Suicide remains one of the leading causes of death among adolescents globally. A population-based study spanning 59 low- and middle-income countries estimated that roughly 17 percent of adolescents aged 13 to 17 report a suicide attempt within a twelve-month window. In China, a meta-analysis covering more than 1.1 million children and adolescents estimated prevalence rates of 15.4 percent for suicidal ideation, 6.4 percent for suicide plans, and 3.5 percent for suicide attempts. Because of the country’s enormous population, even these comparatively moderate figures translate into a serious public health burden, which is precisely why the new cohort’s scale matters so much.

The study drew on a school-based prospective cohort conducted in Yunfu, Guangdong Province, involving 393 primary and secondary schools. Baseline data collection began in October 2022, with follow-up twelve months later in October 2023. After data cleaning and accounting for attrition of about 40 percent, largely driven by school transitions in graduating grades, the final analytic sample comprised 92,120 adolescents, just over half of them female, with a mean age of 13.27 years. Ethical approval came from Tsinghua University, and informed consent was obtained from both the adolescents and their parents or guardians, with referral and psychological support services available to those screened as at risk.

IBS was assessed using a validated five-item, symptom-based screening tool grounded in the Rome III criteria, reviewed by a gastroenterologist before administration. Adolescents reporting abdominal pain or discomfort at least three days per month over the previous three months, without a transient infectious explanation, were classified as having IBS if they also met at least two hallmark criteria: relief after defecation, symptom onset linked to changes in stool frequency, or onset linked to changes in stool form. Suicide risk was measured with the Chinese version of the Suicide Behaviours Questionnaire-Revised, using a cut-off score of seven, while suicide attempts were captured with a direct item asking whether the respondent had attempted suicide in the past twelve months.

At baseline, 7.0 percent of participants met the IBS criteria, 14.7 percent showed elevated suicide risk, and 4.1 percent reported a suicide attempt in the previous year. In the simplest unadjusted models, adolescents with IBS had more than twice the odds of later suicide risk (odds ratio 2.33) and suicide attempts (odds ratio 2.17) compared with their peers. Critically, the association survived a battery of statistical adjustments. In the fully adjusted model, controlling for sex, age, residence, socioeconomic status, psychiatric history, smoking, alcohol use, and baseline suicidality, adolescents with IBS still showed 27 percent higher odds of suicide risk and 21 percent higher odds of suicide attempts at follow-up.

The robustness of the signal was striking. Subgroup analyses found the IBS-suicide risk association significant across every demographic stratum examined: males and females, early, middle, and late adolescence, urban and rural residents, only children and those with siblings, and all socioeconomic groups. For suicide attempts, the association held in nearly all subgroups, with the exceptions of only-child adolescents and the highest socioeconomic group, a pattern the authors interpret as possibly reflecting greater parental attention and healthcare access that buffer the transition from vulnerability to behaviour. Sensitivity analyses restricted to incident outcomes, meaning adolescents without the relevant outcome at baseline, and alternative missing-data strategies including complete-case, k-nearest-neighbours, and random forest imputation, all reproduced the primary findings.

Beyond the headline association, the study probed mechanism through exploratory mediation analyses. Depressive symptoms partially mediated the link between IBS and later suicidality, accounting for roughly 35 percent of the total effect on suicide risk and 50 percent of the effect on suicide attempts; anxiety symptoms mediated approximately 26 percent and 43 percent of the corresponding effects. Notably, the reverse pathway, in which IBS was tested as a mediator between emotional symptoms and suicidality, was not supported. This asymmetry suggests that chronic visceral distress, such as recurrent and unpredictable abdominal pain, may act as a persistent physiological and psychological stressor that erodes interoceptive processing and fuels depressive and anxious states, which in turn elevate suicide vulnerability.

The authors situate these findings within the gut–brain axis framework, a bidirectional communication network linking the enteric and central nervous systems through visceral and interoceptive signalling, operating via inflammatory, neuroendocrine, and neurotransmitter pathways. Adolescence may be a period of particular susceptibility, since the microbiota–gut–brain axis and interoceptive processing remain highly plastic during this developmental window, while adolescents face mounting physical, psychological, and social stressors known to alter gut microbial composition. The adjusted models also revealed familiar demographic gradients: female adolescents carried roughly double the odds of both outcomes, baseline suicide risk strongly predicted follow-up risk, and urban residence and lower socioeconomic status were associated with higher suicide risk but not with attempts, a divergence consistent with ideation-to-action theories holding that volitional factors gate the passage from thought to behaviour.

The clinical and cultural implications may be the study’s most consequential contribution. In contexts such as China, where adolescents are often reluctant to disclose psychological distress but more willing to report physical complaints, IBS symptoms may be more readily noticed by parents and teachers, offering a culturally acceptable entry point for early identification. The researchers argue that mental health screening and suicide risk assessment should be integrated into routine pediatric IBS management, and that population-based strategies should bridge gastroenterological care and adolescent mental health services. Prior adult evidence, including a 2026 meta-analysis estimating 14 percent prevalence of suicidal ideation among adults with IBS, aligns with this framing, suggesting the adolescent findings extend a continuum already visible in older populations.

The authors are candid about limitations. The data were self-reported, IBS was not verified against medical records, and organic gastrointestinal disease or anxiety-related gut symptoms could not be fully excluded. Attrition, residual confounding by factors such as sleep, diet, family environment, and academic stress, and a two-wave design that limits causal inference about mediation all warrant caution, as do the modest effect sizes in fully adjusted models, which the team notes should not be overstated clinically even though small probability differences matter at population scale for low-base-rate, high-severity outcomes. The sample, drawn from a single province, may not generalise to all Chinese adolescents. Even so, the consistency of the association across subgroups and sensitivity analyses marks a genuine advance: a common, identifiable, and treatable somatic condition now stands revealed as a possible beacon for suicide prevention in youth, provided clinicians and schools learn to read the gut’s signals.

Subject of Research: Prospective association between irritable bowel syndrome and suicide risk among adolescents

Article Title: Irritable bowel syndrome and suicide risk among adolescents: a prospective cohort study

Article References: Li, W., Jin, Y., Ye, Z., Ren, Y., Shi, Z., & Chen, R. (2026). Irritable bowel syndrome and suicide risk among adolescents: a prospective cohort study. The Lancet Regional Health – Western Pacific, 74, Article 101984. https://doi.org/10.1016/j.lanwpc.2026.101984

Image Credits: AI Generated

DOI: 10.1016/j.lanwpc.2026.101984

Keywords: irritable bowel syndrome, adolescents, suicide risk, gut-brain axis, prospective cohort, depressive symptoms, anxiety symptoms, China, mental health, suicide prevention, interoception, public health

The instrument in question is the Children’s Response Styles Questionnaire, or CRSQ, a 25-item self-report scale adapted from the adult Response Styles Questionnaire. Developed originally in Canada by Abela and colleagues, the CRSQ asks young people what they typically do and think when they feel sad, with each item rated on a four-point scale from ‘Almost never’ to ‘Almost always’. Items such as ‘Think about how alone you feel’ tap into rumination, while items such as ‘Watch TV or play video games so you don’t think about how sad you are’ capture distraction and problem-solving. Although the questionnaire has been validated in samples from the United States, Turkey, Hong Kong and its native Canada, no one had systematically tested whether it works as intended in British adolescents — a gap that matters, because words and concepts can shift meaning across cultures even within the same language.

To close that gap, researchers led by Olivia Adams of the University of Oxford, together with Patrick Smith of King’s College London, Bao Sheng Loe of the University of Cambridge, and Felicity Waite and Eleanor Leigh of Oxford, pooled data from two school-based surveys conducted in South London. Of 763 young people aged 11 to 16 who took part, 615 provided complete responses to the CRSQ and formed the analytical sample. The group had a mean age of 13.32 years and was composed of 421 boys, 325 girls and 17 participants who did not report their gender. Notably, the sample was ethnically diverse: among those recording ethnicity in the first dataset, more than half identified as Black, Black British, Black Welsh, Caribbean or African — a proportion far exceeding the 4.2 percent recorded in the 2021 England and Wales census — making the study a valuable test of the measure in a population often underrepresented in psychometric research.

The analytical strategy was thorough. Because the questionnaire’s items are answered on an ordinal four-point scale, the team used weighted least squares estimation within a confirmatory factor analysis framework, testing whether the data supported the two-factor structure suggested by earlier studies: one factor for rumination and a combined factor for distraction and problem-solving, abbreviated DPS. Before pooling the two school datasets, the researchers confirmed through multigroup analysis that the questionnaire measured the same constructs in both. Bartlett’s test of sphericity and the Kaiser-Meyer-Olkin index, which came in at a strikingly high 0.91, confirmed the data were well suited to factor analysis in the first place.

The initial two-factor model fit imperfectly, and the refinement process proved revealing. Five items were dropped, all of them from the DPS subscale: ‘Spend a lot of time on schoolwork’, ‘Do something you enjoy’, ‘Do something fun with a friend’, ‘Read a book or magazine’ and ‘Ask a friend, parent or teacher to help you solve your problem’. The researchers suggest a common thread: unlike most other items, these make no explicit reference to feeling sad or to an event that triggered sadness. A conscientious student, for instance, might spend a lot of time on schoolwork for reasons having nothing to do with mood, diluting the item’s connection to the underlying construct. After removing the weak items and allowing two help-seeking items to share correlated error terms, the final model achieved strong fit, with a Comparative Fit Index of 0.985 and a Root Mean Square Error of Approximation of 0.062.

Reliability figures reinforced the asymmetry between the two halves of the scale. The rumination subscale showed excellent internal consistency, with Cronbach’s alpha exceeding 0.9, while the DPS subscale, though still above the conventional 0.7 threshold, was noticeably weaker — unsurprising given that five of its items had been removed. The researchers are candid about the implication: future use of the CRSQ with adolescents may be best restricted to the rumination subscale alone, with any analyses involving the DPS treated as exploratory until the subscale’s structure can be replicated. This echoes earlier concerns that the distraction construct, inherited from the adult questionnaire, may be poorly operationalised, bundling together effective and ineffective coping strategies.

A central question for any questionnaire intended for diverse groups is whether it functions identically across them — the property psychometricians call measurement invariance. Using multiple-group confirmatory factor analysis for categorical data, the team compared boys and girls, and separately compared younger and older participants split at the median age of roughly 13.3 years. For age, full measurement invariance was achieved across configural, threshold, metric and scalar levels, meaning the questionnaire measures rumination the same way whether a respondent is 11 or 15. For gender, the scalar model initially faltered, but freeing thresholds for two items restored adequate fit, yielding partial invariance — sufficient, the authors note, for comparing mean scores between boys and girls without serious concern that the construct itself is being measured differently.

The invariance findings proved substantively interesting in their own right. Girls scored higher than boys on the rumination subscale, replicating a well-documented gender difference that emerges in adolescence and that previous research has linked to the widening depression gap between the sexes. Older participants also ruminated more than younger ones, a statistically significant difference consistent with the idea that rumination intensifies across the teenage years. Meanwhile, the two factors correlated modestly and positively with each other, a pattern the authors attribute to shared variance in mood-related responding rather than any meaningful overlap between dwelling and distraction.

Convergent validity — whether the questionnaire’s scores track related constructs as theory predicts — was examined against four established symptom measures: the Children’s Depression Inventory and the Multidimensional Anxiety Scale for Children in one dataset, and the Mood and Feelings Questionnaire and the Screen for Child Anxiety Related Disorders in the other. Rumination scores correlated moderately and positively with both depression and anxiety symptoms, and crucially, these correlations survived statistical controls. The link between rumination and anxiety remained significant after accounting for depression, and the link between rumination and depression remained significant after accounting for anxiety, with partial correlations of 0.36 and 0.45 respectively. This pattern supports the emerging view of rumination as a transdiagnostic process — a single cognitive habit that feeds into multiple forms of emotional distress rather than depression alone.

The study is not without limitations, which the authors enumerate with unusual thoroughness. The sample came from a single South London borough, so replication across more rural areas and the other UK nations is needed; ethnicity data were missing for one dataset; socioeconomic information was absent; and only binary gender was recorded. The cross-sectional design precludes test-retest and predictive validity checks, and no independent measure of rumination was available to strengthen convergent validity. The combination of two datasets, while statistically justified, means independent replication would be welcome. Most importantly, the sample was a community one, and given rumination’s proposed role as a target for both prevention and treatment, testing the measure in clinical populations — ideally alongside diagnostic interviews rather than self-report alone — is the natural next step.

Even so, the implications are considerable. Depression affects between seven and twenty-four percent of adolescents worldwide, raises the risk of adult depression nearly threefold, and undermines physical health and academic attainment. If repetitive negative thinking is indeed a modifiable mechanism in this cascade, then researchers and clinicians need instruments they can trust to detect it. This study provides the first UK-specific evidence that the CRSQ’s rumination subscale is both reliable and valid in British teenagers, invariant across age and largely invariant across gender, and meaningfully connected to the symptoms it should predict. The caution attached to the distraction and problem-solving subscale is itself a service to the field, steering future studies away from a component that may mislead. As the authors conclude, ensuring that measures are psychometrically robust is a quiet but essential foundation for the larger project of understanding — and eventually interrupting — the cognitive pathways that carry young people into depression.

Subject of Research: Psychometric validation of the Children's Response Styles Questionnaire for measuring adolescent rumination in UK youth

Article Title: Psychometric Validation of the Children's Response Styles Questionnaire in UK Adolescents

Article References: Adams, O., Smith, P., Loe, B. S., Waite, F., & Leigh, E. (2026). Psychometric Validation of the Children's Response Styles Questionnaire in UK Adolescents. Child Psychiatry & Human Development. https://doi.org/10.1007/s10578-026-02084-6

Image Credits: AI Generated

DOI: 10.1007/s10578-026-02084-6

Keywords: rumination, adolescent depression, psychometrics, Children's Response Styles Questionnaire, measurement invariance, UK adolescents, anxiety symptoms, Response Styles Theory, confirmatory factor analysis, mental health, repetitive negative thinking, questionnaire validation

The research, led by Gustavo Garrido and colleagues at the Bahiana School of Medicine and Public Health in Salvador, Brazil, followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Cochrane guidelines, and was prospectively registered in the PROSPERO database under registration number CRD42024579831. The team searched three major biomedical databases—PubMed, Embase, and the Cochrane Library—to identify studies that had examined the relationship between sleep paralysis and anxiety. After screening, eleven eligible studies encompassing a total of 5,568 participants made it into the final analysis. Study quality was appraised with the Newcastle-Ottawa Scale, a standard instrument for judging the methodological rigor of nonrandomized research, and study selection was managed using the Rayyan screening platform.

The statistical approach was deliberately conservative. Rather than pooling data with a single fixed estimate that assumes all studies measure the same underlying effect, the investigators used random-effects models estimated with restricted maximum likelihood (REML), a method that accommodates genuine variability between studies. Two complementary outcomes were examined. The first compared continuous anxiety scores between people who experience sleep paralysis and those who do not, expressed as a standardized mean difference (SMD). The second compared the prevalence of formally diagnosed anxiety disorders between the two groups, expressed as a risk ratio (RR). Heterogeneity among studies was quantified with the I-squared statistic, publication bias was assessed with funnel plots and Egger’s test, and the robustness of the pooled estimates was probed with subgroup and leave-one-out sensitivity analyses. All computations were performed in R version 4.4.1 using the metafor package, a widely respected open-source tool for meta-analytic work.

The headline result concerns anxiety symptoms. Across the pooled studies, individuals who experience sleep paralysis scored significantly higher on measures of anxiety than controls, with a standardized mean difference of 0.35 (95 percent confidence interval 0.27 to 0.44, p < 0.01). In the language of meta-analysis, an SMD of this magnitude represents a small-to-moderate but reliable effect. What makes the finding especially striking is the heterogeneity figure: I-squared was 0.0 percent, meaning the studies were in remarkable agreement despite differences in populations, instruments, and settings. In behavioral science, where pooled effects often scatter wildly, a zero heterogeneity value is rare and lends unusual confidence to the conclusion. Publication bias, the tendency for journals to favor positive findings, was judged to be low, further strengthening the result.

The picture was more nuanced for diagnosed anxiety disorders. When the prevalence of formal anxiety conditions was compared between people with and without sleep paralysis, the pooled risk ratio of 1.28 (95 percent confidence interval 0.59 to 2.77, p = 0.53) did not reach statistical significance. However, the sensitivity analysis told a different story. When a single outlying study was excluded, the estimate shifted to a risk ratio of 1.80 (95 percent confidence interval 1.08 to 2.99), indicating that people with sleep paralysis face nearly double the risk of having an anxiety disorder. The discrepancy between the primary and sensitivity analyses underscores how a single influential dataset can obscure an underlying pattern, and the authors concluded that sleep paralysis appears to occur more frequently among individuals with anxiety-related conditions even though the evidence for diagnosis-level prevalence is less consistent than the evidence for symptom levels.

Understanding why this association exists requires a brief tour of sleep neurobiology. During healthy REM sleep, the brainstem actively inhibits motor neurons, producing the near-total paralysis that prevents us from acting out our dreams. Sleep paralysis occurs when this atonia persists into a state of partial or full wakefulness during the transition between sleep and waking—either while falling asleep (hypnagogic) or upon awakening (hypnopompic). Crucially, the fear circuitry of the brain, including the amygdala, is highly active during REM sleep, and Dreaming during this stage is disproportionately negative in emotional tone. When a person regains awareness while still paralyzed, that raw, unfiltered fear state can merge with waking consciousness, producing the intense terror, chest pressure, and hallucinatory intrusions—shadowy figures, sensed presences, incubi—that define the classic episode. An anxious brain, primed to detect threat, may both experience these episodes more frequently and find them more traumatizing when they occur.

The causal arrow, however, almost certainly points in both directions. Prior research has established that anxiety disorders, post-traumatic stress disorder, panic attacks, and pathological worry are all more common among people who report sleep paralysis, and experimental work has documented objective REM-sleep abnormalities in patients with recurrent isolated sleep paralysis. Chronic anxiety disrupts sleep architecture, fragments sleep, and increases nighttime awakenings—conditions that raise the probability of a REM intrusion into wakefulness. Conversely, the episodes themselves are aversive and frightening enough to generate anticipatory fear of sleep, perpetuating insomnia and reinforcing the anxiety loop. Earlier epidemiological work has also tied sleep paralysis to depression and to poor sleep quality more broadly, and studies of high-stress occupational groups such as firefighters have reported elevated rates of the phenomenon in the context of trauma exposure.

