Mind & body, in brief

Treatments failing fair tests

Blocking small arteries in the knee eased pain no more than a fake procedure after 12 months. [10] A fresh look at a major glaucoma trial found its benefit vanished once every eye was counted. [11] Stem cell injections into knees eased pain across 28 trials but did not rebuild the joint. [12]

Loss harming the body

People who shared a home with someone who died by suicide had a 32% higher risk of dying, in records of 101 million Brazilians. [13] Widowed people had a 67% higher risk of dying of heart and blood vessel disease across 12 studies. [14] Losing a partner raised older people's need for a care home, most of all three months after the death. [15]

Deaths nobody counted

NPR and Boston University estimate heat plays a part in about 9,000 US deaths a year, five times the official count. [16] Of 878 former American football players who died, at least 215 had a brain disease from head hits, and 643 brains were never examined. [17] As many as half of people with glaucoma, an eye disease that slowly blinds, do not know they have it. [18]

Growing tissue back

Three teams paid by ARPA-H, a US health research agency, regrew bone and cartilage in animals. [19] Blocking one enzyme regrew joint cartilage in old and injured mice. [20] Brain areas that shrink in anorexia grew back by 5.8% after about three months of weight regain. [21]

One pressure seen across the day runs through Mind & body too.

• Treatments falling short in careful testsBuilding ↑ · seen in 2 beats→

1. 1Grief is the brain still expecting someone who has died, and for up to one bereaved person in ten it does not ease
2. 2Heat kills five times more than US records show
3. 3At least one in four former American football players had CTE
4. 4Light exercise after concussion shortened recovery
5. 5A suicide at home raises the risk for everyone living there

Prolonged grief disorder is intense grief that does not ease with time and makes it hard to get through the day. [1]

Watch whether the two diagnosis lists let a pet's death count, as Hyland asks, and whether new grief therapies beat today's small-to-moderate effects. [5][9]

NPR and Boston University spent two years building a county-by-county model of deaths in the United States. [16]

They estimate heat played a part in about 9,000 deaths a year from 2018 to 2025. [16]

The official tally kept by the CDC, the US public health agency, has averaged around 1,700 in recent years. [16]

Heat often pushes a weak heart into failure and leaves no mark, so the certificate records a heart attack. [16]

A death with no heat on the certificate never reaches the state and federal tallies. [16]

In Michigan the model found more than 100 heat deaths a year, ten times the official number. [16]

CTE, chronic traumatic encephalopathy, is a brain disease caused by repeated hits to the head. [17] It can only be confirmed by examining the brain after death. [17]

Researchers looked at all 878 former NFL players who died between 2016 and 2021. [17]

At least 215 of them, 24.5%, had CTE, and the true share could be higher because 643 brains were never examined. [17]

The biggest predictor is the total force of hits over a lifetime, and in people without repeated head hits the rate is nearly zero. [17]

For years, young athletes with a concussion were told to rest body and mind until every symptom was gone. [22]

A 2026 review gathered four studies that tried light aerobic exercise instead, kept below the level that brings symptoms on. [22]

In a trial of 118 teenagers, the exercise group recovered in a median of 12 days, against 21 days for a group that only stretched. [22]

Lasting symptoms developed in 9% of the exercise group, against 31% of the stretching group. [22]

MedlinePlus, the US National Library of Medicine's health site, still says the main treatment is physical and mental rest. [23]

The review's authors call the studies small but say they back a shift toward active recovery. [22]

Avoiding reminders of a loss, or resting a concussed brain completely, feels safe now and can make recovery take longer.

Teenagers who exercised lightly recovered in a median of 12 days, against 21 for those who only stretched, which questions the old advice of complete rest.

Shutting feelings off was the symptom that most pushed up other grief symptoms a few hours later.

When helpers kept away, parents closed off too, and those interactions made the distress worse.

The courses went with people pushing painful feelings away less, alongside less grief distress.

Each of them sees only that day's choice, and whether it helped shows up weeks or months later.

