{"paper_id":"b1914c29-585b-4764-9aad-d4e33d58ab82","body_text":"Infertility is defined as “the inability to achieve pregnancy after 12 months or more of unprotected regular sexual intercourse” [ 1 ]. It significantly impacts the lives of couples [ 2 , 3 ] According to the global infertility epidemiology survey, approximately 50 million people worldwide were affected by infertility in 2010 [ 4 ]. In recent years, the number of individuals with infertility has been steadily increasing [ 5 ]. Substantial evidence indicates that infertility and its treatment constitute a profoundly stressful life experience, frequently accompanied by significant psychosocial burdens including anxiety, depression, and perceived distress [ 6–9 ]. These psychological responses significantly diminish overall life satisfaction and well-being among affected individuals [ 10–12 ].\nRecognizing the substantial psychosocial burden of infertility, the World Health Organization (WHO) emphasizes quality of life (QoL) as a critical outcome measure, defining it as “an individual’s perception of their position in life within the context of the culture and value systems in which they live, and in relation to their goals, expectations, standards, and concerns” [ 13 ]. To standardize the assessment of infertility-specific QoL, the European Society of Human Reproduction and Embryology (ESHRE) and the American Society for Reproductive Medicine (ASRM) collaboratively developed the Fertility Quality of Life (FertiQoL) instrument [ 14 ]. FertiQoL has emerged as the gold-standard tool, extensively implemented in global population-based studies to evaluate QoL among individuals experiencing infertility [ 14–16 ]. The FertiQoL tool comprises a core module and a selective treatment module. The core module evaluates emotional, mind/body, relational, and social domains, while the selective treatment module assesses the treatment environment and treatment tolerance domains. A higher total FertiQoL score indicates better QoL for patients.\nAssisted reproduction technologies (ARTs) are treatments widely used for infertile patients. Research has shown that pregnancy outcomes following embryo transfer (ET) are influenced by factors such as age, body mass index (BMI), emotional distress, and lifestyle [ 17–22 ]. However, although QoL has been extensively studied as a core psychosocial indicator among infertility patients, research examining the association between QoL and post-ET pregnancy outcomes remains extremely limited and underexplored. Some studies have found that emotional domain QoL scores were significantly positively correlated with ongoing pregnancy and live birth rates [ 23 , 24 ]. This finding underscores the critical role of the emotional dimension in predicting pregnancy outcomes, yet the comprehensive impact of multi-domain QoL still requires validation in larger samples. Given this gap in the literature and the significance of our cohort, this study aims to prospectively assess the QoL status of infertile patients using the FertiQoL, with a primary focus on analyzing the influence of their QoL scores—including total scores and domain-specific scores—on pregnancy outcomes following ET.\n\nThis prospective cohort study included women who received  in vitro  fertilization (IVF) or intracytoplasmic sperm injection (ICSI) treatment at the Center for Assisted Reproductive Technology (ART) of Northwest Women’s and Children’s Hospital in the People’s Republic of China between January and June 2020. Researchers administered the FertiQoL questionnaire using paper-based instruments. Participants voluntarily completed the survey on-site during pre-treatment consultations under research coordinator supervision. Completed questionnaires were secured immediately upon completion. Subsequent pregnancy and neonatal outcomes were prospectively tracked over two years through electronic medical records and structured telephone follow-ups. A total of 1,500 questionnaires were distributed, with 1,435 completed, yielding a 96% recovery rate. Data from 1,233 patients were included in the final statistical analysis. Inclusion criteria were as follows: (1) Confirmed diagnosis aligning with international IVF/ICSI indications per ESHRE/ASRM guidelines, including tubal factor infertility (bilateral occlusion or hydrosalpinx), severe male factor (sperm concentration <5 million/mL or total motility <10%), ovulatory disorders (PCOS or WHO Group I-II anovulation), endometriosis (ASRM Stage III-IV), or unexplained infertility with ≥3 consecutive failed intrauterine insemination cycles [ 25 , 26 ]; (2) Patients with voluntary participation in the study, including signing an informed consent form and complying with ethical standards. Exclusion criteria were as follows: (1) Patients unable to complete the questionnaire independently; (2) Patients with canceled transfers, missed follow-ups, or data loss; (3) Patients who underwent intrauterine insemination; (4) Patients with incomplete questionnaire responses. The flow chart of the study participants is shown in  Supplementary Figure S1 . This study was approved by the Institutional Review Board of the Northwest Women’s and Children’s Hospital (Assigned Number: 2019-025).