Sex-Differences in COVID-19 Diagnosis, Risk Factors and Disease Comorbidities

doi:10.21203/rs.3.rs-1744101/v1

Introduction: Morbidity and mortality from COVID-19 are greater for males, however, the mechanisms for this difference are unclear. We aim to investigate sex-gender differences in COVID-19 in a large US-based cohort, namely COVID-19 Research Database. Specifically, the objectives of the present study are to explore the socio-economic characteristics of COVID-19 male and female patients and to examine potential sex differences in lifestyle factors and disease comorbidities among diagnosed patients.

Methods: This is a retrospective cohort study comparing males versus females with test-confirmed COVID-19 between January 2020 and December 2021 (N=62,310). The study used secondary data, called Healthjump, which were extracted from patients’ medical claims and electronic health records (e.g., demographics, encounters, medical history, procedures, social history, and vitals).

Results: Significant socio-demographic and comorbidity differences were observed between male and female patients (p<0.05). For example, a significantly higher proportion of males (vs. females) were aged ≥70-year-old (17.04% vs. 15.01%) and smokers (11.04 %vs. 9.24%, p<0.0001). In addition, multiple logistic regression showed that hypertension and diabetes were significantly more frequent in males (AOR=66.19 and AOR=22.90).

Conclusions: Understanding the differences in outcomes between males and females will inform gender equity responsive policy approaches to COVID-19 and improve the effectiveness of health interventions.

Sex differences

COVID-19

Healthjump data

SARS-CoV-2

comorbidities

Since the first appearance of the novel coronavirus disease of 2019 (COVID-19) caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in Wuhan, Hubei Province in China, it has rapidly spread worldwide and has been classified as a pandemic by the World Health Organization on March 11, 2020. As of May 2022, there have been 524,339,768 confirmed cases of COVID-19 including 6,281,260 deaths, globally ¹. The burden of disease has been greatest in the United States (US) with the most number of cases and deaths globally, affecting more than 82 million individuals, and causing more than 900,000 deaths ¹.

Evidence increasingly indicates that male sex is a risk factor for more severe disease and death from COVID-19. Male bias in COVID-19 mortality is observed in nearly all countries with available sex-disaggregated data and the risk of death in males is approximately 1.7 times higher than in females ². The heterogeneity of sex disparities over the course of the pandemic suggests a number of theoretically grounded hypotheses about the role of gender-related social factors including gender-linked health behaviors and occupational exposures, which intersect with other socially-salient variables such as race/ethnicity and socioeconomic status ³.

Recent studies suggest that there are sex differences in the clinical outcomes of COVID-19 ^4–7. Emerging evidence suggests sex-based or gendered differences potentially due to immunological factors ^8–10. Some mechanisms underline the influence of hormonal factors ¹¹, differential distribution of the angiotensin converting enzyme 2 (ACE-2) receptors ⁹, smoking habits ¹², among others ^12–14.

Evidence shows an early indication of sex-specific patterns in diseases across the globe. Based on sex-disaggregated data from the Global Health 50/50 investigators as of May 20, 2021, the infection fatality rate (IFR) in males vs. females showed higher fatality rates among men than among women and it was also higher in men in Brazil, Yemen, Mexico, Ecuador, Scotland, Peru, Guatemala, North Macedonia, and Afghanistan. Overall, men had significant higher odds of death from COVID-19 in 49 countries, compared to women ¹⁵. The latest data from the US show that as of April 9, 2022, 55% of the deceased patients from COVID-19 are men and 45% are women. Important sex and gender differences in the incidence and outcomes of patients with COVID-19 has been a growing concern in research.

Research studies showed that males had higher rates of mortality, hospitalization, and clinical complications from COVID-19 compared to females ^16–18. For instance, male sex was independently associated with in-hospital mortality of COVID-19 patients in China ⁴. A multicenter retrospective cohort study compared males versus females with COVID-19 infections from March 1, 2020, to June 21, 2020, in the Rush University System, Chicago, USA, males were found to have significant higher rates of hospitalization, ICU transfer, vasopressor support, and endotracheal intubation. A multivariable model adjusting for age and disease comorbidities showed a significant association between male sex and mortality in the total cohort of US patients¹⁶.

Using a large US-based cohort, we aim in this research to investigate sex-differences in COVID-19. More specifically, the objectives of the present study are 1) to describe temporal trends in COVID-19 prevalence over time and to summarize age-and sex-distribution of cases among male and female patients, 2) to explore the demographic and socio-economic characteristics of COVID-19 male and female patients, and 3) to examine potential sex differences in lifestyle behaviors, risk factors and disease comorbidities among diagnosed patients. Findings from the present study will help inform governments and global health institutions to consider the sex and gender effects of the COVID-19 outbreak, both direct and indirect, and conduct an analysis of the gendered impacts of the multiple outbreaks, incorporating the voices of women on the front line of the response to COVID-19 and of those most affected by the disease within preparedness and response policies or practices going forward. Sex-disaggregated data will help clinicians to make appropriate patient-tailored medical decisions.

Study design and data source

The present study is a retrospective cohort study comparing males versus females with COVID-19 diagnosis. This study used the patients’ electronic health records (EHRs) registered with the Healthjump database (Healthjump Inc., Philadelphia, PA, USA) and provided by the COVID-19 Research Database consortium (https://covid19researchdatabase.org). All personal information in this database was provided anonymously. Data were extracted by SQL using Snowflake (Snowflake Inc., San Mateo, CA, USA). Data were retrieved from all available departments, including inpatient hospital, outpatient hospital and emergency room, for virtually all types of provided services in each participating hospital. The study sample includes all patients in the source population with test-confirmed COVID-19 (polymerase chain reaction [PCR] + as well as IgG/IgM+) from January 2020 to December 2021. The study is in accordance with relevant guidelines and regulations.

