Influence of Socioeconomic Status on the Association Between Pregnancy Complications and Premature Coronary Artery Disease: Linking Three Cohorts

Background: We hypothesized that there is an influence of socioeconomic status (SES) on association between pregnancy complications and premature coronary artery disease (PCAD) risk. Materials and Methods: This project involved a data linkage approach merging three databases of South Australian cohorts using retrospective, age-matched case–control study design. Cases (n = 721), that is, women aged <60 years from Coronary Angiogram Database of South Australia (CADOSA) were linked to South Australian Perinatal Statistics Collection (SAPSC) to ascertain prior pregnancy outcomes and SES. Controls (n = 194) were selected from North West Adelaide Health Study (NWAHS), comprising women who were healthy or had health conditions unrelated to CAD, age matched to CADOSA (±5 years), and linked to SAPSC to determine prior pregnancy outcomes and SES. This project performed comparative analysis of SES using socioeconomic indexes for areas–index of relative socioeconomic advantage and disadvantage (SEIFA-IRSAD) scores across three databases. Results: Findings revealed that SEIFA-IRSAD scores at the time of pregnancy (p-value = 0.005) and increase in SEIFA-IRSAD scores over time (p-value = 0.040) were significantly associated with PCAD. In addition, when models were adjusted for SEIFA-IRSAD scores at the time of pregnancy and age, risk factors including placenta-mediated pregnancy complications such as preterm birth (odds ratio [OR] = 4.77, 95% confidence interval [CI]: 1.74–13.03) and history of a miscarriage (OR = 2.14, 95% CI: 1.02–4.49), and cardiovascular disease (CVD) risk factors including smoking (OR = 8.60, 95% CI: 3.25–22.75) were significantly associated with PCAD. When the model was adjusted for change in SEIFA-IRSAD scores (from CADOSA/NWAHS to SAPSC) and age, pregnancy-mediated pregnancy complications including preterm birth (OR = 4.40, 95% CI: 1.61–12.05) and history of a miscarriage (OR = 2.09, 95% CI: 1.00–4.35), and CVD risk factor smoking (OR = 8.75, 95% CI: 3.32–23.07) were significantly associated with PCAD. Conclusion: SES at the time of pregnancy and change in SES were not associated with PCAD risk.


Introduction
Coronary artery disease (CAD) is most commonly owing to atherosclerotic plaque development in the coronary arteries obstructing blood flow to the myocardium.It is generally agreed that an obstruction of an epicardial coronary artery by >50% is clinically relevant in the absence of demonstrated ischemia or symptoms. 1Premature coronary artery disease (PCAD) among women is defined as a diagnosis of CAD before 60 years of age as per the Dutch Lipid Clinic Network Score. 2 Traditional cardiovascular disease (CVD) risk factors such as family history, smoking, hypertension, obesity, diabetes, and dyslipidemia do not account for the total observed burden of CAD in women. 3,4][7][8][9][10][11][12][13] The strongest evidence linking pregnancy complications with CAD comes from studies on preeclampsia that has resulted in it being recognized by the American Heart Association as a female-specific cardiovascular risk factor. 9ecurrent miscarriage also portends a doubling of the risk of CAD. 11ost studies have investigated the association between pregnancy complications and either morbidity or mortality because of CAD or the association between pregnancy complications and conventional CVD risk factors.In Australia, the above-mentioned pregnancy complications are common and are prevalent in *25% of all first pregnancies, 14 suggesting these complications are an important cardiovascular red flag.Socioeconomic status (SES) plays an important role in determining perinatal outcomes.In a populationbased study conducted in Nova Scotia, Canada, it was revealed that the prevalence of GDM, small for gestational age (SGA) and infant mortality, were higher among women with low family income or low SES compared with high family income or high SES despite having access to universal health care services. 15nother study from Korea that used a national level database also indicated similar findings that women in the lower SES group had higher rates of abortion, caesarean delivery, preeclampsia, preterm delivery, and obstetric hemorrhage compared with those in the higher SES group. 16ES has a quantifiable and significant impact on cardiovascular health.Women in low SES groups, especially those living in poverty, tend to be disproportionately affected by disparities in education, income, and access to health care resources.This disparity inevitably results in increased risk of ischemic heart disease and rehospitalization rates owing to CVDs compared with women in high SES group. 17,18There is compelling evidence that shows an independent association between SES and cardiovascular outcomes (including mortality) that is comparable in strength and consistency with any major CVD risk factor. 19In addition, a recent study on American adults using data from National Health and Nutrition Examination Survey demonstrated an inverse association (dose-dependent) of family income with the prevalence of CAD. 20he associations between SES and pregnancy complications, and also SES and CVD, are well established.However, the role of SES on the association between pregnancy complications and risk of PCAD has not been explored.To address this gap in knowledge, this study aimed to assess the role of SES on the association between pregnancy complications and CAD risk among women aged <60 years.

