Gender Bias in Diagnosis, Prevention, and Treatment of Cardiovascular Diseases: A Systematic Review

Cardiovascular disease (CVDs) has been perceived as a ‘man’s disease’, and this impacted women’s referral to CVD diagnosis and treatment. This study systematically reviewed the evidence regarding gender bias in the diagnosis, prevention, and treatment of CVDs. Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines were followed. We searched CINAHL, PubMed, Medline, Web of Science, British Nursing Index, Scopus, and Google Scholar. The included studies were assessed for quality using risk bias tools. Data extracted from the included studies were exported into Statistical Product and Service Solutions (SPSS, v26; IBM SPSS Statistics for Windows, Armonk, NY), where descriptive statistics were applied. A total of 19 studies were analysed. CVDs were less reported among women who either showed milder symptoms than men or had their symptoms misdiagnosed as gastrointestinal or anxiety-related symptoms. Hence, women had their risk factors under-considered by physicians (especially by male physicians). Subsequently, women were offered fewer diagnostic tests, such as coronary angiography, ergometry, electrocardiogram (ECG), and cardiac enzymes, and were referred to less to cardiologists and/or hospitalisation. Furthermore, if hospitalised, women were less likely to receive a coronary intervention. Similarly, women were prescribed cardiovascular medicines than men, with the exception of antihypertensive and anti-anginal medicines. When it comes to the perception of CVD, women considered themselves at lower risk of CVDs than men. This systematic review showed that women were offered fewer diagnostic tests for CVDs and medicines than men and that in turn influenced their disease outcomes. This could be attributed to the inadequate knowledge regarding the differences in manifestations among both genders.


Introduction And Background
Cardiovascular diseases (CVDs) represent the leading cause of mortality worldwide accounting for 17.9 million deaths each year [1].CVDs refer to several disorders that affect the heart and blood vessels including coronary heart disease (CHD) also known as ischaemic heart disease (IHD), strokes and transient ischaemic attack (TIA or mini stroke), peripheral arterial disease (PAD) and aortic disease [2].
CVDs have been commonly perceived as 'men's disease' and this misconception has contributed to underdiagnosis and treatment for women worldwide [3].In comparison to men, women are 50% more likely to be misdiagnosed with a heart attack even though they carry the same risk of developing CVDs as men [4].It has been known for over two decades that women experience CVDs differently to men [5,6]; yet both genders are still considered the same by healthcare professionals despite the presence of gender-specific requirements in many guidelines [7][8][9][10][11][12][13].Moreover, mortality linked to CVDs is higher globally in women [14].
Prevention of CVDs is mainly associated with modifying the risk factors including body mass index (BMI), smoking, high blood pressure, high blood cholesterol, lack of physical activity and diabetes [4].Some of these risk factors such as smoking and obesity have been shown to have a greater impact on women than men [15,16].The risk of CVDs is more exponential in women with a sudden increase in risk once they reach around 60 years of age [17,18].
Diagnosis of CVDs is not straightforward in women due to the different and delayed presentation of symptoms when compared to men.Women with acute myocardial infarction (MI) are reported to present with atypical symptoms including abdominal pain, dyspnoea, nausea, back and neck pain, indigestion, palpitations and unexplained fatigue; as opposed to a well-defined chest pain, which is the typical men presentation and often better recognised by doctors [19,20].Furthermore, women are less likely to report their symptoms and know the risks of CVDs [21].Lack of symptom reporting in turn lead to delayed/underdiagnosis and subsequent treatment and that negatively affect the clinical outcomes [17,22].The misconception that CVDs represent men's diseases has led women to believe that they are at a lower risk of CVDs.
The current prevention, diagnosis and treatment approaches of CVDs adopted by physicians failed to consider the two genders (men and women) as physiologically different.These biases have therefore contributed to the lack of awareness of CVDs risk in women contributing to their delayed diagnosis and poor clinical outcomes.Furthermore, the different manifestation of CVDs in men and women together with the limited representation of women in the existing evidence suggest that the current evidence is largely representative of the CVDs in men.This systematic review, therefore, aimed to examine the current evidence related to genderbias in the diagnosis, prevention and treatment of CVDs.