The Healthjump Database

The Healthjump, available through the COVID19 Research Database, extracts the EMR data and includes demographics, appointments, encounters, charges, transactions, medical history, medications, diagnosis, procedures, allergies, immunizations, labs, provider, social history, and vitals. For the present research, we focus on appointments, encounters, medical history, diagnosis, procedures, immunizations, reason for visit, social history, and vitals, all specifically for COVID-19 care. The Healthjump EMR sample we used contains information from inpatient physicians, urgent care and emergency room visits including reason for visits, procedures performed, and laboratory test. A demographic file includes the patient’s date of birth, race, sex, ethnicity, state and the 3-digit zip code of residence.

Data Analysis

Regarding data analysis, the research team selected key variables to assess sociodemographic and lifestyle factors among COVID-19 patients. Data on social history (e.g., education, ethnicity, language), demographic (e.g., age, sex), appointment, immunization (type of vaccine), vitals (e.g., oxygen saturation, BP etc.) and diagnoses (e.g., hypertension, diabetes etc.) were all extracted from the Healthjump. Data was all exported to the statistical software STATA for conducting data analysis and performing some descriptive statistics. Frequency tables were generated to display the information regarding disease comorbidities, symptoms, and other categorical variables. Chi-square tests were used to test for possible statistically significant differences in sociodemographic (e.g., age, education, ethnicity) and lifestyle (smoking, alcohol, BMI) between males and females. Simple logistic regression was conducted to examine sex-differences in demographic and social characteristics, laboratory parameters, vaccination, comorbidities/risk factors/pre-existing conditions as well as primary reason for visit (ICD10) among diagnosed patients. Variables that were found to be significantly associated by sex were all added to the final logistic model. Multiple logistic regression was conducted to examine sex-differences in socioeconomic characteristics, lifestyle factors and disease comorbidities (e.g., obesity, smoking, influenza) among diagnosed patients. Results from the logistic regression models were expressed as adjusted odds ratio (AOR) with 95% CI. All reported p-values were based on two-sided tests and were compared with a significance level of 5%.

Figure 1 displays the overall temporal distribution of daily counts of COVID-19 cases for both male and female patients within the study period. Figure 1 shows for both genders that the temporal evolution of the number of daily COVID-19 cases has followed an increasing trend and reached a daily maximum at the end of December 2020, which was then followed by a slow decrease and then by a rapid decline until the end of June 2021, where confirmed cases went close to near-zero levels. After then, the number of diagnosed patients markedly increase reaching another peak early September 2021. By the end of the study period, the proportion of COVID-19 cases decreased to few cases in December 2021. Figure 2. shows an age-dependent increase in reported cases in both males and females, being patients aged 50–59 years the most affected with cases of 3,628 in men and 6,418 in women.

Figure 3. presents the number of COVID-19 cases among patients with diseases comorbidities by age and sex. A significantly higher proportion of males aged 20–29 years, 60–69 years and > 70 years had diabetes compared to females in the same age group (66.7% vs. 33.33%, 50.30% vs. 49.70% and 50.37% vs. 49.63%, p < 0.05, see Fig. 3a). A higher percentage of females aged 30–39 years, 40–49 years and 50–59 years had diabetes compared to males, however, this difference was not statistically significant (64.58% vs. 35.42%,65.57% vs. 34.43%, and 59.29% vs. 40.71%, p > 0.05). There is an age-dependent increase in metabolic disorders among female patients aged 40–49 years (44.05%), 50–59 years (46.82%), 60–69 years (48.73%) (Fig. 3b). On the other hand, the percentage of males with metabolic disorders decreased by age (55.95%, 53.18% ,51.27% for males aged 40–49 years, 50–59 years, and 60–69 years, respectively. A significantly higher proportion of males aged 40–49 years, 50–59 years, 60–69 years and > 70 years had metabolic disorder compared to females (55.95% vs. 44.05% ,53.18% vs. 46.82%, 51.27% vs. 48.73%, and 52.78% vs. 47.2215, p < 0.05, see Fig. 3b). However, more females aged 30–39 years had metabolic disorders compared to males aged 30–39 years (51.8515 vs. 48.15%, p < 0.05). As shown in Fig. 3c, significantly more females had hypertension within 40–49 years old age group (56.99% vs. 43.01%). No significant differences were observed in abnormal clinical and lab findings between male and female patients by age groups (p > 0.05, see Fig. 3d).

Table 1 presents demographic and social characteristics differences between male and female individuals tested for SARS-CoV-2, 2020–2021. A total of 62,310 patients with a confirmed diagnosis of COVID-19 were identified. Overall, 13% of COVID-19 patients in our sample were below 20 years, 9.39% 20–29 years, 12.32% were 30–39 years, 15.46% were 40–49 years, 17.66 were 50–59 years, 16.33% were 60–69 years and 15.85% were above 70 years (Table 1). Slightly more than half of study sample were white (54.76%) and non-Hispanic (53.08%). With respect to education, 19.63% of study sample had graduate or post-graduate degree, 15.52% had general education or college, 64.16% had high school or below and remaining unknown (0.74%). Majority had transportation (99.40%) and 11.50% lived with family. Significant sex-differences were found in demographic and social characteristics of individuals tested for SARS-CoV-2, 2020–2021 (p < 0.05, Table 1). Males (vs. females) had significantly higher proportion in the 60-69-year-old interval (17.35% vs. 15.60%), and > 70- years (17.04% vs. 15.01%) and predominantly white (56.10% vs. 53.81%, χ2 = 132.2041, p < 0.0001). Consequently, among males (vs. females) there was a greater proportion of individuals of Hispanic ethnicity (23.03% vs. 22.17%, χ2 = 9.5205, p < 0.0001). Slightly higher percentage of male patients had better education level with graduate or post-graduate degree (20.23% vs. 19.31%, χ2 = 22.9419, p < 0.0001).