Study design
This data linkage project used a retrospective, agematched case-control study design by merging three databases of South Australian cohorts: Coronary Angiogram Database of South Australia (CADOSA), South Australian Perinatal Statistics Collection (SAPSC), and North West Adelaide Health Study (NWAHS).This project performed a comparative analysis of SES using socioeconomic indexes for areas-index of relative socioeconomic advantage and disadvantage (SEIFA-IRSAD) scores for the three databases.SEIFA is a product developed by the Australian Bureau of Statistics (ABS) that ranks areas in Australia according to relative socioeconomic advantage and disadvantage. 21SEIFA 2016 is the latest version and has four indexes.One of the indexes is IRSAD, which is an indicator of Australian SES and was analyzed as part of the current article.
IRSAD summarizes information about the socioeconomic conditions of people and households within an area, including relative advantage and disadvantage measures.It is generally measured in decile on a score of 1-10.A low score indicates relatively greater disadvantage and a lack of advantage (e.g., many households with low incomes or many people in unskilled occupations) and a high score indicates relatively greater advantage and a lack of disadvantage (e.g., many households with high incomes or many people in skilled occupations). 21As per the Australian education system's report prepared by the Centre for International Research on Education Systems for the Mitchell Institute, 22 young Australians generally represent three categories of SES: low (25%), medium (50%), high (25%).Taking this as a reference and based on the spread of our data, we categorized IRSAD score of deciles into three: 1 and 2 representing ''low SES,'' 3-7 as ''medium SES,'' and 8-10 as ''high SES.''

Study setting/location
Sites.The data linkage activity was performed independently by SA-NT Datalink, a nationwide consortium operating as part of the Population Health Research Network.De-identified data including SEIFA-IRSAD scores (SAPSC) and postcodes (CADOSA and NWAHS) were provided to the investigator team by SA-NT Datalink.

Study population and setting Data registries
Coronary Angiogram Database of South Australia.CAD-OSA is a statewide clinical quality registry of cardiac catheterization procedures performed in the public tertiary hospitals of South Australia (SA).The CADOSA registry 23 collects data on a standardized case report form and is compatible with the American College of Cardiology CathPCI Registry Ò . 24This registry captures in-hospital data by prospective data collection on: sociodemographic, detailed clinical, angiographic, and outcome data on every SA public hospital patient undergoing coronary angiography/percutaneous coronary intervention.For SES comparative analyses, postcodes in the CADOSA registry were converted to SEIFA-IRSAD scores (deciles) using the 2016 ABS excel spread sheet accessible on their website. 25The data included for these comparative SES analyses was from January 01, 2012 to December 31, 2018.
South Australian Perinatal Statistics Collection.SAPSC is a publicly held state-owned database that records pregnancy and birth outcome data for all births in South Australia since 1981. 26Approximately there are 20,000 births per year in South Australia, and pregnancy data including SES indicators (such as SEIFA-IRSAD scores) are available in the SAPSC for >600,000 births since 1981, which signifies it being a highly robust registry.The data were made available from January 01, 1986 for this specific SES analysis.However, data for body mass index (BMI) were made available from 2007 onward and were limited to very few women.
North West Adelaide Health Study.NWAHS is a longitudinal biomedical population cohort study of *4000 adults, recruited from northern and western regions of Adelaide, South Australia during 1999-2003, with ongoing follow-up of participants. 27In addition, some of the lowest SES suburbs (such as Elizabeth South and Davoren Park) in South Australia, and even in Australia, are part of NWAHS. 21The study focuses on health conditions including asthma, diabetes, chronic obstructive pulmonary disease, arthritis, osteoporosis, mental health, sleep health, and CVD.NWAHS being a multidisciplinary dataset has been involved in several research projects both at international and national level.This is a privately held database.For SES comparative analyses, postcodes in the NWAHS registry were converted to SEIFA-IRSAD scores (deciles) using the 2016 ABS excel spread sheet accessible on their website. 25