Search Strategy
We based our study on the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) (Appendix, Supplementary file S1) .We searched CINAHL, Medline, PubMed, Web of Science, British Nursing Index, Scopus, and Google Scholar by using a combination of MeSH subject headings, keywords and titles, abstract, and full-text keywords, with publication dates between 2000 and 2021.The primary search terms used through the aforementioned databases were 'Cardiovascular disease' and 'Women'.The terms 'Gender bias', 'Prevention', 'Diagnosis', and 'Treatment', and their synonyms were later added to obtain more relevant studies.The search strategy for Medline is specified in Supplementary file S2 (Appendix).No language or publication status limits were applied to the search strategy.

Inclusion and Exclusion Criteria
Studies eligible for inclusion were peer-reviewed articles published between January 2000 and December 2021.We included studies that investigated CVDs among adult women.We excluded studies that did not specify the women-to-men ratio, conference abstracts, and randomised controlled trials where women are often under-represented [24].Moreover, we excluded studies that investigated mainly other conditions than CVDs, such as diabetes and biochemical and/or molecular mechanisms.The title of the study was initially assessed according to the inclusion/exclusion criteria, followed by the abstract and then full text.Figure 1 shows the flow chart regarding the study selection.Any disagreement was resolved by discussion.The extracted studies were organised by study type, and the following characteristics were included for each study: study settings, participants' characteristics, sample size, men-to-women ratio, main diagnosis, treatment, risk factors, and awareness of patients regarding CVDs' preventive measures [33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52] (Table 1).Occurrence of all CV events referred to at least one event of CV death, myocardial infarction, hospitalisation for unstable angina or HF, cerebrovascular accident, or emergency revascularisation.
ACS: acute coronary syndrome, AMI: acute myocardial infarction, CABG/AV: coronary artery bypass graft surgery/atrioventricular, CAD: coronary artery disease, CT: computed tomography, CVD: cardiovascular diseases, CHD: congestive heart disease, HF: heart failure, MI: myocardial infarction, PAD: peripheral artery disease, PCI: percutaneous coronary intervention The Joanne Briggs Institute (JBI) checklist was used to assess the quality of included case-control, case series, and cross-sectional studies [25].In addition, the Critical Appraisal Skills Programme (CASP) checklist was used to assess the cohort studies [26].Each article was subjected to evaluation by two independent reviewers (RB and AA).Any differences in ratings were resolved by consensus among reviewers.All studies were included after passing the evaluation criteria.

Data Analysis
Data were extracted using Statistical Product and Service Solutions (SPSS, v26; IBM SPSS Statistics for Windows, Armonk, NY).The summary statistics were reported for study duration, sample size, and men-towomen ratio.After conducting the test of normality, the data did not show normal distribution, so median and interquartile ranges (IQR) were evaluated.The differences in diagnosis, risk factors, and medicines prescribed for men and women patients with CVDs were also reported.As the included studies had high heterogeneity, the findings were presented as a narrative summary instead of a meta-analysis, using tables and figures to aid presentation where appropriate.
Diagnosis is the process of identifying a disease, condition, or injury from its signs and symptoms [29].Implicit bias operates in an unintentional, unconscious manner, silently exerting its influence on perception, memory, and behaviour [30].Prevention refers to actions taken to lower the risks of a disease occurring [27].Biological sex is assigned at birth, depending on the appearance of the genitals.Gender identity is the gender that a person "identifies" with or feels themselves to be [27].A risk factor is any attribute, characteristic, or exposure of an individual that increases the likelihood of developing a disease or injury [31].Treatment means the management and care of a patient to combat disease or disorder [32].

Study Extraction
Studies included were those focused on gender bias in the diagnosis, prevention, and/or treatment of CVDs.Studies evaluating gender in different contexts to the three aforementioned contexts were excluded.Moreover, studies having CVDs as secondary conditions or those that did not fulfill the ICD-11 criteria for the cardiovascular (CV) condition were excluded.Descriptive studies that had no focus on patients' outcomes were also excluded.
Only peer-reviewed articles were considered; yet, no language restriction was applied.A total of 977,032 articles were screened through the six databases by two reviewers (RB and SA) independently, where 6,400 were obtained after initial screening and removal of duplicates (Figure 1).Out of the 6,400, 6,336 were excluded based on their title considering the inclusion/exclusion criteria.Of the remaining 64 studies, 31 studies were excluded after evaluation of the abstracts according to the inclusion/exclusion criteria.Of the remaining 33 full-text studies, 14 were found to be irrelevant.Consequently, the research resulted in 19 studies that were subject to quality assessment by two reviewers (AA and RB) who scored independently.All of the 19 studies scored high and were included in the review (Figure 1).