Table 1

Demographic and Social Characteristics differences between male and female individuals tested for SARS-CoV-2, 2020–2021 (N = 62,310)
	Total	Female	Male	Significance	Simple logistic regression	Multiple logistic regression
Demographic and Social Characteristics, n (%)					OR (95%CI)	OR (95%CI)
Age				χ2 = 505.0737, p < 0.0001
< 20 y	8,099(13.0)	4,160(11.40)	3,939(15.26)		1.0	1.0
20–29	5,852(9.39)	3,756(10.29)	2,096(8.12)		0.59(0.55,0.63)	0.22(0.006,8.11)
30–39	7,674(12.32)	4,991(13.68)	2,683(10.40)		0.57(0.53,0.61)	1.17(0.05,26.33)
40–49	9,629(15.46)	6,001(16.44)	3,628(14.06)		0.64(0.60,0.68)	2.07(0.09,49.32)
50–59	11,003(17.66)	6,418(17.59)	4,585(17.77)		0.75(0.71,0.80)	1.15(0.05,25.90)
60–69	10,171(16.33)	5,694(15.60)	4,477 (17.35)		0.83(0.78,0.88)	3.51(0.15,81.68)
> 70	9,875(15.85)	5,477(15.01)	4,398(17.04)		0.85(0.80,0.90)	3.00(0.11,79.10)
Race				χ2 = 132.2041, p < 0.0001
Black or African American	6,389(10.30)	4,162(11.46)	2,224(8.66)		1.0	1.0
Asian	618(1.00)	342(0.94)	276(1.07)		1.51(1.28,1.78)	-
White	33,947(54.76)	19,541(53.81)	14,406(56.10)		1.40(1.31,1.46)	2.19(0.25,18.91)
Unknown	21,038(33.94)	12,267(33.78)	8,771(34.16)		1.34(1.26,1.42)	1.05(0.13,8.70)
Ethnicity				χ2 = 9.5205, p < 0.0001
Not Hispanic or Latino	32,492(53.08)	19,214(53.57)	13,278(52.37)		1.0	1.0
Hispanic or Latino	13,791(22.53)	7,953(22.17)	5,838(23.03)		1.06(1.02,1.11)	0.35(0.09,1.38)
Unknown	14,936(24.40)	8,700(24.26)	6,236(24.60)		0.98(0.93,1.02)	0.66(0.23,1.89)
Education				χ2 = 22.9419, p < 0.0001
Graduate/ Post-graduate degree	640(19.63)	412(19.31)	228(20.23)		1.0	1.0
High school or below	2,091(64.13)	1,334(62.51)	757(67.17)		1.03(0.85,1.23)	0.57(0.06,5.79)
General Education/College	506(15.52)	376(17.62)	130(11.54)		0.62(0.48,0.81)	0.12(0.01,1.40)
Unknown	24(0.74)	12(0.56)	12(1.06)		1.81(0.80,4.10)	-
Transportation				χ2 = 0.0522, p = 0.819
No	4(0.60)	2(0.54)	2(0.68)		1.0
Yes	659(99.40)	367(99.46)	292(99.32)		0.80(0.11,5.68)
Living Arrangement				χ2 = 6.1572, p = 0.409
Alone	510(3.8)	294(3.74)	216(3.88)		1.0
Family	1,544(11.50)	915(11.63)	629(11.31)		0.94(0.76,1.15)
Institution	23(0.17)	14(0.18)	9(0.16)		0.88(0.37,2.06)
Friend/Roommate	103(0.77)	65(0.83)	38(0.68)		0.79(0.51,1.23)
Relative	57(0.42)	26(0.33)	31(0.56)		1.62(0.94,2.81)
Spouse	931(6.93)	558(7.09)	373(6.71)		0.91(0.73,1.13)
Unknown	10,258(76.40)	5,993(76.20)	4,265(76.69)		0.97(0.81,1.16)

As shown in Table 2, most of COVID-19 patients had normal oxygen saturation (91.98%) and 8.01% with hypoxemia. In terms of vaccination, few patients had Moderna, US, Inc. (4.34%), Pfizer-BioNTech (2.82%), or other vaccines (e.g., AstraZeneca Pharmaceuticals LP, Novavax, Janssen Products, LP.) (1.29%). Majority of the study sample (90.33%) had routine vaccinations (e.g., chickenpox (Varicella), Hepatitis A, Hepatitis B, Human Papilloma Virus (HPV), Influenza, Measles, mumps, Rubella, and Meningococcal etc.) (Table 2). With regards to comorbidities, 76.53% of COVID-19 patients in the present study were caffeine users, 36.26% do not exercise, 25.67% were alcohol users, 9.98% were smokers, and 5.72% were drug users. More than half of study participants were obese (54.62%), 27.96% were overweight and 15.87% had normal weight. Significant sex-differences were found in laboratory, vaccination, and comorbidities of individuals tested for SARS-CoV-2 (p < 0.05, Table 2). Slightly greater proportion of male patients had mild hypoxemia (9.31% vs. 7.12%, χ2 = 42.9096, p < 0.0001). In terms of risk factors, a higher proportion of males were smokers (11.04%vs. 9.24%, p < 0.0001), caffeine users (77.44% vs 75.97%, p = 0.045), alcohol users (30.72% vs. 22.36%, p < 0.0001) and drug users (6.60%vs. 5.17%, p < 0.0001) compared to females. A higher percentage of females had normal weight (12.48% vs. 18.14%) whereas a higher percentage of male patients were overweight (30.96% vs. 25.95%) or obese (55.13 vs. 54.29%, χ2 = 298.4379, p < 0.0001). No significant sex- differences were observed in transportation (p = 0.819), living arrangement (p = 0.409), exercise (p = 0.814), vaccine (p = 0.334).

