The research, conducted by a team spanning Northwell Health and the Feinstein Institutes for Medical Research in New York and the Medical University of Graz in Austria, followed 117 patients with bipolar disorder, half of them women, for four and a half years. Participants were assessed approximately twice a year, for up to nine timepoints in total, an unusually dense and extended window for a psychosocial study in this population. At baseline, the researchers measured social support using two items from the World Health Organization’s Quality of Life instrument, the WHOQOL-BREF, one capturing satisfaction with personal relationships and one capturing the support patients received from friends. Depressive severity was then tracked repeatedly over the follow-up period using the Hamilton Depression Rating Scale, a clinician-administered measure that remains one of the most widely used standards in mood disorder research.

The analytical approach was designed to exploit that repeated-measures structure. Rather than comparing simple before-and-after averages, the team used Generalized Estimating Equations, a statistical framework built for clustered longitudinal data in which multiple observations from the same individual are inherently correlated. GEE models allowed the investigators to estimate whether baseline social support was associated with depressive symptom severity across the entire follow-up period while controlling for baseline depressive severity, age, and time. This matters because the central methodological trap in studies of social support and mood is reverse causation: people who are depressed may simply rate their relationships as less supportive. By statistically adjusting for how depressed patients were at the start, the researchers could ask whether friendship support carried information about the future beyond what current symptoms already revealed.

The answer was yes, with an important nuance. Across the whole sample, baseline friend support showed a significant negative main effect on long-term depressive symptom severity: patients who reported stronger support from friends at the outset tended to have lower depressive severity over the subsequent years, independent of their starting symptom levels, their age, and the passage of time. Notably, the second WHOQOL-BREF item, satisfaction with relationships in general, did not emerge as the driving predictor. It was specifically the support derived from friends, a distinct social niche separate from partners and family, that carried the predictive signal, suggesting that not all social ties contribute equally to the long-term course of the illness.

When the researchers stratified their analyses by sex, a further layer of complexity appeared. In male patients, the effect was clear and statistically significant: higher baseline friend support predicted lower depressive symptom severity across the 4.5-year follow-up, even after controlling for baseline depression. In female patients, by contrast, the same association did not reach statistical significance. The authors are careful about how this should be interpreted. The sample, while substantial for a longitudinal bipolar cohort, is modest in size, and splitting it by sex halves the statistical power available for each subgroup analysis. A null result in women therefore does not mean friendship is irrelevant to their outcomes; it means the data could not demonstrate the effect in that subgroup, and the finding requires replication in larger samples before firm conclusions can be drawn about sex-specific mechanisms.

Nevertheless, the sex difference is plausible in light of a broader literature on how men and women construct and use social networks. Research in social epidemiology has repeatedly suggested that men tend to rely more heavily on friendships, and particularly on friendships that provide companionship and shared activity, as their primary context for emotional disclosure, while women often maintain more diversified sources of intimate support, including partners and family. If men with bipolar disorder have fewer alternative channels through which they receive emotional support, then the quality of their friendships may carry disproportionate weight in buffering the stress and social withdrawal that accompany depressive episodes. The new findings offer preliminary longitudinal evidence consistent with that framing, though the study was not designed to test the underlying psychological mechanisms directly.

What makes the result clinically interesting is that friend support is a potentially modifiable factor. Much of what determines the long-term course of bipolar disorder, from genetic vulnerability to the illness’s neuroprogressive features, lies beyond a patient’s control, and pharmacological treatment, while essential, leaves many patients with residual depressive symptoms. Psychosocial interventions, by contrast, can in principle be engineered. If friendship-based support genuinely buffers depressive symptoms over years, then interventions that help patients build and sustain friendships, whether through structured group programs, peer support networks, community mental health services, or therapies that explicitly target social functioning, could become a meaningful complement to medication management. The authors frame their finding as preliminary evidence that friend support may be an important buffer against depressive symptoms, particularly in male patients, and emphasize that replication in larger samples is the necessary next step.