Researchers followed 101 million people in Brazil from 2001 to 2018, linked to national death records. [13]

They found 47,982 suicides and studied the people who shared a home with each person. [13] Those household members had a 32% higher risk of dying from any cause. [13]

Their own risk of suicide was more than four times higher, and 44% of those suicides came within two years. [13]

Risk was highest in poorer housing and when the person who died was young or female. [13] The authors say support for the family has to start early. [13]

A 2025 review pooled 12 studies that followed 1,785,857 people over time. [14] Compared with married people, widowed people had a 43% higher risk of dying from any cause. [14]

Their risk of dying from heart and blood vessel disease was 67% higher. [14]

Single and divorced people also had higher risks, and the links differed between men and women and between regions. [14]

It points the same way as the 2014 study of widowed people above: losing a partner shows up in deaths, including from heart disease. [14][8]

Economists at Vrije Universiteit Amsterdam studied what happens to older people after their partner dies. [15]

Partner loss raised the need for care in an institution, peaking three months after the death at 1.5 percentage points. [15] The effect had gone after about ten months. [15]

It was strongest for the very old, the frail, and people with mental conditions of old age. [15]

In a Dutch ageing study the authors analysed, partners gave about 80% of the personal and nursing care older people got at home. [15]

So a death often removes the main carer too.

Swedish researchers used national records from 1990 to 2016 on 12,789 adults who had lost a sibling and later died. [24]

They checked whether those deaths bunched around the anniversary of the sibling's death. [24] Overall they did not. [24]

But men bereaved before age 50 had a 40% higher risk in the 12 days up to the anniversary, while women had a lower risk on the day itself. [24]

The authors say sibling loss can still harm health, but the risk may appear outside the anniversary and differ by sex and age. [24]

A 2026 review looked at people aged 25 or under whose parent or sibling died of cancer. [25]

Pooling five studies, 48% still reported grief symptoms six months or more after the death. [25] The figure was 28% after a parent died and 59% after a sibling died. [25]

Earlier depression, emotional difficulties and poor sleep went with worse symptoms. [25]

No study used a clinical interview, so these are symptoms, not diagnoses, and the estimates are wide. [25]

Drexel University researchers searched US health records and found 2,464 people with a prolonged grief diagnosis. [26] They compared them with 91,868 bereaved people of similar age without it. [26]

Depressive episodes showed up in 81% of the grief group, against 54.3%. [26] Anxiety disorders showed up in 80.7% against 60.8%, and alcohol problems in 24.3% against 18.4%. [26]

Three in four people with the diagnosis were women, with an average age of 59. [26] The authors say little is yet known about who receives this new diagnosis. [26]

In a 2025 study, 228 bereaved adults answered grief questions five times a day for two weeks. [27] The researchers found five different patterns. [27]

The largest group, 46%, had moderate grief that moved up and down a moderate amount. [27]

One in five had moderate grief that swung sharply, while 8% had high grief that barely moved. [27]

People with similar overall grief scores were living quite different days. [27] The authors say seeing those patterns could help fit bereavement care to each person. [27]

Another 2025 study asked 229 adults, who had lost someone about 30 months earlier, to rate 11 grief symptoms five times a day for two weeks. [28]

Yearning and sadness were the most closely linked symptoms. [28] Feeling emotionally numb had the biggest effect on other symptoms at the next check-in. [28]

The authors suggest working on pushing feelings down and on avoiding the reality of the loss. [28]

Shutting a feeling off seemed to feed the grief that came after it, which is the pattern in today's lesson. [28]

A 2026 review found 17 studies of meditation-based programmes for bereaved people. [29] The strongest evidence was for two established courses that train attention on the present moment. [29]

They reduced grief distress and depression and improved how people felt overall. [29] Across programmes, people dwelt less on the loss and pushed painful feelings away less. [29]

Most studies were small, measured different things and had no long follow-up, so the review calls these courses promising add-ons, not proven treatment. [29]

Researchers in Australia interviewed 16 parents whose child had died about the support friends and others offered. [30] The support was rarely neutral. [30]

It either helped parents feel safe, or it made the distress worse through judgement or avoidance. [30]

Parents tended to match the helper, opening up when the helper was emotionally available and closing off when the helper kept away. [30]

The study cites earlier work finding that about a third of bereaved people report being harmed by unhelpful support. [30]

In imagery rescripting, a therapist asks a grieving person to picture a painful memory of the loss and change what happens in it. [9]

Researchers at the University of Twente in the Netherlands plan to add deepfake video, a computer-made likeness, to that exercise. [9]

Nine people with prolonged grief will get three therapist-guided sessions over five weeks. [9] It is the first test of the idea for this disorder. [9]

The team hopes a more lifelike exercise will cut avoidance of the loss, but with nine people it can only show whether a larger trial is worth running. [9]

Glaucoma damages the optic nerve, the cable from the eye to the brain, and is the second-leading cause of blindness. [31]

Pressure inside the eye is usually measured in a clinic once every several months, though it changes through the day. [31]

