Magnesium sulphate given to women at risk of preterm birth for protecting their babies' brains reduces cerebral palsy, and the combined outcome of death or cerebral palsy, in their children up to two years of age, when compared with placebo.

Future research in this area should focus on the effects of treatment:

• on children when they are adolescents and adults; and

• for different groups of women at risk of preterm birth, and with different ways of giving magnesium sulphate.

Magnesium sulphate is a common medicine used across the world for different complications in pregnancy.

Babies born early (preterm, before 37 weeks of pregnancy) have a higher risk of complications including death and disabilities, such as cerebral palsy. In recent years, magnesium sulphate has been given to women who are likely to have their babies preterm (because of spontaneous preterm labour, or a medical indication to plan an induction of labour or caesarean birth early) to help protect their babies' brains and prevent these complications.

We wanted to find out if magnesium sulphate is better than placebo (a 'dummy' treatment that does not contain any medicine but appears identical to the medicine being tested) at protecting the brains of babies likely to be born preterm.

We were interested in the effect of magnesium sulphate on important outcomes, including: death (of the babies, or later as children), cerebral palsy, and major 'neurodevelopmental disability' (which might include serious outcomes like cerebral palsy, blindness, deafness, or global cognitive or intellectual impairment). We were also interested in the effect on important outcomes for women, including serious complications of magnesium sulphate (death, respiratory or cardiac arrest), and stopping treatment because of side effects.

We searched for studies that looked at whether magnesium sulphate caused benefits or harms for women and their preterm babies when compared to placebo or no treatment. We compared and summarised results and rated our confidence in the evidence, based on factors such as study methods and sizes.

We found six studies involving 5917 women at less than 34 weeks of pregnancy and their 6759 babies. The studies were all conducted in high-income countries. The included studies compared magnesium sulphate with placebo.

Compared with placebo, magnesium sulphate in women at risk of having their babies preterm:

• reduces cerebral palsy (evidence from 6 studies with 6107 children) and the combined outcome of death or cerebral palsy (6 studies, 6481 children) for children up to two years of age;

• probably makes little to no difference in death (6 studies, 6759 children), major neurodevelopmental disability (1 study, 987 children), or the combined outcome of death or major neurodevelopmental disability (3 studies, 4279 children), for children up to two years of age;

• may make little to no difference in the above-mentioned outcomes for children at early school age;

• may make little to no difference in serious complications of treatment for women (4 studies, 5300 women), but probably increases women stopping treatment because of side effects (3 studies, 4736 women).

We are confident in our finding that magnesium sulphate reduces cerebral palsy, and the combined outcome of death or cerebral palsy, in children up to two years of age.

We have little confidence in the evidence for outcomes of children at school age, as studies could not provide data for all children, and there are not yet enough studies/data to be certain about the results.

We have little confidence in our finding that magnesium sulphate makes little to no difference in serious complications of treatment for women, as there was only one complication reported in one study. We have moderate confidence in our findings that magnesium sulphate probably increases women stopping treatment because of side effects, as the findings differed across studies, probably because of different decision-making processes for stopping treatment.

The results of further research for the outcomes in which we have limited confidence could differ from the results of this review.

The evidence is current to 17 March 2023.

• Multisensory stimulation (the stimulation of more than one sense at a time) may support growth in preterm infants by increasing weight gain before discharge from the hospital. However, it is unclear whether it improves longer-term developmental outcomes or reduces the risk of death, as the available evidence is limited and uncertain.

• When compared with stimulation of a single sense (touch alone), it is uncertain whether multisensory stimulation is better than touch alone at increasing weight gain during the intervention.

Multisensory stimulation means giving premature babies experiences that activate two or more of their senses. The senses involved and the types of stimulation used include the sense of:

• hearing — a parent’s voice, singing, or calming music;

• touch — massage or passive movements;

• smell and taste — a mother’s scent or breast milk;

• balance — gentle rocking or changing the baby's position;

• sight — soft, cycled lighting and simple shapes, colors, or moving patterns.

During the newborn period, the brain grows very quickly. Providing positive sensory experiences at this time may help premature babies in several ways, such as supporting their memory, helping them regulate emotions, and improving early learning and development.

We wanted to find out whether multisensory stimulation compared with single‑sense stimulation or standard care helps premature babies survive, grow, stay healthy, and develop well. Specifically, we wanted to find out whether multisensory stimulation might:

• reduce the number of babies with severe developmental problems at 18 to 24 months' age, such as cerebral palsy, delay in development, limitations in mental abilities, blindness, or deafness;

• reduce the number of babies who die during the newborn period or before hospital discharge;

• increase the amount of weight the infants gained by the time they left the hospital.

We also looked at other important measures of the babies’ health, including:

• how long they stayed in the hospital or in the neonatal intensive care unit (NICU);

• how long it took before they could feed fully by mouth;

• how many days they needed breathing support.

We searched for studies in which multisensory stimulation began before the baby was discharged from the hospital after birth, and which compared multisensory stimulation with:

• no intervention or standard care;

• single-sense stimulation.

We summarized the results of the studies and rated our confidence in the evidence, based on factors, such as the size of the study (number of babies enrolled) and the methods used.

We included 19 studies (1544 premature babies). Eighteen of these studies compared multisensory stimulation with standard care, and one study compared multisensory stimulation with a single‑sense intervention (touch only).

In 10 studies, the main goal was to examine how multisensory stimulation affects babies’ brain and behavioral development.

The remaining nine studies focused on weight gain during the intervention, weight gain prior to hospital discharge, length of stay in the NICU, total length of hospital stay, time until the baby could feed fully by mouth, number of days the baby needed breathing support, or a combination of these measures.

We also found 13 ongoing studies and 4 studies that have not yet been fully assessed. They may give us more useful information in the future.

None of the studies that compared multisensory stimulation with no intervention reported on the critical outcome major developmental disability at 18 to 24 months' corrected age. Only one study assessed a related outcome, cerebral palsy at 12 months. Because this study was very small, the evidence is very uncertain about whether multisensory stimulation modifies the risk of cerebral palsy.

Multisensory stimulation may result in little or no difference in death before hospital discharge, although it was reported in only one large study. It may increase total weight gain prior to discharge, but the observed effect is small, and its clinical significance remains uncertain.

Compared with single-sense stimulation (touch), we could not tell whether multisensory stimulation helped babies gain more weight during the intervention. The only study we found comparing a multi- to single-sensory intervention did not report the critical outcomes, such as major developmental problems at 18 to 24 months' corrected age, death during the first hospital stay, or total weight gain prior to discharge.

We have limited confidence in these findings because many of the studies were small or not well-designed. Several studies did not report the outcomes we were most interested in or did not provide enough data for us to analyze. Because of these limitations, the true effects of multisensory stimulation may be different from what we found.

The evidence is up-to-date to 28 November 2025.

• In adults with long-lasting (chronic) low back pain, lower-back (lumbar) supports used on their own without additional treatment may not reduce pain or disability.

• In adults with chronic low back pain, pain may be slightly reduced when lumbar supports are used with pain medication compared to using the pain medication on its own. We do not know their effect on disability. We do not know if adding education and exercise to lumbar supports is any better than education and exercise on its own, or whether adding physiotherapy to lumbar supports has any benefit over physiotherapy on its own.

• More research is needed, particularly amongst older adults and people living in low-income and middle-income countries.

Low back pain is pain in the lower part of the back, which may also spread to one or both legs. When symptoms last longer than 12 weeks, it is considered long‐lasting and is referred to as chronic low back pain, which is a common cause of disability impacting people's quality of life across the world. It is estimated to be the leading cause of the need for rehabilitation, making it expensive in terms of healthcare costs and lost working hours.

Lumbar supports include braces, belts, and general supports that help stabilise or support the lower back. Some are made of hard plastic to limit movement. Others are softer, allowing for some movement but still providing support during movement.

Other assistive devices are items that help people with disabilities move around more easily. They include wheelchairs, mobility scooters, tricycles, crutches, walking sticks/canes, and walking frames/walkers.

We wanted to know whether lumbar supports and other assistive devices for adults with chronic low back pain help to reduce pain and disability, and improve health-related quality-of-life. We were also interested in the safety of the treatments, whether people receiving them felt they were helpful, whether they reduced painkiller use and falls, and their effects on anxiety, depression, and taking part in social activities.

We searched for studies where people with chronic low back pain received either a specific back pain treatment or 'usual care' (e.g. painkillers, general physiotherapy, education) or no treatment, and the results of the groups were compared with each other. We combined the results of similar studies and assessed how certain we were about the evidence.

