初產婦睡眠品質對周產期憂鬱之影響

Abstract

背景：全球周產期憂鬱率六年間上升了14%（現已26.3%），其中亞洲地區周產期憂鬱率全球第二。現多有橫斷性研究指出孕期或產後的睡眠品質與憂鬱相關。再者，近年漸有研究將周產期睡眠品質進行趨勢分組，但仍少有研究探討自孕期至產後睡眠品質及其趨勢分組對憂鬱的影響。 目的：了解臺灣初產婦周產期睡眠品質和憂鬱趨勢，並探討周產期睡眠品質及其趨勢分組對憂鬱之影響。 研究方法：本研究採縱貫式研究設計，研究對象為初產婦（N=482），自2022年1月至2023年4月於臺灣北部三家醫學中心，使用結構式問卷於妊娠12-24週、妊娠24週至生產前、產後一個月、產後三個月進行調查。憂鬱以愛丁堡產後憂鬱量表調查，睡眠品質以匹茲堡睡眠品質量表調查，兩者皆是分數越高代表狀態越差，憂鬱和睡眠品質分數分別以12/13和5/6作為切點。以群組化軌跡模式區分睡眠品質趨勢組別，再以廣義估計模式檢驗周產期睡眠品質對憂鬱之影響。 結果：初產婦憂鬱率在第二孕期為12.4%、第三孕期為13.5%、產後一個上升至高峰（27.1%）、產後三個月稍微下降至22.9%，憂鬱分數整體是隨時間呈現緩升趨勢。睡眠品質方面，睡眠品質不良率（PSQI > 5）由第二孕期（57.3%）持續上升至產後三個月的最高點（77.8%）；使用群組化軌跡模式將周產期睡眠趨勢區分為四組， 第一組PSQI穩定低組（52.7%）和第二組PSQI分數逐漸上升組（13.5%）的起始分數相似、且皆呈線性上升，但在第三孕期開始的憂鬱上升程度第二組較為明顯；第三組PSQI穩定中高分組別佔27.9%（y=8.45）和第四組PSQI持續高分組佔5.9%（y=12.14）在第二孕期起始點皆大於6分，且在周產期其他測量時間點PSQI皆與起始點持平。廣義估計模式在控制年齡、教育程度、同住狀態、婚姻滿意度、生產方式、新生兒出生體重後，周產期睡眠品質與憂鬱之間存在顯著相關（β = 0.37, p < 0.001）。若將模型中的周產期睡眠品質連續變項更換為周產期睡眠品質趨勢分組後，與第一組PSQI穩定低組相比，第二組（β = 1.66, p = 0.005）、第三組（β = 2.27, p < 0.001）、第四組（β = 5.12, p < 0.001）之周產期憂鬱症狀皆顯著較高。 結論：臺灣初產婦周產期睡眠品質越差，越容易有較高的周產期症狀。將周產期睡眠品質趨勢分組後，在第二孕期若PSQI>6分時，孕產婦可能屬於第三組PSQI穩定中高分組或第四組PSQI持續高分組，這兩組的睡眠品質在第三孕期以後仍會與起始點持平、維持中高分或高分，因此需要在第二孕期偵測後即早給予介入措施。而若第二孕期睡眠品質起始分數較低，即有可能是第一組PSQI穩定低組或第二組PSQI分數逐漸上升組，需要於第三孕期進行第二次測量，若第三孕期若孕婦的PSQI ≧7分、則該媽媽應屬於第二組PSQI逐漸上升組，偵測到第二組需要特別關注及即早介入，因為該組睡眠品質在周產期的變化是會持續上升的，甚至在產後三個月時會達到11分。整體而言，與PSQI穩定低組（第一組）相比，最憂鬱的組別為PSQI持續高分組（第四組），接著依序為PSQI穩定中高分組別（第三組），及PSQI逐漸上升組（第二組）。建議在第二孕期測量到兩組高分組後即早給予介入措施，並在第三孕期時進一步區分出第一組與第二組，並針對第二組提供評估與進一步介入。