Researchers built a soft lens whose shape shifts with eye pressure and releases eye medicine when pressure rises. [31] It worked in the eyes of rabbits. [31]

Daily eye drops are the usual treatment, and people use them less as time goes on. [31] The lens has not yet been tested in people. [31]

The Collaborative Normal-Tension Glaucoma Study is often cited to show that lowering eye pressure by 30% slows glaucoma in people whose pressure is normal. [11]

Researchers in Switzerland and Sweden re-examined its statistics. [11] The treated group's starting point had been moved after the trial began, and sight lost to cataracts was left out. [11]

Counting every eye from the start, the benefit was no longer significant. [11]

The trial is widely cited as the reason to lower pressure in these patients. [11] The authors call for new trials that count every patient from the day they join. [11]

Eye pressure is measured in millimetres of mercury, and 22 was long treated as the top of normal. [32]

A study of 1.87 million clinic visits by 94,232 glaucoma patients at seven US eye centres checked when doctors started or stepped up treatment. [32] Treatment rose steadily with pressure. [32]

But it jumped most at 22 or higher, even though pressure is now seen as a risk that rises smoothly. [32]

The authors suggest decision tools could help doctors weigh pressure as a sliding risk instead of one line. [32]

A study of health records from 2013 to 2024 covered 322,910 people with open-angle glaucoma, the common slow form. [33] Starting with medicine fell from 91.0% to 85.8%. [33]

Starting with surgery rose from 4.8% to 9.0%. [33]

In 40,119 people with angle-closure glaucoma, where fluid drainage is blocked at the front of the eye, medicine rose from 72.4% to 88.6%. [33]

The authors say studies are needed to judge whether these shifts help patients. [33]

Glaucoma kills the nerve cells that carry sight from the retina to the brain. [34]

Treatment mainly aims to lower pressure inside the eye, but many patients keep losing sight after pressure returns to normal. [34]

A review lists other ways to protect those cells, from drugs that calm inflammation to gene therapy. [34] Most are still in animal or early human studies. [34]

Glaucoma has no early symptoms, and the US National Eye Institute says a dilated eye exam is the only way to check for it. [18]

Osteoarthritis is the slow wearing-down of the cartilage that lines a joint. [35] Doctors in Rotterdam tested genicular artery embolization, a procedure that blocks small arteries around the knee. [10]

Fifty-eight patients got either the real procedure or a fake one. [10] After 12 months both groups had similar lasting pain relief, and inflammation of the joint lining did not change. [10]

The authors say the relief looks mainly like a placebo effect and do not support offering the procedure. [10]

Also seen in Biotech.

Stem cells, often from bone marrow, can be injected into an arthritic knee. [12] A 2026 review pooled 28 trials against control injections. [12]

Pain and some daily-activity scores improved. [12] MRI scans showed no consistent rebuilding of the joint, and injection-site pain and swelling were about two to three times more common. [12]

The review says the injections act on symptoms rather than on the damage itself. [12]

Researchers ran a first-in-human trial of OSCA, an injection combining donor stem-like cells with a material made from cartilage. [36]

Twelve people with moderate knee osteoarthritis got a single injection at one of three doses. [36] No dose caused serious harm, and 3 of 12 had mild-to-moderate problems. [36]

MRI scores of repaired tissue improved in 70% of patients over 24 weeks. [36]

With no comparison group and 12 patients, it shows safety, not proof that the injection works. [36]

ARPA-H, a US federal health research agency, set out three years ago to find a cure for osteoarthritis, which affects 32 million Americans. [19]

Teams at Duke University and the University of Colorado Boulder made injections that regrew bone and cartilage in animals. [19]

A Columbia University team may have found a way to regrow a whole knee. [19] The teams must start testing in patients within 18 months. [19]

A UCLA team in the same programme, awarded up to $33 million, aims for a once-a-year injection and a clinical trial as early as 2027. [37]

Stanford researchers found an enzyme called 15-PGDH rises in the cartilage of old or injured mice. [20]

Blocking it with a small-molecule drug, either through the body or into the joint, regrew cartilage and reduced arthritis pain. [20]

The repair came from existing cartilage cells changing which genes they used, not from stem cells. [20]

The study was in mice, so whether it works in human joints is not known. [20]

Researchers pooled two trials of 375 people with knee osteoarthritis who got exercise with or without hands-on therapy such as massage. [38]

Before the trials, each person said whether they expected such therapy to ease knee pain. [38]