We found eight studies with 501 men and women aged from 25 to 78 years. They had had chronic low back pain for between one and five years. All the studies investigated lumbar supports (for example, braces, corsets). Five of the studies took place in low- and middle-income countries (Bangladesh, Iran, Turkey) and three in high-income countries (Japan, USA).

• Lumbar supports alone, compared to getting no treatment, may have little to no effect on pain and disability after three months (1 study, 107 participants).

• Lumbar supports plus pain medication may result in a small reduction in pain intensity after three to four weeks compared with pain medication alone, but the evidence is very uncertain for disability (2 studies, 149 participants).

• We are very uncertain about the effect of adding lumbar supports to exercise and education (i.e. teaching about the structure and movement of the lower back, how it functions in daily activities, common conditions that can lead to low back pain, and different treatment options to help manage and relieve symptoms) on pain and disability after six weeks, when compared to exercise and education alone (1 study, 24 or 25 participants).

• We are very uncertain about the effect of combining lumbar supports with general physiotherapy (including ultrasound, low-intensity electrical current, heat, exercises to improve flexibility and increase strength) on pain and disability after four weeks, when compared to general physiotherapy alone (1 study, 41 participants).

• Overall, we have limited confidence in the results, and our conclusions might change after further studies are done.

• The most important limitation was that study participants and healthcare providers knew what treatment they were getting. Even though this is impossible to prevent in these types of studies, it may have affected how participants felt about their pain or disability and how they rated it.

• None of the studies reported on the unwanted effects of the treatments.

• The characteristics of the study participants were not usually reported in detail. It is difficult to know how the results should be applied to the broader group of patients with chronic low back pain.

• No studies tested mobility devices like walking aids or wheelchairs, even though these treatments are often prescribed and could potentially affect pain and disability. This gap in the evidence is especially significant for older adults and individuals with disabilities.

This evidence is current to January 2025.

Many people suffer from depression and anxiety. These problems can make people feel sad, scared and even suicidal, and can affect their work, their relationships and their quality of life. Depression and anxiety can occur because of personal, financial, social or health problems.

‘Collaborative care’ is an innovative way of treating depression and anxiety. It involves a number of health professionals working with a patient to help them overcome their problems. Collaborative care often involves a medical doctor, a case manager (with training in depression and anxiety), and a mental health specialist such as a psychiatrist. The case manager has regular contact with the person and organises care, together with the medical doctor and specialist. The case manager may offer help with medication, or access to a ‘talking therapy’ to help the patient get better.

Collaborative care has been tested with patients in a number of countries and health care systems, but it is not clear whether it should be recommended for people with depression or anxiety.

In this review we found 79 randomised controlled trials (RCTs) (90 comparisons) including 24,308 patients worldwide, comparing collaborative care with routine care or alternative treatments (such as consultation-liaison) for depression and anxiety. There were problems with the methods in some of the studies. For example, the methods used to allocate patients to collaborative care or routine care were not always free from bias, and many patients did not complete follow-up or provide information about their outcomes. Most of the studies focused on depression and the evidence suggests that collaborative care is better than routine care in improving depression for up to two years. A smaller number of studies examined the effect of collaborative care on anxiety and the evidence suggests that collaborative care is also better than usual care in improving anxiety for up to two years. Collaborative care increases the number of patients using medication in line with current guidance, and can improve mental health related quality of life. Patients with depression and anxiety treated with collaborative care are also more satisfied with their treatment.

• Taking antibiotics one hour before dental implant placement probably reduces the risk of early implant failure under routine conditions. This benefit is mainly based on studies using a single dose of amoxicillin taken before surgery, most often 2 g one hour before implant placement. Based on the available evidence, about 19 people would need to receive antibiotics to prevent one person from having an early implant failure.

• We do not know whether taking antibiotics for longer periods, taking them before or after surgery, or using other antibiotics other than amoxicillin provides additional benefit. Evidence about unwanted effects is uncertain because they were rare and not always reported.

• Future studies are needed to better understand the effects of giving antibiotics around the time of dental implant surgery, especially when an implant is placed immediately after a tooth is removed. Studies should compare a single dose of antibiotics given before surgery with longer or repeated treatment. It would also be useful to compare commonly used antibiotics, such as amoxicillin and clindamycin.

Dental implants are used to replace missing teeth and support crowns, bridges, or dentures. During implant placement, bacteria can be introduced into the surgical site. If the site around an implant becomes infected, treatment is difficult and removal of the implant is often required.

Antibiotics are sometimes given before or after surgery to reduce the risk of infection. However, they can also cause unwanted effects, ranging from mild symptoms to severe allergic reactions. Also, their widespread use contributes to antibiotic resistance, where bacteria change and grow so that the antibiotics that used to effectively treat them no longer work. For these reasons, the routine use of antibiotics for dental implant placement remains controversial.

We wanted to know whether taking antibiotics around the time of dental implant placement:

• reduces implant failures and artificial teeth failures;

• reduces postoperative infections;

• causes unwanted effects.

We wanted to know if the effects of antibiotics differed according to:

• antibiotics compared to placebo (dummy treatment) or no antibiotics;

• different doses or different durations of the same antibiotic;

• the same dosage of the same antibiotic given at different times;

• different antibiotics.

We searched for studies in which adults receiving dental implants without the addition of bone tissue (bone augmentation procedures) were assigned by chance to receive either antibiotics or no antibiotics, or to receive different antibiotic regimens. We compared and summarised the results of the studies and rated our confidence in the evidence based on factors such as study methods and sizes.

We found 15 studies including a total of 2874 people. The studies were conducted in Europe (Italy, Spain, Belgium, Ireland, and Sweden), the USA, Canada, Brazil, Iran, and Saudi Arabia, and were published between 2008 and 2024.

• People who received antibiotics one hour before dental implant placement were probably less likely to experience early implant failure than those who received placebo or no antibiotics (10 studies, 1919 people). This means that about 80 out of 1000 people without antibiotics experienced implant failure, compared with about 27 out of 1000 people who received antibiotics.

• Antibiotics probably reduce artificial teeth failures (9 studies, 1864 people) and postoperative infections (10 studies, 1919 people).

• There may be little to no difference between taking a single dose of antibiotics and taking multiple doses in implant failures (5 studies, 760 people) or postoperative infections (5 studies, 760 people).

• The evidence for artificial teeth failures is very uncertain.

• The evidence is very uncertain for implant failures, artificial teeth failures, and postoperative infections.

• There may be little to no difference in implant failures, artificial teeth failures, and postoperative infections between amoxicillin and clindamycin (1 study, 81 people).

Unwanted effects were rare and inconsistently reported. Most were mild (such as diarrhoea), but one severe allergic reaction requiring hospital admission was reported. Because few people experienced unwanted effects, we cannot reliably rule out rare but serious harms.

Our confidence in the evidence ranged from moderate to very low because some studies had problems with their methods, and in many cases, the evidence was based on small numbers of events.

This review is current to 19 November 2025.

• If healthcare workers feel they have good support from managers, receive IPC training, and are told about IPC guidelines in a clear and helpful way, they are better able to follow the guidance.

• If there is not enough space where they work, they are too busy, do not have enough personal protective equipment (PPE) and other supplies, or if patients do not follow advice, then it is more difficult for healthcare workers to follow the IPC guidelines.

• Healthcare workers who want to protect themselves, their families, and their patients tend to try to follow the guidelines as much as possible, but the quality and fit of the PPE they have can make it more difficult.

• More research is needed on training, communication and education to make it easier for healthcare workers to follow guidelines, on developing systems to minimise patient contact, and on providing appropriate PPE to enable them to be prepared for pandemics.

COVID‐19 is a respiratory infectious disease that spread worldwide in 2020. During the COVID-19 pandemic, healthcare workers needed to prevent the virus from spreading the disease to other patients, themselves, or people in the public. Infection prevention and control (IPC) guidelines are used to give healthcare workers information about using personal protective equipment (PPE) like masks, gloves, and gowns; separating patients with COVID-19; and stricter cleaning routines. However, some factors make it easier for healthcare workers to follow these guidelines, while others make it harder.

We wanted to find out:

• what made it easier and what made it harder for healthcare workers to follow IPC guidelines for COVID-19;

• compare our findings with our existing Cochrane review, published in 2020, to see what was similar and what was different about COVID-19 compared to other respiratory infectious diseases.

We searched for studies that looked at healthcare workers’ views and experiences of following IPC guidelines for COVID‐19.

• Guidelines could be local, national, or international, for any healthcare setting.

• Studies could take place in hospitals, long-term care places or community health care.

• Healthcare workers could include either clinical or support staff working in the above places.

We then looked for themes (important elements) that came from the data:

• Organisational: supporting safety, communication and training;

• Environmental: workspace, ventilation, facilities, availability of PPE;

• Individual: knowledge, attitudes, use of PPE.