Study objectives
The aim was to assess the role of SES on the association between pregnancy complications and PCAD through proxy indicators such as postcodes and SEIFA-IRSAD scores.

Study procedures Data linkage methodology
Ascertainment of cases.Cases were ascertained from CADOSA.Women <60 years in CADOSA having undergone coronary angiography, with obstructive CAD defined as ‡50% stenosis in one or more coronary arteries (defined as PCAD) were linked to SAPSC database to determine if they had a previous major pregnancy complication.
Pregnancy complications included hypertensive disorders of pregnancy (HDP), which refers to a group of conditions mainly including gestational hypertension and preeclampsia (for this specific SAPSC cohort) that involves high blood pressure ( ‡140/90 mmHg) during pregnancy; preterm birth, which is defined as the spontaneous birth of a baby before 37 weeks of pregnancy; SGA infant, which is defined as having a birth weight <10th percentile for the gestational age; low birth weight (usually a combination of SGA and preterm birth), which is defined as the birth of a baby weighing <2,500 g; GDM, which is defined as glucose intolerance recognized first in pregnancy with revised diagnostic criteria of having fasting glucose of 5.6 mmol/g and/or 8.9 mmol/L in 60 minutes and/or 7.7 mmol/L in 120 minutes on 75 g of oral glucose tolerance test; or a history of previous miscarriage that is defined as loss of pregnancy before 20th week of gestation.Postcodes were also extracted from CADOSA (Fig. 1).
The SAPSC database was established in 1981 with its current dataset commencing in 1986.Therefore, the data were linked from January 01, 1986 to December 31, 2018.The first women recorded in SAPSC for this linkage are now approximately between 55 and 70 years of age.Key variables of interest in SAPSC included maternal age, preexisting medical conditions, socioeconomic indicators, pregnancy history, pregnancy complications, maternal smoking status, and BMI.
Ascertainment of controls.Controls were ascertained from the NWAHS registry.Women having health conditions or symptoms other than CVD/CAD (agematched to CADOSA -5 years) were linked to SAPSC database to identify their pregnancy outcomes (Fig. 1).Important variables extracted from NWAHS included general health and well-being, family history, diabetes, lung function, heart attack, stroke, angina, smoking, alcohol, and postcodes.
Women <60 years who were part of both CADOSA and NWAHS (n = 7) were excluded from the final analyses to prevent inclusion of any cases in the control group (Fig. 1).

Data linkage management
The security practices and protocols at SA-NT Datalink have been developed in accordance with the Australian Government Protective Security Policy Framework.

