Relationship between risk factor profile and prescription of low-dose aspirin for preeclampsia prevention

The purpose of this study was to assess obstetrician-gynecologist utilization of low-dose aspirin for women at increased risk for hypertensive disorders of pregnancy using guidelines developed by the American College of Obstetricians and Gynecologists and supported by the United States Preventive Services Task Force. Further, the study evaluated prescribing practices in relation to specific risk factor profiles to identify which women are at highest risk of not receiving recommended therapy. This was a retrospective cohort study reviewed and approved by the local Institutional Review Board. Electronic health records of women with singleton pregnancies who delivered between February and August 2020 were reviewed to identify risk factors for preeclampsia. Women were eligible for aspirin prophylaxis if they had at least one “high” risk factor or multiple “moderate” risk factors, as defined by the United States Preventive Services Task Force guidelines. Associations of interest were addressed using Pearson Chi-squared tests and multinomial logistic regression. 970 patients were included and 301 pregnant persons (31%) met criteria for low-dose aspirin prophylaxis; of these, 92 (31%) were given this recommendation. Those eligible for prophylaxis by presence of multiple “moderate” risk factors alone are least likely (0–6%) to receive indicated aspirin prophylaxis. Low-dose aspirin is an underutilized tool for preventing preeclampsia. Women with a combination of “moderate” risk factors are most likely to not receive indicated aspirin prophylaxis. Efforts should be made to encourage broader uptake of the recommendations for aspirin prophylaxis among obstetrician-gynecologists. What is already known on this subject? Low-dose aspirin has been shown to reduce preeclampsia risk in pregnant persons. This preventive measure has been recommended by most national and international organizations including the American College of Obstetricians and Gynecologists and the United States Preventive Services Task Force. Yet despite widespread support of this recommendation, uptake is not universal among obstetric care providers. What this study adds? This study identifies those who are most likely to experience a missed opportunity for aspirin prophylaxis, thus providing a suggestion for where provider education or other efforts to increase adherence to this guideline may be most impactful.


Introduction
Preeclampsia is a disorder with an estimated annual incidence of 4.6% worldwide [1]. Although affecting less than 5% of patients, this disease and its long-term consequences have a disproportionate impact on pregnancy-related morbidity and mortality. From 2003 to 2009, hypertensive disorders of pregnancy were responsible for 14% of maternal deaths annually-the number two cause of global maternal mortality [16]. Further, in women affected by preeclampsia, lasting damage to the cardiovascular system likely contributes to morbidity and mortality in future pregnancies and later in life [17].
The potential complications of preeclampsia during pregnancy are well established. These may be severe, including maternal and fetal death [3]; [12]. Given these potentially devastating consequences, there are ongoing efforts to reduce the risk of developing preeclampsia and mitigate its sequelae. This effort begins with early screening and identification of those who are at increased risk of preeclampsia or other hypertensive disorders of pregnancy. Various first trimester screening approaches have been proposed, using a combination of historical factors, serum biomarkers, and patient biometric data; however, the positive predictive value of these methods remain low at 8-33% [9]. The Combined Multimarker Screening and Randomized Patient Treatment with Aspirin for Evidence-Based Preeclampsia Prevention (ASPRE) trial utilized a combination approach, assessing historical risk factors, mean arterial pressure, uterine-artery pulsatility index, and serum biomarkers pregnancy-associated plasma protein A (PAPP-A) and placental growth factor [15]. Using this method, 11% of the 26,941 women screened in the study were deemed high risk for developing preeclampsia and 9% of these high-risk women stratified to the control group ultimately were diagnosed preeclampsia. The primary outcome in the ASPRE trial was the effect of 150 mg of aspirin nightly on the rate of preeclampsia prior to 37 weeks' gestation. The study showed a statistically significant 62% reduction in preeclampsia in the aspirin group versus placebo. Currently in the United States, screening is based solely on patient historical factors as outlined by the United States Preventive Services Task Force (USPSTF), [11] with a goaal of identifying those with at least an 8% absolute risk of developing preeclampsia. Such screening should be performed as early in pregnancy as feasible, so as to facilitate preventive measures such as low-dose aspirin.