Table 2

Laboratory, vaccination, and comorbidities differences between male and female individuals tested for SARS-CoV-2, 2020–2021 (N = 62,310)
	Total	Female	Male	Significance	Simple logistic regression	Multiple logistic regression
Laboratory, vaccination, and comorbidities, n (%)					OR (95%CI)	OR (95%CI)
Laboratory
Oxygen Saturation				χ2 = 42.9096, p < 0.0001
Normal	23,433(91.98)	13,936(92.88)	9,497(90.70)		1.0	1.0
Hypoxemia (Mild, Moderate, Severe)	2,043(8.01)	1,069(7.12)	974(9.31)		1.32(1.21,1.45)	2.21(0.32,4.50)
Vaccination
CVX _code				χ2 = 4.5764, p = 0.334
Other vaccines (e.g., AstraZeneca Pharmaceuticals LP, Novavax, Janssen Products, LP.)	700(1.29)	389(1.22)	314(1.39)		1.0
Moderna, US, Inc.	2,359(4.34)	1,370(4.33)	989(4.36)		0.89(0.75,1.05)
Pfizer-BioNTech	1,532(2.82)	879(2.78)	653(2.88)		0.91(0.76,1.09)
Routine vaccinations (e.g., Hepatitis A, Hepatitis B, Influenza, etc.)	49,078(90.33)	28,633(90.42)	20,445(90.21)		0.88(0.76,1.02)
Unknown	660(1.21)	398(1.26)	262(1.16)		0.81(0.65,1.004)
Comorbidities/Risk Factors/Pre-existing conditions, n (%)
Smoking status				χ2 = 123.3616, p < 0.0001
Non-smoker	17,174(38.56)	10,639(40.58)	6,535(35.66)		1.0	1.0
Smoker	4,445(9.98)	2,422(9.24)	2,023(11.04)		1.36(1.27,1.45)	1.44(0.46,4.54)
Unknown	22,923(51.46)	13,154(50.18)	9,769(53.30)		1.21(1.16,1.26)	2.88(0.70,11.88)
Body Mass Index (BMI) Status				χ2 = 298.4379 P < 0.0001
Normal	6,846(15.87)	4,684(18.14)	2,162(12.48)		1.0	1.0
Overweight	12,066(27.96)	6,703(25.95)	5,363(30.96)		1.70(1.60,1.80)	1.55(0.34,7.08)
Obese	23,570(54.62)	14,021(54.29)	9,549(55.13)		1.44(1.37,1.52)	0.63(0.16,2.53)
Caffeine user				χ2 = 4.0035, p = 0.045
No	3,271(23.47)	2,066(24.03)	1,205(22.56)		1.0
Yes	10,667(76.53)	6,530(75.97)	4,137(77.44)		1.09(1.00,1.18)
Drug user				χ2 = 16.8293, p < 0.0001
No	17,723(94.28)	10,970(94.83)	6,753(93.40)		1.0	1.0
Yes	1,075(5.72)	598(5.17)	477(6.60)		1.30(1.14,1.47)	1.30(0.27,66.23)
Alcohol user				χ2 = 255.8570, p < 0.0001
No	21,694(74.33)	13,676(77.64)	8,018(69.28)		1.0	1.0
Yes	7,493(25.67)	3,938(22.36)	3,555(30.72)		1.54(1.46,1.62)	4.89(1.81,13.23)
Exercise
No	487(36.26)	317(37.60)	170(34.00)		1.0
Yes	856(63.74)	526(62.40)	330(66.00)	χ2 = 1.7636, p = 0.814	1.17(0.93,1.58)

In terms of primary reason for visit, and according to the 10th revision of the International Classification of Disease (ICD-10) (Table 3), 14.29% were primary diagnosed for factors influencing health status and contact with health services (e.g., persons encountering health services for examinations, genetic susceptibility to disease) (n = 3,972), 9.34% for abnormal clinical and lab findings (n = 2,597), 7.41% for diabetes mellitus (n = 2,059), 5.23% for metabolic disorders (n = 1,453), 4.99% for COVID-19 (1,388), 4.16% for Hypertensive diseases (n = 1,156), 3.97% for certain infectious and parasitic diseases (e.g., HIV, TB, etc.) (n = 1,103), 3.61% for diseases of thyroid gland (n = 1,004), 2.50% for anxiety, associative, stress-related and other nonpsychotic mental disorders (n = 696), 2.47% for overweight, obesity and other hyperalimentation (n = 687), 2.29% for diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism (n = 637), 1.46% for injury, poisoning and other external causes (n = 405), 0.85% for mental disorders (e.g., disorders of adult personality and behavior, intellectual disabilities)(n = 237), and 0.76% for influenza and pneumonia (n = 212).

