Telepsychiatry for Proactive Identification of Individuals at High-risk for Postpartum Syndromes: Moving from Symptom Screening to a Preventive Approach at a Rural Hospital

Background: Access to specialized consultation-liaison (CL) and perinatal psychiatry services is a growing challenge. Rural hospitals in health care shortage areas have especially struggled with recruitment and retention. With mental health conditions strongly contributing to pregnancy-associated deaths in the United States,¹ we present a psychiatry-obstetrics (OB) collaboration at a rural hospital which identifies patients at high risk for the development of postpartum syndromes. We implemented a proactive screening on the labor and delivery unit using a risk factor-based algorithm that includes telepsychiatry consultation. We analyzed the algorithm by comparing low- versus high-risk patients.

Methods: All patients admitted for delivery were screened using an algorithm created to identify high-risk women based on published studies of postpartum syndromes: depression/anxiety during pregnancy;* prescription of psychiatric medications during/before pregnancy;* prior history of postpartum depression, anxiety or psychosis;* history of trauma or current intimate partner violence; substance use/medication-assisted treatment; low social support; complications during delivery/baby transferred/fetal demise; Patient Health Questionnaire-2 screen =/ > 3. If patients had one starred item or 2+ non-starred items, a telepsychiatry consult was offered. We compared demographics, socioeconomic status, birth factors (gestational age, route of delivery) and 6-week postpartum EPDS scores of low-risk versus high-risk patients. Sensitivity and specificity of the algorithm was calculated.

Results: 354 women were screened; 128 women were high-risk but declined consult; 65 were considered high-risk and accepted consult; 164 were considered low-risk. Women in the declined group were 28.2 years of age on average compared to 26.6 in the consult group (p=.034). 1.5% of those considered low-risk had a 6-week EPDS ­score > 13 compared to 10.0% of the at-risk accepted and 10.2% of at-risk declined. The algorithm created has a sensitivity of 88.9% and a specificity of 48.2%. In the high-risk group, we found no differences in gravidity (mean 2.9 in declined vs 2.6 in accepted, p=.45) or route of delivery (37.6% C-section in declined, 35.9% in consults, p =.823).

Discussion: Initial analysis of our screening initiative demonstrates differences in postpartum EPDS scores between mothers considered low-risk versus high-risk for postpartum syndromes based on our algorithm. Preliminary results show no demographic, obstetrical, or EPDS differences in high-risk patients who declined consultation versus those who accepted. Our algorithm has high sensitivity but low specificity. Future steps include modifying the screening criteria to optimize sensitivity and considering outreach by the outpatient obstetrical team to high-risk individuals who are declining consultation.

Conclusion: Using risk factors for screening for postpartum syndromes can effectively identify high-risk patients. Utilizing telepsychiatry at time of delivery may help engage vulnerable rural populations using a preventive approach to care.