Statistical methods
The association between pregnancy complications and SEIFA-IRSAD scores from SAPSC was investigated using cross-tabulations with associated chi-square p-values or Fisher's exact test p-values as appropriate.The association between PCAD and SEIFA-IRSAD scores from SAPSC, CADOSA, and NWAHS was investigated using cross-tabulations with associated chi-square p-values.The association between PCAD and change in SEIFA-IRSAD scores was investigated using a cross-tabulation with associated chi-square p-value.Change in SEIFA-IRSAD scores was calculated using the difference in SEIFA-IRSAD scores at the time of either having PCAD (CADOSA) or no PCAD (NWAHS) and SEIFA-IRSAD scores during pregnancy (SAPSC).
Unadjusted binary logistic regression analyses were initially performed with outcome PCAD and predictor: SEIFA-IRSAD scores from SAPSC.Various covariates were then added to the model, one covariate at a time.Including only those covariates with p £ 0.25 on bivariate regression, an initial multivariable binary logistic model was created.Backward elimination was then performed until all covariates had p £ 0.05.This was the final parsimonious multivariable model for SEIFA-IRSAD scores from SAPSC as predictor.
Unadjusted binary logistic regression analyses were initially performed with outcome PCAD and predictor: change in SEIFA-IRSAD scores from CADOSA/ NWAHS to SAPSC.Various covariates were then added to the model, one covariate at a time.Including only those covariates with p £ 0.25 on bivariate regression, an initial multivariable binary logistic model was created.Backward elimination was then performed until all covariates had p £ 0.05.This was the final parsimonious multivariable model for change in SEIFA-IRSAD scores as predictor.

Ethics approvals
Ethics approvals from the following Human Research Ethics Committees (HRECs) were obtained: South Australian Department for Health and Wellbeing HREC, HREC/20/SAH/51 HERC of the University of Adelaide, Approval ID: 34704

Results
The number of records for SEIFA-IRSAD scores for women in SAPSC was n = 914, whereas number of records for SEIFA-IRSAD scores for women in CADOSA and NWAHS was n = 913.

Association between pregnancy complications and SES during pregnancy
Among the placenta-mediated pregnancy complications, preterm birth was independently and significantly associated with low SES as reflected by SEIFA-IRSAD scores at the time of pregnancy ( p < 0.001).None of the other pregnancy complications were associated with SES, although low birth weight was marginally insignificantly associated with low SES ( p = 0.067).Of the CVD risk factors, any cigarette smoking during pregnancy was significantly associated with low SES ( p < 0.001).Of note, women who had preexisting medical conditions during pregnancy were not associated with SES (Table 1).In a nutshell, women who had preterm delivery or were smokers were independently associated with having low SES during pregnancy.

Association between PCAD and SES
When stratified by SES during pregnancy, the prevalence of PCAD was 11.9% in women with high SES compared with 47.1% among those with low SES and this association was statistically significant ( p = 0.005).When stratified by SES at the time of diagnosis of PCAD, the prevalence of PCAD was 16.1% among women with high SES compared with 40.9% among those with low SES but, this association was not significant (Table 2).This meant that women who had low SES at the time of pregnancy were significantly associated with PCAD risk later in life.
Association between PCAD and Change in SES from pregnancy to time of diagnosis of PCAD.Prevalence of PCAD in women with improvement in SES from pregnancy to time of diagnosis of PCAD was 47.6% compared with prevalence of PCAD with decrement in SES was 36.2% and this association was statistically significant ( p = 0.040; Table 3).However, pregnancy complications such as preterm birth (61.3% vs. 38.7%),GDM (88.5% vs. 11.5%), and cardiovascular risk factor, that is, smoking (65.4% vs. 34.6%),were significantly more prevalent in women with improved SES compared with women with decreased SES ( p = 0.039, <0.001, and 0.001, respectively; Table 4).Table 3 concluded that women who had improvement in SES from pregnancy till PCAD diagnosis had higher prevalence of PCAD and Table 4 highlighted that preterm birth, GDM, and smoking were more prevalent in women with improved SES.
Final modeling of PCAD versus SES and change in SES scores PCAD versus SES during pregnancy and significant risk factors.Table 5 shows that low SES during pregnancy was associated with the development of early CAD (unadjusted p = 0.005).However, the final model revealed that the association between low SES during pregnancy and PCAD was not significant after adjusting for confounders such as age, history of miscarriage, preterm birth (i.e., <37 weeks), and any cigarette smoking during pregnancy.For Table 5, SES at the time of pregnancy was independently associated with development of PCAD; however, this association was insignificant when adjusted for confounders mentioned previously.
PCAD versus change in SES scores and significant risk factors.Table 6 demonstrates that an improvement in SES score was associated with the increased risk of development of early CAD (unadjusted p = 0.041).However, the final model revealed that the association between improvement in SES and PCAD was not significant after adjusting for confounders such as age, history of miscarriage, preterm birth, that is, <37 weeks, and any cigarette smoking during pregnancy.For Table 6, improvement in SES (i.e., from pregnancy to PCAD diagnosis) was associated with increased risk of PCAD; however, when adjusted with other risk factors, the association between change in SES scores and development of PCAD was insignificant.

