What follows are highlights from these studies:

Factors Associated with Infant Feeding Behaviors Among Women with Systemic Rheumatic Diseases

While guidelines provide recommendations on which medications for rheumatic diseases are safe during breastfeeding, limited research has explored what factors influence the decision to breastfeed among patients with these conditions. HSS researchers leveraged data from the Rheumatology Women’s Reproductive Health and Wellness Cohort to better understand the range of influences.

The cohort included 158 patients who reported on infant-feeding practices. Among those, 56% had systemic rheumatic diseases. The data showed that among these patients, about two-thirds breastfed at some point, while about one-third exclusively formula fed. The participants also reported their satisfaction levels with infant feeding, with only breast discomfort and lack of milk production being associated with lower satisfaction.

The investigators found that although the medications that the patients were taking for their rheumatic disease influenced their infant-feeding practices, disease-agnostic factors also played an important role. These factors included the experiences of friends and family, social media, and workplace policies, among others.

“It’s important to be mindful that these are things that influence people’s practices,” says Sarah B. Lieber, MD, MS, rheumatologist at HSS.

Dr. Lieber explains that rheumatologists are uniquely positioned to study this topic, as they typically care for patients over many years, including during pregnancy and the postpartum period. “We develop close relationships with our patients and they frequently seek our guidance on a wide range of issues,” she says. “Research like this allows us to expand our focus beyond rheumatology-specific concerns, addressing other important factors that may not have been addressed in other contexts to ensure that we honor their preferences with regard to infant feeding.”

Authors: Sarah B. Lieber, MD, MS; Lucy Masto, BS; Amaya Smole, BA; Ranqing Lan, MA; Michael Parides, PhD; Jonah M. Levine, BA; Bessie Stamm, BA; Caroline H. Siegel, MD, MS; Lisa A. Mandl, MD, MPH; Michael D. Lockshin, MD; Medha Barbhaiya, MD, MPH; and Lisa R. Sammaritano, MD.

Predicting Adverse Pregnancy Outcomes in Women with Systemic Lupus Erythematosus: External Validation of the PROMISSE Model Using Multiple Independent Cohorts

Lupus is associated with severe adverse pregnancy outcomes, including pre-eclampsia, very early preterm deliveries, and in rare cases, fetal death. Fortunately, these occur in only 20% of patients. Early identification of women with SLE who are at high risk of pregnancy complications is vital. A team led by HSS investigators followed more than 400 pregnant patients with lupus — all of whom were enrolled in the PROMISSE Study, a multicenter, multiethnic/racial study — with the goal of using clinical features and laboratory tests to identify those patients at high risk for pregnancy complications. They identified two promising algorithms for predicting poor pregnancy outcomes using PROMISSE data.

The goal of the current work was to externally validate these two models to test their transportability to other populations. In this study, the group obtained data from 966 pregnant lupus patients from six independent cohorts treated at centers in the United States and Europe, a demonstration of global collaboration. Groups that shared data for the validation were from South Carolina, Manhattan, and the Bronx, as well as Germany, Italy, and France. The patient populations varied, with some groups being predominantly white, while others were predominantly Black — either African American or Afro-Caribbean.

The PROMISSE model used statistics and machine learning to create prediction algorithms based on thousands of data points from 400 patients. The new findings with the additional patients confirmed that the algorithms can reliably predict pregnancy outcomes by assessing factors such as the presence of specific autoantibodies, disease activity, blood pressure, medications, and patients’ race and ethnicity. They demonstrate the geographic transportability of the best-performing algorithms. “The study reassured us that this model performs well across a diverse range of patient populations,” says Jane E. Salmon, MD, co-principal investigator and rheumatologist at HSS.

“More than 80% of lupus patients are likely to have an uncomplicated pregnancy, and they should be able to enjoy this time without having to worry. We are now better able to predict risk,” Dr. Salmon adds. “For those at higher risk, we can closely monitor and recommend clinical trials aimed at preventing complications.”

The team also included co-principal investigator and biostatistician Mimi Kim, ScD, and machine learning expert Melissa Fazzari, MS, PhD. Their next step is to develop a publicly available online risk calculator, a tool to allow obstetricians, rheumatologists, and other healthcare providers to better guide their patients based on individualized assessments.

Authors: Melissa Fazzari (Albert Einstein College of Medicine); Jane E. Salmon, MD; Marta Guerra, MS; Nathalie Costedoat-Chalumeau (Inserm; Paris, France); Veronique Le Guern (Cochin Hospital; Paris, France); Gaëlle Guettrot-Imbert (Cochin Hospital, Paris, France); Marta Mosca (University of Pisa; Italy); Dina Zucchi (University of Pisa; Italy); Chiara Tani (University of Pisa; Italy); Rebecca Fischer-Betz (Uniklinik Düsseldorf, Düsseldorf, Germany); Isabell Haase (Heinrich-Heine University Düsseldorf; Düsseldorf, Germany); Anna Broder (Hackensack Network; New Jersey); Navneet Kaur (Touro University Medical Group, Sacramento); Jill Buyon (NYU Grossman School of Medicine); Brooke Cohen (NYU Grossman School of Medicine); Diane Kamen (Medical University of South Carolina); Jessica English (Medical University of South Carolina); Anna Arar (Medical University of South Carolina); and Mimi Kim (Albert Einstein College of Medicine).