We rated our confidence in each study and each finding based on the quality of their methods.

We found 170 studies. To make it easier to handle the data, we selected 26 studies that best represented all the studies. They explored the views and experiences of nurses, doctors, and other healthcare workers dealing with COVID-19 in hospitals and nursing places where people live and receive care.

Several factors influence how well healthcare workers follow IPC guidelines. They reported that:

• it was easier to follow guidelines when they had support from their managers and when guidelines were communicated in a clear and accessible way.

• it was more difficult to follow guidelines if they were very busy, short-staffed, or if patients and visitors refused to wear masks and keep social distancing.

• training was very important. However, they did not all have the chance to attend IPC training.

• they need enough space, PPE and other supplies, hand-washing stations, and showers to follow the IPC guidelines.

• those with good knowledge and attitude towards IPC were more likely to follow the guidelines.

• healthcare workers worried about infecting themselves and other people, but some worried less once people started being vaccinated.

• PPE was uncomfortable to wear for long periods and, if it did not fit properly or was of poor quality, it was more difficult for them to practice IPC.

Many factors affect healthcare workers’ willingness and ability to follow IPC guidelines. For the public and patients attending healthcare settings, this review highlights that there will be better adherence to infection prevention and control in places where staff are supported, trained, and have adequate supplies. However, it is important to note that many of the challenges (e.g. crowded spaces) are structural and not the fault of individual staff.

Our review includes a set of questions based on our findings to help healthcare providers plan, implement, or manage IPC strategies to ensure their workers follow IPC guidelines for diseases like COVID-19, which can spread worldwide and affect many people at once.

We were very confident of some of the evidence, but mainly our confidence in the evidence was limited. This was because no studies explored mental health and issues for children, meaning we are not sure how our findings apply to these groups. Also, the data from the studies were mostly collected at the peak of the pandemic (in 2020), so we have no way of knowing whether IPC guideline practices, training, and PPE evolved at later stages.

This review updates (and extends) our previous review (2020). The evidence is up-to-date to April 2024.

• Physical exercise training may help increase heart and lung function (cardiorespiratory fitness) in people with spinal cord injury up to 6 weeks after training stopped.

• The effects of physical exercise training are unclear on the ability to perform everyday activities, well-being, mood, and any unwanted effects.

• In the future, researchers should focus on large, well-designed studies that address issues that are important to people with spinal cord injury.

The spinal cord is a bundle of nerves and fibres that runs from the brain down the centre of the spine. It carries signals between the brain and parts of the body. A spinal cord injury (as a result of an accident or medical condition) disrupts these signals. This can cause loss of muscle function or movement (paralysis) or weakness, loss of feeling, and issues with basic body functions like bladder and bowel control, sexual function, and blood pressure regulation. The exact location of the injury determines how much of the body is affected. Injuries in the neck can affect the whole body; injuries in the back may only affect the trunk and legs. People with spinal cord injury find everyday tasks and movement difficult. Some need a wheelchair to move about and rely on help from others for daily tasks.

Cardiorespiratory fitness is how well your heart, lungs, and blood vessels work together to supply oxygen to your body during physical activities. The benefits of good cardiorespiratory fitness on general health, daily life, mood and well-being are widely accepted for non-disabled people, but can be challenging to achieve in people who find it hard or impossible to move their limbs. However, people with spinal cord injury may also benefit from good cardiovascular fitness, meaning they could lead more active lives with more independence and better well-being.

We wanted to know whether physical exercise training (compared to no training) improved cardiorespiratory fitness in people with spinal cord injury. We also looked at its effects on:

• functional fitness (the ability to perform daily activities, like push a wheelchair or walk a certain distance);

• well-being:

• mood (depression and anxiety); and

• unwanted effects due to training, such as death, hospitalisation, falls, stroke or heart attack, pain, or injury.

We searched for studies that compared physical exercise training to no training in adults with spinal cord injury. They could have had their injury for any length of time, and it could be from any cause. Training could last for any time and be of any intensity but had to be planned, structured and repetitive, and intended to improve cardiorespiratory fitness. It could include cycling, rowing, walking or swimming. It could take place in hospital, in the community or at home.

We compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as the design and size of studies.

We found 17 studies with a total of 505 people. There were between 10 and 68 people in each study, with a mix of neck and back injuries. They had an average age of 27 to 57 years, and 71% were men. Twelve studies took place in Europe or North America. Training sessions:

• focused on arms only or arms and legs

• typically consisted of using an arm ergometer (stationary arm bike), rowing, wheelchair propulsion, robotic gait training (machine-based, supported walking)

• happened two to five times a week

• ran for between 6 weeks and 9 months, and

• lasted between 30 minutes and 2 hours.

All studies examined the effects of physical exercise up to six weeks after the training had stopped (short-term) and only one study for longer than 6 weeks (long-term).

In the short-term, physical exercise training may help improve cardiorespiratory fitness in people with spinal cord injury (15 studies, 337 participants).

It is unclear whether physical exercise training:

• increases functional fitness (3 studies, 67 people)

• improves well-being (2 studies, 49 people)

• reduces depression and anxiety (1 study, 23 people)

• increases unwanted events (4 studies, 141 people)

Our confidence in the evidence is low or very low for several reasons. Most studies were small and not high quality. For example, many studies failed to present all the data that they collected, failed to provide a plan at the start of the study on how they would analyse the data, and failed to keep most of the participants in the study. Few studies included our most important points of interest for this review.

The evidence is current to 9 December 2025.

• Early treatment of meticillin-resistant Staphylococcus aureus (MRSA) (a type of bacteria which are resistant to some antibiotics) infection in people with cystic fibrosis (CF) seems possible, but we are unsure about the effectiveness of current routine treatment.

• Longer-term implications of treatment are not clear.

• Research is needed to assess MRSA infection and treatment in the age of cystic fibrosis transmembrane conductance regulator (CFTR) modulators (drugs that fix the broken protein made by the CF gene so salt and water can move correctly across cells and help move mucus from the lungs).

CF is an inherited condition which causes thick mucus to build up in the lungs and other organs. It is very difficult for people with CF to cough up this thick mucus, making it an ideal breeding ground for germs (including MRSA), and making them more prone to chest infections. CF worsens over time. Although there is currently no cure, many people with CF can lead active and full lives.

MRSA (meticillin-resistant Staphyloccocus aureus) are bacteria which cause infection that is treated with antibiotics. However, MRSA are resistant to some types of antibiotics, making the infection hard to treat. They are sometimes called 'superbugs' and infection with MRSA is particularly worrying for people with CF, as it is thought that MRSA can cause more damage than other bacteria which are not resistant to antibiotics.

MRSA is usually treated with a combination of antibiotics, taken orally (by mouth) or inhaled, alongside decontamination treatment (antibiotic creams applied to the skin and nasal cavities).

We wanted to know whether treatment with antibiotics:

• can get rid of MRSA infection in the lungs of children or adults with CF;

• does not cause infection with other resistant bacteria;

• does not cause unwanted effects.

We looked for studies that investigated antibiotics (oral or inhaled) to treat MRSA infections in people with CF.

• Studies had to compare antibiotic treatment with either no treatment, placebo (sham treatment) or other type of antibiotics.

• People had to have CF and current infection with MRSA.

We found five studies which recruited 410 people with CF and a diagnosed MRSA infection. Studies took place in Italy or North America and lasted from eight weeks to six months. People in the studies (equal numbers of males and females) were young (average age ranged from 11 to 25 years).

Oral antibiotics compared to no treatment (2 studies, 106 people)

One group of people with CF were treated with oral antibiotics (trimethoprim, sulfamethoxazole and rifampicin or co-trimoxazole and rifampicin), plus decontamination treatment; the other group received no treatment.

• Oral antibiotics may clear MRSA. One trial stopped early due to clear benefits of antibiotics, but some people cleared the infection with no treatment. In six months, there was no difference between groups in the number of people who still had MRSA.

• Antibiotics may increase lung function.

• There may be little to no difference between groups in quality of life, frequency of flare-ups of the disease (although one study reported fewer people taking antibiotics were admitted to hospital in the first six months), unwanted effects of treatment, or changes in weight.

Inhaled antibiotics compared to placebo (2 studies, 251 people)

Both studies compared a standard dose of vancomycin (an inhaled antibiotic) to placebo (a dummy drug) and one study also compared a standard dose of vancomycin to a high dose.

• Inhaled antibiotics probably make no difference to lung function compared to placebo (2 studies), but results from one study indicated that inhaled antibiotics probably lead to a small increase after 20 weeks.