\nThe FertiQoL was developed through collaboration between the ESHRE and the ASRM, along with Merck-Serono S.A., a leading pharmaceutical company ( https://sites.cardiff.ac.uk/fertiqol/ ). It is the First internationally validated instrument for assessing quality of life among Individuals with fertility issues. The FertiQoL tool comprises a core module and a selective treatment module. The Core Module evaluates four critical domains: the Emotional domain assesses affective states including depression, anxiety, jealousy, and anger; the Mind/Body domain measures physical manifestations such as fatigue and pain alongside cognitive-behavioral impairments exemplified by concentration difficulties, daily activity disruption, and delayed life planning; the Relational domain examines consequences for partner dynamics, sexual functioning, and social engagement; while the Social domain appraises environmental stressors encompassing perceived isolation or support from familial, collegial, and social networks. Complementarily, the Treatment Module addresses intervention-related dimensions through its Treatment Environment component, which quantifies satisfaction with clinical infrastructure, healthcare personnel, and informational resources, and its Treatment Tolerability component, which gauges physical burden, treatment-induced discomfort, and functional interference with daily living [ 14 ].\nThe FertiQoL consists of 36 items, which are used to derive six subscales and three total scores: the total FertiQoL score, the core FertiQoL score, and the treatment FertiQoL score (see detailed information at  www.fertiqol.org ). The total FertiQoL score represents the combined QoL from both the core and treatment FertiQoL scores. The core FertiQoL score evaluates fertility-related QoL across the emotional, Mind-body, relational, and social subscales, while the treatment FertiQoL score assesses QoL related to the treatment environment and treatment tolerability. The response format for the 36 items is based on a five-point Likert scale. The FertiQoL total and subscale scores are calculated from 0 to 100, with higher scores indicating better quality of life [ 27 ].\nInformation for each participant was collected. Trained investigators used structured questionnaires to collect baseline data, including sociodemographic characteristics (age, ethnicity, educational status, occupation, annual income, and BMI), as well as the FertiQoL questionnaire. After two rounds of verification, the questionnaire data were collected and entered into EpiData 3.0 using a dual entry method. Data entry was completed by the project team at the Center for ART of Northwest Women’s and Children’s Hospital in the People’s Republic of China. Any errors detected during the data entry process were corrected through telephone follow-up or review of medical records. Baseline data (type of infertility, infertility factors, infertility duration, antral follicle count (AFC), number of oocytes retrieved, fertilization type, type of transfer, ET stage, number of embryos transferred, and quality of ET) and pregnancy outcomes (cumulative live birth rates [CLBRs], biochemical pregnancy, clinical pregnancy, miscarriage, preterm birth, twin birth, mode of delivery, and birth weight) were retrieved and exported from the medical record system.\nOvarian stimulation in ART typically involved the GnRH agonist long protocol or the GnRH antagonist protocol. The choice of protocol and the starting dose relied on the characteristics of patients and the clinician’s experience. Ovulation was triggered using recombinant human chorionic gonadotropin (HCG) when transvaginal ultrasound revealed at least 2 to 3 dominant follicles measuring 17 mm in diameter. Oocyte retrieval was performed 36 h after the trigger injection, guided by ultrasound. Embryos were cultured individually in 50 ml of sequential medium droplets after conventional IVF or ICSI fertilization. The decision to proceed with IVF or ICSI was based on the semen parameters of the male partner. Oocytes were evaluated for successful insemination 16 to 18 h after IVF or ICSI. The choice between fresh ET or frozen embryo transfer (FET) of 1 to 2 cleavage-stage embryos or blastocysts was made according to the patient’s condition and the clinician’s expertise. Luteal support with progesterone supplementation was provided and maintained until 10 weeks of gestation for patients with a normal pregnancy.