To date, the only tool endorsed by the American College of Obstetricians and Gynecologists (ACOG) for primary prevention of preeclampsia is administration of low-dose aspirin in those at increased risk of developing the disease. [3] Studies have estimated a 17-62% reduction in the risk of preeclampsia among such persons when low-dose aspirin is used [4,8,10,15]. Specific dosing continues to be debated as various studies utilizing lowdose aspirin for preeclampsia prophylaxis have used doses ranging 60-150 mg/day, with 81 mg/day the standard dose in the United States, 75 mg/day endorsed by the World Health Organization, and 150 mg/day used routinely by providers around the world. There is a paucity of comparative data to date, and the USPSTF and ACOG only explicitly endorse 81 mg/day dosing, citing insufficient evidence to recommend higher dosing.
When prescribed for preeclampsia risk reduction, initiating aspirin therapy at 12-28 weeks' gestation is preferred by most major organizations as it reflects the timing utilized by the major clinical trials supporting the efficacy of aspirin for this purpose. The risk-reducing effect may be greatest when aspirin is initiated prior to 16 weeks, and starting aspirin prophylaxis between 12 and 16 weeks may be of particular benefit in preventing preterm preeclampsia [7,13,14]. Thus, early initiation of aspirin prophylaxis prior to 16 weeks is encouraged by most professional societies.
In 2018, ACOG and the Society for Maternal-Fetal Medicine (SMFM) issued a joint committee opinion advising the prescription of low-dose aspirin at a dose of 81 mg/day, to be initiated between 12 and 28 weeks' gestation, to anyone deemed at elevated risk for developing preeclampsia, based on the presence of specific risk factors or combinations of risk factors [2]. This approach was also given a grade B level of recommendation by the United States Preventive Services Task Force (USPSTF) [2]. According to all three organizations, pregnant patients are considered eligible for low-dose aspirin prophylaxis if they have at least one "high" risk factor or if they have more than one "moderate" risk factor, as described below.
The goal of the current study is to describe the relationship between individual preeclampsia risk factors and aspirin prescribing practices. Secondary outcomes assessed were the development of hypertensive disorders of pregnancy as well as characterization of risk profile (the various combinations of risk factors) and how these relate to aspirin prescribing practices.

Methods
This single-site retrospective cohort study was performed at a tertiary care, regional-referral center in the Midwest. The study protocol was reviewed and approved by the Advocate Health Care Network Institutional Review Board (IRB) (Downers Grove, IL). A waiver of informed consent was obtained from the IRB. Data were extracted from the electronic medical record (Epic Systems, Madison, WI) of all women ages 18 years and older with singleton gestations, who delivered between February and August of 2020. To improve generalizability of the findings regarding prescribing practices, analysis was restricted to women with singleton gestations. Women with fewer than two prenatal visits prior to twenty weeks gestational age (GA) were also excluded to ensure that the provider had ample time and opportunity to recommend low-dose aspirin prophylaxis. Additional exclusion criteria were uterine evacuation prior to 20 weeks GA, beta-blocker or calcium channel blocker prescription for non-hypertensive indications, and pregnancy complicated by prenatal diagnosis of a life-limiting fetal condition.
were grouped according to their risk of developing preeclampsia, using the classification schema described in ACOG Committee Opinion No. 743: "Low-Dose Aspirin Use in Pregnancy" [2]. They were considered at elevated risk, and thus eligible for low-dose aspirin prophylaxis, if they were determined, on retrospective review, to have one or more of the following "high" risk factors: personal history of preeclampsia, chronic hypertension, type 1 or 2 diabetes, renal disease, or autoimmune disease such as antiphospholipid syndrome or systemic lupus erythematosus; or if they had more than one of the following "moderate" risk factors: nulliparity, obesity defined by body mass index greater than or equal to 30 kg/m 2 , Black race, or age 35 years or older. Due to the retrospective nature of this analysis the following risk factors were not documented consistently or reliably in the available prenatal records and thus were unable to be assessed: family history of preeclampsia in a first degree relative, low socioeconomic status, history of low birthweight or small for gestational age infant, previous adverse pregnancy outcome, or interpregnancy interval greater than ten years. Additionally, women with multifetal gestations were excluded to identify prescribing trends among the most common patients seen by generalist obstetric providers. Of these omitted risk factors, only multifetal gestation is considered a "high" risk factor, the remainder are "moderate" risk factors.