• A standard dose of vancomycin probably makes no difference to the frequency of flare-ups compared to placebo (2 studies);

• One study reported a longer time to the next flare-up with the lower dose of vancomycin;

• One study reported a decrease in MRSA bacteria levels with a higher dose of vancomycin for up to one month;

• Neither study reported any differences in quality of life or unwanted effects.

Oral plus inhaled antibiotics compared to oral antibiotics plus placebo (1 study, 25 people)

One study compared a combination of oral plus inhaled antibiotics to inhaled antibiotics plus placebo.

• We are uncertain if either treatment made any difference to MRSA clearance;

• There may be few or no differences between groups in lung function, quality of life or unwanted effects;

• The study did not report changes in weight or frequency of flare-ups.

Our confidence in the results ranged from high to very low. This was due to potential issues with the trial designs (people knew which treatment they were receiving), or because there were small numbers of people in each trial and many of them dropped out.

This review updates our previous review published in 2022. The evidence is current to 28 April 2026.

• Nicotine electronic cigarettes (e-cigarettes) can help people to stop smoking for at least six months. They work better than nicotine replacement therapy, and probably better than e-cigarettes without nicotine.

• E-cigarettes may work better than no support, or behavioural support alone, and may not be associated with serious unwanted effects.

• However, we still need more evidence, particularly about the effects of newer types of e-cigarettes that have better nicotine delivery than older types, as better nicotine delivery might help more people quit smoking.

E-cigarettes or 'vapes' are handheld devices that work by heating a liquid that usually contains nicotine and flavourings. E-cigarettes allow users to inhale nicotine in a vapour rather than smoke. Because they do not burn tobacco, regulated e-cigarettes do not expose users to the same levels of chemicals that can cause diseases in people who smoke conventional cigarettes.

Using an e-cigarette is commonly known as 'vaping'. Many people use e-cigarettes to help them to stop smoking tobacco. Here we focus primarily on e-cigarettes containing nicotine.

Stopping smoking lowers the risk of many diseases. Many people find it difficult to stop smoking. We wanted to find out if using e-cigarettes could help people to stop smoking, and if people using them for this purpose experience any unwanted effects.

We searched for studies that looked at the use of e-cigarettes for stopping smoking in people of any age.

We were interested in:

· how many people stopped smoking for at least six months; and

· how many people had unwanted effects, reported after at least one week of use.

We found 80 studies including 29,861 adults who smoked. Most studies took place in the USA (40 studies) and the UK (16). Most studies compared nicotine e-cigarettes with:

• nicotine replacement therapy (e.g. patches or gum);

• medicines to help people stop smoking (varenicline and cytisine);

• e-cigarettes without nicotine;

• heated tobacco (products that heat tobacco to a high enough temperature to release vapour, without burning it or producing smoke; these differ from e-cigarettes because they heat tobacco leaf/sheet);

• oral nicotine pouches (pouches that contain no tobacco but release nicotine when kept in the mouth);

• other types of nicotine-containing e-cigarettes (e.g. pod devices, newer devices);

• advice or counselling (behavioural support); or

• no support for stopping smoking.

• More people stop smoking for at least six months with e-cigarettes than with nicotine replacement therapy (11 studies; 4114 people);

• People may be more likely to stop smoking for at least six months with e-cigarettes with nicotine than e-cigarettes without nicotine (7 studies, 1918 people);

• Nicotine e-cigarettes may help more people to stop smoking than behavioural support only or no support (11 studies, 7214 people);

• For every 100 people using nicotine e-cigarettes to stop smoking, 8 to 10 might successfully stop, compared with only 6 of 100 people using nicotine-replacement therapy, 6 of 100 using e-cigarettes without nicotine, or 5 of 100 people having no support or behavioural support only;

• We are uncertain if there is a difference between how many unwanted effects occur using nicotine e-cigarettes compared with nicotine replacement therapy, no support or behavioural support only (although non-serious unwanted effects were more common in groups receiving nicotine e-cigarettes compared to no support or only behavioural support);

• Studies comparing nicotine e-cigarettes to nicotine replacement therapy reported low numbers of unwanted effects, including serious unwanted effects. Compared to nicotine e-cigarettes, there is probably no difference in how many:

  • serious unwanted effects occur in people using nicotine replacement therapy;

  • non-serious unwanted effects occur in people using non-nicotine e-cigarettes.

• We are confident in the evidence that nicotine e-cigarettes help more people to stop smoking than nicotine replacement therapy, but we need more evidence about e-cigarettes without nicotine.

• Studies comparing nicotine e-cigarettes with behavioural or no support also showed higher quit rates in people using nicotine e-cigarettes, but we are more uncertain of the data because the people in the studies were aware of each treatment they were getting.

• Most of our results for the unwanted effects could change when more evidence becomes available, as there were often few cases of effects reported with unclear results.

This review updates our previous review published in 2025. We will update it again when new evidence becomes available. The evidence is up to date to 1 January 2026.

• Taking blood from the umbilical cord or placenta instead of directly from babies born too soon (preterm) likely reduces retinopathy of prematurity, an eye disease that can affect vision, and may reduce complications such as brain bleeding (intraventricular hemorrhage) and bronchopulmonary dysplasia, a type of lung disease.

• We do not know if returning blood after testing or using strict rules to limit blood tests help to minimize blood sampling and the complications associated with it.

• We need large, high-quality studies that follow babies until they are at least three years old to better understand what methods work best and how they affect longer-term development. Future studies should investigate newer approaches, such as taking tiny amounts of blood for testing or using non-blood body fluids, such as saliva and urine, as well as the methods described here.

Babies born before 37 weeks of pregnancy (known as 'preterm') often need many blood tests while they are in hospital. Because these babies are very small, repeated blood sampling can remove a large amount of their total blood volume. This can lead to the need for blood transfusions.

Several methods have been studied to reduce blood loss, such as using blood from the umbilical cord or placenta at birth, using very small blood samples, returning blood to the baby after testing, or following strict rules to limit how often blood is taken. It remains unclear which of these methods is effective and safe.

We wanted to find out whether different strategies to reduce blood sampling in preterm babies:

• improves their survival;

• reduces physical or mental developmental issues, including blindness, deafness, and cerebral palsy (a lifelong condition that affects movement, posture, and muscle control), at 18 months of age and older;

• reduces serious complications, including retinopathy of prematurity (an eye disease that can affect vision), brain bleeding (intraventricular hemorrhage), and bronchopulmonary dysplasia, a type of lung disease.

We searched for studies that compared different blood-saving strategies with the usual care given to preterm babies. We compared and summarized the results of the studies and rated our confidence in the evidence, based on how the studies were designed and how many babies were involved.

We found five studies involving 365 very preterm babies (born more than three months before the due date). Two studies were conducted in India, two in the USA, and one in Israel. The smallest study included 20 babies, and the largest, 102.

One study received funding from the government, a charity telethon, and a technology company. The author of another study was also an owner of the company that made and supplied the blood-returning device used in the study. The three remaining studies either did not provide funding details or stated that there were no conflicts of interest.

None of the studies reported results for physical or mental developmental issues (including blindness, deafness, and cerebral palsy) when the babies were 18 months of age or older.

Using blood from the umbilical cord or placenta at birth, instead of taking blood directly from the baby:

• may reduce brain bleeding during the first week of life: about 12 fewer babies in every 100 may have severe brain bleeding (from 2 to 16 fewer babies; 2 studies, 160 babies);

• may reduce bronchopulmonary dysplasia: about 13 fewer babies in every 100 may develop this lung condition (from 25 fewer to 4 more babies; 2 studies, 152 babies);

• probably reduces retinopathy of prematurity requiring treatment before hospital discharge: about 19 fewer babies in every 100 are likely to develop retinopathy of prematurity (from 6 to 27 fewer babies; 2 studies, 152 babies).

We do not know if using blood from the umbilical cord or placenta improves babies' survival rates.

We also do not know if (1) devices that return blood into the baby's bloodstream after testing, or (2) following strict rules to limit blood sampling make any difference to:

• survival rates;

• serious complications (including retinopathy of prematurity, brain bleeding, and bronchopulmonary dysplasia).

Only one small study looked at each of these approaches.

We are moderately confident that using umbilical cord or placenta blood significantly reduces the number of babies who develop retinopathy of prematurity – our confidence was reduced because relatively few babies were studied. We have little to no confidence in the evidence on survival and other serious complications (brain bleeding and bronchopulmonary dysplasia) because the studies were very small, and in some cases, the results varied.

We are not confident in the evidence for (1) devices that return blood into the baby's bloodstream and (2) following strict rules to limit blood sampling because:

• too few studies investigated these methods;

• the number of babies included was small;

• we had concerns about these studies' methods, reporting of results, or both.

The evidence is current to September 2025.

