In December 2019, a cluster of acute respiratory illness caused by a novel coronavirus named severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) occurred in Wuhan, China. 1 , 2 Epidemiological and clinical characteristics, risk factors for mortality of patients infected with SARS‐CoV‐2, and risk factors in the susceptibility to SARS‐CoV‐2 included age and chronic disease have been reported. 3 , 4 , 5 , 6 However, the use of biological markers to predict the susceptibility to SARS‐CoV‐2 has not been well described. So far, only one study has reported that ABO blood groups were associated with the susceptibility to SARS‐CoV‐2· 7 In the present study, after eliminating other confounding risk factors (including age, gender and comorbidities), we further investigated and confirmed the association of ABO blood groups and risk of SARS‐CoV‐2 pneumonia in patients from the Central Hospital of Wuhan, as well as two hospitals in Wuhan, China. Patients diagnosed with SARS‐CoV‐2 who died or were discharged between February 1 and March 25, 2020, were included in this retrospective cohort study. The study was approved by the Ethics Committee of the Central Hospital of Wuhan, and the need for informed consent was waived. 8 Epidemiological information, clinical data, underlying comorbidities, CT images of lungs, laboratory findings and clinical outcomes were extracted from electronic medical records. The blood group distribution data of the other two hospitals (Wuhan Jinyintan Hospital and Renmin Hospital of Wuhan University) and healthy controls in Wuhan came from the paper published online. 7 Data were expressed as percentages (%). We used chi‐squared tests or Fisher's exact tests in order to compare the various groups. The ABO blood group in 265 patients infected with SARS‐CoV‐2 from the Central Hospital of Wuhan showed a distribution of 39·3 %, 25·3 %, 9·8 % and 25·7 % for A, B, AB and O, respectively (Table I). The proportion of blood group A in patients infected with SARS‐CoV‐2 was significantly higher than that in healthy controls (39·3 % vs. 32·3 %, P = 0·017), 7 while the proportion of blood group O in patients infected with SARS‐CoV‐2 was significantly lower than that in healthy controls (25·7 % vs. 33·8 %, P < 0·01). Table I The ABO blood group distribution in patients infected with SARS‐CoV‐2 and healthy controls in Wuhan. Blood Group A B AB O Controls (Wuhan Area, n = 3694), % 1188 (32·3 %) 920 (24·9 %) 336 (9·1 %) 1250 (33·8 %) Central Hospital of Wuhan (n = 265), % 104 (39·3) 67 (25·3) 26 (9·8) 68 (25·7) χ 2 5·645 0·019 0·152 7·447 P 0·017 0·891 0·696 < 0·01 Age distribution (n = 265), % Less than 40 years (n = 69) 24 (34·8) 17 (24·6) 8 (11·6) 20 (29·0) χ 2 0·213 0·003 0·509 0·714 P 0·644 0·959 0·476 0·398 Between 41–59 years (n = 79) 29 (36·7) 20 (25·3) 8 (10·1) 22 (27·9) χ 2 0·732 0·007 0·099 1·242 P 0·392 0·933 0·753 0·265 Over 60 years (n = 117) 51 (43·6) 30 (25·6) 10 (8·6) 26 (22·2) χ 2 6·752 0·033 0·041 6·871 P < 0·01 0·856 0·839 < 0·01 Gender distribution (n = 265), % Male (n = 113) 48 (42·5) 30 (26·6) 9 (8·0) 26 (23·0) χ 2 5·323 0·158 0·170 5·771 P 0·021 0·691 0·680 0·016 Female (n = 152) 56 (36·8) 37 (24·3) 17 (11·2) 42 (27·6) χ 2 1·462 0·025 0·764 2·521 P 0·227 0·875 0·382 0·112 Chronic disease, % Cerebrovascular disease (n = 55) 19 (34·6) 15 (27·3) 6 (10·9) 15 (27·3) χ 2 0·141 0·162 0·215 1·045 P 0·707 0·687 0·643 0·307 Coronary heart disease (n = 51) 18 (35·3) 14 (27·5) 7 (13·7) 12 (23·5) χ 2 0·226 0·174 1·296 2·393 P 0·634 0·676 0·255 0·122 Heart failure (n = 16) 2 (12·5) 6 (37·5) 1 (6·3) 7 (43·8) χ 2 2·826 1·349 0·000 0·699 P 0·093 0·245 1·000 0·403 Hypertension (n = 115) 48 (41·7) 26 (22·6) 10 (8·7) 31 (27·0) χ 2 4·668 0·315 0·022 2·367 P 0·031 0·575 0·883 0·124 Diabetes (n = 66) 26 (39·4) 19 (28·8) 4 (6·1) 17 (25·8) χ 2 1·552 0·522 0·726 1·895 P 0·213 0·470 0·394 0·169 Digestive disorder (n = 90) 33 (36·7) 26 (28·9) 7 (7·8) 23 (25·6) χ 2 0·816 0·744 0·185 2·700 P 0·366 0·389 0·667 0·100 COPD (n = 11) 4 (36·4) 4 (36·4) 1 (9·1) 2 (18·2) χ 2 0·089 0·769 0·000 0·604 P 0·766 0·380 1·000 0·437 Solid tumour (n = 27) 13 (48·2) 8 (29·6) 1 (3·7) 5 (18·5) χ 2 3·134 0·320 0·405 2·815 P 0·077 0·572 0·525 0·093 Chronic renal disease (n = 41) 15 (36·6) 12 (29·3) 2 (4·9) 12 (29·3) χ 2 0·364 0·708 0·439 0·379 P 0·546 0·400 0·508 0·538 Hepatitis (n = 7) 6 (85·7) 1 (14·3) 0 (0) 0 (0) χ 2 6·883 0·422 0·032 2·224 P < 0·01 0·516 0·858 0·136 Deaths (n = 57), % 20 (35·1) 15 (26·3) 8 (14·0) 14 (24·6) χ 2 0·220 0·060 1·644 2·162 P 0·639 0·807 0·200 0·141 COPD, chronic obstructive pulmonary disease. John Wiley & Sons, Ltd This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency. We next investigated whether age, gender and chronic disease influence the ABO blood group distribution (Table