Background: The global prevalence of perinatal depression has increased by 14% over the past six years, currently reaching 26.3%, with Asia having the second highest rate worldwide. Numerous cross-sectional studies have indicated a correlation between sleep quality and depression during pregnancy or the postpartum period. Furthermore, recent research has begun to categorize perinatal sleep quality into trends, but few studies have explored the impact of sleep quality and its trends from pregnancy to postpartum on perinatal depression. Objective: To understand the trajectories in perinatal sleep quality and perinatal depression among primiparas in Taiwan, and to investigate the impact of perinatal sleep quality and its trend classification on depression. Methods: This longitudinal study involved 482 primiparas and was conducted from January 2022 to April 2023 at three medical centers in northern Taiwan. Structured questionnaires were administered at 12-24 weeks of gestation, 24 weeks of gestation until delivery, one month postpartum, and three months postpartum. Depression was assessed using the Edinburgh Postnatal Depression Scale, and sleep quality was measured using the Pittsburgh Sleep Quality Index, with higher scores indicating worse conditions for both measures. Cutoff points for depression and sleep quality were set at 12/13 and 5/6, respectively. Sleep quality trajectories were classified using the group-based trajectory model (GBTM), and the generalized estimating equation model was used to examine the impact of perinatal sleep quality on depression. Results: The depression rate among primiparas was 12.4% in the second trimester, 13.5% in the third trimester, peaked at 27.1% one month postpartum, and slightly decreased to 22.9% at three months postpartum. Overall, depression scores showed a gradual upward trend over time. In terms of sleep quality, the prevalence of poor sleep quality (PSQI > 5) increased from 57.3% in the second trimester to a peak of 77.8% at three months postpartum. Using group-based trajectory modeling, perinatal sleep quality trends were classified into four groups. The first group, with stable low PSQI scores (52.7%), and the second group, with gradually increasing PSQI scores (13.5%), started with similar scores and both showed a linear increase, but the second group exhibited a more significant increase in depression in the third trimester. The third group, with stable moderate PSQI scores (27.9%, y = 8.45), and the fourth group, with persistently high PSQI scores (5.9%, y = 12.14), had initial scores greater than 6 in the second trimester and remained stable at other perinatal time points. The GEE model, controlling for age, education level, cohabitation status, marital satisfaction, mode of delivery, and neonatal birth weight, found a significant association between perinatal sleep quality and depression (β = 0.37, p < 0.001). When replacing the continuous variable of perinatal sleep quality in the model with the perinatal sleep quality trajectory groups, the perinatal depression symptoms were significantly higher in the second group (β = 1.66, p = 0.005), the third group (β = 2.27, p < 0.001), and the fourth group (β = 5.12, p < 0.001) compared to the first group with stable low PSQI scores. Conclusion: In Taiwan, poorer perinatal sleep quality among primiparas is associated with higher perinatal symptoms. When categorizing perinatal sleep quality trajectories, if the PSQI score exceeds 6 in the second trimester, the pregnant woman is likely to belong to either the third group with stable moderate PSQI scores or the fourth group with persistently high PSQI scores. These groups maintain medium-high or high scores beyond the third trimester, indicating the need for early intervention upon detection in the second trimester. Conversely, if the initial PSQI score in the second trimester is low, the woman may belong to the first group with stable low PSQI scores or the second group with gradually increasing PSQI scores. A second measurement in the third trimester is necessary for these cases. If a pregnant woman’s PSQI score reaches or exceeds 7 in the third trimester, she is likely in the second group with gradually increasing PSQI scores. This group requires special attention and early intervention, as their sleep quality continues to worsen throughout the perinatal period, reaching scores as high as 11 at three months postpartum. In a nutshell, compared to the stable low PSQI group (first group), the group with the highest depression levels is the persistently high PSQI group (fourth group), followed by the stable moderate PSQI group (third group) and the gradually increasing PSQI group (second group). It is recommended to provide early intervention measures upon detecting high scores in the second trimester for the third and fourth groups, and to further distinguish between the first and second groups in the third trimester, with targeted assessment and intervention for the second group.