• Hearing aids probably cause a large improvement in the ability of adults with mild to moderate hearing loss to take part in everyday life and to listen to other people. They may slightly improve everyday health-related well-being but may make little to no difference to loneliness.

• It is unclear whether hearing aids result in a difference in working memory (the ability to hold information in your mind for a short time and act on it) or cause unwanted effects, such as pain or over-exposure to noise, as there was very limited evidence to help us answer this question.

• We need better and longer studies that report the effects of hearing aids separately for different age groups, sexes, levels of hearing loss, types of hearing aids, social and financial situations, and races and ethnicities.

Hearing loss is very common, with most people affected having mild to moderate hearing loss.

• Mild hearing loss means soft sounds are harder to hear and may result in difficulties listening in noisy backgrounds.

• Moderate hearing loss means speech may not be clear and conversations can be difficult to hear, in quiet as well as noisy backgrounds.

The most common causes of hearing loss are ageing and noise exposure.

Hearing aids are the most common technology used to help people with mild to moderate hearing loss and are in widespread use. Hearing aids make sounds louder and aim to make them clearer, so they are easier to hear. This is particularly important for hearing speech. The overall goal of hearing aids is to reduce the impact of hearing loss and improve a person's ability to take part in everyday life.

We wanted to find out if hearing aids were better than other options, such as no hearing aids, placebo (dummy) hearing aids, or education programmes, to improve:

• a person's ability to take part in everyday situations;

• unwanted effects such as pain or those caused by over-exposure to noise;

• well-being related to general health, such as physical and social well-being;

• the ability to listen to other people;

• mental health – we were most interested in loneliness; and

• the ability to think, remember, and learn (cognition) – we were most interested in working memory (holding information in the mind for a short time and acting on it).

We searched for studies that looked at hearing aids for adults with mild to moderate hearing loss compared to other options. We compared and summarised the studies, and then rated our confidence in the evidence based on factors such as study methods, size and duration.

We found 16 studies involving 2261 adults with mild to moderate hearing loss. Their average age was between 58 and 83 years. The studies lasted between four weeks and three years and took place in the USA, Europe, Brazil, Hong Kong, and Australia.

• Hearing aids probably cause a large improvement in the ability of adults with mild to moderate hearing loss to take part in everyday situations (8 studies, 1683 people).

• It is unclear if hearing aids cause unwanted effects. Only one study reported two cases of people feeling pain or discomfort with hearing aids; no studies reported unwanted effects from over-exposure to noise.

• Hearing aids may slightly improve general well-being related to health (4 studies, 1558 people), and probably cause a large improvement in the ability to listen to other people (5 studies, 622 people).

• Hearing aids may result in little or no difference in loneliness (2 studies, 907 people).

• Their effects on working memory are unclear (3 studies, 910 people).

The evidence from this review supports the usual practice of offering hearing aids to adults with mild to moderate hearing loss who seek help for their hearing difficulties.

There were several limitations. It is often difficult to hide whether someone is wearing hearing aids in a study as they are usually visible. This may affect the accuracy of the results. Studies of three months or less may not be long enough to measure the full range of effects from hearing aids. Some effects may only become evident over a longer period. Studies measured the effects of hearing aids differently, which made them difficult to compare.

This is an update of a review first published in 2017. We last searched for evidence in August 2024. We also included published evidence from existing and ongoing studies up to 18 May 2026.

• When a pregnant woman has been given the medicine oxytocin into a vein to start or progress labour, stopping the oxytocin once she is in active labour, rather than continuing oxytocin until the baby is delivered, may make little or no difference to whether a caesarean section (delivery through cut in the tummy) is needed, and it seems to be safe for both mother and baby.

• Stopping the oxytocin when active labour begins may help avoid excessive contractions and reduce unnecessary exposure to medication, but it may mean that labour lasts longer, although the result for duration of labour was very uncertain.

• Decisions about whether to continue or stop oxytocin should be discussed with healthcare professionals and take into account the clinical situation (including labour progress, contractions, current condition of woman and baby), the monitoring that is available, and the woman’s preferences.

In pregnant women, when near the time to deliver their baby ('term', which is after 37 weeks of pregnancy), artificial oxytocin is a medicine that is used to start the womb (uterus) contractions needed for labour or to make them stronger. It is often given through a drip into a vein. Once active labour is established (which is when the cervix has opened (dilated) at least 4 cm and contractions are happening regularly), it is thought that the body produces enough natural oxytocin to keep labour progressing. Continuing to give artificial oxytocin once a woman is in active labour may make contractions too frequent or too strong, which may cause stress to the baby and potentially increase the likelihood of a woman needing to have a caesarean section (cut across the tummy to get the baby out).

We wanted to know whether stopping oxytocin once labour is active decreases the risk of caesarean, and whether it has any benefits or harms for mothers or babies.

We looked for studies that compared the two approaches in women with a term singleton pregnancy (that is, one baby, not having twins or triplets) who were receiving oxytocin for induction or to speed up labour, so that we could find out whether it is better to:

• stop oxytocin once the active phase of labour begins, or

• continue oxytocin until birth?

We searched for evidence up to 17 November 2025. We combined study results where possible to give us the maximum information. We assessed how trustworthy each study is and how confident we can be in the results, based on factors such as the size of the studies, whether they were well designed and carried out, and how precise their results were.

We included nine studies involving 4814 women. The studies took place in hospitals in Denmark, France, India, Iran, Israel, Thailand, Turkey, and the USA. The studies used several different definitions of active labour.

Stopping oxytocin during the active phase of labour:

• may make little or no difference to the number of caesarean births (8 studies, 4710 women);

• makes little or no difference to the number of instrumental births (5 studies, 3776 women); and

• may reduce the risk of contractions that are too frequent or too strong (called uterine tachysystole or hyperstimulation) (3 studies, 1371 women); but

• labour was about 35 minutes longer on average when oxytocin was stopped, though this result was very uncertain (8 studies, 4141 women).

In addition, during the active phase of labour, there is:

• no clear difference between continuation and discontinuation of oxytocin for Apgar scores (a quick assessment of how well a baby is doing after birth) (5 studies, 2096 women);

• no clear difference between continuation and discontinuation of oxytocin for signs of reduced oxygen supply to the baby before birth, measured by testing the blood from the umbilical cord after birth (6 studies 4086 women); and

• probably little to no difference between continuation and discontinuation of oxytocin in admission to the neonatal unit (7 studies, 4606 women).

The reliability of the evidence varied a lot. We were sure about the results for instrumental births, Agpar scores, and reduced oxygen supply to the baby before birth, and quite sure about the results for neonatal unit admission. We were a little unsure about the results for caesarean section and uterine hyperstimulation, and very unsure about the results for the duration of the active phase of labour. Some reasons for uncertainty were that:

• study participants knew which group they were in, that is, whether oxytocin was continued or stopped, and this may have affected the results;

• studies used different definitions for the measures they reported, such as the active phase of labour, uterine tacysystole, and uterine hyperstimulation; and

• how long the pregnant women were in labour varied across the studies.

The evidence is up to date to 17 November 2025.

• Vaccines are a well-researched method of preventing dengue. Most research focuses on whether the vaccines work and are safe. There is less information about how vaccines are offered to people and whether people's behaviour changes when they have been vaccinated.

• Other common areas of research are community education, such as teaching school children about mosquitoes, and changes to the environment, for example, removing areas where mosquitoes might live. Most of this research examines the effects of these methods on mosquitoes. Less research looks at the methods' effects on the spread of the disease.

• This evidence gap map highlights where evidence exists and where important critical gaps remain. Critical gaps include: understanding the type of people who were included in the studies; and how the various methods were used. We need more evidence in these areas to help people make decisions on preventing dengue.

Dengue is a disease caused by a virus that is carried by a specific type of mosquito called 'Aedes'. Dengue spreads when these mosquitoes bite people. It causes flu-like symptoms, leading to more serious illness in about one in 20 people and sometimes even death. In 2023, there were over 6.5 million dengue cases and more than 7300 dengue-related deaths worldwide. Most people who get dengue live in low- and middle-income countries, where health systems may have few resources to prevent dengue. Dengue has spread recently, as a warming climate and changing rainfall patterns have created more areas in which mosquitoes can live.

We wanted to find out what evidence about dengue prevention is available and where important gaps remain. Our goal was to map the evidence that is available for each method of preventing dengue, and see whether some methods and ways of measuring results have been studied more than others. This should tell us where to focus future research efforts to provide the right information for people working to prevent dengue. This evidence gap map does not aim to assess how well any method works.

We searched for evidence on dengue prevention. The evidence could be from ‘primary studies’, where new evidence is gathered directly from a source, for example by comparing two methods, or from ‘systematic reviews’, which combine and analyse the results from more than one primary study. We then mapped the evidence according to:

• the type of prevention method (for example, vaccines, community education, environmental management);

• what the evidence measured (effects on mosquitoes, spread of the disease, behavioural effects, how it was implemented, safety, costs);

• the study design; and

• the countries where the studies took place.

As our aim was to map the number and type of studies available, we did not assess how well the studies were designed and conducted, or how confident we could be in their results. However, where we found large reviews that summarised several studies, we rated our confidence in the evidence and assessed how useful the reviews could be for people making decisions about dengue control.

We asked doctors and people working to control dengue to help us:

• inform decisions about which prevention methods and measurements of impact to include;

• identify additional studies;

• ensure the gap map was usable and relevant; and

• prioritise gaps identified from the map.

We found 550 studies that took place in many different countries, mostly in low- and middle-income areas. We found that there were many more studies for some methods than others. The studies also measured different things. For example, some looked at the number of people getting dengue and some looked at changes in behaviour.

• Vaccines were a commonly researched method, with 99 studies and 4 reviews, mainly focusing on how well the vaccines worked and if they were safe.

• Other methods — particularly community education (187 studies) and environmental management (83 studies) — were mostly assessed by studies measuring effects on mosquitoes or changes in people's behaviour, with far fewer assessing dengue cases, severe disease, or implementation outcomes (these are measures relating to how much and how well a method is use).

• Very few of the studies about interventions other than vaccines measured harms or costs, limiting the relevance of the evidence for real-world programme planning.

• Useful evidence summaries were available for some methods but not for others. We found several reviews of some methods (e.g. vaccines) but other methods only had outdated or low-quality reviews, even though the methods had been tested in many research studies.

The studies did not give much information about the types of people they included or how the different methods were used. This limits our ability to understand how well current research addresses unfair differences in the health of different groups of people (health inequities) and to support government decisions about dengue control in different parts of the world.

The evidence is up to date to August 2025.

• During an operation to widen one or both of the main blood vessels of the neck (carotid endarterectomy), there is probably little or no difference between the use of man-made (synthetic) and natural (biological) patches to close the repair in terms of the risk of stroke or death up to 80 months after the operation. There may be little or no difference between the patches in the risk of stroke on one side of the body (ipsilateral stroke), stroke of any type, or the blood vessel being blocked or narrowing again by more than half its width (arterial occlusion or re-stenosis greater than 50%) up to 80 months after the operation.

• The evidence is very uncertain about the effect of synthetic compared with biological patches on the risk of a stroke on one side of the body or a rupture of the artery (bleeding from the patch repair) during the operation or within 30 days, or a leak from artery repair causing a small blood-filled bulge (pseudoaneurysm formation) up to 80 months after the operation.

• Larger studies with longer follow-up are needed so that we can know for sure whether synthetic and biological patches have similar results, or whether one type is safer or more effective than the other.

Carotid endarterectomy is an operation to remove fatty build-up from one or both of the main blood vessels (carotid arteries) in the neck. Fatty build-up can narrow the artery or block blood flow to the brain, causing a stroke. Removing it helps reduce this risk. After the artery is opened and cleaned, surgeons often sew in a patch to widen the artery and strengthen the repair.

Patches can be made from natural (biological) materials, such as a vein from the patient's own body (autologous vein) or tissue from a cow's heart (bovine pericardium), or from man-made (synthetic) materials, such as PTFE (polytetrafluoroethylene) or Dacron. Different patch materials may affect healing and the risk of infection, blood clots, or the artery narrowing again. These problems may increase the risk of stroke.

We wanted to find out whether the choice of patch material affects the results of surgery, especially the risks of stroke, artery-related complications, and death, and longer-term effects.

We searched for studies that compared different patch materials in people undergoing carotid endarterectomy. We combined patches made using human veins or bovine pericardium into one category (biological patches), and all man-made materials into another category (synthetic patches). We compared and summarised the results of the studies, and rated our confidence in the evidence, based on factors such as study methods and sizes.

From our updated searches, we identified one new study, as well as some long-term results from studies included in previous versions of the review. In total, the review includes 15 studies involving 4971 participants, mainly men in their sixties. The studies were conducted in North America, Europe, and Australia.

Twelve studies (with 2021 carotid endarterectomy procedures) contributed results to our main comparison of synthetic and biological patches.

The evidence is very uncertain about the effects of synthetic versus biological patches on:

• stroke on one side of the body during the operation or within 30 days (5 studies, 797 participants);

• a rupture of the artery during the operation or within 30 days (6 studies, 1068 participants);

• a leak from artery repair causing a small blood-filled bulge up to 80 months after the operation (5 studies, 980 participants).

The use of synthetic patches instead of biological patches may result in little or no difference in:

• stroke on one side of the body up to 53 months after the operation (5 studies, 971 participants);

• stroke of any type up to 53 months after the operation (7 studies, 1167 participants);

• blocking of the artery or narrowing again by more than half the width up to 80 months after the operation (9 studies, 1431 participants).

The use of synthetic patches instead of biological patches probably results in little or no difference in:

• stroke or death up to 80 months after the operation (6 studies, 1115 participants).

We do not know for sure if synthetic and biological patches give similar results for people undergoing carotid endarterectomy, or if one type works better or is safer than the other. We are uncertain because:

• many of the studies were small;

• there were problems with how some studies were carried out and how they reported their results;

• some important complications were rare, which makes it difficult to detect any difference in risk between patch types.

This is an update of a review originally published 30 years ago and last updated in 2021. The evidence is current to 30 April 2025.

• In people with alpha 1 antitrypsin deficiency, delivering additional alpha 1 antitrypsin through a drip into a vein (intravenous augmentation therapy) may lead to a slight increase in the number of exacerbations (worsening of lung disease) people experience each year, but may have little or no effect on the number of people who die. We are unsure whether intravenous augmentation therapy has any effect on the number of people who have serious unwanted events like hospitalisation or permanent damage. Delivering additional alpha 1 antitrypsin through a breathing mask (inhaled augmentation therapy) may have little or no effect on the number of people who have serious unwanted events, and we are unsure whether this treatment has any effect on exacerbations.

• Our review included few studies, and we have little confidence in the evidence.

Alpha 1 antitrypsin (AAT) is a protein that helps protect our organs against damage from our body's own enzymes. AAT deficiency (AATD) is a genetic disease where people do not produce enough AAT. It can lead to lung damage and chronic obstructive pulmonary disease (COPD; a long-term condition that makes it hard to breathe). Symptoms of AATD include shortness of breath, cough, and wheeze. Smokers with AATD have a higher risk of developing severe lung disease. AATD can also lead to liver problems. Some people with AATD can stay healthy throughout their lives.

AAT augmentation therapy aims to raise AAT levels in the bloodstream and lungs using AAT protein taken from the blood of healthy donors. AAT augmentation therapy may help protect lung tissue from damage and slow lung disease progression. It can be given through a drip into a vein (intravenous), injected under the skin (subcutaneous), or inhaled through a breathing mask.

We wanted to find out if topping up AAT levels in people with AATD could reduce the number of exacerbations (worsening of the disease), the number of people who die, and the number of people who have serious unwanted events like hospitalisation or permanent damage. Our review focused on the results that were most important to people with the disease.

We searched for studies that compared AAT augmentation therapy given by any route (intravenous, subcutaneous, or inhaled) against AAT augmentation therapy delivered by another route, a dummy treatment (placebo), or no treatment. We compared and summarised the results and rated our confidence in the evidence based on factors such as study methods and sizes.

We found six studies with 524 participants. Four studies looked at intravenous AAT augmentation therapy, and the other two looked at inhaled AAT augmentation therapy. All six compared AAT augmentation therapy with placebo. The studies were conducted in high-income countries across Europe, North America, and Australia. The smallest study had five participants, and the largest had 180 participants.

Intravenous AAT augmentation therapy versus placebo

• Intravenous AAT augmentation therapy may lead to a slight increase in the number of exacerbations people experience each year. The evidence shows a rate of 1960 yearly exacerbations for every 1000 people receiving intravenous AAT augmentation therapy, compared with 1420 yearly exacerbations for every 1000 people receiving placebo (290 more exacerbations in the AAT group).

• Intravenous AAT augmentation therapy may have little or no effect on the number of people who die compared with placebo.

• We do not know whether people who receive intravenous AAT augmentation therapy are any more likely to have one or more serious unwanted events than those who receive placebo.

Inhaled AAT augmentation therapy versus placebo

• We do not know whether inhaled AAT augmentation therapy has any effect on the number of exacerbations people experience each year compared with placebo.

• No studies reported deaths.

• Inhaled AAT augmentation therapy may make little or no difference to the number of people who have one or more serious unwanted events compared with placebo.

