMI.

3. The study found a U-shape, which is interesting, especially for the underweight group. Table 1 shows that young people have the highest proportion of underweight people; however, young people have a low risk of fatal illness. So the results are a ‘paradox’. It’s good to report and check the distribution of continuous BMI by fatal illness status.

4. The results for the underweight group could be driven by the small size of this group (so the confidence interval is very wide). Conclusion based on this small number could be misleading. The authors should acknowledge this limitation and be cautious about their conclusion.

5. The authors interestingly used overweight group, rather than normal weight group, as the reference in their logistic regression. They justified this by stating that the overweight group had the lowest fatality rate. However, this is not a good justification, and the resulted ORs have little practical usage (because they are not compared with normal weight). Weight group should be the reference out of nature and intuition. The study should use the normal weight group as reference and report new results.

Reviewer #2: Please see below for my general comments on the Manuscript: Body mass index and fatality from coronavirus disease 2019: A nationwide epidemiological study in Korea. I have provided detailed feedback and questions regarding the methods, results, and interpretation of this analysis - both general comments/questions as well as line-specific. If my suggested analytical updates are made and thorough copy editing is completed, I believe this could add to the body of evidence around BMI and COVID-19 severity and fatality.

General comments/questions:

1. The manuscript has poor English grammar, and requires copy-editing for greater clarity.

2. The authors should doing a sensitivity analysis with the global BMI cut-points for underweight, normal weight, overweight and obesity. It would be interesting to see if there is a difference, given these Asia-Pacific population cut-points are relatively new and less widely used (which they discuss in the discussion but don't analyze)

3. The authors should provide definitions of each of the co-morbid conditions (self-report, taking medications), particularly for cancer – which cancers did they include?

Statistical Analysis: general comments/questions:

1.How did the authors account for the correlation between the various risk factors as well as potential mediating affects i.e. BMI as a mediator of diabetes)?

2.Why did the authors decide to categorize rather than use continuous variables (i.e. BMI, age as continuous variables)? Such multivariable analyses using continuous variables would add to the literature which typically categorizes BMI.

3.The authors report missing data in the Results (Table 1) but don’t explain how they handle missing data in the methods. Please clarify there.

4.Typically, the internal reference group for a cohort study should be the group with the least exposure (i.e. normal weight). While there is sufficient sample size to use the overweight as the reference, I am worried that using overweight instead of normal weight as the internal reference group makes the results confusing to interpret both within the study and in external comparison to other studies that use normal weight as reference. The OR for critical and fatal illness are both non-significant may be interpreted as a higher risk for critical and fatal illness unless very clearly discussed in the abstract, results and discussion. This could paint an impression that overweight is protective, when this may not be the case. For these two reasons, I’d suggest changing the reference group to normal weight. The authors provide some discussion on overweight BMI as a healthy in older adults. However, they do not do any further analysis to test this hypothesis. If they want to keep overweight as the reference category, an analysis of BMI - COVID19 severity/fatality by age (i.e. effect modification by age), would be helpful to determine if there is indeed this age effect.

Results general comments/questions:

1. At present the tables and figures do not provide sufficient information/footnotes to stand on their own for interpretation. For instance, in Table 2, the footnote needs to indicate that age is for every 10 years (categorical instead of continuous) and provide details on how critical and fatal illness are defined, how co-morbid conditions are defined, and what covariates were used in analysis

2. Figure 2: It is surprising to me that these two have the exact same p-interaction value (for female/male and critical illness, fatal illness) to the 0.0001 confidence level. I’d advise the authors double check this and confirm, especially considering how different the trends/figures look.

3. Include 95% CI in the narrative text in addition to the tables and figures

4. Improve interpretation sentences for odds ratios.

Discussion general comments/questions:

1. The authors include new results from the sub-group analyses in their discussion instead of summarizing and interpreting findings from their results.

