Identifying clusters from multidimensional symptom trajectories in postpartum women

Area of Science:

• Maternal and Child Health
• Psychiatry
• Perinatal Mental Health

Background:

• Perinatal women often experience depressive symptoms, stress, fatigue, and sleep deprivation.
• These symptoms can negatively impact maternal and child health outcomes.
• Understanding symptom trajectories is crucial for timely intervention.

Purpose of the Study:

• To explore the evolution of perinatal symptoms and identify distinct groups of women based on symptom severity and patterns.
• To classify women using a model-based clustering approach for multidimensional symptom trajectories.

Main Methods:

• An observational study followed 151 pregnant women from the 3rd trimester through 6 months postpartum.
• Data collected included depressive symptoms (Edinburg Postnatal Depression Scale - EPDS), perceived stress (Perceived Stress Scale - PSS), sleep duration, and fatigue.
• Model-based clustering was used to analyze symptom severity and trajectories.

Main Results:

• Two distinct clusters were identified: Cluster 1 (n=43) with fewer depressive symptoms, less stress, less fatigue, and longer sleep duration.
• Cluster 2 (n=108) exhibited higher EPDS and PSS scores, increased fatigue, and shorter sleep duration.
• Pro-inflammatory markers showed associations with sleep duration and depressive symptoms.

Conclusions:

• Clustering multidimensional symptom trajectories can identify distinct groups of women in the perinatal period.
• This approach may help predict women at risk for severe symptoms, enabling targeted clinical interventions.
• Early identification and intervention are key for improving maternal and child health outcomes.