The study also illustrates the value of long follow-up windows in psychiatric research. Cross-sectional studies can show that socially isolated people are more depressed, but they cannot disentangle cause from consequence. By following patients across nine assessment points and modeling depressive severity as a continuous longitudinal outcome, the researchers could show that the association persisted across years rather than reflecting a snapshot in which low mood and low perceived support simply co-occur. The use of a validated quality-of-life instrument for the exposure and a clinician-rated scale for the outcome further strengthens the design, reducing reliance on self-reported mood for both sides of the equation. All participants provided informed consent, and the study was overseen by the ethics committee of the Medical University of Graz.

Cautions remain. Two questionnaire items are a coarse measure of something as multidimensional as social support, which encompasses emotional closeness, practical help, and perceived availability. The cohort was drawn from a single clinical region, and patients willing to participate in repeated research assessments over four and a half years may differ systematically from the broader population of people with bipolar disorder. The GEE models control for baseline severity, age, and time, but other unmeasured variables, from illness duration to medication regimens to personality traits, could partly explain the association. And because the sex-stratified analysis rests on smaller subgroups, the male-specific effect should be treated as a hypothesis-generating result rather than a settled fact.

Even with those caveats, the study adds weight to a growing recognition that the social world is not merely background context for severe mental illness but an active ingredient in its course. For clinicians, the practical message is that asking patients with bipolar disorder about their friendships may be more than a courtesy; it may offer a genuine window into their long-term prognosis and a target for intervention. For patients and families, the findings are a reminder that maintaining friendships through the disruptions of the illness is not a luxury. As the authors conclude, friend support may serve as an important buffer against depressive symptoms, and if larger studies confirm the effect, one of the most accessible protective factors in bipolar disorder may turn out to be the company we keep.

Subject of Research: Longitudinal association between friend-based social support and depressive symptom severity in bipolar disorder

Article Title: Social support from friends predicts lower long-term depressive symptom severity in bipolar disorder: a 4.5-year follow-up study

Article References: Social support from friends predicts lower long-term depressive symptom severity in bipolar disorder: a 4.5-year follow-up study. (n.d.). https://doi.org/10.1186/s12888-026-08673-y

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08673-y

Keywords: bipolar disorder, depression, social support, friendship, longitudinal study, sex differences, WHOQOL-BREF, Hamilton Depression Rating Scale, psychiatry, mental health, psychosocial factors, GEE models

The trial, registered prospectively on ClinicalTrials.gov as NCT05408377, enrolled 108 adult women with cyclic mastalgia across three outpatient centers in China. Participants were randomized in a one-to-one ratio to receive either manual acupuncture or sham acupuncture, and both groups followed an identical schedule: 18 sessions delivered over 12 weeks. That design choice matters. By matching the two groups for visit frequency, practitioner attention and the ritual of treatment, the investigators attempted to isolate the specific effect of needling at defined acupuncture points from the considerable placebo and context effects that are known to accompany any hands-on therapy. Of the 108 women who entered the study, 94—87.0 percent—completed the full protocol, a retention rate that lends weight to the results.

The primary endpoint was the change from baseline to week 12 in the Visual Analogue Scale for Breast Pain, or VAS-BP, a standard instrument in which patients rate their pain intensity on a continuous scale. In the intention-to-treat analysis, which included all 108 randomized participants regardless of whether they completed every session, the manual acupuncture group showed a between-group difference of minus 2.0 points at week 12, with a 95 percent confidence interval of minus 2.5 to minus 1.4. In plain terms, women receiving real acupuncture reported substantially less breast pain than women receiving the sham procedure, and the statistical uncertainty around that estimate did not come close to crossing zero. What makes the study unusual is what happened next: at week 36, twenty-four weeks after treatment had ended, the between-group difference was minus 2.2 points, with a confidence interval of minus 2.8 to minus 1.6. The benefit had not faded; it had, if anything, deepened slightly.