Table 3

Primary reason for visit (International Classification of Diseases 10th Revision (ICD-10)) differences between male and female individuals tested for SARS-CoV-2, 2020–2021 (N = 62,310)
	Total	Female	Male	Significance	Simple logistic regression	Multiple logistic regression
					OR (95%CI)	OR (95%CI)
Primary Reason for Visit (ICD10)				χ2 = 600.9711, p = 0.017
COVID-19	1,388(4.99)	775(4.69)	613(5.44)		2.65(2.16,3.25)	2.23(0.06,86.36)
Abnormal clinical and lab findings	2,597(9.34)	1,501(9.08)	1,096(9.73)		2.45(2.02,2.96)	13.82(1.19,159.92)
Mental, Behavioral and Neurodevelopmental disorders
Anxiety, dissociative, stress-related, and other nonpsychotic mental disorders	696(2.50)	477(2.88)	219(1.94)		1.54(1.21,1.95)	5.51(0.03,902.75)
Mental disorders (e.g., disorders of adult personality and behavior, intellectual disabilities)	237(0.85)	112(0.68)	125(1.11)		3.74(2.74,5.09)	89.72(3.34,2413.21)
Diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism	637(2.29)	432(2.61)	205(1.82)		1.59(1.25,2.02)	4.42(0.29,68.07)
Diseases of the circulatory system
Hypertensive diseases	1,156(4.16)	660(3.99)	496(4.40)		2.52(2.04,3.11)	22.90(2.17,241.09)
Endocrine
Diabetes Mellitus	2,059(7.41)	1,066(6.45)	993(8.82)		3.12(2.57, 3.79)	19.97(1.96,203.84)
Metabolic disorders	1,453(5.23)	716(4.33)	737(6.54)		3.45(2.82,4.22)	4.65(0.29,73.54)
Disorders of thyroid gland	1,004(3.61)	723(4.37)	281(2.49)		1.30(1.04,1.63)	21.40(0.81,2871079)
Overweight, obesity and other hyperalimentation	687(2.47)	409(2.47)	409(2.47)		2.28(1.80,2.87)	23.51(1.26,439.98)
Certain infectious and parasitic diseases (e.g., HIV, TB, etc.)	1,103(3.97)	674(4.08)	429(3.81)		0.94(0.84,1.05)	1.12(0.05,26.75)
Diseases of the respiratory system
Influenza and pneumonia	212(0.76)	108(0.65)	104(0.92)		1.16(0.91,1.40)	66.19(1.02,4,288.9)
Factors influencing health status and contact with health services (e.g., persons encountering health services for examinations, genetic susceptibility to disease)	3,972(14.29)	2,412(14.58)	1,560(13.85)		2.17(1.80,2.61)	3.55(0.,36.9734)
Injury, poisoning, and other external causes	405(1.46)	231(1.40)	174(1.54)		2.52(1.94,3.28)

Significant sex-differences were found in primary reason for visit (International Classification of Diseases 10th Revision (ICD-10)) of individuals tested for SARS-CoV-2 (p < 0.05, Table 3). A higher proportion of male patients had abnormal clinical and lab findings (9.73% vs. 9.08%), hypertensive diseases (4.40% vs. 3.99%) and diabetes (8.82% vs. 6.45%) compared to female patients (χ2 = 600.9711, p = 0.017, Table 3). Whereas a higher proportion of female patients had factors influencing health status and contact with health services (e.g., persons encountering health services for examinations, genetic susceptibility to disease) (14.58% vs. 13.85%), diseases of thyroid gland (4.37% vs. 2.49%) and anxiety, dissociative, stress-related, and other nonpsychotic mental disorders (2.88% vs. 1.94%) (p = 0.017).

Simple logistic regression showed significant sex-differences for age, race, ethnicity, education, laboratory parameters, smoking status, BMI status, caffeine user, drug user, alcohol user, primary reason for visit (except certain infectious and parasitic diseases) (Tables 1–3). For example, a greater proportion of males identifying with Asian race (OR = 1.51; 95% CI: 1.28,1.78), White race (OR = 1.40; 95% CI:1.31,1.46) and Hispanic or Latino ethnicity (OR = 1.06; 95% CI:1.02,1.11) compared to females (Table 1). As compared to females, a lower proportion of males had general or college education (OR = 0.62; 95% CI: 0.48,0.81). Hypoxemia was 32% more likely among male patients in comparison to female patients (OR = 1.32; 95% CI:1.21,1.45, see Table 2). Male patients had a significantly higher likelihood of smoking as compared to females (OR = 1.36; 95% CI:1.27,1.45). Male COVID-19 patients were 40% more likely to be obese and 70% more likely to be overweight compared to females. In addition, males had significantly higher risk of drug and alcohol use (OR = 1.30; 95% CI:1.14,1.47 and OR = 1.54; 95% CI:1.46,1.62, Table 2). Furthermore, males had significantly higher odds of diseases and related health problems such as abnormal clinical and lab findings (OR = 2.45; 95% CI:2.02,2.96), hypertensive diseases (OR = 2.52; 95% CI :2.04,3.11) and metabolic disorders (OR = 3.45; 95% CI:2.82,4.22) (Table 3).

Findings from multiple logistic regression showed sex-differences in COVID-19 for alcohol use and primary reason for visit (ICD10). Abnormal clinical and lab findings were significantly more frequent in males, with associated ratios of 13.82 (95% CI: 1.19,159.92) (Table 3). Influenza and pneumonia were more likely among male patients in comparison to female patients (OR = 66.19; 95% CI:1.02,288.9). Men significantly suffered more from mental disorders (e.g., disorders of adult personality and behavior, intellectual disabilities) than women (OR = 89.72; 95% CI:3.34,24113.21, Table 3). Male COVID-19 patients showed high frequency of underlying comorbidities including hypertensive diseases (OR = 22.90; 95% CI: 2.17,241.09) and diabetes (OR = 66.19; 95% CI:1.02,4228.9), even after adjusting for significant covariates such as age, education, and ethnicity.

Using a big data approach and from a population perspective, we have identified important sex-dependent differences in the risk factors and disease comorbidities associated with COVID-19. In particular, male COVID-19 patients showed high frequency of underlying comorbidities including abnormal clinical and lab findings, hypertensive diseases, diabetes, whilst adjusting for significant covariates such as age, education, and ethnicity.