2. The authors focus their results on the literature around BMI and all-cause mortality. This is somewhat related, but it would bet better to review literature more specific to BMI and infection-related fatality.

3. No mention of alternative exposure measures for overweight/obesity like waist circumference or waist to hip ratio. Even if not possible with this data set, it would be good to mention limitations of BMI as an exposure measure.

4. Need more discussion on small sample size for underweight and obese II and how miss affected the certainty in the odds ratio estimates.

I suggest the authors add into their discussion and intro this new U.S. BMI and COVID-19 severity and death, which found a similar non-linear relationship between BMI and COVID-19 severity/death (increased risk at underweight and obese level BMIs)

Kompaniyets L, Goodman AB, Belay B, et al. Body Mass Index and Risk for COVID-19–Related Hospitalization, Intensive Care Unit Admission, Invasive Mechanical Ventilation, and Death — United States, March–December 2020. MMWR Morb Mortal Wkly Rep. ePub: 8 March 2021. DOI: http://dx.doi.org/10.15585/mmwr.mm7010e4external icon.

**************

Line specific Comments:

Abstract:

32- 34: No mention of OR for normal weight in the abstract.

36-37: Conclusion explains underweight and obese have increased risk for critical/fatal illness than normal and overweight, but OR is only for overweight as reference.

Introduction:

45 : poor grammar – “as the increasing period of social distancing”, “more like to eat unhealthy food”

48-50: Its unclear why the author is mentioning U-shaped curve for BMI and all cause, CVD mortality, when the outcome here is COVID-19 related severe illness and fatality. Better to remove and cite literature on the association between BMI and flu-like virus severe illness and fatality (which you do a bit later).

51-57: There is now far more global data that can be cited regarding cohorts of COVID-19 severity and hospitalization with BMI as a risk factor. Consider citing others (particularly within Asia region for comparability)

58: clarify if cardiac disease means Cardiovascular disease (CVD)?

65-66 : bad English grammar

68: “associated with COVID-19…” and “obesity as a risk factor for COVID-19” Be more specific. Are you referring to COVID-19 infection or COVID-19 severe illness/fatality? I don’t know of any evidence suggesting obese have a higher risk for being INFECTED, though there is substantial data for SEVERE ILLNESS/FATALITY?

74-75: this line about Korean vs. Westerns having lower obesity and higher underweight prevalence belongs earlier in the introduction. Move and add more about the prevalence (cite national level stats for Korea)

Data Source and Study Population

83: “almost all confirmed cases” – what does this mean? Can you provide an estimate of the percentage of tracking currently occurring?

88-91: how has the exclusion of 1487 patients affected representativeness of the data? In particular, are those individuals without available BMI values systematically different than those with BM data? This could lead to serious selection bias. Include tables/data on representativeness of selected data vs. missing data.

Statistical Analysis

118: incorrect use of the term multivariate- this is a multivariable** logistic regression. Multivariate refresh to modeling of data that are often derived from longitudinal studies (repeated measures)m which this is not.

118: should be COVID-19 critical illness and COVID-19 fatal illness

127- 131: language around the stratified analyses is unclear at present. It sounds like the authors did an overall multivariable logistic regression, controlling for all potential confounders and then did stratified analyses by sex and each co-morbid condition separately. Add in language around stratified analysis for greater clarity.

129-131: this last statement is also confusing. It is unclear from what the authors are saying whether, when stratifying by each comorbidity, they are calculating the odds ratios only for patients who do vs. do NOT have the target disease?

Table 1

- Include percentage of total population for each BMI (i.e. 1741 is 42% of total patients) as you have written this in your narrative results.

- Is the p-value for the statistically significant differences within a given covariate (age group, sex, etc.) across BMI categories? Or is it a p-value for statistically significant differences within a BMI category across covariates (across age groups, sexes, etc.). For the results text it seems to be the former, but please clarify in methods and in table footnotes what the p-value refers to

161: start new paragraph when start talking about the association/odds ratio results. “Fatal illness was significantly related to the combined comorbidity” – unclear what this means as written.