Discussion
To the best of our knowledge, this is the first study to report the role of SES on the association between pregnancy complications and PCAD.This data linkage study showed that SES at the time of pregnancy and change in SES when adjusted for pregnancy complications and smoking were not significantly associated with PCAD.This translates that a woman with low SES who is a nonsmoker (living a healthy lifestyle) and also had a totally uncomplicated pregnancy does not have an increased risk of PCAD which seems clinically quite relevant.
9][30] A longitudinal cohort study conducted in Southern Alberta, Canada, showed that residence in a neighborhood with low SES plays an important role in accessing cardiac health services leading to adverse cardiac outcomes, especially in women compared with men. 31 Similar to the findings from the literature, our data linkage study demonstrated that SES score at the time of pregnancy was significantly associated with PCAD risk.
Furthermore, when change in SES scores was considered, a woman with an improved SES score (i.e., moving from a low SES postcode area to a high SES postcode area) still had high prevalence of PCAD later in life.This appears to be associated with a high prevalence of the unmodifiable risk factors of pregnancy complications including preterm birth and GDM and her persistent unhealthy lifestyle such as smoking when moving to a high SES suburb.
Women who give birth preterm are at twice the risk of CAD compared with those who give birth at term. 5 The four main pathways leading to preterm birth include ascending genital infections (disturbed microbiome), placental dysfunction, stress, and excessive uterine stretch. 32,33In particular placental dysfunction and ascending infections/disturbed microbiome can be associated with inflammatory cytokines and dyslipidemia, both of which are associated with atherosclerosis and endothelial dysfunction leading to increased CVD risk. 34A study from Korea utilizing their national health insurance database revealed that mothers in the low SES group had higher rates of preterm delivery compared with those in the middle or high SES groups. 16Extended working hours and workrelated fatigue have been identified as risk factors for preterm delivery. 35Our findings are consistent with those of the Korean study and demonstrate that in both the final models women who deliver preterm are at four-fold increased risk of PCAD after adjusting for SES scores at the time of pregnancy.For these cohorts, preterm birth underlines the risk for PCAD and moving to a higher SES area does not mitigate the risk.This may be because of the fact that these women had higher prevalence of pregnancy complications such as HDP, low birth weight, GDM, and had preexisting diabetes (type 1 or 2), were smokers and obese, despite moving to a higher SES area.
Miscarriage is one of the most common adverse pregnancy outcome defined as the loss of pregnancy before 20 weeks of gestation, 36 majority being in the first trimester of pregnancy. 37][40][41][42] Regarding the association between history of miscarriage and CVD, some shared risk factors such as smoking, excessive alcohol intake, and obesity play important roles as confounders in the causal pathway.After adjusting for these risk factors, a stronger evidence for an association between miscarriage and future CVD risk in mothers was found. 43n a survey conducted in Lahore, Pakistan, the study concluded that having low SES was indirectly associated with miscarriage, because improper diet was the main reason for pregnancy loss before 20 weeks, the latter being associated with poverty among pregnant women. 44Similarly, a Korean Prenatal Diagnosis Study conducted in Seoul from 2016 to 2018 revealed that there was an inverse association between the risk of fetal chromosomal abnormality leading to miscarriage and low level of household income in a prospective cohort of pregnant women. 45Our current data linkage study (for both the final models) also showed that women with a history of miscarriage are twice as likely to experience PCAD compared with women without any pregnancy loss after adjusting for SES scores at the time of pregnancy.
GDM is strongly associated with the lifetime development of CAD even in the absence of overt type 2 diabetes mellitus. 46,47A systematic review of observational studies found that women with a history of GDM have a 45% increased risk of CAD compared with women without GDM. 480][51] This relationship tends to become stronger in the presence of advanced maternal age and increase prepregnancy BMI. 51The findings of our study revealed that GDM is associated with an increased risk of PCAD and even moving to a higher SES suburb does not reduce that risk.This can be owing to increased prevalence of pregnancy complications and preexisting diabetes (type 1 or 2) in these women as mentioned previously.
Smoking is considered as a major risk factor for influencing the pattern (location of the damaged artery) and severity of CAD. 52Cigarette smoking causes a wide range of vascular abnormalities including dyslipidemia, endothelial dysfunction, defects in coagulation and fibrinolysis, and platelet dysfunction. 53Maternal smoking during pregnancy has proven to be associated with an increased risk of maternal CVD and the risk was higher for mothers who smoked during their last pregnancy. 54n addition, smoking during pregnancy has not only been linked to adverse pregnancy outcomes such as low birth weight, SGA, and preterm birth but also has an inverse association between SES and the frequency of maternal smoking during pregnancy. 55,56Income level has been significantly associated with CVD risk and a large study representing United States and Finland found a similar association with an increased risk of myocardial infarction and cardiac mortality in lowincome cohorts after adjusting for smoking. 57indings of our study are similar to the preexisting literature: mothers who smoke during pregnancy are eight times more likely to experience PCAD (as shown in both the final models) compared with mothers who are nonsmokers after adjusting for SES scores at the time of pregnancy.One of the reasons could be that many women who smoke at conception give up at least while they are pregnant but those who continue to smoke during pregnancy are more likely to continue smoking in to later life, even if these women move to a higher SES area, indicating it is a strong behavioral risk factor.