\nThe primary outcome measure was the cumulative live birth rate (CLBR). Secondary outcomes included the biochemical pregnancy, clinical pregnancy, miscarriage, preterm birth, twin birth, mode of delivery, and birth weight. Definitions used were as follows: (1) Cumulative live birth: Delivery of at least one live-born infant (≥ 22 weeks gestation) during an ovarian-stimulated fresh or subsequent frozen-thawed cycle. (2) Biochemical pregnancy: Serum β-hCG level ≥ 50 IU/L 14 days after ET. (3) Clinical pregnancy: Ultrasound evidence of one or more gestational sacs in the uterus at 6–8 weeks after ET. (4) Miscarriage: Loss of the clinical pregnancy before 22 weeks of gestation. (5) Preterm birth: Delivery occurring between 22 and 37 weeks of gestation. Clinical outcomes were defined according to the 2017 consensus by the ASRM [ 28 ].\nLatent profile analysis (LPA) was conducted using Mplus 8.3 to identify distinct FertiQoL subgroups. Model fit was evaluated using the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), where smaller values indicate superior fit. Classification accuracy was assessed  via  Entropy (range: 0–1), with higher values reflecting greater precision. The optimal number of latent classes was determined using the Lo-Mendell-Rubin likelihood ratio test (LMRT), where a significant p-value ( p  < 0.05) indicated that the k-class model outperformed the (k-1)-class model. Participants were assigned to classes based on maximum posterior probabilities.\nThe means and standard deviations of continuous variables are presented as means, while the rates of categorical variables are presented as rates. Continuous variables were analyzed using Analysis of Variance (ANOVA), and categorical variables were assessed using the chi-square test. When the conditions for the chi-square test were not met, the Fisher’s exact test was employed. Pairwise comparisons among multiple groups were conducted using the Bonferroni method. Multiple regression analyses were used to explore the relationship between the total FertiQoL score and CLBRs. The total FertiQoL score was analyzed both as a continuous variable and as categorical variables. Crude and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. A generalized additive model was applied for smoothed curve fitting, and threshold effects were assessed to determine the cutoff value of the total FertiQoL score that impacts CLBRs. Bootstrap resampling with 500 iterations was performed to internally validate the optimal cutoff values for CLBRs and cesarean section ( Supplementary Table S1, Supplementary Figure S3 ). We applied a directed acyclic graph ( Supplementary Figure S2 ) combined with previous literature to guide the selection of potential confounders.including the woman’s age, the man’s age, ethnic, educational status, annual income, smoking history, gravidity, parity, BMI, type of infertility, infertility factors, infertility duration, AFC Statistical analyses were performed using EmpowerStats ( https://www.empowerstats.net/cn/ ) and R software ( http://www.r-project.org ). A  p -value of <0.05 was considered statistically significant. For pairwise comparisons among multiple groups, a  p -value of <0.0167 was considered significant to account for multiple testing.\n\nThis study included 1233 infertility patients who were classified into three distinct subgroups through LPA of FertiQoL dimensions. Models with 2–6 classes were evaluated, with the 3-class solution selected as optimal based on fit indices and practical significance ( Table 1 ). This model demonstrated strong classification accuracy (Entropy = 0.830), statistically significant improvement over the 2-class model (LMRT,  p  < 0.0001), and an acceptable smallest class size (22.5%). Although 4–6 class models yielded lower AIC and BIC values, they showed reduced entropy and included classes with limited clinical utility. The final classification comprised Low FertiQoL (L-FertiQoL,  n  = 278), Medium FertiQoL (M-FertiQoL,  n  = 676), and High FertiQoL (H-FertiQoL,  n  = 279) groups ( Supplementary Figure S1 ). Significant gradients were observed across all FertiQoL domains ( Figure 1 ). The total FertiQoL score and individual module scores for the three groups are presented in  Table 2 .