Female-Specific Factors Don’t Enhance Common Risk Calculators

Known risk factors for CVD among women, when added to commonly used risk calculators, don’t improve the tools’ ability to tease out which female patients need to start an antihypertensive or lipid-lowering medication, according to new data from the UK Biobank.

The study, published online recently in Circulation: Cardiovascular Quality and Outcomes, considered a slew of potential contributors—both together and individually—to CV risk: early menarche, preeclampsia, and endometriosis, among others. Yet none further honed the calibration and discrimination of the Pooled Cohort Equation - Atherosclerotic Cardiovascular Disease (PCE-ASCVD), QRISK2, and PREDICT models.

Lead author Jenny Doust, PhD (The University of Queensland, Herston, Australia), told TCTMD that this was a bit unexpected given the long history of research showing a link between these risk factors and subsequent disease.

“Several studies have shown that women who have a history of the female-specific risk factors considered in our study are at increased risk of CVD,” she wrote in an email. But because these studies didn’t include clinical values for blood pressure and cholesterol, they can’t be used as evidence to guide treatment.

While women with a history of these female-specific conditions “are at an increased risk of CVD relative to other women,” what their own findings suggest is that “having these risk factors doesn’t add to your estimate of risk in the next 5-10 years if you are using a risk calculator such as the PCE-ASCVD,” said Doust.

She emphasized that risk factors unique to female patients should still “absolutely” be considered in clinical settings. “It should be routine for clinicians to specifically ask women about their reproductive history [and] use the presence of these risk factors as an early warning of the risk of later potential risk of CVD and a marker of overall risk of CVD,” Doust advised.

Notably, the current study did not include an evaluation of PREVENT, a calculator released in late 2023 that creates sex-specific estimates of 10- and 30-year CVD risk in patients as young as 30 years, an age when the many pregnancy-related risk factors may be at play; however, PREVENT does not include these female-specific risk factors.

UK Biobank Data

Doust and colleagues turned to the UK Biobank to identify 135,142 women ages 45 to 69 years (mean 57.5 years) who were free from CVD at baseline (2006-2010) and followed through the end of 2019.

The researchers first created Cox proportional hazards models based on the risk factors already included in three contemporary calculators, the PCE-ASCVD, developed in 2013, as well as the 2008 QRISK2, and the 2018 PREDICT.

They then added various female-specific risk factors to the calculations: early menarche (< 11 years); endometriosis; excessive, frequent, or irregular menstruation; miscarriage; number of miscarriages; number of stillbirths; infertility; preeclampsia or eclampsia; gestational diabetes (without subsequent type 2 diabetes); premature menopause (<40 years); early menopause (<45 years); and natural or surgical early menopause (menopause <45 years or timing of menopause reported as unknown and oophorectomy reported at age <45).

During follow-up, the incidence of CVD was 5.3 cases per 1,000 person-years with PCE-ASCVD and PREDICT and 5.2 cases per 1,000 person-years with QRISK.

At the outset, the c-indices for the PCE-ASCVD, QRISK2, and PREDICT models were 0.710, 0.713, and 0.718, respectively. When adding all the female-specific risk factors together to the models, the c-indices were 0.712, 0.715, and 0.720, respectively.

It should be routine for clinicians to specifically ask women about their reproductive history [and] use the presence of these risk factors as an early warning of the risk of later potential risk of CVD and a marker of overall risk of CVD. Jenny Doust

Considered individually, the female-specific risk factors failed to make a difference in c-index, net reclassification index, integrated discrimination index, slope of the regression line for predicted versus observed events, or Brier score for any of the calculators.

“It seems that the female-specific risk factors are early markers of the risk of developing more intermediate risk factors for cardiovascular disease: higher blood pressure, cholesterol, and glucose levels,” said Doust. “Once you include these in the risk prediction, the female-specific risk factors do not add to our ability to predict who will and won’t develop overt CVD.”

An editorial by Setareh Salehi Omran, MD, and Michelle Leppert, MD (both from University of Colorado Anschutz Medical Campus, Aurora), like Doust, also cautions against dismissing female-specific risk factors in clinical practice or in the creation of future scores.

While the characteristics “may not improve CVD risk calculators, these risk factors still play an important role in comprehensive cardiovascular risk stratification in women,” they point out. “Many of these risk factors occur earlier in life and upstream of the causal pathway leading to traditional risk factors and then CVD, which represents a valuable opportunity for earlier intervention in women.” Premature menarche has been tied to later development of hypertension, hypercholesterolemia, and diabetes, for instance, as have adverse pregnancy outcomes.

Moreover, the relationship between these risk factors and CVD may be influenced by more-traditional risk factors, Omran and Leppert explain. “Because these mediators are already included in the risk model, the upstream effects of the exposure may be masked. However, this does not mean that female-specific risk factors do not ultimately affect CVD occurrence.”

Instead of abandoning these female-specific factors, it makes more sense to see them as an early chance to identify and prevent the traditional risk factors before they occur, they suggest, adding closer monitoring and lifestyle changes could be protective. “Clinicians should continue to obtain a comprehensive obstetric and gynecological history when assessing CVD risk in women,” the editorialists urge.

Doust said she agrees with these takeaways. “These risk factors [need] to be considered as early markers of CVD,” she noted. Also, “women with a history of these conditions [need] to be closely monitored for the development of later hypertension, hypercholesterolemia, and diabetes.”