A patient was considered having been prescribed lowdose aspirin if there was mention of the recommendation for low-dose aspirin documented her electronic medical record or if aspirin was included on medication reconciliation at the time of her admission to Labor and Delivery. Specific dosing ranging from 81 to 162 mg/day was left to the individual provider's discretion. Analysis was performed using an intention-to-treat methodology and patient adherence to prescribed therapy was not assessed. Women who met eligibility criteria and who had a low-dose aspirin prescription were classified as having been "appropriately prescribed" whereas eligible women without evidence of an aspirin prescription were deemed "missed opportunities." The primary outcome of interest was the rate of "missed opportunities" for aspirin prescription. Secondary outcomes included the specific risk factor profiles corresponding to missed opportunities for prophylaxis, and rates of development of hypertensive disorders of pregnancy among highrisk patients prescribed and not prescribed low-dose aspirin.
Analyses were performed using SPSS (IBM Inc, Armonk, NY). Data are presented as mean ± standard deviation for continuous variables or number (percentage) for categorical variables. All categorical variables were compared using Pearson Chi-square tests and a two-tailed p value of 0.05 to be considered statistically significant. The association between individual risk factors for preeclampsia and likelihood of low-dose aspirin prescription was addressed using multinomial logistic regression. Results were reported as  Fig. 1 Study population, preeclampsia risk classification, and aspirin prescription. This figure illustrates study selection criteria and breakdown of preeclampsia risk classification and aspirin prescription odds ratios (OR) or adjusted odds ratios (aOR) with 95% confidence interval (CI).

Results
There were a total of 1138 charts reviewed for study inclusion; of these, 168 patients were excluded, leaving a total of 970 patients included in the analysis. A vast majority [n = 164 (98%)] were excluded due to not having available data from two prenatal visits prior to twenty weeks GA. There was one patient taking propranolol prenatally for portal hypertension and there were three lethal fetal diagnoses. 301 patients (31%) were classified as being at increased for developing preeclampsia, and thus were eligible for low-dose aspirin prophylaxis. Of these, only 92 (31%) were actually prescribed low-dose aspirin (Fig. 1).
This low rate of appropriately prescribed low-dose aspirin was driven primarily by missed opportunities among pregnant patients who qualified for prophylaxis on the basis of having multiple "moderate" risk factors without comorbid "high" risk factors. There were a total of 179 such patients, making up 59% of the population eligible for low-dose aspirin prophylaxis, but only ten (6%) of them were recommended aspirin. These patients were 70% less likely to receive indicated low-dose aspirin prophylaxis compared to with at least one "high" risk factor [OR 0.3 (95% CI 0.1-0.6), p < 0.001], who were appropriately prescribed lowdose aspirin 67% (n = 82/122) of the time. Table 1 details the rate of low-dose aspirin prescription among patients with varying combinations of risk factors. Figure 2 depicts appropriate prescriptions and missed opportunities, stratified by individual risk factors, among women who were eligible for prophylaxis.
Three of the four "moderate" risk factors studied were associated with a decreased rate of appropriate aspirin prescribing compared to the overall rate. Missed opportunities occurred most frequently in nulliparous patients who had at least one additional risk factor, thus qualifying them for preeclampsia prophylaxis. These at-risk persons were the least likely to receive an aspirin prescription, at 12% (n = 16/129). Eligible patients age 35 years or greater were appropriately prescribed aspirin prophylaxis in 26% (n = 52/199) of cases and 54 of 186 (29%) eligible obese persons had a low-dose aspirin prescription. Black race was the only "moderate" risk factor for which aspirin prescription was given at a higher rate [n = 7/13 (54%)] than the overall appropriately prescribed rate.