We have little confidence in the evidence because few studies were available, the studies we found included few participants, and the results only covered some of the people we were interested in.

We included evidence published up to August 2025.

• Brain–computer interface (BCI) training may slightly improve arm movement, but may make little to no difference to leg movement. We are very uncertain whether it helps with everyday activities.

• Compared with a 'sham' (fake) BCI, BCI training may make little to no difference to movement or everyday activities.

• Larger, well-designed studies are needed to confirm the true benefits and harms of BCI training.

A brain–computer interface (BCI) is a technology that creates a direct link between the brain and an external device. It usually uses sensors placed on the scalp to detect brain signals when a person imagines or attempts a movement.

After a stroke, people often experience muscle weakness or loss of movement. BCI training reads the brain signals produced when a person intends to move, and turns them into real-time feedback. For example, when the patient thinks about moving their hand, the BCI detects this and triggers a robotic glove to move their hand, or delivers electrical stimulation to the muscles. This feedback happens at the same moment as the person's effort, which is thought to help "rewire" the brain and promote recovery of movement.

We wanted to find out whether BCI training is better than conventional therapy, other active treatments, or a sham (fake) BCI at improving arm and leg movement, and the ability to carry out everyday activities in people with stroke. We also wanted to find out if BCI training is linked to any unwanted effects.

We searched for studies looking at the effects of BCI training in adults who had had a stroke and had difficulty moving as a result. The people in these studies had had either type of stroke (one caused by a blocked blood vessel or one caused by bleeding in the brain), and most were in the months following their stroke, once their condition had stabilised. We then compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as the methods used and the number of people studied.

We found 43 studies involving 1628 adults who had had a stroke. Here is what the evidence suggests:

• When BCI training was compared with conventional therapy, it may give arm movement a small boost (10 studies, 611 people), but may make little to no difference to leg movement (1 study, 64 people). We cannot tell whether it helps with everyday activities (5 studies, 282 people), as the evidence is very uncertain.

• When compared with other active treatments, BCI training may give balance a small boost (6 studies, 165 people), but may make little to no difference to leg movement (4 studies, 130 people). For arm movement (14 studies, 331 people) and everyday activities (5 studies, 163 people), we are not confident enough to say either way.

• When it was compared with a fake or 'sham' BCI, which looks and feels like the real thing but does not give genuine brain-signal feedback, BCI training did not seem to make much difference to arm (9 studies, 279 people) or leg movement (3 studies, 106 people), or to everyday activities (1 study, 28 people).

We also looked at a few other things, such as muscle tightness, muscle strength, and overall recovery of the nervous system. The evidence on these was too uncertain for us to draw conclusions.

We also wanted to know about unwanted effects, but many studies simply did not mention them. In the studies that did, the effects were usually mild, such as sore muscles or feeling tired. Because so few studies reported this properly, we cannot be sure whether BCI training carries any extra risk.

We have little to very little confidence in the evidence, for several reasons:

• Most of the studies were small and involved only a few people.

• Many were carried out in ways that could have influenced their results. For example, the people taking part often knew which treatment they were receiving.

• The findings also varied widely from one study to another.

• Finally, some studies that found no benefit may not have been published, which could make BCI training appear more helpful than it really is.

The evidence is up to date to October 2025.

• Only one study looked at whether giving medicines that stimulate breathing, like caffeine, to babies born a few weeks early can help prevent or reduce problems with low oxygen levels or breathing. This means that we have very little evidence about their use.

• Caffeine may help reduce the number of times late preterm babies have low oxygen levels, but we do not know if it reduces the need for breathing (respiratory) support (such as ventilators); and we are not sure if caffeine causes unwanted effects.

• More research is needed to know if medicines that stimulate breathing are helpful and safe for babies born a few weeks early. We found some studies that will be published soon that will help to update the evidence.

Babies born a few weeks early, between 34 and 36 weeks of pregnancy, are called ‘late preterm’. These babies are not as fully developed as babies born at full term (after 37 weeks, or 8.5 months), and their lungs and other organs may be weaker.

Sometimes, a person's oxygen levels can drop for short periods of time (a condition called intermittent hypoxia). When this happens, it is difficult to breathe. Breathing difficulties can cause long-term problems by damaging the brain.

Methylxanthines are medicines which make breathing easier by relaxing the muscles we use to breathe. Caffeine is the most commonly used type of methylxanthine because it has fewer unwanted effects than other types.

Although methylxanthines like caffeine are often used, successfully, to treat breathing problems in very early (premature) babies (born before 32 weeks of pregnancy), they are used less often in late preterm babies, and it is not clear if this treatment can help them. We wanted to know if methylxanthines are better than no treatment or a dummy treatment (placebo) for preventing or treating problems with low oxygen levels or breathing in late preterm babies. We also wanted to know if these medicines cause any unwanted effects.

We searched for studies that compared methylxanthines with placebo or no treatment in late preterm babies. We looked for studies that measured:

• how often babies had low oxygen levels;

• whether they needed help to breathe;

• how long they stayed in the hospital;

• if the treatment had any unwanted effects;

• if the babies survived until leaving the hospital.

We looked at the results of the only study we found. We summarized the results, and rated our confidence in the evidence based on factors such as the number of babies in the study (study size), and methods used when conducting the study.

We found one study from New Zealand with 132 late preterm babies. The study compared different doses of caffeine with a placebo and had the following results.

• Caffeine may reduce the number of times babies have low oxygen levels over one week from starting treatment.

• It is not clear if caffeine helps with other problems, like needing help to breathe, or improves the chances of being able to leave hospital earlier or surviving until leaving the hospital.

• It is not clear if caffeine causes unwanted effects.

• The study did not report on other important outcomes, like long-term development or the number of times a baby stopped breathing (apnea).

We found no studies that looked at using methylxanthines to treat, rather than prevent, breathing problems in these babies.

The biggest limitation is that we only found one study. This means we cannot be sure about the results. The study did not report on all the outcomes we wanted to know about, and some results were not clear. Our confidence in these findings is very low because they are based on only one small study, and we had some concerns about how the study was carried out and reported. Future research may change these results.

The evidence is current to December 2025. There are five studies in progress and when they are complete, their results may help us better understand if methylxanthines are helpful and safe for late preterm babies.

• In people with obesity and type 2 diabetes, gastric (stomach) bypass (i.e. Roux-en-Y gastric bypass and one-anastomosis gastric bypass) may be most beneficial for improving body weight, waist circumference, type 2 diabetes status and long-term blood sugar levels at least five years after surgery, but for one-anastomosis gastric bypass, we are very uncertain about the results.

• It is unclear whether the different surgical procedures have an effect on risks such as serious unwanted health effects and death from any cause at least five years after surgery.

• Larger and well-designed studies may help reduce the existing uncertainties and increase our confidence in the findings.

What are obesity and type 2 diabetes?
Obesity is a condition with too many fat deposits that can damage health. In type 2 diabetes, the body has difficulty controlling blood sugar levels because the hormone insulin does not work properly. Type 2 diabetes often occurs in people with obesity, and both conditions increase the risk of premature death and other health problems, such as cardiovascular disease.

Lifestyle interventions such as nutrition, exercise, and behavioural programmes, along with medicines, are often used first for weight loss and improvement of diabetes status. When these do not work well, metabolic and bariatric surgery may be considered as an additional treatment option. There are a number of different types of bariatric surgery:

• gastric (stomach) banding;

• gastric sleeve (removal of a part of the stomach);

• Roux-en-Y gastric bypass (RYGB);

• one-anastomosis gastric bypass, or mini-gastric bypass;

• gastric plication (folding to make the stomach smaller);

• biliopancreatic diversion with duodenal switch.

• Which of the different types of surgery are the safest and most beneficial ones by comparing them with each other and with lifestyle and medication therapy regarding their long-term (at least 5 years follow-up) benefits and risks.

• Whether these surgical procedures improve body weight, waist circumference, type 2 diabetes status, long-term blood sugar levels, and quality of life at least five years after surgery.

• Whether surgery affects serious unwanted health effects, death from any cause, and type 2 diabetes complications over this period.

What did we do?
We searched for studies involving people with obesity and type 2 diabetes. We included studies that compared at least two different surgical procedures, or surgery with lifestyle and medication treatment. We summarised and compared all treatments using a method called network meta-analysis, and we rated our confidence in the results based on factors such as study methods and sample size.

What did we find?
We found 18 studies, including 15,282 people with obesity and type 2 diabetes. In 13 of these studies (including 911 people), participants were assigned randomly to two or more treatment groups. This is the best way to ensure that groups of people are similar. Five studies (with 14,371 people) used ‘real-life’ observational data. The studies investigated surgical procedures such as gastric band, gastric sleeve, Roux-en-Y gastric bypass, one-anastomosis gastric bypass, and gastric plication. We found that:

• Roux-en-Y gastric bypass, may improve body weight and likely reduce waist circumference compared with lifestyle and medication treatment in the long term.