\nDistribution of characteristics of three potential categories of QoL in infertility patients.\nComparison of fit indicators for latent profile analysis models with 2 to 6 classes.\nNote : AIC, Akaike Information Criterion; BIC, Bayesian Information Criterion; LMRT, Lo-Mendell-Rubin Likelihood Ratio Test.\nComparison of the distribution of FertiQoL score among the three groups.\nSignificant differences were observed in several baseline characteristics across the FertiQoL groups. Compared to other groups, the H-FertiQoL group had older male partners ( p  = 0.007), lower proportions of university-educated patients ( p  < 0.001), and higher rates of secondary infertility ( p  = 0.005). Annual income distributions differed significantly ( p  = 0.001), with both the L-FertiQoL and H-FertiQoL groups showing higher proportions of low-income patients (≤50,000 yuan/year) than the M-group. The L-FertiQoL group exhibited lower gravidity and parity, along with fewer fresh embryo transfers. Ethnicity distributions also varied. Other baseline variables did not show significant differences between groups. Details are presented in  Table 3 .\nBaseline characteristics of infertility patients by FertiQoL subgroup.\nNote : *Fisher’s exact test.\nData are presented as n (%) or mean ± SD. SD, standard deviation; BMI: body mass index; AFC: antral follicle count; IVF:  in vitro  fertilization; ICSI: intracytoplasmic sperm injection; ET: embryos transfer; FET: frozen-thawed embryo transfer; Good quality cleavage-stage embryos were defined as those derived from normally zygote with an embryonic cell number of 7–9 and a fragmentation level of less than 10% on day 3 after fertilization; high-quality blastocysts were defined as those with a score of better than 4CC.\nSignificant differences in CLBRs were observed across FertiQoL groups ( p  = 0.002). The M-FertiQoL group demonstrated the highest CLBR (77.96%), significantly exceeding the L-FertiQoL group (67.63%). Miscarriage rates also differed significantly ( p  = 0.016), with the M-FertiQoL group showing the lowest rate (7.36%). Other pregnancy outcomes, including the biochemical pregnancy, clinical pregnancy, preterm birth, twin birth, mode of delivery, and birth weight did not show statistically significant differences across the three groups. The detailed results are presented in  Table 4 .\nComparison of pregnancy outcomes across FertiQoL profile groups.\nNote : Data are presented as  n  (%) or mean ± SD. SD, standard deviation.\nIn multivariable regression analyses, the total FertiQoL score was positively associated with CLBRs after adjustment (adjusted OR: 1.012, 95% CI: 1.000–1.023,  p  = 0.044), while no significant associations were observed between continuous FertiQoL scores and the remaining pregnancy outcomes ( p  > 0.05). When analyzed categorically, the M-FertiQoL group exhibited significantly higher CLBRs (adjusted OR: 1.704, 95% CI: 1.217–2.380) and lower miscarriage risk (adjusted OR: 0.513, 95% CI: 0.304–0.875) compared to the L-FertiQoL group. The medium group also demonstrated increased cesarean delivery odds (adjusted OR: 1.728, 95% CI: 1.168–2.548). No significant associations were found for biochemical pregnancy, clinical pregnancy, preterm birth, twin birth, or birth weight across FertiQoL profiles. The detailed results are presented in  Table 5 .\nMultivariable regression analysis of FertiQoL score on pregnancy outcomes.\nNote : *Adjusted for age of the woman, age of the man, ethnic, educational status, annual income, smoking history, gravidity, parity, BMI, type of infertility, infertility factors, infertility duration, AFC. OR, odds ratio; CI, confidence interval.\nAnalysis of FertiQoL subscales revealed that higher Mind/Body scores were significantly associated with higher CLBRs (adjusted OR: 1.013, 95% CI: 1.004–1.022) and higher clinical pregnancy rates (adjusted OR: 1.012, 95% CI: 1.002–1.022). Similarly, higher Social scores were significantly associated with higher CLBRs (adjusted OR: 1.010, 95% CI: 1.001–1.019) and higher clinical pregnancy rates (adjusted OR: 1.011, 95% CI: 1.000–1.021). Higher Relational scores were associated with lower twin birth risk (adjusted OR: 0.971, 95% CI: 0.952–0.989). No significant associations were observed between any subscales and biochemical pregnancy, miscarriage, preterm birth, cesarean section, or birth weight outcomes after adjustment. Details are presented in  Table 6 .