175 -176: unclear what the authors mean by “did not change the aspect of underweight”. This is not common statistical language.

Figure 2: It is surprising to me that these two have the exact same p-interaction value (for female/male and critical illness, fatal illness) to the 0.0001 confidence level. I’d advise the authors double check this and confirm, especially considering how different the trends/figures look.

Discussion:

203-209: This review of sub-group analyses should be moved to the results section. The authors should only include a short summary of this analysis and its implications in the discussion section.

223-224: It is not clear what the authors mean by “contradictory results were reported in obesity according to other studies”. I suggest they make this clear.

225-244: Remove or reduce the detailed comparison to UK literature on BMI and overall mortality (non-specific to infection). A mention of the overall association between BMI and mortality is ok, but not a detail comparison as the authors here are specifically looking at COVID-19 severe illness and fatality. Better to compare to literature on similar viral infections or COVID19 studies.

245-255: same is true here. The discussion should be focused more on literature about the actual exposures and outcomes of interest in this paper, not a more general topic like all cause mortality.

256: “The BMI value for healthy weight varied with age” – is this a finding of the analysis or something gleaned from the literature. As written, it is confusing whether this is a general statement or a specific finding.

258-259: This is an interesting statement. Did the authors do any analysis to confirm whether there was effect modification by age for the OR in overweight vs. normal weight in the study. Would be necessary to confirm this with data.

266-268: poor English grammar, consider revising.

266-284: discussion around Asian specific BMI cut points – did the authors consider doing a sensitivity analysis with the general WHO BMI cut-points? This would be a helpful comparison to see how much these differential BMI cut-points made a difference in the reported odds ratios.

279: typo – should be 25.0-29.9 kg/m2, not 2.5-29.9 kg/m2

286-287: mention of how obesity potentiates multiple cardiovascular risk factors, but they did not discuss how they addressed this in their methods. See general statistical analysis question above

315: the authors looked at the relationships between BMI and critical and fatal illness from COVID-19 (not just fatality from COVID-19). Revise how this is written.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Jun 22;16(6):e0253640. doi: 10.1371/journal.pone.0253640.r002

Author response to Decision Letter 0

26 May 2021

We deeply appreciate the Reviewer for his/her time and input. We have carefully considered the reviewer’s comments and have addressed them point-by-point.

We attached this as a file.

Attachment

Submitted filename: 1. response to reviewers_ver 4.2.docx

Click here for additional data file. (272.7KB, docx)

Decision Letter 1

Jie V Zhao

10 Jun 2021

Body mass index and severity/fatality from coronavirus disease 2019: A nationwide epidemiological study in Korea

PONE-D-21-03568R1

Dear Dr. Kang,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jie V Zhao

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Acceptance letter

Jie V Zhao

14 Jun 2021

PONE-D-21-03568R1

Body mass index and severity/fatality from coronavirus disease 2019: A nationwide epidemiological study in Korea

Dear Dr. Kang:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jie V Zhao

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Table. Baseline characteristics of the study population according to disease severity.

    (DOCX)

    Click here for additional data file. (18.8KB, docx)
    S1 Fig. Number of critical and fatal illnesses, and fatalities according to body mass index.

    The left y-axis presents bar graphs showing the number of critical and fatal illnesses, and the right y-axis presents as a line graph showing fatality.

    (TIF)

    Click here for additional data file. (793.7KB, tif)
    Attachment

    Submitted filename: 1. response to reviewers_ver 4.2.docx

    Click here for additional data file. (272.7KB, docx)

    Data Availability Statement

    Data cannot be shared publicly for protecting personal information by the Korea Disease Control and Prevention Agency (KDCA). Broad information regarding Korean COVID-19 statistics are released daily on a public web site (http://ncov.mohw.go.kr/en/) and through the media by KCDA. You may contact KDCA for the detailed data in current study.

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES