Abstract
Authors: Mia Charifson, Timothy Wen, Sara Sauer, Isabel Fulcher
Conference: Society for Maternal and Fetal Medicine Annual Pregnancy Meeting 2026
Objective: The connection between pregnancy symptoms and adverse pregnancy outcomes has been understudied, largely due to lack of data infrastructure to measure symptoms longitudinally. This research uses longitudinal symptom data from a digital pregnancy care platform to investigate the association between pregnancy symptoms and preterm birth, a common adverse pregnancy outcome.
Study Design: Pregnant individuals from four prenatal clinics were enrolled in a mobile application (2022-2025) and were instructed by their care team to track weekly symptoms from a list of 100 symptoms (including ‘No symptoms’). Individuals who tracked symptoms for more than 5 gestational weeks were included in the analysis. Principal component analysis and k-means clustering identified and classified patients into three distinct clusters of symptom reporting throughout gestation. Multivariate logistic regression was conducted to estimate the association between symptom cluster and preterm birth (delivery <37 weeks gestation), adjusting for primiparity, pre-pregnancy body mass index, primary language, and prenatal clinic.
Results: 718 individuals enrolled in the digital health platform, delivered and reported symptoms for at least 5 gestational weeks. Three symptom clusters were identified (1) few to no symptoms (n=286, 39.8%), (2) common symptoms (e.g., fatigue, back pain) (n=355, 49.4%), (3) neurological/psychological symptoms (e.g., headaches, dizziness, memory changes, stress) (n=77, 10.7%). Adjusting for potential confounders, the neurological/psychological symptom cluster was associated with significantly higher odds of preterm birth (aOR=2.34 [1.11, 4.81]) compared to the few to no symptoms group. The effect estimate for the common symptom cluster suggested higher odds of preterm birth, albeit with a wide confidence interval (aOR = 1.44 [0.86, 2.48]).
Conclusions: This study demonstrated a meaningful association between symptom clusters and preterm birth. Advancements in digital tools to track symptoms throughout pregnancy open new avenues to monitor and potentially identify patients at risk of adverse outcomes.
Abstract
Authors: Mia Charifson, Timothy Wen, Sara Sauer, Isabel Fulcher
Conference: Society for Maternal and Fetal Medicine Annual Pregnancy Meeting 2026
Objective: The connection between pregnancy symptoms and adverse pregnancy outcomes has been understudied, largely due to lack of data infrastructure to measure symptoms longitudinally. This research uses longitudinal symptom data from a digital pregnancy care platform to investigate the association between pregnancy symptoms and preterm birth, a common adverse pregnancy outcome.
Study Design: Pregnant individuals from four prenatal clinics were enrolled in a mobile application (2022-2025) and were instructed by their care team to track weekly symptoms from a list of 100 symptoms (including ‘No symptoms’). Individuals who tracked symptoms for more than 5 gestational weeks were included in the analysis. Principal component analysis and k-means clustering identified and classified patients into three distinct clusters of symptom reporting throughout gestation. Multivariate logistic regression was conducted to estimate the association between symptom cluster and preterm birth (delivery <37 weeks gestation), adjusting for primiparity, pre-pregnancy body mass index, primary language, and prenatal clinic.
Results: 718 individuals enrolled in the digital health platform, delivered and reported symptoms for at least 5 gestational weeks. Three symptom clusters were identified (1) few to no symptoms (n=286, 39.8%), (2) common symptoms (e.g., fatigue, back pain) (n=355, 49.4%), (3) neurological/psychological symptoms (e.g., headaches, dizziness, memory changes, stress) (n=77, 10.7%). Adjusting for potential confounders, the neurological/psychological symptom cluster was associated with significantly higher odds of preterm birth (aOR=2.34 [1.11, 4.81]) compared to the few to no symptoms group. The effect estimate for the common symptom cluster suggested higher odds of preterm birth, albeit with a wide confidence interval (aOR = 1.44 [0.86, 2.48]).
Conclusions: This study demonstrated a meaningful association between symptom clusters and preterm birth. Advancements in digital tools to track symptoms throughout pregnancy open new avenues to monitor and potentially identify patients at risk of adverse outcomes.
Abstract
Authors: Mia Charifson, Timothy Wen, Sara Sauer, Isabel Fulcher
Conference: Society for Maternal and Fetal Medicine Annual Pregnancy Meeting 2026
Objective: The connection between pregnancy symptoms and adverse pregnancy outcomes has been understudied, largely due to lack of data infrastructure to measure symptoms longitudinally. This research uses longitudinal symptom data from a digital pregnancy care platform to investigate the association between pregnancy symptoms and preterm birth, a common adverse pregnancy outcome.
Study Design: Pregnant individuals from four prenatal clinics were enrolled in a mobile application (2022-2025) and were instructed by their care team to track weekly symptoms from a list of 100 symptoms (including ‘No symptoms’). Individuals who tracked symptoms for more than 5 gestational weeks were included in the analysis. Principal component analysis and k-means clustering identified and classified patients into three distinct clusters of symptom reporting throughout gestation. Multivariate logistic regression was conducted to estimate the association between symptom cluster and preterm birth (delivery <37 weeks gestation), adjusting for primiparity, pre-pregnancy body mass index, primary language, and prenatal clinic.
Results: 718 individuals enrolled in the digital health platform, delivered and reported symptoms for at least 5 gestational weeks. Three symptom clusters were identified (1) few to no symptoms (n=286, 39.8%), (2) common symptoms (e.g., fatigue, back pain) (n=355, 49.4%), (3) neurological/psychological symptoms (e.g., headaches, dizziness, memory changes, stress) (n=77, 10.7%). Adjusting for potential confounders, the neurological/psychological symptom cluster was associated with significantly higher odds of preterm birth (aOR=2.34 [1.11, 4.81]) compared to the few to no symptoms group. The effect estimate for the common symptom cluster suggested higher odds of preterm birth, albeit with a wide confidence interval (aOR = 1.44 [0.86, 2.48]).
Conclusions: This study demonstrated a meaningful association between symptom clusters and preterm birth. Advancements in digital tools to track symptoms throughout pregnancy open new avenues to monitor and potentially identify patients at risk of adverse outcomes.
Abstract
Authors: Mia Charifson, Timothy Wen, Sara Sauer, Isabel Fulcher
Conference: Society for Maternal and Fetal Medicine Annual Pregnancy Meeting 2026
Objective: The connection between pregnancy symptoms and adverse pregnancy outcomes has been understudied, largely due to lack of data infrastructure to measure symptoms longitudinally. This research uses longitudinal symptom data from a digital pregnancy care platform to investigate the association between pregnancy symptoms and preterm birth, a common adverse pregnancy outcome.
Study Design: Pregnant individuals from four prenatal clinics were enrolled in a mobile application (2022-2025) and were instructed by their care team to track weekly symptoms from a list of 100 symptoms (including ‘No symptoms’). Individuals who tracked symptoms for more than 5 gestational weeks were included in the analysis. Principal component analysis and k-means clustering identified and classified patients into three distinct clusters of symptom reporting throughout gestation. Multivariate logistic regression was conducted to estimate the association between symptom cluster and preterm birth (delivery <37 weeks gestation), adjusting for primiparity, pre-pregnancy body mass index, primary language, and prenatal clinic.
Results: 718 individuals enrolled in the digital health platform, delivered and reported symptoms for at least 5 gestational weeks. Three symptom clusters were identified (1) few to no symptoms (n=286, 39.8%), (2) common symptoms (e.g., fatigue, back pain) (n=355, 49.4%), (3) neurological/psychological symptoms (e.g., headaches, dizziness, memory changes, stress) (n=77, 10.7%). Adjusting for potential confounders, the neurological/psychological symptom cluster was associated with significantly higher odds of preterm birth (aOR=2.34 [1.11, 4.81]) compared to the few to no symptoms group. The effect estimate for the common symptom cluster suggested higher odds of preterm birth, albeit with a wide confidence interval (aOR = 1.44 [0.86, 2.48]).
Conclusions: This study demonstrated a meaningful association between symptom clusters and preterm birth. Advancements in digital tools to track symptoms throughout pregnancy open new avenues to monitor and potentially identify patients at risk of adverse outcomes.
Abstract
Authors: Mia Charifson, Timothy Wen, Sara Sauer, Isabel Fulcher
Conference: Society for Maternal and Fetal Medicine Annual Pregnancy Meeting 2026
Objective: The connection between pregnancy symptoms and adverse pregnancy outcomes has been understudied, largely due to lack of data infrastructure to measure symptoms longitudinally. This research uses longitudinal symptom data from a digital pregnancy care platform to investigate the association between pregnancy symptoms and preterm birth, a common adverse pregnancy outcome.
Study Design: Pregnant individuals from four prenatal clinics were enrolled in a mobile application (2022-2025) and were instructed by their care team to track weekly symptoms from a list of 100 symptoms (including ‘No symptoms’). Individuals who tracked symptoms for more than 5 gestational weeks were included in the analysis. Principal component analysis and k-means clustering identified and classified patients into three distinct clusters of symptom reporting throughout gestation. Multivariate logistic regression was conducted to estimate the association between symptom cluster and preterm birth (delivery <37 weeks gestation), adjusting for primiparity, pre-pregnancy body mass index, primary language, and prenatal clinic.
Results: 718 individuals enrolled in the digital health platform, delivered and reported symptoms for at least 5 gestational weeks. Three symptom clusters were identified (1) few to no symptoms (n=286, 39.8%), (2) common symptoms (e.g., fatigue, back pain) (n=355, 49.4%), (3) neurological/psychological symptoms (e.g., headaches, dizziness, memory changes, stress) (n=77, 10.7%). Adjusting for potential confounders, the neurological/psychological symptom cluster was associated with significantly higher odds of preterm birth (aOR=2.34 [1.11, 4.81]) compared to the few to no symptoms group. The effect estimate for the common symptom cluster suggested higher odds of preterm birth, albeit with a wide confidence interval (aOR = 1.44 [0.86, 2.48]).
Conclusions: This study demonstrated a meaningful association between symptom clusters and preterm birth. Advancements in digital tools to track symptoms throughout pregnancy open new avenues to monitor and potentially identify patients at risk of adverse outcomes.
Abstract
Authors: Mia Charifson, Timothy Wen, Sara Sauer, Isabel Fulcher
Conference: Society for Maternal and Fetal Medicine Annual Pregnancy Meeting 2026
Objective: The connection between pregnancy symptoms and adverse pregnancy outcomes has been understudied, largely due to lack of data infrastructure to measure symptoms longitudinally. This research uses longitudinal symptom data from a digital pregnancy care platform to investigate the association between pregnancy symptoms and preterm birth, a common adverse pregnancy outcome.
Study Design: Pregnant individuals from four prenatal clinics were enrolled in a mobile application (2022-2025) and were instructed by their care team to track weekly symptoms from a list of 100 symptoms (including ‘No symptoms’). Individuals who tracked symptoms for more than 5 gestational weeks were included in the analysis. Principal component analysis and k-means clustering identified and classified patients into three distinct clusters of symptom reporting throughout gestation. Multivariate logistic regression was conducted to estimate the association between symptom cluster and preterm birth (delivery <37 weeks gestation), adjusting for primiparity, pre-pregnancy body mass index, primary language, and prenatal clinic.
Results: 718 individuals enrolled in the digital health platform, delivered and reported symptoms for at least 5 gestational weeks. Three symptom clusters were identified (1) few to no symptoms (n=286, 39.8%), (2) common symptoms (e.g., fatigue, back pain) (n=355, 49.4%), (3) neurological/psychological symptoms (e.g., headaches, dizziness, memory changes, stress) (n=77, 10.7%). Adjusting for potential confounders, the neurological/psychological symptom cluster was associated with significantly higher odds of preterm birth (aOR=2.34 [1.11, 4.81]) compared to the few to no symptoms group. The effect estimate for the common symptom cluster suggested higher odds of preterm birth, albeit with a wide confidence interval (aOR = 1.44 [0.86, 2.48]).
Conclusions: This study demonstrated a meaningful association between symptom clusters and preterm birth. Advancements in digital tools to track symptoms throughout pregnancy open new avenues to monitor and potentially identify patients at risk of adverse outcomes.
Abstract