Our findings are in line with studies of COVID-19 in US, Europe and China that found that the disease disproportionally affects men and women. Initial data show that men suffer from more severe disease and have higher mortality than women ^19–21. After adjusting for several key socio-demographic, clinical, and hospital course variables, results from a US-based cohort study of male and female patients showed a clear and strong independent association between male sex and higher COVID-19 susceptibility, greater likelihood of ICU admission, utilization of mechanical ventilation and longer length of stay-all clinical indicators of higher severity of the COVID-19 disease ²². A recent meta-analysis of 229 studies covering over 10M patients, men had a higher risk for infection with COVID-19 than women, and when infected, they also had a higher risk for hospitalization, higher risk for severe COVID-19, higher need for Intensive Care, and higher risk of death ²³.

In our study population, a greater proportion of male COVID-19 patients were alcohol users and had multiple comorbidities such as diabetes, hypertension, and metabolic disorders in the adjusted model. The presence of comorbidities increases the risk of adverse outcomes to COVID-19, and males are overrepresented compared to females in patients with the common comorbidities for COVID-19. Hypertension is often listed as the most common comorbidity in patients hospitalized with COVID-19, and a study has shown that males under the age of 65 have higher levels of hypertension than females²⁴.

Sex disparities in severity and mortality were also attributed to a higher rate of risky-behaviors and higher presence of comorbidities (i.e., cardiovascular disease, hypertension, diabetes, and chronic lung disease) in men ^4,19. Men are more involved in various risky- behaviours, such as smoking and alcohol consumption. ^4,19,25. Smoking has been linked to adverse outcomes of COVID. For example, smokers were 1.4 times more likely than non-smokers to experience severe COVID-19 symptoms, according to the combined results of five studies ^5,26. Research has shown that smoking increases the expression of the SARS-CoV-2 receptor, ACE2, in the lungs, which increases the possible sites of viral entry, which could explain why this subset of patients has a higher prevalence of COVID-19 ²⁷. Patterns from the United States, Italy, and China—the most affected nations— show that men have higher rates of smoking than women ^25,28. This trend is depicted globally, as well, which can provide additional rationale for the gendered differences in COVID-19 outcomes. Furthermore, the above-mentioned behavioral factors, such as smoking and alcohol intake, predispose men to comorbid conditions including respiratory disease, hypertension, and cardiovascular disease, all of which are risk factors for death ²⁹. This could also explain why men have a greater overall death rate ^25,30,31.

Sex differences are also intertwined with differences in gender roles socially, which also influence COVID-19 incidence and outcomes. Men are commonly working in basic sectors and occupations that require them to continue being active as well as to interact with other people even during the containment phase (e.g., food or pharmacy manufacturing and sales, agriculture or food production and distribution, transportation, and security). As a result, most males do not stay at home and instead gather with others, removing their masks to drink and smoke. This increased level of exposure predisposes men to a high risk of contracting COVID19. For example, men account for 76% of agricultural workers and for 96% of construction, maintenance, and repair workers in the U.S ³². U.S. women hold the majority of administrative, secretarial and teaching roles that were likely restructured for remote working during the pandemic, although women are also more likely to provide paid and unpaid cross-household domestic and care-giving labor that increases risk of exposure ^32,33. Studies have also found women and girls more likely to report mask wearing, hand washing, and compliance with other public health and social distancing recommendations ^3,34. Furthermore, there exist some social norms that discourage men from seeking treatment or physician consultation. As a result, men often delay seeking treatment, which can increase the probability of adverse outcomes after COVID-19 infection ²⁴.

In addition, there are also possible biological mechanisms of male sex bias that affect the severity of COVID-19, particularly with respect to immune responses. Men and women are also known to respond differently to foreign and self-antigens, and sex differences in the immune response are well-documented ^35,36. Possible explanation for the observed sex biases included that male patient had higher plasma levels of innate immune cytokines such as IL-8 and IL-18 along with more robust induction of non-classical monocytes ^35,36. By contrast, female patients had more robust T cell activation than male patients during SARS-CoV-2 infection³⁶. Notably, it was found that a poor T cell response negatively correlated with patients’ age and was associated with worse disease outcome in male patients, but not in female patients ³⁵. Additionally, studies have shown that the production of Nitric oxide (NO) is enhanced by estrogen via estrogen receptor mediated transcriptional upregulation of endothelial nitric oxide synthase (eNOS) ³⁷. The impact of estrogen in females on NO may provide a further layer of protection against COVID-19, and the role of NO as virus replication inhibitor can explain why females tend to experience less severe outcomes of COVID-19 infection ³⁸.Also, plasma concentration of testosterone, which decreases with age, is reported to be reduced by some comorbidities, such as diabetes, obesity, and obstructive sleep apnea^39,40. Evidence has shown that these comorbidities are common in patients with COVID-19 ³⁹. In our study, the higher cases observed in males compared to females may also be due to higher number of male patients being diabetic, obese and had hypertension, especially older age males (> 50 years).

Age could also partially explain the stark differences in risk of COVID-19 reported in the present study. Males (vs. females) patients in our study sample were significantly older with higher proportion aged 60–69, and > 70- years. Previous studies reveal the mortality and fatality rates increased with age and are predominant in men 50 years of age or older ^5,15. The vast majority of the deaths in COVID-19 occurred in patients with advanced age (> 50years), and the gender-dependent risk of poor outcomes after COVID-19 was more pronounced with advancing age, the risk of mortality in patients with advanced age was higher compared to a matched group of females of similar age ¹⁹.