Reported CVDs in Women Compared to Men
Reported CVDs differed between men and women that depended on gender, age, comorbidities, and lifestyle.In certain conditions related to atherosclerosis, women showed milder symptoms than men; yet, they were at higher risk of developing multiple comorbidities than men [42].Moreover, in CVDs such as ACS, women were more likely to be admitted to hospital at older age, where old age represented an additional risk factor for ACS [52].Women admitted at older age to hospital could further explain the higher rates of stroke and heart failure among women [34][35][36]40,46].
Nonetheless, the prevalence of ACS and its manifestations (ST-elevated myocardial infarction (STEMI), non-ST elevated myocardial infarction (NSTEMI), and unstable angina) has been reported higher among men in four studies [35,37,49].For stable angina, the prevalence among genders varied between men and women depending on the severity when assessed by the Canadian Cardiovascular Society Grading Scale [53].In this respect, women showed more class II and less class I symptoms in mild-to-moderate angina [44].It is worth mentioning that class II symptoms were more severe than class I symptoms.Patients with class II symptoms experienced limitations in carrying out normal everyday activities such as walking or taking stairs [53].
Differences between women and men in the prevalence of ACS and stable angina in terms of severe outcomes in women indicated underdiagnosis of women's symptoms.This was confirmed in additional studies that reported higher rates of PAD and chronic kidney disease (CKD) among women where both PAD and CKD represented modifiable risk factors for ACS [35,40,41,47].Moreover, two of the aforementioned studies reported higher rates of CAD among women [35,47] contrary to the literature where men showed higher rates of CAD than women [34,40,41,46].This could be related to CV symptoms being misdiagnosed among women as anxiety or gastrointestinal-related symptoms [39].

CV Risk Factors in Women Compared to Men
Women were less likely to have their CV risk factors assessed than men and that impacted CVD prevention among women [45,51].Eight risk factors were mentioned in this sense: body mass index (BMI), hypertension (HTN), hypercholesterolaemia, hyperlipidaemia, hypertriglyceridemia, smoking, and family history of CVDs (Table 2).BMI was reported significantly higher in women than men in four studies [34,35,43,49], significantly lower in women than men in two studies [39,47], and not significantly different between both genders in two studies [37,46].Similarly, diabetes was reported significantly higher in women than men in five studies [35,36,40,43,44], significantly lower in women than men in six studies [37,39,45,[47][48][49], and not significantly different between both genders in two studies [46,50].

Assessment and Management of CVDs in Women
Women had less assessment of essential and complementary examination of their medical condition in chest pain, angina, and CHD.When complaining of chest pain, women received an equal assessment of the essential examination (i.e., chest pain severity) but were offered less complimentary examination and were referred less to cardiologists or hospitalisation [33,49].On the other hand, men received more complementary examinations, including coronary angiography, ergometry, ECG, and cardiac enzymes (creatinine kinase and troponin) [33,49].Similarly, in the initial assessment of angina, exercise ECG and coronary angiography were performed less for women [44].
In CHD patients, women were 2.5 times less likely to be referred to a cardiologist than men and that indicated stronger gender bias in the management of the condition [33].In the latter study, referral of women to cardiologists was less encountered when male physicians undertook the initial diagnosis rather than women physicians.Subsequently, women overall were less likely to achieve treatment targets in CHD [43].
Management of CVDs was performed less in women whether admitted to hospital with ACS post-diagnosis of MI.After admission to a hospital with ACS, women were less likely to receive coronary intervention regardless of the type of ACS [52].In addition, in-hospital mortality following concomitant coronary artery bypass graft (CABG) was significantly higher in women than in men [38].Furthermore, post-diagnosis of MI, women had longer waiting times to get treatment and undertook less frequent chemical/mechanical thrombolysis than men [48].Moreover, invasive coronary angiography and revascularisation after follow-up post-CAD were performed significantly less in women than in men [39,44,49].This could be partly due to significantly lower levels of coronary artery calcium in women [49].

Prescription of Medicines to Women
Six medicines/medicine classes were reported: antihypertensive, antiplatelets (including aspirin), betablockers and lipid-lowering agents (including statins), oral anticoagulants, and unspecified antianginal drugs (Table 3).It is noteworthy to mention that aspirin was often reported separately from the remaining antiplatelets, and statins were reported separately in studies from the remaining lipid-lowering agents.