• One-anastomosis gastric bypass and, to a lesser extent, gastric sleeve, may also improve body weight and waist circumference, but we are very uncertain about the results.

• Roux-en-Y gastric bypass, one-anastomosis gastric bypass, gastric sleeve, and gastric band may improve type 2 diabetes status, but we are very uncertain about the results.

• It is unclear whether the types of surgery have an effect on serious unwanted health effects or death from any cause.

• Roux-en-Y gastric bypass may reduce type 2 diabetes complications in the long term compared to non-surgical treatments, but we are very uncertain about the results.

• Roux-en-Y gastric bypass may improve blood sugar levels compared to non-surgical treatments.

• One-anastomosis gastric bypass and gastric sleeve may also improve blood sugar levels, but we are very uncertain about the results.

• RYGB and SG may lead to a greater increase in health-related quality of life than non-surgical treatments, but again, we are very uncertain about the results.

What are the limitations of the evidence?
Our confidence in the evidence is often low to very low because the studies included only small numbers of people, and the results for some outcomes, especially serious unwanted health effects, death from any cause, and type 2 diabetes complications, were based on very few events. Moreover, for several surgical procedures, there were not enough studies to be certain about the outcomes. It is important to note that rating our confidence in findings was based only on studies in which participants were randomly assigned to groups and did not include observational studies.

How up to date is this evidence?
This evidence is up to date as of June 2024.

• Any type of physical activity may slightly reduce body weight, waist size, and body fat percentage in adolescents with obesity, compared to no physical activity.

• We do not know if physical activity has an effect on other measures of weight, body size, or health problems due to obesity, compared to no physical activity.

• We need larger, longer-lasting, and better-quality studies to be more certain about the effects of physical activity on adolescents with obesity.

Obesity is when a person has too much body fat, which can harm their health. Worldwide, obesity amongst adolescents (aged 10 to 19 years) has tripled since 1990, and more than 300 million children and adolescents may be living with obesity by 2050.

Obesity during childhood and adolescence can negatively affect physical and mental health, and may increase the risk of having cancer, liver disease, or kidney disease in adulthood.

Weight loss is the best way to reduce the health risks caused by obesity. This is usually done through lifestyle changes such as doing more physical activity (e.g. swimming, running, weight-lifting) and eating healthier foods, or with surgery or medication.

We wanted to find out if, compared to no physical activity, physical activity helps adolescents with obesity:

• lose weight and reduce their BMI (the 'body mass index' is used to check if someone is a healthy weight; their weight is divided by their height multiplied by itself), BMI z-score (how someone's BMI compares to BMIs of others of the same age and sex), waist size, and body fat percentage;

• lower their blood sugar levels and insulin resistance (insulin is a hormone that helps to control blood sugar levels).

We also wanted to find out if one type of physical activity was better than another type at producing these effects, and if the physical activity programmes led to any unwanted, harmful events.

We searched for studies that compared participating in physical activity (such as doing aerobic exercise, weight-lifting, or playing football) to:

• no physical activity;

• another type of physical activity (for example, high-intensity exercise compared to moderate-intensity exercise).

The adolescents needed to do the physical activities for at least 12 weeks. We compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as study methods and number of participants.

We found 30 studies including 1508 adolescents. The studies were small: the smallest included only 21 adolescents and the largest, 106. Just under half of the adolescents involved were female (46%). Their average age ranged from 10 years to nearly 17 years. The studies took place in 15 countries: 5 in Tunisia; 4 each in China and Iran; 3 in Brazil; 2 each in Canada, Spain, and the USA; and 1 each in Australia, Estonia, India, Mexico, South Korea, Sweden, Taiwan, and Thailand.

Most studies lasted 12 or 13 weeks; the longest study lasted 32 weeks.

We organised the studies' results into two broad comparisons:

• any type of physical activity (e.g. aerobic exercise, active video-gaming) compared to no physical activity (22 studies);

• one type of physical activity compared to another type (17 studies).

Compared to no physical activity, any type of physical activity may slightly reduce:

• body weight: adolescents in the active groups weighed on average 1.88 kg less (17 studies, 749 adolescents);

• waist size: in the active groups, adolescents' average waist size was 2.88 cm smaller (10 studies, 476 adolescents);

• body fat percentage: in the active groups, adolescents' average body fat percentage was 2.96% lower (9 studies, 313 adolescents).

We are very uncertain about the effect of any type of physical activity on:

• BMI;

• BMI z-score;

• blood sugar levels;

• insulin resistance;

• unwanted, harmful events.

We are very uncertain about the effect of one type of physical activity compared with another on any of the points we were interested in.

We have limited confidence in the evidence because:

• the studies were small;

• the types of physical activities varied;

• the studies' reporting of their methods and findings was poor;

• some studies did not report all their results;

• no studies reported on unwanted, harmful events.

The studies comparing two different types of physical activity were quite varied, which meant we were mostly unable to combine their results to produce strong evidence. Therefore, we do not know whether one type of physical activity is better than another.

This evidence is current to December 2025.

• This updated review assessed special solutions (culture media) that are important for helping embryos develop in assisted reproduction technology (ART) treatments used for women who cannot get pregnant naturally. Currently, we do not know which culture medium gives the best chance of giving birth to a live baby. One study showed that an embryo culture medium called G5 probably leads to more live births than one called HTF, but the other studies did not show clear differences in success rates between different culture media.

• The evidence is limited because the studies tested different culture media and their results could not be combined. Many studies had weaknesses, such as small numbers of people taking part. More, well-designed studies are needed to find out which embryo culture medium works best.

A woman who cannot get pregnant naturally may use assisted reproductive technology (ART) treatments, such as IVF (in vitro fertilization), in which eggs are fertilized by sperm outside the body. Supporting the embryo this creates to develop (culture) is a very important part of IVF treatment. The embryo is placed in a special solution called an embryo culture medium. This medium provides nutrients and helps the embryo grow in the laboratory until it is transferred to the womb (uterus) or frozen for later use. There are many different embryo culture media.

There is currently no clear agreement on which type of embryo culture medium gives the best chances of having a live baby. The available research does not give enough reliable information to clearly choose one culture medium over another. This makes it difficult for doctors and patients to know which option to use.

We wanted to know whether different types of human embryo culture media affect the chances of having a baby or an ongoing pregnancy (pregnancy confirmed by heartbeat on ultrasound) for women undergoing IVF treatments.

We searched for studies that compared different embryo culture media available for use during IVF treatments. We looked at the results of studies for all the measurements we think are important. We considered the reliability of each study based on how it was designed and conducted. We made a judgement about how certain we were about the evidence for each important measure.

We found 26 relevant studies, eight of which evaluated the most important measures, such as live birth or ongoing pregnancy. Another five studies are in progress, and there are 60 studies that we have not been able to include because they are only published as short summaries without enough information. It was not possible to combine any of the studies' results because no two studies compared the same culture media.

Eight studies (3315 women) of the 26 studies included in the review reported live birth or ongoing pregnancy.

Most studies did not find a clear difference between the culture media they compared. However, the two largest studies did find some differences that could be important.

One study showed that there was probably a 6% higher chance of having a live birth with G5 compared to HTF (836 women). Another study showed that there may be a 6% higher chance of having a live birth with EmbryoAssist plus GM-CSF (with low HSA (a low concentration of a protein in the blood called albumin)) compared to a version of EmbyroAssist without the GM-CSF (743 women), but we are very uncertain about this result, especially because the treatment was changed during the study.

The same studies found differences in ongoing pregnancy rates. One suggested that G5 probably increases the chance of ongoing pregnancy compared with HTF (836 women). The other suggested that EmbryoAssist plus GM-CSF (low HSA) may increase the chance of ongoing pregnancy compared with EmbryoAssist without GM-CSF (743 women), but we are very uncertain about this result.

The evidence is very limited and unclear, and we could not combine results from different studies. Further well-designed studies are needed to confirm the findings and determine if any specific culture medium improves treatment success.

• Many studies included only a small number of women.

• Some studies had problems with how they were designed.

• A few studies assigned eggs or embryos (rather than women) to the groups being compared, or transferred embryos from different study groups to the same woman.

• Most studies did not clearly describe the ingredients in the culture media.

• Many studies did not report all important measures.

Overall, the evidence was mostly very uncertain because:

• the studies could not be combined with each other;

• the studies' results were not precise; and

• the studies' results were not always worked out in the right way.

This review updates one published in 2015. The evidence is current to 28 August 2025.