Sleep disorders have emerged as a prominent public health challenge, exerting widespread impacts on both individuals and society. Globally, insufficient sleep is highly prevalent across all age groups, with approximately 30–40 % of adults suffering from insomnia, among whom 10–15 % meet the diagnostic criteria for chronic insomnia (undefined, u. and u. undefined, 2024; Børge et al., 2021). Notably, sleep disturbances exhibit distinct patterns across demographic groups. Adolescents are more prone to sleep deprivation and circadian rhythm disruptions (CRDs), whereas women demonstrate higher susceptibility to insomnia, and men show increased risk of obstructive sleep apnea (OSA) (Valeria et al., 2024; Elizabeth et al., 2024; Olaf et al., 2024). Beyond impairing physical and mental health, sleep disorders are closely associated with various chronic conditions, including cardiovascular diseases (CVDs), metabolic disorders, and psychiatric conditions such as depression and anxiety (Jaspan et al., 2024; Pikula et al., 2024; Singh et al., 2025). Moreover, sleep disturbances contribute to reduced work efficiency, heightened traffic accident risks, and substantial socioeconomic burdens (Rebecca et al., 2022).

Sleep quality demonstrates a robust association with mental health (Cheng et al., 2018; Liu et al., 2024; Yin et al., 2021). Empirical evidence indicates that poor sleep quality impairs emotional regulation capacity and increases susceptibility to negative affect, thereby exacerbating anxiety and depressive symptoms (Whiting et al., 2023). Sleep deprivation (SD) has been shown to disrupt normative levels of key neurotransmitters involved in mood regulation, including serotonin and dopamine (Singh et al., 2025). Furthermore, sleep restriction (SR) activates the stress response system, leading to elevated cortisol levels that may exert deleterious long-term effects on mental health (Messa et al., 2024). Notably, improving sleep efficiency can alleviate clinical symptoms in patients with major depressive disorder (MDD) through neuromodulatory mechanisms, partially attributable to the functional reorganization of prefrontal cortex-limbic system (PFC-limbic) connectivity (Cheng et al., 2018). Neuroimaging studies have provided mechanistic insights into this relationship. Functional magnetic resonance imaging (fMRI) investigations have demonstrated significant associations between sleep dysfunction and aberrant resting-state functional connectivity (rsFC) in brain regions implicated in emotion regulation. Specifically, aberrant rsFC in the hypothalamic-insula pathway has been found to mediate the impact of sleep quality on anxiety and depressive symptoms (Li et al., 2023). Further evidence indicates that low sleep efficiency is associated not only with altered functional connectivity between the cuneus and temporal cortex but also with impaired white matter integrity in the corona radiata and posterior limb of the internal capsule (Zhu et al., 2020; Yang et al., 2020). Moreover, rapid eye movement (REM) sleep abnormalities, a hallmark feature of depression, are correlated with decreased voxel-mirrored homotopic connectivity (VMHC) in the precentral gyrus and inferior parietal lobe as well as multimodal brain alterations in sensorimotor and visual processing regions (Liu et al., 2025; Zhang et al., 2024). Collectively, these findings reveal that sleep disturbances contribute to the development and maintenance of depressive and anxiety symptoms through their effects on specific neural circuit functions.

Sleep quality demonstrates a robust association with mental health (Cheng et al., 2018; Liu et al., 2024; Yin et al., 2021). Empirical evidence indicates that poor sleep quality impairs emotional regulation capacity and increases susceptibility to negative affect, thereby exacerbating anxiety and depressive symptoms (Whiting et al., 2023). Sleep deprivation (SD) has been shown to disrupt normative levels of key neurotransmitters involved in mood regulation, including serotonin and dopamine (Singh et al., 2025). Furthermore, sleep restriction (SR) activates the stress response system, leading to elevated cortisol levels that may exert deleterious long-term effects on mental health (Messa et al., 2024). Notably, improving sleep efficiency can alleviate clinical symptoms in patients with major depressive disorder (MDD) through neuromodulatory mechanisms, partially attributable to the functional reorganization of prefrontal cortex-limbic system (PFC-limbic) connectivity (Cheng et al., 2018). Neuroimaging studies have provided mechanistic understanding into this relationship. Functional magnetic resonance imaging (fMRI) investigations have demonstrated significant associations between sleep dysfunction and aberrant resting-state functional connectivity (rsFC) in brain networks implicated in emotion regulation, such as the default mode network (Willoughby et al., 2024; Datta et al., 2023), the prefrontal-limbic circuit (Cheng et al., 2018), and the hypothalamic-insula pathway (Li et al., 2023; Kreutz et al., 2021). Alterations in these neural circuits, including those related to circadian regulation (Wang et al., 2022), are suggested to mediate the impact of sleep quality on anxiety and depressive symptomatology.