\nMultivariable regression analysis of FertiQoL subscales and pregnancy outcomes.\nNote : Core subscales: Emotional, Mind/Body, Relational, Social. Treatment subscales: Environment, Tolerability.\n*Adjusted for age of the woman, age of the man, ethnic, educational status, annual income, smoking history, gravidity, parity, BMI, type of infertility, infertility factors, infertility duration, AFC. OR, odds ratio; CI, confidence interval.\nPiecewise linear regression revealed significant threshold effects for two outcomes ( Table 7 ,  Figure 2 ). For CLBRs, a cutoff of 74.306 was identified (log-likelihood ratio test  p  = 0.054 for model fit). Below this threshold, each unit increase in FertiQoL score was associated with 2.1% higher CLBRs (adjusted OR: 1.021, 95% CI: 1.006–1.036), while no association existed above it (adjusted OR: 0.980, 95% CI: 0.947–1.014). Bootstrap internal validation with 500 resampling iterations confirmed the stability of this cutoff ( Supplementary Table S1, Supplementary Figure S3 ). Similarly, for cesarean section, a cutoff of 62.5 was significant (log-likelihood ratio test  p  = 0.039 for model fit), with scores below this threshold showing 3.2% increased odds per unit (adjusted OR: 1.032, 95% CI: 1.005–1.060). No significant threshold effects were detected for biochemical pregnancy, clinical pregnancy, miscarriage, preterm birth, twin birth, or birth weight.\nThreshold effect of total FertiQoL score on cumulative live birth rates and cesarean section.\nNote : Adjusted for age of the woman, age of the man, ethnic, educational status, annual income, smoking history, gravidity, parity, BMI, type of infertility, infertility factors, infertility duration, AFC.\nThreshold effects of FertiQoL scores on pregnancy outcomes using piecewise linear regression.\nNote : *Adjusted for age of the woman, age of the man, ethnic, educational status, annual income, smoking history, gravidity, parity, BMI, type of infertility, infertility factors, infertility duration, AFC. OR, odds ratio; CI, confidence interval; K, optimal cutoff value determined by likelihood ratio test.\n\nThis prospective cohort study employed LPA to classify 1,233 infertility patients into three distinct FertiQoL subgroups (Low, Medium, High). Our primary finding indicates that women with medium FertiQoL scores exhibited significantly higher CLBRs and lower miscarriage risk compared to the Low-FertiQoL group, even after rigorous adjustment for clinical confounders. Crucially, threshold analysis identified 74.306 as the cutoff for the total FertiQoL score, below which each unit increase was associated with significantly higher CLBRs (2.1% per unit). These results reveal a non-linear relationship between QoL and live birth outcomes, suggesting that optimal pregnancy success is associated with moderate QoL levels.\nThere has been limited research on the relationship between QoL in infertility patients and pregnancy outcomes after ET. Santoro et al. investigated the QoL in patients with unexplained infertility (UI) and polycystic ovary syndrome (PCOS) and found that the score of the emotional subscale of the FertiQoL was positively correlated with live birth rates and conception chances in PCOS patients, though the significance was unclear in UI patients [ 29 ]. The inconsistent results led Santoro et al. to suggest that QoL might not be a crucial predictor of pregnancy outcomes [ 29 ]. In contrast, Yoko Urata et al. found that in women undergoing IVF treatment, a higher FertiQoL treatment module score was associated with a higher rate of positive gestational sac detection after embryo transfer [ 30 ]. This study also found a significant positive correlation between the total FertiQoL score and CLBRs. Given that low-QoL in infertility patients is often associated with severe emotional distress, including stress, anxiety, and depression [ 31–33 ]. Notably, our results align with longitudinal studies by Wu et al. where higher emotional domain scores predicted increased ongoing pregnancy and live birth rates in IVF patients [ 24 ], underscoring the critical role of psychological well-being.