Pain intensity is only part of the story for cyclic mastalgia, and the trial’s secondary outcomes addressed the condition’s broader footprint. The number of days of breast pain fell more in the acupuncture group, with a between-group difference of minus 2.0 days at week 12 and minus 3.0 days at week 36. Quality of life, measured with the World Health Organization’s WHOQOL-BREF questionnaire, improved significantly more in the acupuncture group, with a total score advantage of 13.5 points and a physical health domain advantage of 1.0 points. Perhaps most striking for clinicians is the response rate: at week 12, 72.2 percent of women in the manual acupuncture group—39 of 54—were classified as responders, compared with just 22.2 percent—12 of 54—in the sham group. A threefold difference in responder rates between an active treatment and a matched sham control is a large effect by the standards of pain research.

The statistical machinery behind these conclusions was designed to handle the realities of longitudinal clinical data. The investigators used generalized estimating equations, a framework well suited to repeated measurements collected at multiple time points, and they analyzed the data under both intention-to-treat and per-protocol assumptions. Reporting both analyses is important because intention-to-treat preserves the benefits of randomization even when participants drop out, while per-protocol analysis asks what happened among those who actually received the full intervention. The consistency of the results across these frameworks, and the narrow confidence intervals around the primary estimates, suggest the findings are not an artifact of how missing data were handled.

Safety is always a central question for needle-based therapies, and here the trial’s record was reassuring. Three mild adverse events were reported in the manual acupuncture group, and none were reported in the sham group. Mild adverse events in acupuncture trials typically include minor bruising or transient soreness at needle sites, and the low event rate in this study is consistent with the generally favorable safety profile documented for the technique when performed by trained practitioners. For a condition in which standard drug options can carry hormonal side effects, a low-risk non-pharmacological alternative is an attractive proposition, provided the efficacy evidence is credible.

Why might needling at specific points influence cyclic breast pain at all? The trial’s authors are affiliated with acupuncture and moxibustion departments, including the Beijing Key Laboratory of Acupuncture Neuromodulation, and the study sits within a growing research program that seeks mechanistic explanations in neuroscience rather than traditional metaphysics. Pain of hormonal origin involves fluctuating levels of estrogen, progesterone and prolactin acting on breast tissue, and the trial recorded these hormones—estradiol, progesterone and prolactin—among its study variables. Acupuncture is known from other lines of research to engage endogenous opioid and descending pain-modulatory pathways, and neuromodulation at spinal and supraspinal levels can alter how painful signals from the chest wall are processed. Whether such mechanisms fully explain a benefit that persists for six months after treatment ends remains an open question, and the study was not designed to answer it definitively.

The sham-controlled design deserves particular attention, because it is where acupuncture trials most often stumble. Sham acupuncture in such trials typically mimics the sensory experience of treatment—insertion at non-acupoint locations or the use of blunt, retractable needles—so that patients cannot easily tell which arm they are in. The large gap between the two groups in this study, with responder rates of 72.2 percent versus 22.2 percent, suggests the sham procedure did capture much of the ritual of care without reproducing the therapeutic effect. That said, blinding in acupuncture is never perfect: patients may guess their assignment from the presence or absence of deqi sensations, and practitioners cannot be blinded. The authors’ use of a sham rather than a waitlist control strengthens the causal interpretation considerably, but readers should keep these structural limitations in mind.

There are also limits of scope to weigh. The trial was conducted entirely in China, in settings where acupuncture is an established part of routine care, and it is not yet clear whether the same effect sizes would appear in health systems where patient expectations differ. The sample of 108 participants, while adequate for the primary endpoint, is modest, and the study focused specifically on cyclic mastalgia—the cyclical, hormone-linked form of breast pain—rather than non-cyclic pain, which has different underlying mechanisms and may respond differently. The published version notes that the article was shared early as an accepted manuscript subject to further editorial processing, so some details may be refined in the final version of record. None of these caveats, however, undermines the core result of a durable, statistically robust separation between real and sham treatment.

For the millions of women who experience cyclical breast pain severe enough to seek care, the practical message is cautiously encouraging. The trial suggests that a 12-week course of manual acupuncture—18 sessions in all—can deliver relief that outlasts the treatment itself by at least half a year, alongside measurable gains in quality of life, with only mild side effects. It also offers a methodological template: a prospectively registered, sham-controlled, multicenter design with long-term follow-up is exactly the standard that complementary medicine trials need to meet if their results are to change clinical guidelines. Whether acupuncture earns a place alongside analgesics and hormonal agents in the mastalgia treatment algorithm will depend on replication in other populations and on cost-effectiveness analyses, but this study moves the needle—literally and figuratively—toward taking the therapy seriously for a condition medicine has long under-served.

Subject of Research: Long-term efficacy of manual acupuncture versus sham acupuncture for cyclic mastalgia in a multicenter randomized clinical trial

Article Title: The long-term effects of acupuncture for cyclic mastalgia: a multicenter randomized clinical trial

Article References: Yu, C., Shen, B., Li, X., Fang, Y., Xu, L., Han, P., Wang, X., Gao, Y., Zhang, Y., & Guo, J. (2026). The long-term effects of acupuncture for cyclic mastalgia: a multicenter randomized clinical trial. BMC Complementary Medicine and Therapies. https://doi.org/10.1186/s12906-026-05622-w

Image Credits: AI Generated

DOI: 10.1186/s12906-026-05622-w

Keywords: acupuncture, cyclic mastalgia, breast pain, sham-controlled trial, randomized clinical trial, pain management, quality of life, VAS-BP, WHOQOL-BREF, complementary medicine, neuromodulation, women's health

The research, conducted by Subha Jayaram and Kalpana B. Kosalram of the SRM School of Public Health at SRM Institute of Science and Technology in Tamil Nadu, India, set out with a deceptively simple objective: to assess the levels of depression, anxiety, stress, and quality of life among primary caregivers of adults with locomotive disability. While the burden of disability itself is routinely measured in clinical settings, the psychological state of the unpaid relatives who carry out the daily work of caregiving is far less frequently quantified, particularly in low- and middle-income countries where formal respite services are scarce. The study was designed to fill that gap with validated psychometric instruments rather than anecdote.

To do so, the team gathered data from 271 primary caregivers through a structured interview schedule. Each participant completed the Depression, Anxiety and Stress Scale, known as DASS-21, a widely used 21-item instrument that partitions emotional distress into three distinct subscales and assigns severity ratings ranging from normal to extremely severe. Quality of life was measured with the World Health Organization Quality of Life BREF, or WHOQOL-BREF, a condensed version of the WHO’s longer instrument that evaluates physical health, psychological well-being, social relationships, and environment. Both tools have been validated across numerous cultural contexts, which strengthens the comparability of the results with the broader international literature on caregiver burden.

The results paint a sobering picture. A major portion of the caregivers surveyed showed moderate depression and moderate anxiety, but the standout finding concerned stress: the largest share of participants fell into the extremely severe stress category. In practical terms, this means that the people shouldering responsibility for bathing, feeding, transporting, and monitoring adults with impaired mobility were not merely feeling occasional strain — many were operating at stress levels that clinicians would classify as clinically significant. Prolonged stress of this magnitude is associated with elevated blood pressure, disrupted sleep, immune dysregulation, and increased risk of cardiovascular disease, which means the findings have implications well beyond mental health.