Sleep quality serves as a crucial indicator for assessing mental health status, and the establishment of its multidimensional evaluation system holds significant clinical value for early identification and intervention of psychiatric disorders (Chauhan et al., 2024; Appleton et al., 2022). The Pittsburgh Sleep Quality Index (PSQI), currently recognized as the international gold-standard assessment tool, provides a comprehensive framework for evaluating sleep quality through seven core dimensions: subjective sleep quality, sleep onset latency, sleep duration, sleep efficiency, sleep disturbances (SDi), use of sleeping medication (USM), and daytime dysfunction (DD) (Buysse et al., 1989). Sleep efficiency reflects the capacity to maintain sleep continuity and has been identified as an important clinical marker for screening sleep disorders (Jennifer et al., 2022; Andreas et al., 2017). Accumulated evidence demonstrates significant intercorrelations between sleep efficiency and other PSQI dimensions. For instance, several studies have found positive correlations between sleep efficiency and subjective sleep quality, while negative correlations exist between sleep efficiency and SL (María del Carmen et al., 2019). However, factors such as individual psychological state and sleep expectations may influence subjective sleep quality assessments, whereas objective sleep efficiency measurements rely more on actual physiological data. This discrepancy may lead to negative correlations between the two measures (Masaki et al., 2025; Pierson-Bartel and Ujma, 2024). Furthermore, the relationship between sleep duration and sleep efficiency follows an inverted U-shaped curve. Optimal sleep efficiency is achieved with moderate sleep duration, while both insufficient and excessive sleep duration result in decreased sleep efficiency (Yazdanpanah et al., 2020; Chih Chiang Benjamin et al., 2024).

Sleep efficiency, traditionally referred as Habitual Sleep Efficiency (H-SE), is widely used in sleep medicine and psychological research that to some extent reflects the relationship between sleep quality and mental health. H-SE is calculated as the ratio of Sleep Duration to Time in Bed (H-SE = Sleep duration/ Time in Bed × 100 %). However, accumulating evidence has revealed limitations in conventional H-SE calculations, particularly regarding their applicability to modern sleep behaviors. Contemporary populations frequently engage in wakefulness-promoting pre-sleep activities—including electronic device usage, reading, and video viewing—which substantially inflate Time in Bed measurements (Reed and Sacco, 2016; Willoughby et al., 2024). Pre-sleep electronic device use has been demonstrated to prolong sleep onset latency, thereby artificially reducing calculated H-SE values, though this phenomenon may predominantly reflect behavioral habits rather than genuine sleep disturbances (Willoughby et al., 2024). The inclusion of extensive non-sleep activities in Time in Bed systematically contaminates the denominator in H-SE calculations, resulting in underestimated efficiency values that risk misclassifying behavioral issues such as bedtime procrastination as physiological insomnia, ultimately compromising diagnostic accuracy and therapeutic efficacy (Reed and Sacco, 2016). Furthermore, H-SE fails to account for key determinants of sleep quality, including sleep environment comfort, pre-sleep activity profiles, and sleep duration effects (Datta et al., 2023), as well as critical sleep architecture parameters such as sleep depth and nocturnal awakening frequency (Kreutz et al., 2021). To address these limitations, this study innovatively proposed an adjusted sleep efficiency metric (A-SE) with a novel algorithm: A-SE = [Sleep duration/ (sleep duration + sleep onset latency)] × 100 %, which operationally distinguishes "sleep preparation time" from "effective sleep time" through mathematical reformulation. Given the predominant reliance on cross-sectional designs in existing literature, this study employed longitudinal datasets to comprehensively investigate bidirectional temporal relationships between sleep efficiency, depression, and anxiety, while systematically accounting for the moderating effects of sleep duration and demographic covariates (gender and age). Furthermore, we conducted comparative analyses of H-SE and A-SE associations with other sleep quality dimensions and rigorously evaluated their respective goodness-of-fit within longitudinal modeling frameworks. These advancements establish an empirical foundation for developing precision sleep interventions targeting both sleep efficiency optimization and sleep duration adequacy.