By contrast, those with the "high" risk factors of diabetes [n = 11/15 (73%)], personal history of hypertensive Table 1 Appropriately prescribed low-dose aspirin according to risk profile Data are reported as n (%). "NA" = not applicable. a "Moderate" risk factors include nulliparity, obesity defined by body mass index greater than or equal to 30 kg/m 2 , Black race, or age 35 years or older a "High" risk factors include personal history of preeclampsia, chronic hypertension, type 1 or 2 diabetes, renal disease, or autoimmune disease such as antiphospholipid syndrome or systemic lupus erythematosus , and autoimmune disorder [n = 1/3 (33%)] were all appropriately prescribed aspirin therapy more often compared to the overall rate of 31%. This is consistent with findings of the multivariate logistic regression model used to determine independent associations between individual risk factors and having a low-dose aspirin prescription (Table 2). According to the model, only the "high" risk factors were independently associated with having an aspirin prescription, a finding that is also intuitive given that, by definition, more than one "moderate" risk factor is required to qualify a patient for prophylaxis, whereas a single "high" risk factor is needed for eligibility. Table 3 illustrates the impact of aspirin prescription on rates of hypertensive disorders of pregnancy.
Among those deemed at increased risk of developing preeclampsia, in our cohort those who were not prescribed low-dose aspirin appeared to be 50% less likely to develop any hypertensive disorder of pregnancy compared to those who did receive an aspirin prescription [OR 0.5 (95% CI 0.3-0.9), p = 0.013]. More specifically, eligible patients who did not receive aspirin prophylaxis appeared to have a 60% decreased odds of developing preeclampsia with severe features [OR 0.4 (95% CI 0.2-0.8), p = 0.007].

Conclusions for practice
Low-dose aspirin prophylaxis is currently the only intervention promoted by the American College of Obstetricians and Gynecologists, the Society for Maternal-Fetal Medicine, and the United States Preventive Services Task Force  for the primary prevention of preeclampsia, a potentially devastating condition for both mother and fetus. The recommendation to prescribe low-dose aspirin to pregnant persons at elevated risk of preeclampsia was first made by the USPSTF in 2014 [10]. And yet, more than six years later, there is little known about the impact of these guidelines on physician prescribing practices. The few studies that have been performed have analyzed only prescription rates among those with the well-recognized "high" risk factors. We are not aware of studies assessing prescribing practices for those eligible for prophylaxis by multiple "moderate" risk factors.
Our study provides insights into both provider familiarity with identifying patients at increased risk for preeclampsia and knowledge of the recommendation to prescribe low-dose aspirin prophylaxis for such individuals.
One study that examined aspirin prescribing practices, performed by Banala et al. focused on the impact of the ACOG committee opinion recommending aspirin use among patients with chronic hypertension [5]. The authors found that aspirin was offered to these individuals at a rate of 7% prior to the publication of the guideline and a rate of 70% following publication. Importantly, the authors found that outcomes of preeclampsia, small for gestational age neonate, and preterm birth were not affected by this change in prescribing practice.
A second study, by Boelig and colleagues, assessed provider adherence to aspirin recommendations before and after implementation of a preeclampsia risk assessment tool designed to help identify at-risk pregnancies [6]. The study analyzed only prescribing rates for patients with one or more "high" risk factor, and did not assess rates for those with multiple "moderate" risk factors. They found that prior to implementation of the screening tool 74% of eligible patients were recommended aspirin prophylaxis, compared to 95% following this intervention.
Unlike prior analyses, our study considered both "high" and "moderate" risk factors when determining which patients ought to have been prescribed low-dose aspirin prophylaxis. We found that up to 69% of those deemed atrisk for preeclampsia by the ACOG, SMFM, and USPSTF guidelines were not given a recommendation for aspirin prophylaxis. Despite having the same risk for developing preeclampsia as women with a single "high" risk factor, pregnant persons who qualified for aspirin prophylaxis solely on the basis of having multiple "moderate" risk factors were significantly less likely to be prescribed low-dose aspirin. But even among those with a single "high" risk factor, aspirin was prescribed at best 75% of the time. The improved prescription rate among person with "high" risk factors may be due to greater recognition of the association between those factors, such as history of preeclampsia or chronic hypertension, and the risk of developing preeclampsia.