On the contrary, only few studies showed that females were at a higher risk for developing long-term post-COVID symptoms including anxiety, depression, or poor sleep quality than males ^41,42. It is also possible that other factors such as higher psychological stress could play a role in the development of post-COVID symptoms. It is possible that COVID-19 outbreak surrounding factors such as depression, sleep deprivation, isolation, stress, inactivity, could also promote the development of more post-COVID symptoms in females.

Strength and limitations

This is the first study to examine sex-differences in COVID-19, underlying risk factors and health conditions in a large and consistent sample covering U.S. population. The main strengths of our proposed research include immediacy, big data approach, and direct access to real-world evidence. Additionally, our methodology ensures absence of any bias in patient selection. There are some limitations to this study which should be noted. Case ascertainment is subject to sensitivity and specificity of the employed PCR testing platforms, a small proportion of individuals who were tested multiple times could have been misclassified based on approach of using first encounter information. Furthermore, our research focused on evaluating the comorbidity that may underlie observed sex differences. Although strong evidence in favor of the role of biological pathways has been presented, further research is also needed to investigate how socio-behavioral factors might affect health outcomes.

In conclusion, sex-based differences exist in high-risk behavior and comorbidities among a large, US-based cohort. This gender-specific risk is especially more pronounced in advanced age. This study suggests that more attention should be paid to sex as a variable for the interpretation of COVID-19 data. Sex-disaggregated data will help clinicians to make appropriate patient-tailored medical decisions. Identifying the gender differences in COVID-19 outbreak would be a fundamental tool to understand the effects of a health emergency on individuals and communities as well as to carry out effective and equitable policies, public health measures and targeted solutions. Further research is needed to understand the causes of this disparity and may be beneficial for decisions in both clinical and policy-making realms.

Acknowledgments: The authors would like to thank the Health Care Cost Institute, Datavant and the COVID-19 Research Database Grant Accelerator Program, Bill & Melinda Gates Foundation, for funding this study. The authors would also like to thank Ms. Alex Muir and Ms. Aleah Peffer for their time and effort in resolving technical issues and other related issues in data management and analysis of the Healthjump COVID-19 Research Database. The database and analysis environment used for the study were provided by the COVID-19 Research Database consortium (https://covid19researchdatabase.org).

Data availability: The data that support the findings of this study are available from Healthjump database and the COVID-19 Research Database consortium but restrictions apply to the availability of these data, which were used under licence for the current study, and so are not publicly available. Data are however available from the leading author upon reasonable request and with permission of Healthjump database and the COVID-19 Research Database consortium.

Author contributions: SK participated in the conceptualization of the idea, investigation and Project administration. SK & MD contributed to the data curation, formal analysis, methodology, software, validation, visualization, manuscript drafting and the final review of the manuscript. All authors have read and approved the final manuscript.

Funding: The present study is funded by the Health Care Cost Institute, Datavant and the COVID-19 Research Database Grant Accelerator Program, Bill & Melinda Gates Foundation (Grant number 104136).

Competing interests: The authors declare no competing interests.