The clinical implications of the new synthesis are tangible. Lifetime prevalence of sleep paralysis in the general population has been estimated at roughly 8 percent in earlier systematic reviews, with substantially higher rates among students and psychiatric populations, meaning the phenomenon is far from rare. Sleep medicine specialists emphasize that isolated sleep paralysis is benign in itself, but the new findings suggest that clinicians encountering patients with recurrent episodes should screen for anxiety symptoms rather than treating the parasomnia in isolation. Because the effect on anxiety scores was consistent across all included studies, even modest elevations may accumulate into meaningful distress, particularly given that anxiety disorders impose substantial burdens on quality of life worldwide and their global prevalence has been rising. Behavioral interventions—improving sleep hygiene, regularizing sleep schedules, addressing sleep deprivation, and treating underlying anxiety—may simultaneously reduce both the frequency of episodes and the emotional suffering surrounding them.

The researchers are careful to note the limitations inherent to observational synthesis. The included studies were largely cross-sectional, so the data cannot determine whether anxiety causes sleep paralysis, sleep paralysis fuels anxiety, or shared vulnerabilities—such as genetic predisposition, trauma history, or disrupted REM regulation—drive both. Self-reported sleep paralysis also depends on accurate recall and honest disclosure of an experience many people find embarrassing or frightening to describe. Still, with eleven studies, nearly 5,600 participants, zero heterogeneity in the symptom analysis, and low publication bias, the association itself now rests on a firmer statistical foundation than ever before. For the millions who wake in the small hours, frozen and afraid, the message of this research is oddly comforting: the night-mare is not a supernatural visitation but a measurable intersection of REM physiology and emotional health—and one that science is steadily learning to name, explain, and treat.

Beyond the pooled statistics, the composition of the underlying evidence base offers useful perspective on how the association was detected. Several of the included studies focused on university students, a population in which sleep paralysis is consistently reported at elevated rates, while others examined clinical groups such as outpatients with panic attacks or anxiety disorders. Work among Egyptian college students, for example, linked sleep paralysis to trait anxiety, pathological worry, and post-traumatic stress symptoms, and a survey of Polish students identified comparable psychological risk factors. Studies of African American samples with panic disorder contributed some of the earliest systematic documentation of isolated sleep paralysis in psychiatric settings, and a large international survey has since catalogued the clinical features and coping strategies people use to disrupt episodes, such as attempting small movements or focusing on breathing.

The measurement instruments themselves matter for interpretation. Because sleep paralysis is typically ascertained through retrospective self-report questionnaires rather than overnight laboratory recording, reported prevalence can vary with how questions are framed and how vividly episodes are remembered. One case-control study that did use polysomnography found objective differences in REM sleep characteristics among people with recurrent isolated sleep paralysis, lending physiological credibility to accounts that might otherwise be dismissed as exaggerated. Standardized tools such as the Unusual Sleep Experiences Questionnaire have been developed specifically to capture the phenomenology of these episodes, separating paralysis from the hallucinatory and fear components that often dominate the experience.

For future research, the authors’ findings point toward longitudinal designs that can disentangle temporal ordering, as well as interventions that target both sleep continuity and anxiety. If disrupted or irregular sleep increases the likelihood of REM intrusions, then trials testing whether anxiety treatment reduces episode frequency would provide a direct test of the mechanistic pathway suggested by this meta-analysis.

Subject of Research: The association between sleep paralysis and anxiety examined through a systematic review and meta-analysis

Article Title: Association between sleep paralysis and anxiety: a systematic review and meta-analysis

Article References: Garrido, G., Donato, K., Gonzalez, J. V., Brito, G. N. E., Romeo, B., Machado, V., Guimarães Lopes, L., & Salles, C. (2026). Association between sleep paralysis and anxiety: a systematic review and meta-analysis. Journal of Clinical Sleep Medicine, 22(1), Article 165. https://doi.org/10.1007/s44470-026-00190-9

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00190-9

Keywords: sleep paralysis, anxiety, anxiety disorders, meta-analysis, systematic review, REM sleep, parasomnia, sleep disorders, mental health, sleep medicine, sleep-wake transition, anxiety symptoms