I). The results showed that, among blood group A (43·6 % vs. 32·2 % in controls, P < 0·01) and blood group O (22·2 % vs. 33·8 % in controls, P < 0·01), patients over 60 years of age were consistent with all the above patients. Similarly, we also found that A (42·5 % vs. 32·2 %, P = 0·021) and O (23·0 % vs. 33·8 %, P = 0·016) distribution of blood groups in male patients was consistent with all the above patients. In all chronic diseases, we found that the proportion of hypertension (41·7 % vs. 32·2 %, P = 0·031) and hepatitis (85·7 % vs. 32·2 %, P < 0·01) in blood group A was much higher than that in the control group; however, there is currently no literature supporting that hypertension and hepatitis increase the risk of infection of SARS‐CoV‐2. In dead patients, we found no differences between blood types. Finally, we integrated the data of the three hospitals in Wuhan for analysis (Table II). 7 We still find that the proportion of blood group A in patients infected with SARS‐CoV‐2 was significantly higher than that in healthy controls (38·0 % vs. 32·2 %, P < 0·001), while the proportion of blood group O in SARS‐CoV‐2 infected patients was significantly lower than in healthy controls (25·7 % vs. 33·8 %, P < 0·001). The distribution ratio of blood type A and O between various ages and genders was almost consistent with the trend of all patients. Table II The ABO blood group distribution in patients infected with SARS‐CoV‐2 from three Wuhan hospitals. Blood Group A B AB O Controls (Wuhan Area, n = 3694), % 1188 (32·2) 920 (24·9) 336 (9·1) 1250 (33·8) Three Wuhan Hospitals (n = 2153), % 819 (38·0) 561 (26·1) 219 (10·2) 554 (25·7) χ 2 20·859 0·953 1·833 36·445 P <0·001 0·329 0·176 <0·001 Age distribution (n = 2153), % Less than 40 years (n = 342) 124 (36·3) 95 (27·8) 29 (8·5) 94 (27·5) χ 2 2·395 1·372 0·145 5·688 P 0·122 0·241 0·704 0·017 Between 41–59 years (n = 784) 304 (38·8) 196 (25·0) 79 (10·1) 205 (26·2) χ 2 12·739 0·003 0·740 17·439 P <0·001 0·956 0·390 <0·001 Over 60 years (n = 1027) 391 (38·1) 270 (26·3) 111 (10·8) 255 (24·8) χ 2 12·617 0·818 2·749 30·034 P <0·001 0·366 0·097 <0·001 Gender distribution (n = 2153), % Male (n = 1143) 451 (39·5) 305 (26·7) 110 (9·6) 277 (24·2) χ 2 20·749 1·461 0·291 37·271 P <0·001 0·227 0·590 <0·001 Female (n = 1010) 368 (36·4) 256 (25·4) 109 (10·8) 277 (27·4) χ 2 6·549 0·082 2·664 14·878 P 0·010 0·774 0·103 <0·001 Three Wuhan hospitals: the Central Hospital of Wuhan, Wuhan Jinyintan Hospital and Renmin Hospital of Wuhan University. John Wiley & Sons, Ltd This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency. In this study, we demonstrated that blood group A patients were at higher risk of hospitalization following SARS‐CoV‐2 infection, while blood group O patients had lower risk, which suggested that the ABO blood type could be used as a biomarker to predict the risk of SARS‐CoV‐2 infection. Coincidentally, previous studies found that ABO blood type distribution also had significant differences in other viral infections. Chen et al. reported that blood group O individuals were less likely to become infected by SARS coronavirus, 9 Batool et al. found that blood group O might have some influence in protecting against blood‐transmitted infection, and people having blood group A were more prone to contract hepatitis B and HIV. 10 Jing et al. found that blood group B was associated with a lower risk of HBV infection. 11 Guillon et al. reported that the S protein/angiotensin‐converting enzyme 2‐dependent adhesion of these cells to an angiotensin‐converting enzyme 2 expressing cell line was specifically inhibited by human natural anti‐A antibodies, which might block the interaction between the virus and its receptor. 12 This could explain why blood group A is susceptible, while blood group O is not. However, there may be other factors that need further study. In summary, based on our research, and confirmed by reported data, people with blood group A had a significantly higher risk of SARS‐CoV‐2 infection, whereas blood group O had a significantly lower risk of SARS‐CoV‐2 infection. People with blood type A should strengthen protection to reduce the risk of infection; however, people with blood type O should not take the virus lightly, and must still take precautions to avoid increasing the risk of infection. The underlying molecular mechanism of our findings will need further study. Funding information This study was supported by the Health and Family Planning Commission of Wuhan City (WX18M02). Conflict of interest No reports. Authors' contribution Conceived and designed the experiments: J.L., M.Y. and A.D. Performed the experiments: J.L., X.W. and A.D. Analysed the data: J.L., X.W., J.C. and A.D. Wrote the paper: J.L. J.L. and X.W. contributed equally.