At nine weeks, high or low expectations made no difference to how much the hands-on therapy helped. [38]

The authors call the finding exploratory. [38]

Anorexia nervosa is an illness of food restriction and very low body weight. [39]

A study of 547 girls and young women aged 12 to 30, including 168 with anorexia, scanned brain areas for reward and appetite. [21]

During the illness those areas were 5.3% smaller on average. [21] After about three months of weight regain they grew by 5.8%, except parts of the hypothalamus, the brain's hunger centre. [21]

Rising leptin, a hormone from fat, explained much of the regrowth. [21]

The authors say brain changes in anorexia are larger than in other mental disorders but largely reversible. [21]

A study compared 79 women with acute anorexia against 79 healthy women. [40] Before weight regain, levels of several hormones tied to energy balance and reproduction were abnormal. [40]

After short-term refeeding they partly or fully recovered, and they were back in 35 women who had recovered for more than six months. [40]

Healthy women taking the contraceptive pill showed some of the same hormone shifts, so the pill has to be allowed for when reading these tests. [40]

Researchers found women with anorexia had high cortisol, the stress hormone, at their first medical visit. [41] Their immune cells were fewer but more active. [41]

In rats with an anorexia-like condition, blocking the cortisol receptor calmed their overactivity. [41] The authors say these changes may keep people vulnerable after weight is back to normal. [41]

It challenges using weight regain alone as the sign that someone is in remission. [41]

Two small pilot trials gave women with anorexia gut bacteria from healthy donors' stool, mostly as capsules. [42][43]

In one trial of 18 young women, 15 completed treatment and gained an average of 38 new bacterial species. [42]

Donor strains made up 41% of their gut bacteria three weeks later and were still there at 12 weeks. [42]

In the second trial, 18 of 22 adult women completed treatment with no serious harms reported. [43]

Neither trial had a comparison group, so they show the transplants can be given safely, not that they help. [42][43]

Even after weight is restored, fear of eating and worry about body shape often last in anorexia, and relapse is common. [39]

A small trial gave 22 adults a weight-maintaining ketogenic diet, very low in carbohydrate, for 14 weeks. [39]

Eighteen finished, and at the end 72% had eating-disorder and depression scores in the normal range. [39] Three months later, scores had risen in 64% of those who stopped the diet. [39]

There was no comparison group, and a review proposes ketones as another fuel for the starved brain, which is still a hypothesis. [39][44]

Researchers in Germany and the US offered online eating-disorder prevention courses to 3,654 women from the general public. [45]

Each version was pitched at a different level of risk and ran four to twelve weekly lessons based on cognitive behavioural therapy. [45]

Weight concerns fell in four of the five versions among those who finished, and some gains lasted 12 months. [45] But only 18% to 44% completed a whole course. [45]

Online programmes like these work in trials but are rarely used outside them. [45] Here, between 61% and 78% of women did not fill in the final questionnaire. [45]

Emma Dyer, 40, from Nottinghamshire in England, bought weight-loss injections online for 115 pounds, despite a history of anorexia and bulimia. [46]

The website asked only for her body mass index, which she could lie about, and within days she collapsed. [46]

About 1.6 million UK adults used weight-loss injections in the past year, researchers at University College London estimate. [46] In the US, 18% of adults have taken one of these medicines. [47]

About 26% of US users get them from online providers or beauty clinics, and psychologists warn the medicines can hide or worsen eating-disorder symptoms. [47]

The UK charity First Steps ED says its adult referrals rose 57% in a year. [46]

Bipolar disorder brings swings between very high and very low moods. [48] Lithium is the standard long-term treatment, but many people do not respond or cannot tolerate it. [48]

Researchers followed 165,555 people in Sweden and Finland for about nine years. [48]

Compared with lithium alone, clozapine, an antipsychotic medicine, went with a 39% lower risk of a psychiatric hospital stay. [48] Clozapine with aripiprazole went with a 58% lower risk. [48]

Among 20,645 people who stopped lithium, several combinations also went with fewer hospital stays. [48] The authors offer these as possible options when lithium fails or is stopped. [48]

The NHS in England has begun inviting almost 50,000 adults with bipolar disorder, schizophrenia, psychosis or major depression into a study called GlobalMinds. [49]

Volunteers give blood or saliva for DNA, fill in online questionnaires and share their medical records. [49]

About 2,000 had enrolled when it was announced, and 1,000 people with dementia will also join. [49] It is run by the company Akrivia Health with Cardiff University. [49]