\nThis study found that while the M-FertiQoL group exhibited significantly higher CLBRs and lower miscarriage risk compared to the L-FertiQoL group, no such advantages were observed in the H-FertiQoL group despite their superior FertiQoL scores. We propose two mechanistic explanations for this phenomenon. First, patients in the M-FertiQoL range may experience moderate stress levels that enhance physiological adaptation (e.g. optimized HPA axis activity promoting oocyte maturation [ 34 ], whereas L-FertiQoL individuals endure overwhelming distress that suppresses reproductive function through cortisol dysregulation [ 9 ]. Conversely, H-FertiQoL patients’ minimized distress may reflect avoidant coping strategies—such as disengagement from treatment realities—potentially compromising treatment adherence or self-advocacy. Second, the M-FertiQoL group’s relatively higher socioeconomic status likely facilitates resource access (e.g. timely appointments, nutritional support), while their balanced psychological engagement promotes treatment compliance [ 6 , 7 , 35 ]. In contrast, H-FertiQoL patients’ heightened sense of control may lead to underestimation of medical risks, mirroring findings that extreme optimism correlates with poorer ART adherence [ 36 ].\nEmotional distress, such as stress, pain, and anxiety, has been shown to impact pregnancy outcomes following ET [ 18 , 31 , 32 ], and biomarkers of stress, such as salivary α-amylase, are negatively associated with fertility [ 33 ]. As shown in network analyses of Chinese infertility patients, symptoms like “social withdrawal” and “self-devaluation” strongly correlate with reduced FertiQoL scores [ 37 ]. Previous studies suggest that negative emotions may activate the hypothalamic-pituitary-adrenal (HPA) axis [ 38 ], leading to the release of stress hormones, such as cortisol, which inhibit the reproductive system. The activation of the HPA axis could impair estradiol production, damage granulosa cells, disrupt follicular maturation, and reduce the number of oocytes [ 34 ], ultimately affecting pregnancy outcomes. Therefore, this study suggests that emotional distress may contribute to lower CLBRs in patients with poorer QoL. Our threshold effect suggests that mitigating distress below this critical point may optimize physiological conditions for pregnancy. Furthermore, controlled ovarian stimulation used in ART is associated with the risk of ovarian hyperstimulation syndrome (OHSS), a serious iatrogenic complication, and oocyte vitrification has been proposed as an effective strategy for fertility preservation that may also reduce this risk [ 39 , 40 ]. Nutraceutical supplementation with inositols has also been reported to exert beneficial biological effects on oocyte quality and reproductive outcomes in infertile patients [ 41 ].\nThis study had several strengths. Firstly, the questionnaire was administered to patients before ART treatments, which might effectively reduce recall bias. Secondly, the study included a relatively large sample size, accounted for multiple confounding factors, and employed both continuous and categorical analyses of the total FertiQoL score. Smoothed curve fitting and threshold effect analyses were used to determine the cutoff value for CLBRs, enhancing the accuracy of the results. Third, using data directly from the clinic minimized observational bias.\nHowever, several limitations warrant consideration. First, despite adjusting for numerous confounders through multivariable regression analyses, unaccounted factors—including lifestyle behaviors (e.g. sleep patterns [ 22 ] and psychosocial confounders—may introduce residual bias. Notably, our exclusion criteria omitted patients with diagnosed mental disorders (e.g. major depression or anxiety disorders) or those experiencing major traumatic life events (e.g. bereavement, job loss, or recent failed ART cycles/pregnancy loss) within 6 months preceding QoL assessment. These factors are established determinants of psychological well-being and QoL metrics, potentially confounding observed associations. Second, while we adjusted for broad infertility etiologies (e.g. tubal, male, ovulatory), granular data regarding prognostically significant diagnostic specifics were unavailable (e.g. endometriosis staging, uterine pathology severity). Additionally, prior ART cycle history—including the number of failed embryo transfers—was not documented, despite its known influence on cumulative live birth rates and potential interaction with QoL. Third, persistent baseline differences between FertiQoL groups (e.g. male partner age, education level, infertility type, income) after statistical adjustment raise concerns about residual confounding, potentially affecting association interpretability. Finally, the exclusive focus on pregnancy outcomes precluded examination of neonatal/perinatal outcomes or pregnancy complications. As a single-center investigation, generalizability requires validation across diverse clinical settings.","source_license":"CC-BY-4.0","license_restricted":false}