WHO. WHO Coronavirus (COVID-19) Dashboard < https://covid19.who.int/> (2022).
Scully, E. P., Haverfield, J., Ursin, R. L., Tannenbaum, C. & Klein, S. L. Considering how biological sex impacts immune responses and COVID-19 outcomes. Nature Reviews Immunology 20, 442–447 (2020).
Danielsen, A. C. et al. Sex disparities in COVID-19 outcomes in the United States: Quantifying and contextualizing variation. Social Science & Medicine, 114716 (2022).
Sha, J. et al. Sex differences on clinical characteristics, severity, and mortality in adult patients with COVID-19: A multicentre retrospective study. Frontiers in Medicine 8, 123 (2021).
Abate, B. B., Kassie, A. M., Kassaw, M. W., Aragie, T. G. & Masresha, S. A. Sex difference in coronavirus disease (COVID-19): a systematic review and meta-analysis. BMJ open 10, e040129 (2020).
Ya'qoub, L., Elgendy, I. Y. & Pepine, C. J. Sex and gender differences in COVID-19: More to be learned! American Heart Journal Plus: Cardiology Research and Practice 3, 100011 (2021).
Jun, T. et al. Analysis of sex-specific risk factors and clinical outcomes in COVID-19. Communications Medicine 1, 1–8 (2021).
Chen, N. et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. 395, 507–513 (2020).
Takahashi, T. et al. Sex differences in immune responses to SARS-CoV-2 that underlie disease outcomes. (2020).
Arnold, C. G., Libby, A., Vest, A., Hopkinson, A. & Monte, A. A. Immune mechanisms associated with sex-based differences in severe COVID-19 clinical outcomes. Biology of sex Differences 13, 1–5 (2022).
Grandi, G., Facchinetti, F., Bitzer, J. J. T. E. J. o. C. & Care, R. H. The gendered impact of coronavirus disease (COVID-19): do estrogens play a role? 25, 233–234 (2020).
Leung, J. M. et al. ACE-2 expression in the small airway epithelia of smokers and COPD patients: implications for COVID-19. 55 (2020).
Liu, S. et al. Prevalence and patterns of tobacco smoking among Chinese adult men and women: findings of the 2010 national smoking survey. 71, 154–161 (2017).
Gargaglioni, L. H. & Marques, D. A. J. J. o. A. P. Let’s talk about sex in the context of COVID-19. 128, 1533–1538 (2020).
Ramírez-Soto, M. C., Ortega-Cáceres, G. & Arroyo-Hernández, H. Sex differences in covid-19 fatality rate and risk of death: An analysis in 73 countries, 2020–2021. Infezioni in Medicina, 402–407 (2021).
Gomez, J. M. D. et al. Sex Differences in COVID-19 Hospitalization and Mortality in Chicagoland. 142, A17393-A17393 (2020).
Ancochea, J., Izquierdo, J. L., Group, S. C.-R. & Soriano, J. B. J. J. o. W. s. H. Evidence of gender differences in the diagnosis and management of coronavirus disease 2019 patients: An analysis of electronic health records using natural language processing and machine learning. 30, 393–404 (2021).
Mo, P. et al. Clinical characteristics of refractory COVID-19 pneumonia in Wuhan, China. (2020).
Singh, S., Chowdhry, M., Chatterjee, A. & Khan, A. Gender-Based Disparities in COVID-19 Patient Outcomes: A Propensity-matched Analysis. medRxiv, 2020.2004.2024.20079046 (2020). https://doi.org:10.1101/2020.04.24.20079046
Guan, W.-j. et al. Clinical Characteristics of Coronavirus Disease 2019 in China. New England Journal of Medicine 382, 1708–1720 (2020). https://doi.org:10.1056/NEJMoa2002032
Jin, J. M. et al. Gender Differences in Patients With COVID-19: Focus on Severity and Mortality. Front Public Health 8, 152 (2020). https://doi.org:10.3389/fpubh.2020.00152
Vahidy, F. S. et al. Sex differences in susceptibility, severity, and outcomes of coronavirus disease 2019: Cross-sectional analysis from a diverse US metropolitan area. PloS one 16, e0245556 (2021).
Pijls, B. G. et al. Temporal trends of sex differences for COVID-19 infection, hospitalisation, severe disease, intensive care unit (ICU) admission and death: a meta-analysis of 229 studies covering over 10M patients. F1000Research 11, 5 (2022).
Gebhard, C., Regitz-Zagrosek, V., Neuhauser, H. K., Morgan, R. & Klein, S. L. Impact of sex and gender on COVID-19 outcomes in Europe. Biology of sex differences 11, 29–29 (2020). https://doi.org:10.1186/s13293-020-00304-9
Cai, H. Sex difference and smoking predisposition in patients with COVID-19. The Lancet Respiratory Medicine 8, e20 (2020).
Vardavas, C. I. & Nikitara, K. COVID-19 and smoking: A systematic review of the evidence. Tobacco induced diseases 18 (2020).
Zou, X. et al. Single-cell RNA-seq data analysis on the receptor ACE2 expression reveals the potential risk of different human organs vulnerable to 2019-nCoV infection. Frontiers of medicine 14, 185–192 (2020).
Walter, L. A. & McGregor, A. J. Sex- and Gender-specific Observations and Implications for COVID-19. The western journal of emergency medicine 21, 507–509 (2020). https://doi.org:10.5811/westjem.2020.4.47536
Yang, J. et al. Prevalence of comorbidities in the novel Wuhan coronavirus (COVID-19) infection: a systematic review and meta-analysis. Int J Infect Dis 10 (2020).
Granö, N., Virtanen, M., Vahtera, J., Elovainio, M. & Kivimäki, M. Impulsivity as a predictor of smoking and alcohol consumption. Personality and individual differences 37, 1693–1700 (2004).
Wong, D. R., Willett, W. C. & Rimm, E. B. Smoking, hypertension, alcohol consumption, and risk of abdominal aortic aneurysm in men. American journal of epidemiology 165, 838–845 (2007).
U.S. Bureau of Labor Statistics. Labor Force Statistics from the Current Population Survey., < https://www.bls.gov/cps/cpsaat11.htm> (2020).
Robertson, C. & Gebeloff, R. How millions of women became the most essential workers in America. The New York Times 18 (2020).
Chen, J. T. & Krieger, N. Revealing the unequal burden of COVID-19 by income, race/ethnicity, and household crowding: US county versus zip code analyses. Journal of Public Health Management and Practice 27, S43-S56 (2021).
Wilkinson, N. M., Chen, H.-C., Lechner, M. G. & Su, M. A. Sex Differences in Immunity. Annual review of immunology 40 (2022).
Park, M. D. Sex differences in immune responses in COVID-19. Nature Reviews Immunology 20, 461–461 (2020). https://doi.org:10.1038/s41577-020-0378-2
MacRitchie, A. N. et al. Estrogen Upregulates Endothelial Nitric Oxide Synthase Gene Expression in Fetal Pulmonary Artery Endothelium. Circulation research 81, 355–362 (1997). https://doi.org:10.1161/01.RES.81.3.355
Mukherjee, S. & Pahan, K. Is COVID-19 Gender-sensitive? Journal of neuroimmune pharmacology 16, 38–47 (2021). https://doi.org:10.1007/s11481-020-09974-z
Dana, P. M. et al. An insight into the sex differences in COVID-19 patients: what are the possible causes? Prehospital and disaster medicine 35, 438–441 (2020).
Bhasin, S. et al. Testosterone therapy in men with hypogonadism: an Endocrine Society clinical practice guideline. The Journal of Clinical Endocrinology & Metabolism 103, 1715–1744 (2018).
Iqbal, F. M. et al. Characteristics and predictors of acute and chronic post-COVID syndrome: A systematic review and meta-analysis. EClinicalMedicine 36, 100899 (2021).
Fernández-de-Las-Peñas, C. et al. Female Sex Is a Risk Factor Associated with Long-Term Post-COVID Related-Symptoms but Not with COVID-19 Symptoms: The LONG-COVID-EXP-CM Multicenter Study. Journal of Clinical Medicine 11, 413 (2022).

No competing interests reported.

Sex-Differences in COVID-19 Diagnosis, Risk Factors and Disease Comorbidities

Status:

Version 1

Abstract

Figures

Introduction

Methods

Study design and data source

The Healthjump Database

Data Analysis

Results

Discussion

Strength and limitations

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1