The statistical analysis also revealed that distress was not distributed evenly across the caregiver population. The researchers found a significant difference in depression according to marital status, and significant differences in both anxiety and stress according to age and marital status. Although the study’s abstract does not detail the direction of every comparison, the pattern indicates that demographic position — whether a caregiver was married or unmarried, younger or older — shaped how intensely the caregiving role registered psychologically. This kind of stratification matters for intervention design: a one-size-fits-all counseling program is unlikely to serve a young unmarried caregiver and an older widowed one equally well, and the data suggest screening efforts should be targeted accordingly.

Quality of life fared no better. Nearly two thirds of the caregivers reported a low level of quality of life on the WHOQOL-BREF, a striking figure given that the instrument captures not just emotional states but also physical health, social functioning, and material circumstances. Caregiving for someone with a locomotive disability is physically demanding — lifting, transferring, and assisting a person with limited mobility takes a toll on the body — and it is socially demanding, often forcing caregivers to withdraw from employment, friendships, and community life. The finding that such a large proportion rated their overall quality of life as low suggests these pressures compound rather than offset one another.

Perhaps the most consequential result is the relationship between psychological distress and quality of life. The study found that depression, anxiety, and stress among the caregivers were significantly and negatively correlated with their quality of life: as each form of distress rose, quality of life fell. Correlation alone cannot establish causation, and it is plausible that poor quality of life feeds back into distress in a vicious cycle. But the strength and consistency of the association across all three DASS-21 subscales underscores that mental health and life quality in this population are tightly coupled, and that addressing one likely requires addressing the other.

To probe which factors actually predicted quality of life, the researchers turned to regression analysis, and the outcome was telling. Education emerged as a positive and significant predictor of quality of life — caregivers with more schooling reported better overall life quality, possibly because education confers greater access to information, resources, coping strategies, and income. Stress, by contrast, was a significant negative predictor: higher stress reliably forecast lower quality of life even when other variables were accounted for. The juxtaposition is instructive. It suggests that while some determinants of caregiver well-being, such as educational attainment, are slow-moving and structural, the single most modifiable lever identified by the data is stress reduction itself — through respite care, peer support, financial assistance, or psychological intervention.

The study was approved by the independent review committee of the Tamil Nadu Directorate of Public Health and Preventive Medicine and conducted in accordance with the Declaration of Helsinki, with written informed consent obtained from all participants and anonymity maintained throughout. The authors received no external financial support for the research itself, and open access publishing costs were borne by the SRM Institute of Science and Technology. The methodological transparency — validated instruments, a defined sample of 271 caregivers, and explicit statistical tests — gives the findings a solidity that much of the caregiver-burden literature, which often relies on smaller or less systematically sampled groups, has lacked.

What the study ultimately argues, implicitly but forcefully, is that caregiver mental health is not a private family matter but a measurable component of public health. When nearly two thirds of a caregiver population reports low quality of life and a major share reports extremely severe stress, the consequences ripple outward: caregiver burnout degrades the quality of care received by adults with disabilities, increases the likelihood of institutionalization, and imposes costs on health systems that are rarely attributed to their true source. The authors’ data point toward concrete policy responses — routine psychological screening of caregivers, education-linked support programs, and targeted interventions for the demographic groups shown to be most affected. As populations age and disability prevalence rises worldwide, the number of people occupying this caregiving role will only grow. The message of this research is that their distress is quantifiable, predictable, and therefore preventable — if health systems choose to look.

Subject of Research: Depression, anxiety, stress, and quality of life among primary caregivers of adults with locomotive disability

Article Title: An assessment of depression, anxiety, stress and quality of life among primary caregivers of adults with disabilities

Article References: Jayaram, S., & Kosalram, K. B. (2026). An assessment of depression, anxiety, stress and quality of life among primary caregivers of adults with disabilities. Discover Psychology. https://doi.org/10.1007/s44202-026-00873-7

Image Credits: AI Generated

DOI: 10.1007/s44202-026-00873-7

Keywords: primary caregivers, locomotive disability, depression, anxiety, stress, quality of life, DASS-21, WHOQOL-BREF, mental health, caregiver burden, disability studies, public health