Strengths and limitations of this study
The use of three robust, comprehensive databases as part of this linkage project provided a unique opportunity to assess the role of SES on the association between pregnancy complications and PCAD in Australian women.This data linkage activity is the first of its kind in Australia using a control cohort and was performed independently by a third party, SA-NT Datalink.Researchers were involved in data analyses and interpretation and had access to de-identified data only, thereby minimizing selection bias.One-to-one comparison of women representing SEIFA-IRSAD scores (in deciles) across all three databases, served as ideal ''Cases'' and ''Controls.''One of the limitations of this study was owing to it being a retrospective case-control study; causality could not be established between pregnancy complications (including other CVD risk factors) and PCAD after adjusting for SES, although temporal inference could be drawn.Second, because we were interested in one-to-one comparison of women across three databases, women with multiple pregnancies, having maximum number of pregnancy complications in a single pregnancy were included, while omitting other pregnancies, reducing our overall sample size for this data linkage activity.In addition, data for BMI were collected from 2007 onward in SAPSC database, which restricted our numerator (to n = 69) of the total women included in this cohort.Third, we had the issue of missing data in SAPSC database because some of the risk factors for both pregnancy complications and CVD were included in the database across time rather than from the onset of data collection.
Fourth, SEIFA-IRSAD scores are used as proxy indicator for SES within Australian communities and they represent population level SES characteristics (ranking areas into relative socioeconomic advantage and disadvantage) instead of individual level data.These proxy indicators might tend to misclassify SES at the individual level.Fifth, all the three databases (CADOSA, NWAHS, and SAPSC) might have had the issue of reporting bias (or recall bias) because self-reported data such as ''average number of cigarettes smoked per day'' was collected from respective study partici-pants.Sixth, SAPSC database did not differentiate spontaneous from iatrogenic preterm birth and also gestational hypertension from preeclampsia.As a result, our findings are limited to women with preterm birth and HDP in general.