Quality Improvement Initiative to Increase Contraception Screening and Documentation for Reproductive-Aged Women Prescribed Teratogenic Medications in an Academic Rheumatology Center

Contraception use among patients with rheumatologic diseases who are taking medications that may cause fetal defects is low. They are not always counseled on contraceptive options or prescribed effective contraception. In addition to primary care providers and gynecologists, rheumatologists play an important role in contraception counseling for patients with rheumatic disease. However, there are several barriers to rheumatologists providing contraception counseling to patients. Documentation of contraception use in patients with rheumatic disease is poorly standardized nationally. Researchers from HSS, who previously looked at developing a more streamlined process for documentation of sexual activity and contraception in these patients, expanded their work in this new study to improve documentation in this patient population.

In the study, they implemented a pop-up alert in the electronic health record across all ambulatory areas. Over a period of 24 weeks, the alert was set to trigger during clinical visits for female patients between the ages of 18 and 50 who were taking one of these high-risk medications. The alert reminded the nurses to ask about patients’ sexual activity and contraceptive use and document it in the medical record.

Despite the initiative, the investigators found that for the most part, documentation rates did not increase significantly over the time period, except in the unit where the project had previously been piloted. “In that unit, we had nursing champions who were invested in the process and understood the importance of asking these questions,” says Nancy Pan, MD, rheumatologist at HSS. “They can encourage their colleagues to help, including by suggesting ways to normalize what is often a sensitive topic with patients yet is critical to providing the best medical care possible for them.” She adds that despite these results, she and her colleagues learned from this program and that their findings are prompting new interventions to optimize contraception documentation, counseling, and ultimately referrals to women’s health providers.

“When patients choose to take these medications, we always make sure to counsel them on contraception,” Dr. Pan says. “However, as a patient’s personal circumstances can change, it’s essential that we reassess their needs regularly. Optimizing documentation of contraception is an important first step to achieving improved health outcomes in these patients.”

Authors: Caroline H. Siegel, MD, MS; Avi Mikhaylov; Emily Wu; Deanna Jannat-Khah, DrPH; Erika Abramson (Weill Cornell Medicine), Lisa R. Sammaritano, MD; and Nancy Pan, MD.^.

“We have been interested in studying UCTD because it’s commonly encountered in clinical practice but not well characterized,” said lead study author Caroline Siegel, MD, a rheumatology fellow at HSS. “UCTD is inherently challenging to define, but it’s very important to understand what features differentiate patients with UCTD from those with defined connective tissue diseases (CTD) such as lupus so that we can better manage patients with this condition and ultimately predict who may go on to develop more severe clinical manifestations.”

The study included 89 patients aged 18 years and older treated at HSS between 2018 and 2022 who had rheumatologist-diagnosed UCTD with positive antinuclear antibody and at least one sign or symptom of CTD. The researchers reviewed medical records to identify those who fulfilled established classification criteria for systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), systemic sclerosis, primary Sjögren’s, idiopathic inflammatory myopathy, or antiphospholipid syndrome (APS). The researchers compared sociodemographic, clinical, serologic, and treatment variables between 59 patients with UCTD and 30 patients who fulfilled criteria for CTD.

Patients in both groups had similar non-criteria manifestations, most commonly joint pain and fatigue. However, patients with UCTD were less likely to have arthritis, criteria-based hematologic manifestations, anti-double stranded DNA or anti-Smith antibodies, and hypocomplementemia. The frequency of RA, Sjögren’s, and APS-related serologies did not differ between groups. Compared to those with CTD, UCTD patients were less likely to have ever received systemic corticosteroids.

“Our main findings were that patients who met a strict definition of UCTD versus those who met classification criteria for other CTDs were less likely to have clinical and laboratory features such as arthritis and hematologic abnormalities, as well as certain autoantibodies that are commonly seen in lupus,” said Dr. Siegel. “They were also less likely to have received treatment with steroids,” she added.

The results reveal that even among patients diagnosed by their treating physician with UCTD, some may meet classification criteria for other rheumatic diseases. Within this group, patients with strictly defined UCTD may differ in meaningful ways.

“We are starting to recognize features of strictly defined UCTD that will allow us to better identify and manage this condition,” said senior study author Medha Barbhaiya MD, MPH, rheumatologist at HSS. “Patients struggle with the uncertainty of a UCTD diagnosis and our study aimed to demonstrate that UCTD may be a distinct entity from other rheumatic conditions. These findings lend support to the claim that patients who have UCTD require separate study, and their disease manifestations warrant further understanding.”