TABLE 3: Medications prescribed to men and women CVD patients
-= not reported.% refers to the percentage of male and then female participants separately.*significant male-to-female difference p = <0.001Specified antihypertensive medicines in Lee et al.'s [47] study include short-acting nitrates, angiotensin-converting enzyme inhibitors (ACEIs), and angiotensin receptor blockers (ARB).Specified antihypertensive medicines in Mariani et al.'s [36] study include ARB, ACEIs, and calcium channel blockers (CCB).Millett et al. [37] reported antianginal: nitrate (in men 9.8 and women 12.4%).The study by Verani et al. [41] reported statins and high-intensity statins.Antihypertensive medicines reported by Zhao et al. [43] include total antihypertensive, ACE, ARB, and CCB.Zhao et al. [43] reported antianginal nitrates (30.5% in men and 37% in women).Statins reported in Rachamin et al.'s [51] study include low intensity, medium intensity, and high intensity (as primary and secondary prevention of CVDs); the antihypertensive medicines reported were ARB and CCB as primary and secondary prevention of CVDs, respectively.
Women were less likely to be prescribed CV medications than men in the same age group [43,45].However, that was not the case for antianginal nitrates and antihypertensive treatment.Antianginal nitrates were more significantly prescribed in women than men in two studies [36,43].Similarly, antihypertensive medicines were more prescribed in women than men in most studies and could be related to the higher prevalence of HTN among women [34,36,39,45,51].Not all studies reported the specific class of antihypertensive reported, but where reported, angiotensin-converting enzyme inhibitors [36,43,47], angiotensin II receptor blockers [43,47,51] and calcium channel blockers have been used [36,43,51].Where reported, beta-blockers were more prescribed in women than men in three studies [34,45,51] and less prescribed in five studies [36,40,43,44,47].
For antiplatelets (including aspirin), women were less often prescribed antiplatelets than men after initial assessment of angina and hence were more likely to suffer MI or death in the one-year follow-up after angina [44,45].
Lipid-lowering agents (including statins) were more significantly prescribed in men [41,44,45,51] despite that women showed high levels of low-density lipoprotein [51].Following hospital discharge of MI, women were prescribed significantly less high-intensity statins (atorvastatin and rosuvastatin) than men, which was not related to differences in sociodemographic factors, comorbidities, and healthcare utilisation [40].

Women Patients' Awareness Towards Preventive Measures
Only two studies indicated women patients' awareness towards CVDs [35,49].The aforementioned two studies found that men patients were more likely to consider themselves at risk of heart disease and consult with a primary care physician and/or a cardiologist than women patients.On the contrary, women patients did not always consider their risk factors as related to CVDs nor were told by healthcare professionals that they were at risk and/or how to minimise the risk [35].