Conclusion
This data linkage study found that women's SES at the time of pregnancy was not associated with PCAD.In addition, women diagnosed with PCAD have conventional CVD risk factor such as smoking and some that are uniquely related to their pregnancy (maternal placental syndromes) 58 such as preterm birth and miscarriage in the presence of maternal age and SES.This calls for improved recognition by the medical and midwifery community of these clinical risk factors for future PCAD as part of preventative cardiology practice.These women may benefit from postpartum follow-up to increase their awareness of the risk of PCAD, encourage healthy lifestyle, increase access to health care for women, and manage CVD risk factors as they appear in the future to reduce PCAD risk. 59Moreover, these findings also provide evidence for the need for targeted postpartum interventions for women who experience pregnancy complication such as preterm delivery.

Fig. 1 .
Fig. 1.Case-control study design including SES comparative analyses-Diagrammatic illustration of the linkage process.SES, socioeconomic status.

a
SES divided into high, medium, and low reflective of change in SEIFA-IRSAD scores from SES at the time of having PCAD (CADOSA) or no PCAD (NWAHS) minus the SES at the time of pregnancy (SAPSC).b Presence of risk factors for premature coronary artery disease including pregnancy complications, medical conditions during pregnancy, and other cardiovascular risk factors during pregnancy.c p £ 0.05 was taken as being statistically significant.

Table 1 .
Independent Association Between Pregnancy Complications and Socioeconomic Status During Pregnancy £ 0.05 was taken as being statistically significant.BMI, body mass index; CVD, cardiovascular disease; SEIFA-IRSAD, socioeconomic index for areas-index of relative socio-economic advantage and disadvantage; SES, socioeconomic status.
a SES divided into: high, medium, and low reflective of SEIFA-IRSAD scores.b Presence of risk factors for premature coronary artery disease including pregnancy complications, medical conditions during pregnancy, and other cardiovascular risk factors during pregnancy.c p

Table 3 .
Association Between Premature Coronary Artery Disease Versus Change in Socioeconomic Status Scores IRSAD scores from SES at the time of having PCAD (CADOSA) or no PCAD (NWAHS) minus the SES at the time of pregnancy (SAPSC).
a SES divided into high, medium, and low reflective of change in SEIFAb p £ 0.05 was taken as being statistically significant.

Table 2 .
Association Between Premature Coronary Artery Disease and Socioeconomic Status During Pregnancy and at the Time of Premature Coronary Artery Disease Diagnosis £ 0.05 was taken as being statistically significant.CADOSA, Coronary Angiogram Database of South Australia; NWAHS, North West Adelaide Health Study; PCAD, premature coronary artery disease; SAPSC, South Australian Perinatal Statistics Collection.Khoja, et al.; Women's Health Reports 2024, 5.1 http://online.liebertpub.com/doi/10.1089/whr.2023.0092 b p

Table 4 .
Association Between Pregnancy Complications and Change in Socioeconomic Status from Pregnancy to Time of Diagnosis of Premature Coronary Artery Disease

Table 5 .
Association of Premature Coronary Artery Disease and Socioeconomic Status During Pregnancy Including Significant Risk Factors a Modeling the probability of having premature coronary artery disease using multivariable binary logistic regression analysis.b p £ 0.05 was taken as being statistically significant.

Table 6 .
Association of Premature Coronary Artery Disease and Change in Socioeconomic Status Scores Including Significant Risk Factors Khoja, et al.; Women's Health Reports 2024, 5.1 http://online.liebertpub.com/doi/10.1089/whr.2023.0092 a Modeling the probability of having premature coronary artery disease using multivariable binary logistic regression analysis.b p £ 0.05 was taken as being statistically significant.