Another potential explanation is that these patients are perhaps more likely to be referred to Maternal-Fetal Medicine specialists, who may be more familiar with the guidelines and more comfortable recommending aspirin therapy.
One curious finding that deserves additional explanation is the apparent reduction in risk of preeclampsia or gestational hypertension and of preeclampsia with severe features in women not taking aspirin. The authors feel that this is not so much a finding refuting the benefits of low-dose aspirin, but rather highlighting the observation that only the patients at very highest risk of developing preeclampsia-those with two or more "high risk" factors-were reliably prescribed aspirin in our cohort. Thus, the finding is interesting and worth noting for completeness, however is not thought to be an accurate representation of the effectiveness of low-dose aspirin.
This study is limited by the retrospective nature of its analysis. Because of this, not all preeclampsia risk factors were able to be reliably assessed when classifying patients and determining their eligibility for aspirin prophylaxis. The impact of these missing risk factors can be estimated and accounted for by the observation that there were 16/669 "low risk" patients who received an aspirin prescription. All "high-risk" risk factors were assessed in this group ultimately determined to not be candidates for aspirin prophylaxis by the factors examined in this study, so these sixteen women were either prescribed low-dose aspirin for indications beyond the specific recommendations of ACOG and the USPSTF, or they could theoretically have been identified through possession of two or more of the "moderate-risk" risk factors unable to be assessed in this study. Even if all sixteen of these patients did truly meet criteria for low-dose aspirin, the key finding of this study that those with multiple "moderate-risk" risk factors are often overlooked as candidates for preeclampsia prophylaxis would remain, as the proportion of patients without any "high-risk" risk factors who received an indicated low-dose aspirin prescription would still be only 26/195 (13%) versus 82/122 (67%) among persons with at least one "high-risk" risk factor. Thus, we believe the results of this study are still meaningful because expanding the review to include additional risk factors would only increase the number of eligible pregnant persons, and thus would likely only enhance our current findings.
Second, the timing of aspirin initiation was not captured in this retrospective study. As the study focused primarily on physician knowledge and practical application of published guidelines, this variable was not assessed. Given that evidence exists that the effectiveness of aspirin prophylaxis may be greatest when initiated between 12-and 16-weeks' gestation, this is an important marker for future study. Thus, for future pragmatic studies aimed at determining the effectiveness of real-world aspirin prescribing on preeclampsia risk reduction, we would recommend including timing of aspirin initiation as an important sub-analysis or comparator.
These findings indicate a need for further education among generalist obstetrician-gynecologists about screening for risk factors associated with preeclampsia, with a focus on identifying pregnant persons with multiple "moderate" risk factors. Prospective studies specifically designed to assess aspirin utilization practices are needed to completely define the problem and identify the most critical areas for improvement. Future studies that seek to understand factors motivating prescriber practices will also be important to inform interventions aimed at increasing uptake of lowdose aspirin for preeclampsia prophylaxis. Finally, it is critical to study the impact of these efforts on the outcome of interest-preeclampsia and its sequelae-to determine the clinical consequences of fully implementing these guidelines and to assess real-world efficacy of preeclampsia risk factor screening and low-dose aspirin prophylaxis.
Funding There was no financial support obtained for the conduct of this research or preparation of this article for publication.
Data availability Data were obtained from the local electronic medical record, inquiries regarding data can be directed to the corresponding author.

Conflict of interest
The authors report no conflicts of interest.

Ethical approval
The project was reviewed and approved by the local Institutional Review Board (IRB) prior to commencement of the project.
Informed consent Waivers of informed consent or consent for publication were obtained from the IRB due to the retrospective nature of this chart review and since no patient-identifying information will be published.