Discussion
The findings of this review suggested the presence of gender bias against women in relation to CVDs' diagnosis, prevention, and treatment.The studies included in the review mainly reported under consideration of gender-specific requirements.Hence, studies either did not consider CV risk factors, diagnostic tests, and/or medicines among women or considered them and found them more tailored to men.Therefore, this review highlighted gender bias encountered in the diagnosis, prevention, and treatment of CVDs among women.Two previous systematic reviews assessed the risks of CVDs in women patients; however, they did not address the diagnosis and/or treatment of CVDs [52,54].Similarly, two other reviews were narrative in nature and only assessed the gender bias specific to IHD in women [16,17].Consequently, by including further diagnosis and treatment, this review added to the previous literature and provided useful insight for both patients and healthcare professionals working with CVDs.It is noteworthy to mention that the review did not show selection bias in the women-to-men ratios in the individual studies that showed almost equal ratios with a median of 53:47 (IQR 44:36-64:56).
With regards to CVD risk factors, HTN was most reported in women, while men showed higher rates of smoking [39,47,55].Women only started heavy smoking in the US and the UK in the 1940s [55].Smoking increases the risk of atherosclerosis and its manifestations (IHD and PAD) [56].Hence, over-considering the aforementioned risk factors that are more prevalent in men could lead to unconscious misdiagnosis of CVDs among women.
In relation to CVD diagnosis, women received fewer diagnostic tests than men, which could be explained by the differences in sensitivity towards the tests between both genders.This could be explained by the different manifestations of CVD in women when compared to men [17,57,58].Women show a delayed onset of CVDs than men, and women's symptoms are often described as atypical, leading to underappreciating the severity of CVDs [33].A review by Canto et al. demonstrated that women with ACS are more likely to present with no chest discomfort than men and that is attributed to the underdiagnosis of CVDs in women [59].Men are known to have larger more visible blockages in coronarys compared with women [16].This could further explain why women experience worse clinical outcomes than men [17,22,60], with the disease not being diagnosed and treated as efficiently as it could be.In other studies, physicians misdiagnosed CHD symptoms in women as gastrointestinal or mental health conditions' symptoms [61,62].Physicians-patient interaction has been shown to differ when interacting with men versus women patients where three main factors play a role, including physician's prejudices, symptom perception, and gender-specific description [63].Though guidelines exist regarding differences in genders [8][9][10][11][12][13], physicians may not be aware of them or may think that CVD is a man's disease [64,65].Hence, gender stereotyping and bias represent potential reasons for gender disparities in the diagnosis of CVD [11,66,67].These findings suggest the need for more intensive investigation for women patients [16,17].
This bias not only affected diagnostic tests but also other procedures relating to the assessment and management of CVDs.For instance, men received more revascularisation than women despite their unwillingness to undertake invasive procedures [68].Hence, men were seven times more likely than women to refuse invasive investigation [69].
Furthermore, women received fewer prescribed medicines when compared to men.Men were reported to have better access to four out of six key medicines/classes (antianginal, antiplatelets, oral anticoagulants, and statins) in comparison to women [17,44,60].This was consistent with studies in the literature that reported women were more likely to receive fewer medicines for CVDs than men [60,70].Even when prescribed statins, women are more likely to be prescribed lower doses of statins than men [41].This was encountered more with high-intensity statins despite that the American College of Cardiology/American Heart Association (ACC/AHA) cholesterol guideline did not make a differentiation by gender in relation to statin prescription [12].
Patient awareness of CVDs was underreported in the included studies.However, it was stated that women had considered themselves at less risk of CVDs and were less likely to visit a primary care physician [71.
However, men considered their symptoms more seriously and were more likely to visit a cardiologist [71].In similar reviews, women were found to be more likely to call 911 for a friend experiencing heart attack symptoms than for themselves and less likely to report their own symptoms [7,21,[72][73][74][75].The latter finding was confirmed by a survey that interviewed 1,000 American women who had reported their lack of awareness of CVDs being the main cause of death among women [74].The lack of awareness could be partly caused by stress and partly socioeconomic factors where more women live in poverty than men and that contributes to higher rates of anxiety that is strongly linked to adverse CV events [75].Future work to improve this awareness is necessary to ensure women patients have access to healthcare before conditions worsen.
There were limitations in this review.Although rigorous and systematic, the reviewers did not include unindexed and unpublished research.Furthermore, it had strict selection criteria that affected the number of studies yielded and generalisability of the findings.Thus, the review had a low number of studies (n = 19), and that was related to the definitions and the inclusion criteria used to increase sensitivity.Moreover, there was an imbalance in the CVD evaluated in the studies.Thus, only three studies evaluated CVDs in general, and the remaining studies explored ACS, CAD, CHD, MI, MVD, or PVD.This introduced heterogeneity in the studies.Hence, the included studies were of different findings and variable quality with considerable heterogeneity between the included studies particularly with variation in reporting statistics.Owing to the significant heterogeneity, a meta-analysis was not conducted.As the spread of the studies occurred over the last 20 years, this affected the generalisability of the findings.Furthermore, no clear definition of gender bias was used by any of the included studies with inconsistencies in the term used.There was no way of assessing the severity of the CV conditions reported and that affected the outcomes.In addition, the review was limited to quantitative studies, which could be enhanced by the addition of qualitative studies that may further uncover barriers to women's access to CV diagnosis and therapy.However, this was limited considering the low availability of the published literature regarding women's perspectives regarding CVDs.Nonetheless, the studies in this review included a significant number of women sufficient to provide evidence of women's experience with CVDs.

FIGURE 1 :
FIGURE 1: Extraction of studies considering the PRISMA guidelines (Supplementary files S1 and S2) , South, and Southwest China).(4000 residents in 39 communities) Adults ≥45 years old who have been living in their area for more than 6 months whether urban or rural of median age in the range of 59 -62 years old CVDs

Term Synonym
Cardiovascular disease*Cardiovascular disorders or heart diseases or heart conditions or cardiovascular defects